Abstract Book
Canadian Society for Brain, Behaviour and Cognitive Science
York University · Toronto, Ontario · June 1–3, 2026
15:45 – 18:00Award Lectures · Second Student Centre · Rooms A, B & C (combined) · Plenary
Vincent Di Lollo Early Career Award15:45 – 16:45
Dr. Jeffrey Wammes
Queen's University
▶ Read bio
Adapted from a nomination letter submitted by Monica Castelhano
Dr. Jeffrey Wammes has established a highly productive and independent research program that is internationally recognized for its rigor, innovation, and breadth. His research sits at the intersection of learning, attention, and memory, with a clear mechanistic focus on representational change in the brain and in behavior. A unique aspect of his work is the connection he draws between laboratory measures of basic cognitive processing and naturalistic cognition. Dr. Wammes integrates these two lines of work by combining behavioral measures, neuroimaging, and computational analyses.
A primary component of Dr. Wammes' work addresses how experience reorganizes neural representational spaces. Working at the cutting edge of computational neuroimaging, Dr. Wammes applies manifold analysis and machine learning to fMRI data to reveal how learning reshapes the geometry of cortical representations. His most recent work in this area—a 2025 publication in the Journal of Neuroscience—provides a unique whole-brain perspective on the neural basis of statistical learning, demonstrating that statistical learning engages a far more distributed neural architecture than previously understood, with changes propagating across cortical and subcortical networks to reshape their low-dimensional structure. This was the first application of whole-brain manifold analysis to statistical learning, and it fundamentally advances our understanding of the spatial scale over which learning-related neural reorganization operates.
In a separate and equally impactful line of work, Dr. Wammes has conducted seminal studies establishing the role of drawing and motor enactment on memory performance. He has demonstrated repeatedly that drawing improves learning and retention relative to canonical study strategies—a finding anchored by his highly cited 2016 paper in the Quarterly Journal of Experimental Psychology, which remains one of the most-read articles in that journal's history. Dr. Wammes' contributions to science are highly impactful, offering innovative ways to understand and bolster human cognition. As an early career researcher, he is already moving the field of cognitive neuroscience forward in unique and important ways. His growing stature is further reflected in his service as a nominated Early Career Researcher representative for the Canadian Brain Research Strategy (CBRS)—a recognition by the national neuroscience community that he is among the most promising voices of his generation.
In addition to his highly productive research career, Dr. Wammes has an individualized and student-centered approach to mentorship which makes him a highly sought-after supervisor. He has been particularly supportive of students who were struggling to complete their degrees for personal or professional reasons. Under his supportive mentorship thoughtfully tailored to each individual trainee, his students have won a remarkable array of competitive awards, including the D.O. Hebb Award for best student presentation, the CPA Certificate of Academic Excellence, and the Andrew McGhie Prize among others. One of his former students is now a McCall MacBain Postdoctoral Fellow—one of Canada's most prestigious postdoctoral fellowships.
D.O. Hebb Award Lecture16:45 – 18:00
Dr. Penny Pexman
University of Calgary
▶ Read bio
Adapted from a nomination letter submitted by D. Titone
Dr. Pexman is a world leader in psycholinguistics whose scholarship has advanced our fundamental understanding of the human capacity for language. She has made substantial contributions to advancing our understanding of the processes by which we derive meaning from language and how those processes develop and are shaped by experience. Her work has been instrumental in building theory and methodology and in engaging children, their caregivers, and the public in the cognitive science of human language.
By applying methods of experimental psychology, Dr. Pexman carried out the first studies investigating figurative language development and thus published some of the earliest work in the field that is now known as experimental pragmatics. Since then, using novel eye-tracking and action-tracking methodologies, she has shown how children use their emerging cognitive skills to solve the interpretive problem of ironic language. Dr. Pexman also proposed the only theory that accounts for children's acquisition of verbal irony understanding, and for the social-cognitive skills that are related to irony processing and understanding. This theoretical framework fundamentally shifted attention in the field to developmental questions and interventions for those who struggle with irony comprehension.
Sydney Gets Sarcastic, her research-informed storybook for children, is used by teachers and speech-language pathologists as a tool in their work with children and is now available in 15 languages with worldwide downloads exceeding 30,000 copies. Her research has been featured in media and podcast interviews, with outlets including NPR, BBC, and CNN, and the magazines Discover, Popular Mechanics, and Scientific American. The ten pieces she has written for The Conversation Canada have together garnered over 500,000 reads.
For more than two decades Dr. Pexman's research has been continually and simultaneously funded by both NSERC and SSHRC, including in 2008 a prestigious NSERC Discovery Accelerator Supplement. These accolades include Fellow status in the Canadian Psychological Association, in the Association for Psychological Science, and in CSBBCS. She received the Richard C. Tees Distinguished Leadership Award from the CSBBCS in 2016 and the inaugural Mid-career award from CSBBCS in 2020.
Dr. Pexman is the recipient of nine awards for mentorship and teaching. She has served as President of CSBBCS, as the Governing Board Chair of the Psychonomic Society, and as a Governing Board member of the Cognitive Science Society. Long an advocate of equity, diversity and inclusion, Dr. Pexman received the NSF-Sponsored Women in Cognitive Science Mentorship Award (2005). As the first woman editor-in-chief of the CPA's Canadian Journal of Experimental Psychology in its 65-year history, she brought new open science practices to the journal. In 2016, she co-founded Women in Cognitive Science – Canada, a national organization dedicated to building a more inclusive discipline.
09:00 – 10:00Session 1 · Attention (Visual Search) · Memory (Encoding) · Cognitive Neuroscience (Visuomotor) · Perception (Predictive) · Symposium I: Task-Optimized ANNs
Talk SessionS1A · HNE Rm 035 · Attention – Visual Search & Feature SelectionRoom 035▼ Show
Using a single item prime method to explore "priming of pop-out search"
▶ Show abstract
When the targets of two consecutive singleton searches repeat, response time is faster than when the search targets switch. This effect is commonly referred to as “priming of pop-out”. A prominent theoretical account of priming of pop-out proposes that attention to a search boosts the activation of feature representations of the search target, with that activation carrying over and speeding search for an identical target on the next trial. Interestingly, when a single item, a target without distractors, serves as the “prime”, the priming of pop-out effect is significantly less robust, or often not present at all. This lack of single item priming does not align with the activated feature representation account. In a series of experiments, we compared the priming of pop-out effect for search primes and single item primes, and whether there are conditions in which single item primes produce robust priming of pop-out. We describe the results from these experiments while highlighting the implications for the activated feature representation account and other accounts of the priming of pop-out effect.
Learned Attentional Control: Mitigating Capture by Irrelevant Salience While Preserving Capture by Relevant Salience
▶ Show abstract
Salient yet distracting information often captures attention, disrupting goal-directed behaviour. Attentional capture can be studied using the additional singleton paradigm, wherein observers search for an odd-shaped target while a salient odd-colored distractor is either absent (color-distractor absent trials) or present (color-distractor present trials). Typically, responses are slower and more prone to error on color-distractor present trials than on color-distractor absent trials, indicating attentional capture by the salient odd-colored distractor. Here we explored whether experience with the salient odd-colored distractor can mitigate attentional capture. During training, observers encountered either only color-distractor absent trials or only color-distractor present trials. During test, both trial types were randomly intermixed to assess the impact of training on attentional capture. Attentional capture was reduced following color-distractor present training relative to color-distractor absent training. Critically, these findings are novel in showing that observers can selectively suppress salience from an irrelevant dimension (e.g., color) even when they must remain sensitive to salience from a relevant dimension (e.g., shape) to identify the search target. We discuss the specificity of this learned suppression effect and present evidence showing that it generalizes beyond learned feature values (e.g., suppression that is not tied to a particular distractor color).
Visual search leaves a neural trace: ERP evidence for lingering effects of previously attended locations
▶ Show abstract
Visual search can leave lingering spatial biases, but the neural time course of these effects remains unclear. We examined probe-locked ERPs following search to test whether recently attended locations are processed differently from uncued locations. Participants searched for a target in an array of arrows, returned fixation to the center following a cue-back signal, and then responded to a red probe that appeared either at the previously fixated target location (cued) or at the horizontally opposite location (uncued), while eye position and EEG was recorded throughout. Behaviourally, reaction times did not differ reliably between cued and uncued probes, suggesting that any inhibition of return like aftereffect was offset by competing facilitatory influences. Despite this behavioural null result, posterior ERPs revealed clear cueing effects. At occipital electrodes (O1/O2), early P1 amplitudes were reduced for cued relative to uncued probes, with the strongest modulation observed ipsilaterally. N1 amplitudes showed a similar but weaker pattern. A later sustained negativity (Nd) was also more negative for cued than uncued probes at both contralateral and ipsilateral sites. Together, these findings suggest that visual search leaves a measurable neural trace that modulates subsequent probe processing even when behavioural cueing effects are absent.
Motion Impairs Detection for Spatial but not Non-Spatial Features
▶ Show abstract
While research on change detection has primarily relied on static stimuli, prior research suggests that motion impairs both attention and memory – two processes necessary for successful change detection. Therefore, the goal of the present research was to investigate whether motion disrupts change detection. Across two experiments, participants viewed an array of task-relevant triangles and task-irrelevant circles that were either stationary, moving synchronously, or moving asynchronously. In each trial, one task-relevant stimulus gradually changed while participants attempted to identify the change target. In Experiment 1, detection for Colour and Orientation changes was tested. While motion impaired detection for Orientation changes under high load, producing both a Pure Motion Effect (Synchronous > Static) and a Chaotic Motion Effect (Asynchronous > Synchronous), motion had no impact on colour detection, although detection for both features declined with increasing load. Therefore, Experiment 2 tested whether motion selectively impairs detection for spatial features by comparing detection for size versus luminance changes. Indeed, motion impaired detection for Size changes, while having no impact on detection for Luminance changes. These results challenge the assumption that motion impairs change detection by disrupting attention and memory, instead supporting a feature-specific mechanism whereby motion selectively impairs detection for spatial features.
Talk SessionS1B · HNE Rm 037 · Memory – Encoding, Production & Strategic ProcessingRoom 037▼ Show
Reading words aloud organizes memory retrieval by encoding context: Evidence from part-list cuing
▶ Show abstract
Almost all research on the production effect—the memory advantage for words read aloud versus silently—has emphasized how production influences memory for the studied items themselves. But does production shape the broader dynamics of retrieval? To examine these retrieval dynamics, we used the part-list cuing paradigm, asking whether items studied aloud versus silently differ in their influence as retrieval cues. In two experiments, participants studied either pure lists (all aloud or all silent; Experiment 1) or mixed lists (both aloud and silent; Experiment 2) and were assigned to a retrieval cue condition at test: no cues, aloud cues, or silent cues. In Experiment 1, production improved overall recall but did not modulate the magnitude of the part-list cuing impairment indicating that, without within-list distinctiveness, production does not alter cue-driven retrieval disruption. In Experiment 2, aloud cues selectively impaired recall of aloud items and silent cues tended to selectively impair silent items. This pattern suggests that production organizes items into encoding-defined subsets that constrain accessibility at retrieval, challenging a pure strength account of the production effect. Production does not merely determine how well individual items are remembered but also shapes how those items are represented and accessed during retrieval.
Resilience of the production effect under divided attention: Converging evidence from a spiking neural network
▶ Show abstract
Encoding under divided attention (DA) negatively affects memory relative to full attention (FA), although engaging in an encoding technique might reduce this negative effect. We looked for converging evidence between behavioural experiments and neural network modelling. Human participants were visually presented with two 20-item word lists, one under FA and another under DA, at encoding. Under DA, participants concurrently performed an auditorily presented digit-monitoring task. Encoding technique was manipulated within subjects such that half of the words in each condition were read aloud (produced) or silently. Memory was assessed via typed free recall after each condition. We created a spiking neural network using Nengo to model human performance; “read aloud” words were reinforced at encoding by increasing item strength by 90%, while words “read silently” were not reinforced. We modelled DA by degrading the encoded stimulus. Recall was determined by neural populations competitively accumulating evidence of word representations to a threshold. Memory performance following 50 model simulations and in 50 human participants was compared. In both, recall was higher for words read aloud than silently, with the pattern preserved even under DA. Results suggest a limited role of attention on production benefits to memory, with converging evidence from two methodologies.
Offload your information and remember it too: Supplemental images improve memory performance even when individuals offload memory demands
▶ Show abstract
Offloading to-be-remembered information to external stores can save us the effort of having to remember that information ourselves. But at the same time, we can incur a mnemonic cost if we suddenly lose access to this externally stored information. We examined whether the presentation of supplemental images can help individuals remember this lost information even when they offload memory demands. Across three experiments, we provided participants with the opportunity to offload lists of words to an external store. In Experiment 1, for half of the study items, we presented representative images after participants saved each study word to their external store. For the control items, we presented blank screens after each saved word. In Experiment 2, we displayed supplemental words as our control items. Finally, in Experiment 3, we implemented a between-subject design and measured participants’ subjective encoding effort. Overall, when participants lost access to their external stores, study words that were followed by representative images were recalled significantly more often than control items. Additionally, the supplemental images did not have a significant effect on encoding effort. These findings suggest that the presentation of supplemental images can improve memory while maintaining the effort savings associated with cognitive offloading.
Memory Signals Underlying Conjunction Errors: Retrieval Increases Familiarity and Recollection-Based False Recognition
▶ Show abstract
Retrieval practice is widely used to enhance learning because successful retrieval strengthens memory for previously studied information. However, retrieval is also a reconstructive process that can reshape memory representations and sometimes introduce systematic distortions. One such distortion is the formation of memory conjunction errors, in which features from multiple studied items are recombined and later falsely recognized as previously encountered. Previous work has shown that engaging in retrieval can increase false alarms to conjunction lures compared to restudy, but the relative contributions of recollection and familiarity to these errors remained unclear.The present study examined whether retrieval practice influences the memory signals underlying memory conjunction errors. Participants studied compound words and then completed either retrieval practice or restudy during a review phase. Memory was later assessed using a recognition test that included targets, recombined conjunction lures, and novel items. Participants made Remember/Know/New judgments and provided graded confidence ratings to examine recollection- and familiarity-based recognition. Results showed that retrieval practice increased false alarms to conjunction lures relative to restudy. Importantly, both familiarity-based and recollection-based false recognition were elevated following retrieval compared to restudy.
Talk SessionS1C · SC Room B · Cognitive Neuroscience – Visuomotor & ActionRoom B▼ Show
Cortical mechanisms of human eye-hand coordination: Beyond summation of saccade and reach networks
▶ Show abstract
Eye-hand coordination underpins everyday behaviour, yet most neuroimaging work has examined eye and hand movements in isolation. The aim of this study was to determine whether eye-hand coordination is simply the additive combination of saccade and reach-only neural processes, or whether it recruits distinct neural mechanisms.Twenty adults performed an fMRI task, consisting of three phases, isolating neural responses to visual input, movement planning, and execution of saccades, reaches, and eye-hand coordination (EHC). Behaviour was measured using MRI-compatible eye-tracker and LED-touchscreen. Analyses included, 1) permutation testing to identify task-related activation, and 2) parametric conjunction and ROI analyses to compare EHC activity to saccades and reaches (all p<0.05, FWE-corrected). Planning and execution revealed dissociable, effector-specific patterns. Planning: Reaches / EHC engaged extensive occipito-parieto-frontal areas compared to saccades. Execution: Saccades produced extensive medial / lateral frontal and occipito-parietal activation, while reaches / EHC shifted from frontal to more extensive primary sensorimotor areas. Conjunction and ROI analyses demonstrated that planning and executing EHC required greater activity than saccades and reaches alone in visual, superior parietal, and premotor dorsal areas. This demonstrates distinct organization of eye and hand signals, and additional neural processes dedicated to eye-hand coordination, highlighting potential candidate mechanisms supporting its function.
Shared and distinct cortical manifold structure for error- and reinforcement-based motor learning
▶ Show abstract
Humans can adapt their motor behaviours using different feedback signals, ranging from sensory predictions to scalar reward outcomes. While prior neuroscientific models distinguish error-based (EL) and reinforcement-based (RL) learning, the extent to which they recruit shared versus distinct large-scale neural changes remains unknown. Here, we addressed this question by characterizing whole-brain functional connectivity with lower-dimensional manifolds, within the same participants (N=37) completing separate EL and RL tasks. Behaviourally, learning was captured by a single component reflecting domain-general learning ability, where individuals with high scores performed well across both tasks. At the neural level, we found changes in manifold organization were driven primarily by learning-stage (Baseline, Early, Late-learning), not feedback type (EL, RL). Across tasks, Early-learning was characterized by widespread network segregation, while Late-learning involved reintegration of higher-order transmodal systems, including limbic, default-mode, and attentional networks. Notably, individual learning ability was not predicted by task-specific networks, but by reconfigurations of limbic and attentional systems. That is, strong learners showed early segregation of limbic regions at Early-learning, followed by reintegration with task-relevant networks as performance stabilized. Together, our findings reveal a shared network architecture supporting distinct forms of motor learning and highlight how distributed brain systems reorganize to support general learning.
Apraxia Disrupts the Spatiotemporal Structure of Grasping, Despite Preserved Aperture Scaling
▶ Show abstract
Limb apraxia is a neurobehavioural consequence of stroke, traditionally thought to selectively disrupt high-level action representations. This view relies on peak-derived metrics, like maximum grip aperture (MGA), which may overlook deficits in spatiotemporal movement structure.In Experiment 1, we assessed apraxic patients (n=25) and age-matched controls (n=25) on a width-scaling task. In Experiment 2, we compared younger (n=25) and older adults (n=25) to determine whether similar changes in visuomotor control are observed in healthy aging.Consistent with previous findings, apraxic patients successfully scaled MGA with object size; nonetheless exhibiting slower movement times. To assess spatiotemporal movement structure, we applied a Dynamic Time Warping (DTW) analysis across all object sizes, yielding a similarity matrix of all pairwise comparisons that was correlated with a theoretical size-scaling model. Apraxic patients showed significantly weaker correspondence with the size model, even with their ipsilesional hand. In contrast, in Experiment 2, Bayesian analysis confirmed that younger and older adults showed equivalent trajectory organization, with both groups outperforming apraxic ipsilesional hands.These findings reveal spatiotemporal deficits in apraxic reach-to-grasp that are dissociable from MGA-based measures, irreducible to healthy aging, and previously uncharacterized, suggesting that apraxia disrupts the kinematic architecture of grasping beyond high-level action representations.
Cerebellar Modulation of Motor Preparatory Dynamics Using Low-Intensity Focused Ultrasound Stimulation
▶ Show abstract
Hypothesis/Research Question: The cerebellum, particularly the dentate nucleus (DN), is recognized for its role in motor planning and execution. I hypothesized that low-intensity transcranial focused ultrasound (TUS) targeting the DN would modulate motor preparatory signals via dentato-thalamo-cortical (DTC) pathways, as indexed by the Bereitschaftspotential (BP).Materials and Methods: Ten healthy adults participated in EEG experiments across one session (active TUS and sham). Participants performed self-paced wrist extensions after offline TUS to the DN. EEG-derived BP was aligned to EMG onset and analyzed for peak negativity, slope, and area under the curve, each capturing a distinct aspect of the pre-movement preparatory buildup.Results: Dentate TUS altered BP dynamics: peak negativity was reduced by 6.07 µV, slope decreased by 2.44 µV/s, and area under the curve was reduced by 7.13 µV·s, indicating suppressed motor preparatory activity. Cortical effects extended across Cz, Fz, C4, and C3, suggesting network-level modulation.Conclusion: Cerebellar TUS targeting the DN can modulate motor preparatory potentials prior to voluntary movement. These findings support a causal role for the cerebellum in shaping cortical excitability through DTC pathways, with potential implications for conditions in which preparatory motor signaling is disrupted.
Talk SessionS1D · SC Room C · Perception – Predictive Processing & AudiovisualRoom C▼ Show
Deep Neural Networks as Models of Human Lightness Perception
▶ Show abstract
Human vision rapidly estimates surface reflectance despite large variations in illumination, yet image-computable models of lightness perception have remained difficult to develop. Recent advances in deep learning provide a promising framework for modeling this ability. In our previous work, we introduced a convolutional neural network trained to estimate surface reflectance directly from luminance images and showed that it exhibits several classical lightness illusions similar to those perceived by human observers. Building on this, her we evaluate whether such behavior generalizes across diverse deep learning architectures and examine the computational strategies underlying these effects. Multiple convolutional and transformer-based networks were trained to estimate reflectance from rendered three-dimensional scenes containing objects under varied illumination. All architectures successfully recovered surface reflectance and showed susceptibility to a range of classic lightness illusions, suggesting that these perceptual phenomena can emerge from optimization for reflectance estimation. Gradient-based saliency analyses further revealed convergent spatial patterns across models, highlighting image regions that most strongly influenced predicted reflectance and illusion magnitude. Together, these results suggest that deep networks trained for reflectance estimation can serve as useful computational models of human lightness perception and provide insight into the image structures and computations that give rise to lightness illusions.
Dissociating Predictive and Postdictive Audiovisual Inference: A Unified Cue-Weighting Principle
▶ Show abstract
Perception is often modelled as probabilistic inference, integrating sensory evidence with prior expectations. We asked whether reconstructing the past depends on shared mechanisms, and whether context is weighted along a common axis of sensory uncertainty, allowing the brain to handle different forms of degraded input with the same predictive strategy. We tested these possibilities in an audiovisual working memory task that orthogonally manipulated cue timing and visual uncertainty.Participants encoded faces morphed along a female-to-male continuum and later reproduced them by searching for the target in face space. Gendered voice cues were presented either before (predictive) or after (postdictive) face onset. We parametrically manipulated visual uncertainty in the predictive condition by degrading faces via resolution loss or occlusion.Voice cues biased search toward the cued gender in both pre- and post-cue conditions. However, cue effects were uncorrelated across individuals, demonstrating that predictive and postdictive mechanisms are dissociable. During prediction, cue weighting increased with visual uncertainty, regardless of degradation type, revealing a general mechanism for weighting context, independent of the source of sensory uncertainty. Additionally, cue weighting was stronger when participants saw the target face during search than when they did not, showing that context flexibly fine-tunes perception under uncertainty.
Predictive Constraints Shape Object Recognition Across Fixations
▶ Show abstract
Scene context strongly influences object recognition and allows observers to anticipate where objects are likely to appear. Predictive processing frameworks propose that such expectations arise from learned regularities between objects and scenes. However, objects differ in how reliably they are associated with regions of a scene. The Spatial Certainty Index is a measure of how consistently objects appear in specific regions and was used to manipulate object-location regularities. High-certainty objects are associated with one region (e.g., chandelier-ceiling), whereas Low-certainty objects appear across multiple (e.g., plant-hanging, desktop, floor). Although such regularities are known to guide attention, it is less clear how they constrain object recognition and the integration across fixations.We examined how spatial certainty shapes object processing using a gaze-contingent preview paradigm. Behavioral measures revealed an interaction between Spatial Certainty and Placement, whereas eye-movement measures only showed independent effects of Placement and Certainty. This dissociation suggests spatial expectations constrain later decision stages more than early perceptual processes.Together, these findings indicate spatial regularities function as predictive constraints on object recognition, operating in a graded and stage-dependent manner. Rather than acting as cues, spatial expectations shape how visual evidence is interpreted and used in decision-making across fixations.
Evidence for auditory narrative lingering of target attended speech amidst competing distractor speech
▶ Show abstract
Speech can linger in our thoughts after listening. Conversational speech is immersive but often occurs amidst competing talkers, requiring attentional resources to resolve target speech from background speech. The extent to which speech lingers in people’s thoughts under such conditions and whether residuals of competing speech also linger, is unknown. The current study examined whether varying levels of competing speech influence narrative lingering. Twenty-four younger adults (18-46 years, Mage=27.05 years, Nfemale=19) listened to two narratives simultaneously, attending to one and ignoring the other. Narratives were presented in +2dB and -6dB Target-to-Masker Ratios (TMRs), and speech-in-quiet as baseline. Participants completed a free association task before and after listening. Analyses compared the degree of semantic similarity between participants’ pre- and post-listening words with the direct and thematic words extracted from both stories. Compared to pre-listening words, post-listening words were more similar to attended-story direct and theme words, indicative of narrative lingering; however, this effect was not moderated by TMR, suggesting that lingering persists even under challenging listening conditions. No lingering was observed for ignored stories. EEG analyses reveal the attentional mechanisms underlying narrative lingering. The current findings suggest that selective attention during speech-on-speech listening promotes narrative lingering, regardless of competing speech levels.
SymposiumSymposium I · SC Room A · Agentic Contributions to Memory and AttentionRoom A▼ Show
What Role Does Agency Play in Attending, Learning, and Remembering?
▶ Show abstract
Experimenter-controlled studies of human cognitive processes have undoubtedly yielded important insights into our understanding of the mechanics of thought. However, more recent work within the fields of human attention and memory has begun to foreground the role of human agency. To introduce this symposium, I will review some recent work that illustrates how agentic considerations can open new avenues of inquiry within each field, and offer new insights into how memory and attention processes operate. Through this lens, we can begin to understand both how agents express their autonomy, as well as the efficacy of this autonomy in areas such as inattention, learning, and remembering. For example, it has been shown that moments of inattention traditionally viewed as reflecting involuntary “lapses” in attention, can instead reflect strategic tuning of attention to the demands of the environment. Similarly, memory processes have been shown to be critically shaped by strategic mechanisms that curate both what is encoded and what is retrieved. Moreover, affording agents autonomy often yields performance gains in both attention and memory tasks. We believe this emerging perspective holds promise in furthering our understanding of how memory and attention operate in naturalistic contexts.
The Importance of Agency in AI-Assisted Learning
▶ Show abstract
Advancements in large language models (LLMs) and their increasing use in learning contexts have raised important questions about how cognitively offloading to LLMs shapes later memory. Although offloading can improve efficiency, relying on LLMs to produce study materials may reduce the active engagement that supports strong encoding. We examined whether offloading mnemonic generation to ChatGPT impairs memory relative to self-generation, and whether greater agency during LLM use can attenuate that cost. Previous work in our lab has shown that offloading to LLMs impairs both recognition and cued-recall memory. To further investigate this disadvantage, Experiment 1 tested whether allowing participants to interact iteratively with ChatGPT would decrease the cost of offloading. The cost of relying on LLMs was dependent on the extent to which participants engaged in back-and-forth mnemonic refinement. Critically, self-generation still produced the best recognition and cued-recall memory overall. Experiment 2 tested whether differences in generated-mnemonic memorability contribute to this disadvantage by presenting self-generated and LLM-generated mnemonics to a new sample. Recognition was better for acronyms paired with human-generated rather than LLM-generated mnemonics, whereas cued recall did not differ by source. Together, these findings suggest that agency partly offsets, but does not eliminate, the memory costs of offloading.
When Study Decisions Matter: Evidence of Decision Driven Metacognitive Reactivity in Memory
▶ Show abstract
Previous research on metacognitive reactivity, the finding that requiring individuals to provide a metacognitive judgment (e.g., restudy decisions) changes memory performance, has considered conditions in which individuals are and are not required to provide metacognitive judgments. It is unknown whether the contents, and future ramifications of the judgment made may also induce decision-driven metacognitive reactivity. The current series of experiments sought to answer this question and shed light on reasons for this type of decision-driven metacognitive reactivity. Across our experiments participants were presented with a list of to-be-remembered words, and were required to select half of the words to later restudy. Critically, only half of the items selected were ever re-presented. Additionally, half of the items that were not selected were also re-presented. Our results show that memory for the non-selected items that were never re-presented was better than selected items that were not re-presented. We provide evidence that this decision-driven reactivity effect is not due solely to item-difficulty confounds, nor is it merely due to learners’ expectations of future learning opportunities. Instead we show that this effect may be dependent on which information learners choose to focus on or offload, highlighting the strategic and agentic aspects of memory encoding.
Strategically distracted: What volitional media multitasking reveals about how we regulate our attention
▶ Show abstract
Research on volitional media multitasking, wherein one chooses to concurrently engage in multiple media-based tasks rather than focus on a single task, has provided valuable insights into how individuals intentionally regulate their attention over time. Studies investigating temporal changes in attention commonly find declines in attention with time-on-task, an effect that has traditionally been attributed to a failure of the cognitive system to maintain attentional control. However, this traditional framework fails to acknowledge that reductions in task engagement can reflect deliberate decisions to engage one’s attention elsewhere rather than mere failures of control. This deliberate allocation of attention is particularly evident in volitional media multitasking. Across several studies, we examined temporal changes in volitional media multitasking by having participants complete an attention-demanding task while given the option to media multitask by simultaneously playing a task-irrelevant video. Consistent with prior findings that attention declines with time-on-task, media multitasking increased over time. Moreover, individuals appeared to modulate their media multitasking behaviours in response to numerous factors, including boredom, motivation, and task difficulty. I will discuss how these findings provide evidence that individuals dynamically and strategically allocate their attention in response to changing motivational priorities.
12:45 – 14:15Mel Goodale Career Celebration Symposium · Special Plenary · Second Student Centre · Rooms A, B & C (combined) · Plenary
Special Plenary12:45 – 14:15
Mel Goodale Career Celebration Symposium
A tribute to Mel Goodale's foundational contributions to visual neuroscience and perception-action research · 5 invited speakers
Jonathan Cant
University of Toronto Scarborough
Just an average talk: Investigating the influence of ensemble statistics on visually guided grasping
▶ Show abstract
In this talk I will first review findings from research conducted with Mel Goodale during my PhD studies showing that different neural substrates are involved in the perception of the form (lateral occipital cortex) versus the surface properties (parahippocampal cortex) of objects. Subsequent neuroimaging research revealed that parahippocampal cortex is also involved in ensemble perception, which involves the extraction of statistical features from groups of objects (e.g., average size or orientation). Internally generated ensemble representations have a multifaceted influence on behaviour, from biasing the remembered features of objects held in visual working memory towards the ensemble average, to the deployment of spatial attention. It is unclear, however, if ensemble representations influence visuomotor control. To investigate this, I drew influence from Mel's work on dissociations between perception and action. We show that participants could perceive differences in the ensemble statistics (i.e., average size and orientation) of a display composed of real-world objects, but these same statistics did not influence grasping movements towards an object embedded in these displays. Thus, visually-guided actions towards real-world objects are not biased by internally generated ensemble representations, adding additional evidence to Mel's influential research on dissociations between the cognitive and neural mechanisms mediating perception versus action.
Matthias Niemeier
University of Toronto Scarborough
Task Optimization Gives Rise to Streams for Action and Perception
▶ Show abstract
Primate vision is organized into dorsal and ventral pathways, canonically associated with vision for action and vision for perception. Whether this division reflects intrinsic architectural priors or emerges from task demands remains unresolved. We trained a single convolutional network to jointly perform 3D object classification and grasp prediction without imposing modular structure. A dual-stream topology with functionally distinct visuomotor and perceptual pathways emerged spontaneously, alongside rich cross-talk. Shapley value analyses revealed a gradual functional differentiation of features for action and perception across network depth, consistent with task-driven hierarchical organization. Time-resolved EEG showed that model activity aligned with dissociable temporal components of human cortical processing: ventral-aligned signals appeared early and late, with a mid-latency interval in which dorsal- and ventral-aligned responses converged. These findings demonstrate that optimization for perception and action alone can recapitulate core properties of dorsal-ventral organization, supporting the view that vision-for-action and vision-for-perception arise from shared feedforward computations rather than hard-coded architectural separation.
Robert L. Whitwell & Angus Lau
Department of Physiology & Pharmacology · Western University
Dual adaptation dissociates the effects of haptic object size and orientation on grasp aperture and orientation
▶ Show abstract
When we reach for an object, visual input about the object's size and orientation informs the hand's in-flight grasp aperture and orientation. The visuomotor relationships between these object properties and their corresponding grasp features rely on distinct but overlapping areas of intraparietal cortex. Nevertheless, vision is not the sole contributing sensory domain. A persistent absence of a tangible, real object induces variable and exaggerated scaling of grasp aperture to object size in normally-sighted participants, and in the case of 'DF', who has visual form agnosia, an absence of grip scaling altogether. Are size and orientation coded independently in the haptic domain like they are in the visual domain? We used a dual adaptation paradigm to address this question. Participants (N=96) reached for visual virtual targets with real (haptic) sizes and/or orientations that were congruent or incongruent with their visual counterparts. Adaptation to incongruent haptic object orientation and size was assessed by measuring aftereffects on grasp orientation and aperture, respectively. Using the logic of dual adaptation, when motor updates require changes to the same underlying circuit, interference ensures one or both aftereffects are null. By this same token, independent circuitry affords both compensatory updates to occur and, accordingly, aftereffects for both manipulations. Consistent with independence, haptic changes in target size and orientation induced aftereffects on grasp aperture and orientation, respectively. Furthermore, aftereffects extended to non-trained visual target sizes and orientations. Overall, these findings indicate that haptic object size and orientation contribute independently to updating motor plans for grasp orientation and aperture.
Jody C. Culham
Department of Psychology; Neuroscience Program; Centre for Brain and Mind · Western University
Recognizing and Grasping the Contributions of Mel Goodale
▶ Show abstract
Mel Goodale is widely recognized for his innovative theoretical contributions to cognitive science, most notably his influential proposal that visual processing proceeds along two streams specialized for visual recognition vs. visually guided actions. Mel has also made vital contributions to the flourishing of cognitive neuroscience in Canada, including his roles as co-founder and inaugural president of CSBBCS. As an early postdoc and long-time collaborator, I will provide a personal perspective on Mel's scientific and leadership contributions. This will include his formulation with David Milner of the two visual streams hypothesis. It will also highlight how the theory has evolved through new approaches like neuroimaging, as well as new data and perspectives. I will also honour Mel's contributions to cultivating cognitive neuroscience in Canada through his leadership and mentorship.
Mel Goodale
University of Western Ontario
14:30 – 15:30Session 2 · Attention (Eye Movements) · Memory (Recognition) · Social Cognition (AI & Trust) · Development · Symposium II: Sensory Processing in Autism
Talk SessionS2A · HNE Rm 035 · Attention – Eye Movements, Gaze & Dynamic ScenesRoom 035▼ Show
Eye movement patterns underlying the search advantage for facing social groups
▶ Show abstract
Human perception appears fine-tuned for detecting social groups, with faster search times for groups with members facing toward vs. away from each other. Here, we investigated the mechanisms underlying this facing advantage by analyzing eye movements during visual search for facing and non-facing three-person groups. In separate blocks, they searched for a facing group among non-facing distractors, and vice versa. When search displays were upright (Experiment 1), significantly fewer fixations were needed to find the facing (vs. non-facing) targets. Further, when targets were found on the first fixation, this fixation was longer for facing targets. No differences in the number of fixations for detecting facing and non-facing groups were found when search displays were inverted in Experiment 2, suggesting that the facing advantage is driven by group processing rather than lower-level features. Analyses of fixations on specific body regions (head, torso, feet) of group members showed that in both Experiments, participants most frequently looked at the region that contained the members' heads. Thus, eye movements associated with search advantages for facing social groups demonstrate specific patterns when groups are shown in a typical upright orientation, reflecting a specialization for the extraction of social dynamics rather than responses to directional information.
Gaze Behaviour in Multiple Object Tracking with Heterogeneous and Identical Items
▶ Show abstract
Multiple object tracking involves keeping track of several moving targets among distractors. Tracking performance is better when items are heterogeneous, such as 16 unique combinations of four colours and four shapes, compared to when items are identical (standard MOT). The largest improvement over standard MOT occurs when targets share a common feature such as colour (feature share condition), but there is even a benefit when targets do not share any features with each other (no share condition). To better understand the mechanisms behind these benefits, we measured gaze behaviour in Experiment 1 to test whether improvements in MOT are associated with less target-directed gaze and more centroid-directed gaze (imaginary centre of the four targets), suggesting reduced need for overt refreshing. Contrary to our predictions, there were no differences in gaze behaviour across feature share, no share, and standard MOT conditions. To further investigate this, we restricted eye movements in Experiment 2 to test whether these benefits require overt gaze. Restricting eye movements impaired MOT performance overall, but size of the benefits remained consistent with those observed in Experiment 1. We discuss the implications of these results for models of MOT and attentional control.
Saccadic Remapping in Dynamic Natural Scenes: Evidence From Road Videos
▶ Show abstract
Saccadic remapping supports perceptual stability by providing precise information about saccade targets before saccade onset. We examined whether remapping helped drivers localize immediate hazards in dynamic road scenes. Eleven licensed drivers completed a dashcam-based hazard localization task. They initially fixated 13º below video-centre then saccaded towards the video following an auditory cue. Videos were shown peripherally for 1-3s and were removed from the screen either when at the cue (covert attention), on saccade initiation (saccadic remapping) or 100 ms after saccade completion (first fixation). After removal, video was replaced with a rectangular frame, and participants clicked within to report hazard location. Click errors per-trial were calculated as smallest Euclidean distance from click to annotated hazard. For each condition and participant, we computed the mean error, expressed as z-score relative to a permuted null. We observed a significant main effect of condition (p<.001), with the lowest accuracy in the covert attention condition (mean z =-3.257, SD = 0.98), followed by the saccadic remapping condition (mean z =-3.85, SD =1.19), and highest accuracy in the first fixation condition (mean z =-5.375, SD =0.84). Our results demonstrate that saccadic remapping contributes to perceptual stability even in dynamic scenes.
Eye Movements Across Languages: Universals in Temporal and Spatial Patterns during Reading
▶ Show abstract
An important question in the study of reading is which aspects of reading behavior are language-specific and which reflect universal constraints of the human visual and oculomotor system. This talk analyzes 28 datasets of text reading from typologically diverse languages in the Multilingual Eye Movement Corpus (MECO). The first study shows that gaze transitions between lines are cross-linguistically stereotypical. Extreme fixations cover roughly 80% of each line, and fixation durations display consistent positional patterns: shorter near line endings and longer at line beginnings. These effects are not significantly influenced by linguistic properties, script characteristics, or reading direction, suggesting that return-sweep behavior is largely universal and driven primarily by visuo-oculomotor mechanisms. The second study investigates the relationship between saccade length and subsequent fixation duration. Fixation durations increase monotonically with saccade length up to a critical point, after which they plateau or decline for longer saccades. Importantly, the saccade length associated with peak fixation duration in each language closely matches independently established perceptual span boundaries. Thus, the interaction between spatial and temporal characteristics of fixation–saccade sequences reflects systematic cross-linguistic variation. Together, these results refine our understanding of the tight coordination between cognitive, linguistic, visual, perceptual and physiological systems during reading.
Talk SessionS2B · HNE Rm 037 · Memory – Recognition, Familiarity & Computational ModelsRoom 037▼ Show
Collective Memory Modeling: Using Individualized Representations to Drive Processing Models
▶ Show abstract
Computational models of memory aim to explain both the processes that operate during memory retrieval and the representations over which those processes act. Despite their theoretical interdependence, most computational modeling work has focused on processing mechanisms while assuming fixed representations. This assumption is particularly evident in models of false memory in the Deese–Roediger–McDermott (DRM) paradigm where it is necessary to combine processing models with semantic representations of word meanings to explain pertinent data. The present study examines whether variability in semantic representations alone can produce differences in episodic memory performance. Individualized semantic representations were constructed from large corpora of language produced by individual users on social media and integrated with the MINERVA 2 model of recognition memory. Process parameters were held constant across models so that differences in predicted behavior could be attributed solely to representational variability. The results demonstrate that meaningful variation in performance and item-level recognition rates emerges across individualized models. These findings highlight the importance of incorporating representational variability into computational theories of memory and serves as an introduction to a new paradigm for accounting for item-level variability in episodic memory.
Dissociation of Neural and Behavioural Perceptual Discrimination Following Lesions to Hippocampal Dentate Gyrus
▶ Show abstract
The hippocampal dentate gyrus (DG) supports pattern separation of overlapping representations at encoding. Emerging evidence suggests its role in complex perceptual discrimination of novel objects. We examined whether perceptual deficits following DG damage are explained by degraded neural discrimination at the sensory level, and whether such deficits extend to the auditory domain. BL, an individual with selective bilateral DG lesions, and controls completed a passive auditory oddball paradigm while the mismatch negativity (MMN), an event-related potential indexing discrimination in auditory sensory memory, was recorded. Memory for encoded oddball stimuli was subsequently tested against similar lures and foils, alongside a same-different perceptual discrimination task. BL showed a selective deficit for complex discrimination of similar tone sequences despite showing comparable MMN responses to controls. Although explicit memory discrimination of encoded targets compared to lures and foils showed no difference, BL tended to misattribute lures as “old” compared to controls. Findings suggest a dissociation whereby DG damage does not disrupt automatic neural discrimination at the sensory level, but instead impairs the orthogonalization of overlapping sensory inputs into precise perceptual representations that support explicit behavioral discrimination. Results extend the role of the DG beyond memory, highlighting contributions to perceptual discrimination of overlapping auditory features.
Knowing What We Don't: What drives the mirror effect in recognition memory?
▶ Show abstract
The mirror effect in recognition memory describes the pattern whereby conditions that increase hit rates simultaneously decrease false-alarm rates. Two major explanations have been proposed: criterion shift, which attributes the effect to changes in decision bias, and differentiation, which attributes it to differences in memory strength distributions. Disentangling these accounts is challenging because many manipulations are transparent to participants, who may strategically shift their criterion — confounding the underlying process. This project induces a metacognitive illusion using an errorful generation task to address this issue. Errorful generation strengthens memory encoding without a corresponding increase in confidence, such that participants consistently expect read-only items to be better remembered than error-gen items. This discrepancy between actual memory performance and subjective confidence creates a misalignment between criterion placement and true memory strength, effectively disabling the criterion shift mechanism. Consequently, any mirror effect observed under these conditions can be more directly attributed to differentiation. Results showed that the mirror effect remained intact following the successful induction of the metacognitive illusion, a pattern consistent with differentiation as the primary driver. These findings are discussed within a signal detection framework, with implications for understanding the mechanisms underlying recognition memory.
Reminding improves discrimination of similar items by promoting recollection rejection
▶ Show abstract
Similarities between experiences elicit spontaneous remembering, where the current experience reminds us of previous, similar experiences. Recently, we found that remindings, when later recollected, can improve discrimination between past experiences and novel, similar lures (Churey & Fiacconi, 2026). However, further investigation is required to better understand the mechanisms underlying this benefit of reminding. On the one hand, it has been suggested that false memories of new, similar items can arise due to recollection of similarities to previously studied items (false recollection) or if they simply feel familiar enough (false familiarity). Alternatively, these false memories can be avoided if the similarity between new and old items elicits recollection of details that disqualify the new item as being old (recollection rejection; Parks et al., 2026). Using this logic, the current study sought to quantify how much of the mnemonic benefit of reminding reflected a recollection rejection process using a novel receiver operating characteristic (ROC) approach. Across two experiments, we found reminding at study facilitated greater use of recollection rejection to reduce overall false recognition of similar lures at test, relative to when reminding was not reported. Importantly, this research underscores the importance of similarity-driven remindings in the accurate discrimination of similar experiences.
Talk SessionS2C · SC Room B · Social Cognition – AI, Technology & TrustRoom B▼ Show
Mind Attribution in Synthetic Media: An Extension of the Medusa Effect
▶ Show abstract
Across history, humans have readily attributed mental states to others. This tendency extends to pictures and images, though in diminished form. Prior work shows that mind perception is further reduced for people depicted within pictures inside other pictures, a phenomenon known as the Medusa Effect. The present study examined whether the Medusa Effect extends to AI generated content. Undergraduates viewed a series of AI generated images and rated depicted individuals on three dimensions of mind: agency (ability to do), experience (ability to feel), and realness (of the natural world). At the end of the study, participants estimated the percentage of images they believed were AI generated. Results demonstrate that the Medusa Effect replicates with AI-generated stimuli. Critically, even when participants reported that most images were AI generated, mind ratings were not significantly altered. These findings suggest that the Medusa Effect is robust across media formats and persists despite participants’ beliefs about AI generated content. Together, these results highlight the stability of mind attribution processes in the context of synthetic media.
Trusting the machine: How anthropomorphism impacts epistemic trust in generative AI
▶ Show abstract
As generative artificial intelligence (GenAI) becomes more sophisticated and its use becomes more widespread, understanding how its perceived humanness shapes trust is increasingly important. Previous work has primarily focused on trust in AI agents as advisors; however, it remains unclear whether anthropomorphism also influences epistemic trust in content produced by GenAI agents. Across two experiments, we tested whether anthropomorphic cues affect trust in a GenAI author and epistemic trust of a blogpost attributed to that author. Participants either read about (Experiment 1; N = 270) or interacted with (Experiment 2; N = 144) a GenAI author presented with varying levels of anthropomorphic cues. These cues included agent descriptions suggesting curiosity and emotion versus purely algorithmic processing (Experiment 1) and conversational responses that varied in warmth and empathy (Experiment 2). All participants read the same science communication blog post about a fictional medical breakthrough, then rated both author and content trust. Author trust and content trust were strongly positively correlated; however, increasing anthropomorphic cues directly increased trust in the author but did not directly increase trust in the content. Overall, these findings suggest that anthropomorphic cues primarily influence trust in the source rather than trust in the informational content itself.
Suspicion and Shared Trauma: Asymmetric Drivers of Post-Election Extremism in the American Electorate
▶ Show abstract
Do democratic elections channel political attitudes into meaningful expression, or can they serve as catalysts for extremism? This research examines how the ballot box moves the mind within the highly polarized, volatile context of the United States. Utilizing a longitudinal pre- and post-election design (N = 534), we tracked shifts in behavioral endorsements of left-wing (LWA) and right-wing authoritarianism (RWA) and investigated how epistemic orientations—specifically conspiracy mentality and existential nihilism—interact with electoral outcomes to drive radicalization.Our findings reveal asymmetric drivers of radicalization between electoral winners and losers. For the political Right, victory weaponized existing suspicion, with conspiracy mentality driving increased RWA behavioral endorsement. Conversely, for the political Left, the shared trauma of electoral defeat drove radicalization via existential nihilism. Furthermore, we identify how trust in democracy fundamentally moderates these effects to shape distinct voter typologies that dictate how citizens justify authoritarian responses. Finally, we demonstrate that the protective buffer of general societal confidence weakens significantly following an election, increasing political distress. Ultimately, this research highlights that, despite their democratic necessity, elections in polarized climates can become radicalizing events.
Long-term attitude change from AI dialogue using political in-group disagreement and out-group agreement
▶ Show abstract
This study investigates whether AI-driven conversations can produce durable attitude change toward controversial issues using in-group and out-group political messaging. Participants engaged with an AI chatbot in which they stated their attitudes and explained why they held them, while the AI responded using arguments from the participant’s in-group that disagreed with them or from their out-group that agreed with them. Attitudes were measured before, immediately after, and one week after interacting with the AI. We found that both in-group and out-group messaging were effective in producing durable attitude change, with neither approach outperforming the other. We also found that the perceived vividness of memory for the AI interaction predicted positive attitude change, whereas accurate memory of the interaction had no effect. These findings suggest that conversational AI can produce long-term attitude change through active cognitive processing, highlighting its potential as a scalable tool for shifting public attitudes on complex issues through interactive engagement.
Talk SessionS2D · SC Room C · Development – Statistical Learning, Infants & ChildrenRoom C▼ Show
Behavioural and computational evidence that category learning benefits from delayed exceptions in adults, but not children
▶ Show abstract
Grouping information into meaningful categories allows us to generalize past knowledge to new experiences. But not everything fits within existing categories (“exceptions”; e.g penguins are birds but cannot fly). Prior research shows that category learning in adults benefits from delayed introduction of exceptions, perhaps because they can encode exceptions as separate from established categories. However, slow maturation of memory systems across childhood may mean that children struggle to learn exceptions, even when categories are well-established. In this study, children (7-9 years) and adults (18-23) completed a rule-plus-exception category learning task in which exceptions were introduced either at the start of learning (early) or only after many repetitions of category-consistent items (late). We found developmental differences in exception performance, with adults but not children benefitting from late exception introduction. In a network model of category learning (SUSTAIN), superior learning of late-introduced exceptions in adults was associated with recruitment of more clusters to represent exceptional items. Developmental differences across model parameters indicate that children may instead fall into a “learning trap”, failing to disengage from category rules to create separate representations for exceptional items. More broadly, these findings underscore the importance of aligning a learning curriculum to the learner’s neurodevelopmental state.
A Unified Statistical Learning Account of Early Face Representation in Infancy
▶ Show abstract
How does the developing visual system build stable representations of the social world? While infants' ability to discriminate individual faces is well-documented, real-world exposure involves navigating complex sets of faces. We propose a unified statistical learning framework: the infant brain actively employs divergent computational strategies to represent face sets based on visual experience.Across seven within-subject, cross-cultural experiments testing a massive sample of over 250 infants in Canada and China, we examined two foundational mechanisms of face-set representation: extracting central tendency (prototype formation) and extracting structural variance (principal component/dimension learning).We discovered a striking, experience-dependent double dissociation. Infants consistently extracted summary prototypes for unfamiliar other-race faces, a categorization bias that strengthened with age. By contrast, using a novel principal component learning paradigm, we found that infants successfully extracted the underlying dimensions of facial variation exclusively for familiar own-race faces.Theoretically, these contrasting trajectories reveal an active, adaptive visual system that optimizes processing based on environmental statistics. By specializing variance extraction for in-groups (facilitating individuation) and generic summary averaging for out-groups, this early computational bifurcation provides a powerful, mechanistic perceptual foundation for the developmental origins of social stereotyping.
The impact of sensory environment on sensorimotor integration in Autism and ADHD across development
▶ Show abstract
Reports from lived experiences of neurodiverse individuals (ADHD and Autistic) suggest that their sensory environment impacts sensorimotor integration. We used a VR environment to manipulate sensory load and rhythmicity. We assessed sensorimotor integration by measuring gait cadence, controlling for cognitive load. Participants included adults and children with ADHD (Ns=33 and 19, respectively) and Autism diagnoses (Ns=15 and 16, respectively), or neurotypical (Ns=33 and 26, respectively), with data collection ongoing. We observed differences in sensorimotor integration across diagnosis and development. Adults with ADHD showed a greater decrease in cadence in high versus low sensory-load relative to their neurotypical peers. This pattern was not observed in ADHD children. Autistic individuals, across development, did not show significant differences in the effect of sensory load from neurotypical adults. Entrainment to the sensory environment differed by diagnostic group but not by age. Autistic and neurotypical participants showed consistent entrainment, but both ADHD groups showed the opposite effect. In ADHD specifically, moving from a low to a high sensory-load environment prompted a slower cadence, and when that same sensory-load environment is made predictable in the rhythmic condition, gait cadence increases back to its original pace.
The influence of emotional consistency on toddlers' social behaviors
▶ Show abstract
Toddlers are highly attuned to others’ emotional expressions. However, emotions are not just isolated signals; their consistency over time provides critical information about a social partner’s predictability. Yet, how toddlers monitor and utilize this emotional stability remains largely unexplored. Across three studies, we bridge this gap by revealing that toddlers actively track the temporal consistency of others' emotions to guide their social and cognitive behaviors.Across four studies, 208 toddlers (aged 12-36months) were first exposed to informants who displayed either consistent or inconsistent emotional reactions toward objects. We found that emotional consistency drove social engagement: toddlers were significantly more likely to follow the gaze of informants with stable emotional responses. Strikingly, however, emotional inconsistency facilitated learning. In a word-learning study, toddlers acquired novel labels better from informants whose emotional responses varied across events. Finally, a live-interaction study extended these screen-based findings to reveal a developmental shift in social preference. While younger children preferred interacting with emotionally consistent individuals, older toddlers increasingly favored partners displaying greater emotional variability.Together, these findings establish that toddlers actively monitor the temporal stability of others’ emotions, flexibly deploying this information to optimize both social engagement and early learning.
SymposiumSymposium II · SC Room A · Sensory Processing in AutismRoom A▼ Show
Structural and functional Sensory Phenotypes in Autism and ADHD
▶ Show abstract
Sensory issues in Autism and ADHD cluster together, forming sensory phenotypes that predict clinical profiles. We mapped out structural and functional brain differences across sensory phenotypes using data from the Province of Ontario Neurodevelopmental Network (Autistic N=1388, ADHD N=1266). We used K-means cluster analyses to group individuals into five phenotypes. We analyzed cortical thickness and surface area from 76 brain regions (ASD N=180, ADHD N=126), and used a linear discriminant analysis to predict phenotype. Structural features predicted phenotype with 60% and 55% accuracy, respectively (chance=20%), with the most significant contributions in bilateral inferior and middle temporal gyri, left temporal pole, and inferior and superior parietal.We also analyzed resting-state data in Autistic participants (N=147) and calculated graph-theoretical functional-connectivity matrices for Yeo parcellation networks. We used a support vector classifier to identify regions that best differentiated phenotypes. Our classifier significantly predicted phenotype at 56.3%. Differences in network connectivity were observed across phenotypes, including in limbic, default-mode, visual, and sensorimotor networks.Sensory phenotypes are associated with broad differences in the brain’s functional and structural architecture not only in low-level sensory networks, but also networks associated with higher-level cognitive processes. This reflects findings showing that sensory differences cascade to influence higher-level cognitive development.
Enhanced Metacognition in Autism for Sensory-Based Perceptual Decisions
▶ Show abstract
Atypical sensory processing is increasingly recognized as a core feature of autism. Previous research has largely focused on overall performance differences, yielding inconsistent findings of both enhanced and reduced sensitivity. However, such approaches do not disentangle the multiple computational processes underlying perception. Here, we adopt a perceptual decision-making framework to decompose these processes, distinguishing between first-order decisions (stimulus categorization) and second-order processes (confidence judgments reflecting metacognition).
Using a Bayesian framework, we examined how autistic and non-autistic adults integrate three key sources of information: sensory uncertainty, prior expectations, and reward, across a series of orientation categorization tasks. Participants reported both their perceptual decisions and confidence, enabling independent assessment of perceptual inference and metacognitive evaluation.
Across experiments, autistic participants showed typical integration of sensory evidence, priors, and reward during first-order decisions. In all conditions, both groups deviated similarly from optimal Bayesian performance, indicating suboptimal but comparable inference strategies. Importantly, autistic participants flexibly updated their behavior in response to changes in prior probabilities, challenging accounts of reduced prior use.
In contrast, group differences emerged at the metacognitive level. Autistic individuals exhibited enhanced metacognitive sensitivity when decisions relied primarily on sensory evidence, suggesting more precise access to internal sensory uncertainty. However, metacognitive performance was comparable to controls when prior information or reward was integrated into decision-making.
These findings suggest that atypical perception in autism may arise not from differences in basic sensory integration, but from higher-order processes involved in evaluating perceptual decisions. By combining psychophysics with computational modeling, this work provides a unified framework for characterizing perceptual variability in autism and highlights the importance of dissociating perceptual inference from metacognitive evaluation.
Reduced Functional Specialization in Autism As Seen Through Contextual Influences on Perception and Action
▶ Show abstract
Altered sensory and perceptual processing is widely documented in autism, yet the mechanisms underlying these differences, specifically how contextual information shapes perception and action, remain poorly understood. One influential framework of the visual system, the Two-Visual Pathways Hypothesis, proposes a division of labour between the ventral stream (supporting perception) and the dorsal stream (supporting action), such that contextual information typically affects perceptual judgments but has limited influence on visually guided grasping. In this talk, I will present evidence from a series of experiments manipulating spatial context, stimulus history, and irrelevant object dimensions, demonstrating that while non-autistic individuals show the expected perception–action dissociation, autistic individuals exhibit contextual influences across both perceptual estimation and grasping tasks, as well as reduced differentiation in movement trajectories. I will discuss how these findings support a reduced functional specialization account and what they reveal about how contextual information is integrated across perceptual and motor systems in autism.
Utilitarian Attention: A Strengths-Based Approach to Studying Styles of Attending in Autism
▶ Show abstract
Autism, a condition historically recognized within a medical deficit framework, is increasingly understood as reflecting alternative ways of being (Pellicano & den Houting, 2022). Consistent with the neurodiversity movement, we consider the central role of attention research in the evolving discourse on potential strengths and distinctive styles of information processing in autism. Autistic individuals appear less bound by automatic semantic associations and generalizations, and instead rely more on information available in the immediate context (Brosnan et al., 2016; Pellicano & Burr, 2012). Accordingly, we propose that they use a utilitarian style of processing, prioritizing task-relevant, data-driven information (Brodeur et al., 2018; Burack & Brodeur, 2020).
We review the notion of utilitarian attention in autism along with evidence for it. We present findings from 31 autistic and 31 nonautistic participants on a correlated flanker task assessing adaptation when task performance depends on information that is counterintuitive to prior experiences. Autistic participants showed more rapid improvements than nonautistic participants in using predictive incongruent flanker cues (β = –16.52, SE = 7.55, p = .029), demonstrating enhanced sensitivity to task-specific information over time. Consistent with a utilitarian attentional style, they appear to capitalize on context-relevant cues, while remaining sensitive to prior semantic associations.
17:15 – 18:15Mid-Career Award Lecture · Second Student Centre · Rooms A, B & C (combined) · Plenary
Mid-Career Award Lecture17:15 – 18:15
Dr. Frank Russo
Toronto Metropolitan University
▶ Read bio
Adapted from a nomination letter written by Penny Pexman and Debra Titone
Dr. Russo received his Ph.D. in Brain, Behavior and Cognitive Science from Queen's University in 2002 and has since established an internationally recognized and highly interdisciplinary research program that has significantly advanced our understanding of auditory cognition, music perception, communication, and healthy aging. Dr. Russo's work is distinguished by its unusual breadth, theoretical coherence, and consistent translational impact. Across his career he has addressed fundamental questions about how humans perceive and use complex auditory signals—music, speech, and emotional vocalizations—while simultaneously shaping methodologies and technologies that allow these questions to be studied in ecologically valid contexts.
In music cognition, Dr. Russo has advanced a multimodal and embodied theoretical framework for the perception of singing and emotional speech, emphasizing the role of motor simulation and cross-modal integration. Equally important is his foundational program of research on vibrotactile music perception, which demonstrated that musical structure and affect can be conveyed through touch. By showing how complex temporal and spectral information can be represented in the deaf brain through a process of neural entrainment measurable using EEG, Dr. Russo helped redefine the sensory boundaries of music perception and provided a mechanistic framework for inclusive musical communication.
In the domain of hearing science and auditory aging, Dr. Russo has produced a sustained body of work that has influenced both theory and technology. His studies have generated key insights into how hearing aids process music and emotional speech and have helped shift the field away from speech intelligibility as the sole outcome measure toward metrics that capture listening effort, sound quality, affective communication, and real-world communicative success. These contributions have informed the development and evaluation of next-generation hearing-aid algorithms.
Dr. Russo's research has received coverage in major international and national media outlets, including CNN, The Guardian, and The Globe and Mail. He has been invited to deliver over 20 Keynotes and Plenaries. Dr. Russo has an exceptional and sustained record of training highly qualified personnel—in recognition of his outstanding mentorship and teaching, he received the Dean's Teaching Award in 2019. His service includes roles on the organizing committee for the 2014 annual meeting, as a member of the CSBBCS Executive (2019–2021), and on the Editorial Board of the Canadian Journal of Experimental Psychology (2012–2024). He received CSBBCS's Vincent Di Lollo Early Career Award and was elected a Fellow of the Society.
09:00 – 10:00Session 3 · Attention (Reading) · Memory (Episodic) · Perception (Faces) · Decision Making · Symposium III: Signaling Function of Social Gaze
Talk SessionS3A · HNE Rm 035 · Attention – Reading, Letters & Conscious AwarenessRoom 035▼ Show
Measuring Conscious Perception Interferes with the Allocation of Attention to Distractors
▶ Show abstract
A topic of interest in the attention capture literature is the interaction between attention and conscious perception. The purpose of this study was to determine whether the assessment of conscious perception of the distractor impacts the allocation of attention to it. We investigated whether measuring conscious perception of a distractor increases its relevance to the subject, causing attention to be allocated to the distractor voluntarily rather than captured involuntarily. To examine this possibility, subjects completed a Rapid Serial Visual Presentation (RSVP) task in which we measured two effects indicative of attention capture, the contingent capture effect and a spatial distance effect. When conscious perception was measured distractors captured attention irrespective of whether the distractor and target colour matched, but when it was not measured a robust contingent capture effect was consistently observed. Thus, we concluded that measuring conscious perception of a distractor interferes with the allocation of attention to the distractor, masking the hallmarks observed when attention is captured. The present findings suggest that previous claims made about the interaction between conscious perception and attention may be premature, necessitating either development of a new measure, or adjustments to existing measures.
The Surprise Signal: Attentional Reallocation During Category Rule Violations
▶ Show abstract
During new concept learning, attention is tuned to relevant features and systematically shifts when learning goals change. However, less is known about how attention operates when new information must be integrated into existing concepts. Using a delayed rule-plus-exception paradigm and eye-tracking measures, we investigated how attention is tuned and reallocated in response to learning conflicting information. Learners first acquired a general category rule before learning to categorize exceptions, stimuli that share features with members from the opposite category making them highly confusable. Eye movements revealed that attention was quickly directed towards diagnostic features during rule-learning and expanded to additional features when exceptions were introduced. After errors were encountered, participants broadened attention, however the degree of this adaptive reallocation of attention was related to end of learning performance. Additionally, participants whose focus was too broad early on in learning showed worse accuracy at test. These findings advance our understanding of how attention, learning, and memory interact to support flexible reconfiguration of existing concepts in memory.
Typo detection as a window into interactions between visual search and readability
▶ Show abstract
The interaction between stimulus properties and visual search remains understudied. In three experiments, we used typo detection to explore how factors like text appearance and attentional guidance influence search. In Experiments 1 and 2, participants (N = 17 each) completed a single-target search task, scanning “pseudoparagraphs” of random words for typos written in hard- or easy-to-read fonts (with stroke contrast manipulated in Experiment 1 and width, in Experiment 2). Though accuracy was unaffected, hard font slowed response times. Drift diffusion modelling isolated this change to slower evidence accumulation; nondecision time and decision threshold were unaffected. Evidently, manipulations to text appearance interrupt processing efficiency. Experiment 3 (ongoing) used a typo foraging task (0–7 typos/pseudoparagraph) with imperfect “spellchecker” annotations to explore how attentional guidance impacts search. Annotation validity was manipulated between participants. While annotations improved speed and accuracy regardless of validity, eye tracking data indicated a shift in strategy—with annotations, participants had shorter fixations and fixated fewer words; this effect was modulated by validity. This work highlights how stimulus properties influence search, with poor readability hampering evidence accumulation and reliance on automated aids impacting quitting threshold, pointing to shared processes across different forms of visual search.
The revised attentional disengagement model of the missing-letter effect
▶ Show abstract
When participants read text while searching for a target letter, they omit more targets and have longer reaction times (RTs) for high-frequency function words than for low-frequency content words. This missing-letter effect (MLE) has been used to investigate the processes involved in reading, but previous accounts have struggled to explain its empirical signatures. We present a computational version of the Attentional Disengagement (AD) model, which attributes the MLE to a faster disengagement of attention from function words during reading. This model was introduced by Roy-Charland et al. (2007), but previous studies were not designed to allow a systematic test of its predictions. Therefore, we created the longest text ever used to assess the MLE to generate reliable RT distributions. In two experiments, 80 participants completed a letter-search task during reading. Experiment 1 used rapid serial visual presentation to eliminate parafoveal processing, whereas Experiment 2 used a more typical text presentation with eye-movement processing. Both experiments revealed the expected results, and we successfully fit the AD model using a single set of parameters for each experiment. The AD model stands out as the first model to account for the patterns of omissions and RTs in the MLE using two different procedures.
Talk SessionS3B · HNE Rm 037 · Memory – Episodic, Emotional & AutobiographicalRoom 037▼ Show
From Scary Movies to Haunted Houses: Validating the Severity of Horror Media-induced Intrusive Memories (SHMIM) Scale
▶ Show abstract
Horror media-induced intrusive memories (involuntary thoughts or images that “pop” into mind following a horror movie) offer a controlled context for studying analogue trauma. This research introduces the Severity of Horror Media-induced Intrusive Memories (SHMIM) scale, designed to quantify these experiences. In Study 1 (N = 318), exploratory factor and graph analyses of the 36-item pool yielded a five-factor (Cognitive Distress, Physiological Reactions, Vividness, Intentionality, Perceived Control), 15-item solution with strong reliability (ω > .80). Scores correlated with self-reported retrospective frequency and duration of horror media-induced intrusive memories and were higher among horror non-fans than fans. In Study 2 (N = 333), confirmatory factor analysis supported the five-factor model with excellent fit and high standardized factor loadings (.76 to .90), alongside strong reliability (ω > .87). In Study 3 (N = 240), a Danish translation of the SHMIM administered to haunted house participants showed excellent fit with substantial factor loadings (.58 to .86). SHMIM scores, scare level, and days since the attraction predicted the frequency of intrusive memories of the haunted house. These findings indicate that the SHMIM has a robust five-factor structure, strong reliability, cross-linguistic generalizability, and can predict intrusive memories following a real-life frightening experience (i.e., haunted house).
Controlling Unwanted Memories: A Multisite Registered Replication of the Think/No-Think Effect
▶ Show abstract
The Think/No-Think (TNT) paradigm is an influential approach to studying the control of unwanted memories. Participants first learn a series of cue-target pairs (BEACH-AFRICA) until the cue reliably elicits the target; then for a subset of the cues (BEACH) they practice avoiding retrieval of the target (AFRICA). Evidence from this paradigm suggests that intentionally not retrieving unwanted experiences renders those experiences less accessible at test. This is referred to as the Suppression-Induced Forgetting (SIF) effect and has been linked to a variety of conditions involving unwanted past or future experiences. The SIF effect obtained with independent cues has been suggested to provide compelling evidence for the existence of cognitive inhibition. However, some researchers have questioned its replicability. The current multisite replication project clarifies these concerns by re-examining the TNT paradigm using two common instructional approaches (Thought Avoidance, Thought Substitution) and testing procedures (same-probe, independent probe) across 60+ sites, 11 languages and over 3000 participants. The SIF effect was observed to be small (d ~ .2) for all conditions excepting Thought Substitution when tested using the same cue at test (d ~ .8). Minimal variation was observed across sites. Future work will explore factors impacting the magnitude of the SIF effect.
Oculomotor and neural mechanisms of vividness in autobiographical memories
▶ Show abstract
Vivid recollection is theorized to be supported by the visual and oculomotor systems, which could help reinstantiate perceptual or scene-based details and the subjective sense of reliving. Past studies have independently found relationships between vivid memory and eye movements (e.g., more fixations), and between vivid memory and visual brain regions (e.g., precuneus, V1). Here, we examined relationships between all three: vivid memory, oculomotor activity, and neural activity. To do so, we reanalyzed data from a study in which 44 participants silently retrieved recent vs. remote autobiographical memories (AMs) during fMRI, and provided ratings of vividness/reexperiencing (Palombo, 2013). Eyetracking data during free-viewing AM retrieval was decoded using DeepMReye (Frey et al., 2021), a deep learning-based method of estimating gaze position via BOLD from eye voxels. Results indicate significantly higher fixation rate and higher gaze dispersion during recent AMs, despite being matched with remote AMs for visual content (verbal retrieval cue). Interestingly, eye movements during resting state were comparable to AM retrieval and significantly higher than a control task (odd-even number detection), suggesting overlap between these states. Our findings support the roles of oculomotor and visual systems in vivid AMs, and potentially, resting state as a proxy for AM retrieval.
Event Tagging: A Novel Intervention to Remediate Age Differences in Episodic Memory
▶ Show abstract
We experience the world as a continuous flow of information, but segment it into discrete events to be encoded into long term memory. While younger adults have been shown to build stronger associations within than between events, some work has shown that less distinct event boundaries may contribute to episodic memory differences in older adults. Leveraging this information, we developed an intervention aimed at improving event memory in older adults by making boundaries more distinct and reinforcing within-event associations. A large online sample of younger (N=181) and older (N=196) adults watched a movie and came up with keywords to describe what they just saw at event boundaries. Across age groups, memory for a film was benefited by this ‘event-tagging’ intervention compared to a low-level control task and normal, continuous viewing. This benefit was greater for within- than between-event memory. The intervention benefited the number of episodic details recalled by a subset of participants (N=55 younger, 65 older) in a free recall task 24 hours later. This suggests that our intervention may be helpful in reinforcing associations within events while maintaining distinct boundaries between them.
Talk SessionS3C · SC Room B · Perception – Face, Identity & Ensemble ProcessingRoom B▼ Show
Cultural Variations in the Visual Information Supporting Face Recognition
▶ Show abstract
Face perception is central to human social interaction. Within the first year of life, faces already capture infants’ attention, and adults spend approximately 12 minutes of every waking hour looking at faces. Given this high level of exposure and their profound social significance, faces are often considered a “special” class of stimuli, one for which humans develop exceptional perceptual expertise. Surprisingly, however, similar levels of expertise can be achieved while relying on different visual strategies. We describe a series of studies demonstrating that the core visual information individuals use to recognize faces varies with cultural background. Using complementary methods, including eye tracking (n=148) and data-driven classification image techniques applied on diverse samples ranging from Canadian and Chinese university students tested both in laboratory (n=48) and online (n=108) settings to participants from eight world regions recruited via Prolific (n=570), we show that individuals from Asian cultural contexts (including East Asia, South Asia, and the Middle East) rely more heavily on lower-granularity visual information (i.e., lower spatial frequencies) than participants from other regions of the world. We argue that a community-wide effort to increase diversity in participant samples is necessary to understand the cognitive processes linked to face perception.
How do we organize facial identity in memory?
▶ Show abstract
Our ability to recognize faces has important implications for social interactions, security systems, or eyewitness testimonies. Face recognition, however, is a perceptually difficult problem to solve. On one hand, all faces look similar to each other, sharing the same basic configuration of eyes, nose, and mouth (between-people homogeneity). On the other hand, one person's face can look very different across changes in viewing conditions such as lighting, makeup, or hairstyle (within-person variability). According to the face space theory, the visual system deals with between-people homogeneity by encoding faces in a multidimensional space based on their relative similarity. It is unclear, however, how within-person variability is represented in the face space. Observers (N = 331) provided similarity ratings to pairs of naturalistic images depicting familiar and unfamiliar faces. These ratings were analyzed using multidimensional scaling to visually construct a 3-dimensional face space. Analyzing the coordinates derived from this space, we found that familiarity decreased the distance between images of the same person. Familiarity, however, did not increase the distance between identities. We argue that the face space theory can provide a unifying account for the organization of facial identity in memory.
Reconstructing the Self-Face: Appearance and Belief in Perceptual and Memory Representations
▶ Show abstract
Self-representations are fundamental to cognition, yet the structure and accuracy of the images we carry of ourselves remain unclear. We combine behavioral and computational methods to estimate and evaluate self-face representations from both perception and memory. Participants provided pairwise similarity judgments for self, familiar, and unfamiliar faces, which we compared to representational spaces from artificial neural networks (ANNs). Further, we used behaviour-based image reconstruction to visualize internal representations. Discriminative ANNs matched self-face similarity judgments more closely than generative models, suggesting that self-representations emphasize higher-level, identity-invariant features rather than lower-level visual detail. Image reconstruction showed that self-face representations are recoverable from perception and memory, but with reduced fidelity relative to familiar and unfamiliar faces. Further analyses indicated that self-beliefs modulate these representations: higher self-concept clarity predicted greater reconstruction accuracy, and higher self-attractiveness ratings predicted more attractive self-representations, as revealed by reconstruction. To characterize heterogeneity in distortion across participants, we applied clustering across reconstructions and found four stable, anatomically interpretable bias modes (e.g., distortions across periorbital/eyebrow, nasal, and perioral regions). These results evinced structured but heterogeneous distortions rather than a single belief-linked axis. Together, our findings provide insights into visual self-representations and a novel framework for quantifying self-image content and systematic distortions.
The obligatory mean: Default ensemble encoding under uncertain task demands
▶ Show abstract
The efficiency of ensemble encoding (the synthesis of summary information from sets of items, e.g., mean orientation) is well-established. However, the instinctive and reflexive nature of the ensemble summary and the role of task instruction and feedback requires further examination. Here, participants were shown an ensemble of eight triangles with varied orientations and subsequently asked to choose between a target and distractor. The target was always the mean, and the distractor was either (1) an item from the set, (2) an item outside the set’s range, or (3) a mean from a different set. Participants were given ambiguous instructions to choose the “correct” item, and across six blocks of trials, some were given accuracy feedback (n = 39), and some received no feedback (n = 43). Interestingly, participants chose the mean as the “correct” item more than the distractors, and accuracy did not vary across block or feedback condition. However, participants receiving feedback showed a sharp decrease in response time across blocks compared to those without feedback. Moreover, participants were more sensitive to means from other sets compared with distractors outside the set range. Altogether, these results argue for a default, obligatory reliance on the mean during ensemble perception.
Talk SessionS3D · SC Room C · Decision Making – Judgment, Moral Cognition & BiasRoom C▼ Show
Crime and Punishment: Heuristic Retribution Supports Rational Deterrence
▶ Show abstract
Deterrence and retribution are common justifications for punishing transgressors. Based on the assumption of rational criminal behaviour, Deterrence Theory prescribes larger punishments given lower apprehension rates, and vice-versa. In four pre-registered, incentivized public goods game experiments (total N = 1250), we tested the effect of apprehension rate (between-subjects, low vs. high) on punishment assignments and free-riding behaviours. Studies 1-3 find punishers do not intuitively incorporate apprehension rate into their punishment assignments, even when they stand to gain from deterring free-riding and even when prompted to think about the deterrent effect of their punishments. Instead, we find that optimally deterring punishments arise as an emergent consequence of how players interact: Potential offenders rationally incorporate apprehension rate into their free-riding decisions (Study 4), while punishers incrementally increase punishments for repeat-offenders. A mathematical model shows that, under plausible assumptions, this dynamic approximates optimal deterrence over the course of multiple rounds. These results demonstrate that economically optimal outcomes can emerge as a consequence of heuristic retributive motives used by boundedly rational players.
Who Should Get Help? Moral Norms for Partiality are Influenced by Relationship Expectations
▶ Show abstract
Dominant ethical traditions praise impartiality as a moral ideal. Yet morality serves a socio-relational role in sustaining social ties. We argue that within close relationships, there is a strong moral preference for partiality. Study 1 (N = 565) featured sacrificial dilemmas in which actors either sacrificed one individual to save five others or vice versa. Actors were either partial, saving a close friend, or impartial, saving strangers. Impartial actors were judged as less moral and less predictable compared to partial actors, suggesting a moral preference for partiality. We also show that positive moral perceptions of partiality are mediated by reputational mechanisms such as predictability, as favouring close others can signal being a predictable cooperation partner. Study 2 (N = 1017) featured vignettes in which actors were either partial and helped a close other (relative, friend, or community member) or impartial and helped a stranger. As in Study 1, impartial actors were perceived as less moral and less predictable, suggesting a moral preference for partiality. A significant relationship-by-partiality interaction revealed that these effects were amplified for closer relationships: actors helping a stranger over a relative were judged particularly harshly. These findings have implications for understanding cooperative behaviour and charitable giving.
Fooled or Enlightened by Frequency: When Frequency Tricks Decision-Making
▶ Show abstract
Humans tend to prefer an option that resulted in more frequent rewards due to more frequent sampling (e.g., 7 rewarding outcomes in 10 samples = 70% reward rate) over an option that exhibited a higher chance of reward in less frequent sampling (e.g., 3 rewarding outcomes out of 4 samples = 75% reward rate). In two experiments (total n = 171), we tested whether this frequency bias reflects an overestimation or more accurate estimation of reward rate for the frequently sampled option. When the overall reward rate was high (i.e., > 50%, Experiment 1), participants preferred a frequently sampled but less rewarding option over an infrequently sampled but more rewarding option. Critically, although frequent sampling increased the estimated reward rate towards the actual reward rate, it never surpassed the actual reward rate. Furthermore, when the overall reward rate was low (i.e., < 50%, Experiment 2), even though the frequently sampled option resulted in more frequent rewards than the infrequently sampled counterpart, individuals preferred the infrequent option due to more accurate reward rate estimation for the frequent option and inflated estimates for the infrequent option. Taken together, the frequency bias may reflect a rational use of information that is frequently sampled.
Truthiness Effects: Looking Under the Hood
▶ Show abstract
The “truthiness effect” refers to the finding that obscure claims (e.g., “The navy bean is also known as the sieva bean”) are more likely to be judged to be true if they are presented alongside a related image (e.g., a photo of some navy beans) than if they are presented without a picture (Newman et al., 2021). Our mega-analysis applying item-level analyses to data from six truthiness experiments (N = 1,300) suggests that there are two qualitatively different mechanisms at play. One mechanism is that non-probative related photos facilitate the processing of claims and that fluent processing of claims contributes to a feeling of familiarity or correctness, producing a small increment in the probability of judging the claim to be true. The second mechanism is that subjects sometimes perceive the picture as providing evidence as to the truth or falseness of the claim. The 56 photos used in our studies were intended be non-probative, but our analyses indicate that 12 of them were perceived as providing evidence of truth and 5 of them were perceived as providing evidence of falseness.
SymposiumSymposium III · SC Room A · Signaling Function of Social GazeRoom A▼ Show
Minds in motion: Inferring mental states from observed kinematics
▶ Show abstract
A central challenge for vision is identifying important stimuli in the environment, and we often rely on a simple cue: others’ gaze. Indeed, observers automatically orient their attention in the direction others look. But there is more to gaze than direction: people look with different intentions. For example, head turns can be driven by internal goals (“endogenous”; e.g., turning to reach for something) or external stimuli (“exogenous”; e.g., turning toward a sudden noise). Can observers discriminate these intentions when observing others’ turns? In Experiment 1 (N=100), participants watched videos of actors turning their heads and judged each turn as endogenous or exogenous. While observers were unable to discriminate actors’ intentions above chance, their judgments were biased by turn speed: slower turns were perceived as endogenous and faster ones as exogenous. In Experiment 2 (N=50), we replicated this effect with a direct manipulation of turn speed. Finally, in Experiment 3 (N=30), we examined whether perceived intentions influence observers’ attention. Participants localized targets appearing along vs. away from actors’ turn directions, and we examined response times for endogenous vs. exogenous turns at various speeds. This work shows how observers perceive intentions underlying human movement, uncovering the fundamental building blocks of social perception.
The meaning of gaze depends on social roles
▶ Show abstract
Eye gaze plays a central role in social interactions, with its communicative value often depending on the social roles individuals occupy. Here, we investigated how group leadership structure influenced gaze dynamics during naturalistic interactions. While wearing dual mobile eye-trackers dyads performed a joint task with a group leader either assigned or allowed to emerge naturally. We examined mutual and unidirectional gaze patterns during group interactions as well as partner-rated leadership impressions ratings. Group leadership ratings differed across groups and the leadership structure modulated gaze behaviours. Groups with an assigned leader showed a strong bias toward eye-to-mouth mutual looks, whereas groups with emerging leaders distributed mutual gaze more evenly across eye-to-eye, eye-to-mouth, and mouth-to-mouth looking combinations. The group dynamics in mutual looks did not directly predict individual leadership judgments. Instead, leadership impressions were selectively linked to unidirectional looks within groups with assigned leaders, with group members who looked more to their partner’s eyes and body, and less to the mouth, rated as more leader-like. Together, these findings show that gaze is a socially role-dependent signal: its social meaning is not intrinsic to specific gaze patterns but emerges from dynamics within a given group structure.
Thinking or Looking? Humans are sensitive to others’ mental states underlying gaze aversion
▶ Show abstract
Two people in conversation typically alternate between periods of mutual gaze and periods of gaze aversion. Gaze aversion can reflect different mental states, sometimes occurring in order to inspect the environment (looking gaze aversion) and other times to disengage from mutual gaze while thinking (thinking gaze aversion). Correctly inferring the underlying mental state may help interlocutors coordinate their attention. However, it remains unclear whether observers can accurately distinguish these two types of gaze aversion and which visual cues support such judgments. In this study, observers viewed short, muted video clips of actors averting their gaze during conversation and rated whether the actors were thinking or looking. Analyses of actors’ eye and head movements confirmed distinct kinematic patterns for these two types of gaze aversion. Moreover, observers were able to reliably distinguish them based on specific eye and head movement signals. Thinking judgements were associated with more blinks, fewer and shorter fixations, less head movement, larger eye movements, and greater head-eye decoupling than looking judgments. These findings show that gaze-aversion behavior contains rich, naturally occurring kinematic signals associated with underlying mental states, and that these signals can be accurately perceived by independent observers.
Covert orienting: The dark matter of social attention
▶ Show abstract
Although eye and head tracking are widely used to study attention across domains such as psychology, medicine, and emerging VR/AR technologies, they rest on a largely unexamined assumption: that gaze reflects attention. This "eye-mind" hypothesis simplifies measurement but overlooks a critical reality -- that attention is often covertly directed away from where gaze is directed.
This mismatch is especially pronounced in social contexts, where gaze serves not only perceptual but also communicative functions. People actively manage where they look to signal intentions or avoid scrutiny, while simultaneously attending to other, unseen targets. As a result, covert attention becomes a functional necessity, enabling individuals to gather information without revealing their focus.
I argue that this hidden layer -- the "dark matter of social attention" -- is routinely neglected in current research. Methods relying solely on visible gaze risk producing incomplete or misleading accounts of cognition and behavior, particularly in naturalistic and interactive settings.
Understanding real-world social behavior therefore requires moving beyond eye and head movements to incorporate covert attentional dynamics. Until we confront this unseen force, our models of behaviour will remain incomplete shadows of the truth.
12:35 – 13:35Session 4 · Attention (Social/Clinical) · Memory (Metamemory) · Social Cognition (Gaze) · Learning · Symposium IV: Agentic Contributions
Talk SessionS4A · HNE Rm 035 · Attention – Social, Sustained & ClinicalRoom 035▼ Show
Weapons, Faces, and Attention: Understanding the Cognitive Mechanisms Underlying the Weapons Focus Effect
▶ Show abstract
Eyewitness testimony is critical in legal contexts but is often compromised by attentional and emotional biases. The Weapons Focus Effect (WFE) occurs when a weapon captures attention and reduces memory for peripheral details, such as a perpetrator’s face. This Registered Report study investigated the cognitive mechanisms underlying this effect, and moreover whether individual differences in social avoidance intensify attentional disruption. Using a within-subjects visual search paradigm, participants detected a neutral target (a butterfly) among distractors that vary in salience: neutral objects, faces, and weapons. Reaction time and accuracy indexed attentional capture, and a subsequent surprise face-recognition test measured memory encoding. Participants also completed the Social Avoidance and Distress Scale (SADS) to examine how trait social avoidance predicts performance costs in weapon-present trials. Results indicated that both faces and weapons independently slowed visual search, suggesting that each type of stimulus captured attention. Face recognition accuracy was significantly lower for faces encoded alongside weapons compared to faces encoded alone, consistent with the Weapon Focus Effect. Additionally, higher SADS scores were associated with improved face memory. These findings suggest that weapon presence disrupts memory for faces, even when attentional effects during visual search are not strongly pronounced.
Why Some Minds Read Minds Better: Gradient Reconfigurations Reveal Individual and Age Differences in Mentalizing
▶ Show abstract
Mentalizing—the capacity to infer others’ mental states (e.g., thoughts, feelings)—is essential for adaptive social functioning and varies substantially across individuals and the lifespan. However, most neuroimaging studies focus on isolated brain regions or networks, overlooking how the brain’s hierarchical organization supports the integration of distributed processes underlying this capacity. We applied a system-level approach based on gradients of large-scale brain organization to investigate individual and age-related differences in mentalizing. We analyzed fMRI data from two large datasets (n=252; n=260) in which participants made social (mentalizing) and factual (control) inferences. Task-evoked activation patterns were projected onto principal gradients of cortical organization derived from independent Human Connectome Project data. Mentalizing elicited systematic shifts along multiple large-scale gradients, including reconfiguration along Gradient 3, which differentiates the default mode network from frontoparietal control systems. The magnitude of this reconfiguration predicted individual differences in mentalizing ability. Late adulthood (ages 65-77) was associated with reduced mentalizing-related reconfigurations along gradients, indicating diminished neural flexibility with age. Our findings suggest that mentalizing involves coordinated brain-wide shifts aligned with the brain’s principal topographical gradients. Gradient-based markers of large-scale neural reconfiguration reveal how brain organization supports mentalizing and may help identify individuals at risk for social dysfunction.
The Effects of Prescription Stimulants on Mind-Wandering in Individuals with ADHD
▶ Show abstract
Prescription stimulants are a frontline treatment for ADHD, yet their effects on the cognitive mechanisms underlying attention remain poorly understood. Given elevated rates of mind-wandering observed in individuals with ADHD, and the role of mind-wandering in disengaging attention from tasks, examining how stimulants influence mind-wandering may provide key insights into their therapeutic mechanisms. Using a within-subject design, young adults with ADHD taking either Adderall XR or Vyvanse completed the Metronome Response Task (MRT) before and after medication in a laboratory setting. Consistent with prior work, participants completed a baseline block followed by blocks instructing them to mind-wander 20% or 80% of the time. Stimulant use was associated with significant reductions in both spontaneous and deliberate mind-wandering, and with improved task performance. Notably, despite stimulants reducing overall mind-wandering rates, the characteristic effects of mind-wandering instructions on both reports and task performance were substantially attenuated and largely unaffected by medication, suggesting that individuals with ADHD may have a fundamental difficulty in voluntarily regulating their mind-wandering that persists independent of pharmacological treatment. We also observed that deliberate mind-wandering was more strongly associated with response timing variability than spontaneous mind-wandering. These findings illuminate the cognitive mechanisms by which stimulants support attention regulation in ADHD.
The Effect of a Brief Mindfulness Intervention on Negative Affect and Sustained Attention
▶ Show abstract
Sustained attention is aversive, and brief mindfulness-based interventions (BMBIs) have been associated with improvements in sustained attention. We propose that BMBIs support sustained attention by making it less aversive, specifically by fostering better state emotion regulation and subsequent levels of negative affect. To test this, we recruited 191 undergraduate participants and randomized them to a BMBI condition or an audiobook control condition. They completed a 20-minute continuous performance task (CPT) to measure sustained attention, with measures of state emotion regulation and negative affect administered between CPT blocks. Findings revealed that BMBIs were associated with fewer omission errors and commission errors at later time points in the CPT, suggesting that BMBIs can help to buffer against a performance decrement. The BMBI condition was also associated with lower negative affect earlier on in the CPT. A serial mediation model testing the effect of condition on CPT performance through the path of state emotion regulation and negative affect was not supported. Condition did not predict state emotion regulation, and negative affect did not predict CPT performance. Condition and state emotion regulation independently predicted negative affect during the CPT. Interpretations and limitations of the study will be discussed.
Talk SessionS4B · HNE Rm 037 · Memory, Metamemory & HealthRoom 037▼ Show
Cancer Treatment and Memory in Breast Cancer Survivors: Differential Effects on Detailed Episodic and Gist-Like Memory Recall Over Time
▶ Show abstract
Breast cancer survivors (BCS) frequently report cognitive difficulties following chemotherapy, yet standard episodic memory assessments may underestimate real-world memory impairments. This study examined the long-term impact of cancer treatment on naturalistic episodic memory using ecologically valid paradigms and standard cognitive measures. Participants included BCS at least six months post-treatment and age-matched non-cancer controls. In one study, participants viewed short film clips, with recall tested immediately and after a 7-day delay, assessing central (gist-like) and peripheral (episodic) details. BCS recalled fewer central and peripheral details than NCs, with pronounced deficits in encoding and early retrieval of sensory and contextual features. In a complementary study, participants completed the Taler Stories and Verbal Paired Associates (VPA) tasks. BCS exhibited impairments in both gist and verbatim recall for Taler Stories, persisting at delayed testing, but performed comparably to controls on the VPA task. BCS also reported higher cognitive workload and greater subjective difficulties with episodic and semantic memory. Together, these findings indicate that chemotherapy-treated BCS experience subtle but meaningful impairments in complex, naturalistic event memory that are not fully captured by standard memory assessments. Ecologically valid paradigms may therefore offer a more sensitive approach to detecting treatment-related cognitive changes.
The Impact of Mental Health on Relational Memory Performance
▶ Show abstract
Broad memory challenges arise for people with clinical mental health disorders, partly due to changes in hippocampal structure and function. However, memory impairments are also evident in individuals with subclinical symptoms, even without a formal diagnosis. The present study examined how mental health symptoms influence performance on two relational memory tasks that depend on hippocampal function. Participants across the adult lifespan completed a transverse patterning task (n=622) for which they learned directional relations among abstract objects. In a separate experiment, participants (n=295) completed a transitivity task, learning relations between object pairs and making inferences across them. All participants completed the Depression, Anxiety and Stress Scale (DASS-42). Logistic regressions examined performance on each task and interactions among predictors. Greater anxiety symptoms were associated with reduced accuracy on both tasks. Depression symptoms were associated with accuracy declines only when participants also reported greater stress levels. Older participants performed less accurately on both tasks, with even greater deficits on the transitivity task for those experiencing stress and depression. These findings demonstrate that even sub-clinical levels of depression, anxiety, and stress can negatively impact hippocampal-dependent memory, underscoring the importance of early mental health support for maintaining cognitive and brain health across the lifespan.
Examining Temporal and Spatial Components of Episodic Metamemory
▶ Show abstract
Episodic metamemory, the ability to accurately monitor and interpret one’s episodic memory performance, has important functional implications, such as informing the use of external memory aids and internal strategies. However, it is unclear what factors influence our accuracy in judging our memory performance for past events. Previous research has focused on controlled experimental tasks which may not capture the richness of mental representations constructed when remembering past episodes. The current study explored the influence of different elements of episodic memory (space vs. time) and stimuli richness (wordlist vs. video-clips) on metamemory monitoring performance. Participants aged 18-28 years completed spatial (n=52) or temporal (n=52) memory tasks. First, participants encoded a series of counterbalanced video clips and wordlists. They next performed spatial or temporal associative recognition tasks, which involved indicating which side of space images from clips or words from lists appeared on (spatial), or which of two words or images came first during encoding (temporal). After each recognition trial, participants made a metamemory judgement, rating their confidence in their memory performance. Confidence ratings significantly predicted memory performance overall, but individuals were particularly overconfident in their ability to recognize spatial elements from videos, suggesting rich spatial information may preferentially influence memory confidence.
Metacognitive Effort Regulation in Problem-Solving
▶ Show abstract
People are often described as cognitive misers, avoiding effortful tasks when given a choice. While effort avoidance has been studied through task choice, it is unclear whether these findings extend to in-the-moment decisions, in which people can opt out of effortful tasks. We propose that both forms of avoidance are driven by the metacognitive monitoring of effort demands from salient task features (Ackerman & Thompson, 2017). Study 1 (N=50) establishes effort avoidance in a problem-solving context using an adapted demand selection task (DST; Dunn et al., 2016), in which participants rate the relative difficulty of anagram pairs varying in length and pronounceability and then choose which anagram they would prefer to solve. We show that anagrams are avoided based on perceived difficulty, with harder items avoided in favour of easier ones. In Study 2 (N=100), participants first complete the DST, then attempt to solve anagrams, with the option to opt out at any time. We predict that if prospective and in-the-moment avoidance correspond, anagrams avoided in the DST should also be opted out of more frequently during solving. These studies extend our understanding of effort avoidance to real-time problem-solving contexts.
Talk SessionS4C · SC Room B · Social Cognition – Face, Gaze & Social InteractionRoom B▼ Show
You Move Like Me: Stable Motor Compatibility Between Partners Persists Across Social Contexts
► Show abstract
Daily activities often unfold alongside others, yet whether simultaneous co-action promotes interpersonal motor alignment without a shared goal remains unclear. Pairs of young adults (n = 48, 24 dyads) grasped objects in a fixed sequence, first individually then face-to-face, with no instruction to coordinate. Hand kinematics were recorded via motion capture, and Dynamic Time Warping (DTW) quantified trajectory similarity between partners. Co-presence did not enhance dyadic kinematic similarity. In incongruent trials, where partners grasped different-sized objects, dyadic similarity trended toward greater dissimilarity, suggesting possible coordination interference rather than facilitation. Critically, dyadic similarity was strongly predicted by baseline similarity measured when partners acted alone, across both congruent (r = 0.706) and incongruent (r = 0.768) trials, indicating an object-independent effect. Permutation testing confirmed this relationship is pair-specific and not reducible to individual motor traits — within-subject consistency did not predict dyadic DTW. Furthermore, within-subject movement variability remained stable across individual and dyadic contexts, suggesting co-action did not destabilize individual kinematics. Together, these findings indicate that interpersonal motor alignment during reach-to-grasp co-action reflects stable kinematic compatibility between specific partners rather than socially induced entrainment.
Multimodal dyadic communication behaviour: a database
▶ Show abstract
Dyadic communication contains a wellspring of data involving tightly coupled head, face, and gaze behaviours, that complement and validate communicated verbal information. We introduce a new dataset capturing over 100 dyadic conversation sessions (50+ hours of data) between interlocutors that engage in 25-30 minutes of unscripted dialogue. The dataset combines audio and video recordings with high precision motion capture (OptiTrack Motive 3), wearable eye tracking (Pupil Labs Neon) and mobile tracking (iPhone 15 Pro TrueDepth camera). Multimodal, accurately synchronized recordings include 6 DOF head and eye pose, fixations, saccades, blinks, pupillometry, facial motion (in the form of 52 ARKit blendshapes), audio, and egocentric and allocentric video streams.Our current use cases include the assessment of social signalling processes including interpersonal synchrony as a means to study communication efficiency across different communication platforms, the training of a generative foundation model to apply visual chatbots with social competence, and to directly assess consumer grade head and gaze tracking systems against high-fidelity ground truth.The dataset continues to grow. As we are adding new behavioural paradigms and diversifying the participant population, we are also inviting collaborations from both social psychologists and computer engineers to further exploit the full potential of the database.
Perspective taking in real life
▶ Show abstract
While laboratory research has established that humans readily adopt the mental perspective of others, it remains unknown whether this behavior also occurs in natural environments. Here, we examined this question using a novel paradigm in which two participants completed a visual perspective-taking task together. Dyads were seated at a square table either next to one another (“Shared” mental perspective) or across from one another (“Non-shared” mental perspective). On each trial, a target with multiple possible spatial representations was projected onto the center of the tabletop. Participants identified the target in alternating sequence using either their own or the other participant’s mental perspective. Consistent with past work, participants were slower when responding from the perspective of the other person. However, this effect emerged only when mental perspectives were non-shared and not when they were shared. Furthermore, when perspectives were non-shared, participants were faster when responding from their own perspective. Together, these results demonstrate that perspective taking can be measured in naturalistic tasks, but that the dynamics of mental state attribution may be more complex than those previously found in passive laboratory tests.
Talk SessionS4D · SC Room C · Learning – Category Abstraction & Cognitive MechanismsRoom C▼ Show
Gist-first Learning Facilitates Category Abstraction Across GAN-generated Scene Space
▶ Show abstract
Organizing information into categories is a powerful cognitive ability that supports later generalization in unfamiliar contexts. While such category learning is often probed using simple stimuli, less is understood about the learning conditions that best promote parsing and generalizing categories from continuous stimuli that approach real-world complexity. To test this, we used an approach pioneered in previous work (Son et al., 2024) that generates a multidimensional space of continuously varying naturalistic stimuli, using embeddings from a generative adversarial network. We arbitrarily bisected this space with a 'decision boundary', creating novel categories on either side. Participants were divided into a “hard” group that was trained on disproportionately more exemplars close to the decision boundary than a complementary “easy” group. After learning, we tested participants' ability to generalize their learning beyond the space of the trained exemplars. While the “easy” group unsurprisingly performed better, this was unexpectedly driven by the hardest exemplars, despite the group having experienced them dramatically less. The easy group also extrapolated better, implying that they extended their enhanced knowledge of coarse category distinctions to guide their judgments about even the most distant cases. Together these results indicate that generalization of category learning benefits from simplicity, not difficulty.
Category-exemplar learning: Dissociations in memory and metamemory
▶ Show abstract
Repetition and variability support learning and memory in distinct ways. Repetition strengthens memory for specific items whereas exposure to related information promotes encoding variability, often improving generalization. Yet learners often misunderstand these benefits, overvaluing repetition because repeated information feels more fluent. We examined how repetition versus exemplar change influences both memory performance and beliefs about memory in a category-exemplar learning paradigm. In Experiment 1, participants studied 30 category-exemplar pairs: 10 appeared once (e.g., FRUIT-peach), 10 were repeated identically (e.g., ANIMAL-goat…ANIMAL-goat), and 10 were presented with two different exemplars (e.g., OCCUPATION-doctor…OCCUPATION-pilot). At test, participants recalled either the category names or the exemplars. Our results showed a clear dissociation: Recall of categories was greatest when exemplars changed whereas recall of exemplars was greatest for repetitions, demonstrating the selective benefit of variability and repetition, respectively. In Experiment 2, participants read a description of one of the learning conditions from Experiment 1 and reported their beliefs regarding memory performance under that condition. Despite the memory advantage of variability for category recall observed in Experiment 1, participants predicted better performance for repeated pairs under both learning conditions. These findings highlight robust metacognitive biases favouring repetition even when variability more effectively supports learning and memory.
Event Segmentation During Lecture Videos and Its Relation to Learning and Thought Content
▶ Show abstract
Students experience lectures as continuous streams of information, but their comprehension may depend on how this information is segmented into meaningful units. Event segmentation theory proposes that people naturally parse ongoing experience into discrete events, supporting comprehension and memory. However, little is known about the downstream impact of event segmentation in educational contexts. In this study, undergraduate participants watched an educational video (evolutionary psychology or economics; ~30min) and indicated when they perceived event boundaries throughout. Before and after the lecture, participants completed a novel free association word generation task, designed to capture spontaneous thoughts that suggest lecture-related concepts were lingering in mind. Last, participants completed a comprehension test. Preliminary analyses suggest that participants who identified more event boundaries tended to have lower comprehension scores, suggesting that excessive segmentation may disrupt a higher-order conceptual understanding of the content. In addition, free association responses became more semantically aligned with lecture themes after viewing the lecture, which revealed stronger conceptual representations and the lingering influence of lecture content that may not be directly reflected in comprehension scores. These findings extend event segmentation research into educational contexts and highlight free association as a promising tool for capturing task-unconstrained conceptual traces left by learning experiences.
Evidence of allocentric learning in male rats with large lesions of the hippocampus
▶ Show abstract
The hippocampus (HPC) is the neural substrate of allocentric spatial representation–viewpoint-invariant cognitive maps. HPC lesions disrupt performance on allocentric tasks such as the Morris Water Task (MWT), in which rodents learn the location of a platform submerged within a circular pool. Navigation from varied start points requires integrating multiple information sources, including allocentric representations of the environmental cue configuration. Rats with HPC lesions, however, may learn to navigate by using a less effective alternative strategy; a body-centered (egocentric) cue-based search strategy. Here, we investigated whether HPC lesion size predicts the shift in reliance from egocentric to allocentric strategy use in the MWT. Control and HPC groups showed evidence of learning, with the control rats adopting an allocentric strategy after a few trials. The HPC rats, in contrast, showed more heterogeneous strategy use with 19% adopting a consistent egocentric strategy and notably, 15% adopting a consistent allocentric strategy. Performance did not correlate with lesion size. These findings suggest that rats with large HPC lesions acquire and use allocentric spatial information in the MWT, highlighting the potential contributions of other brain regions to spatial learning and memory.
SymposiumSymposium IV · SC Room A · Using Task-Optimized ANNs to Understand Brain & BehaviourRoom A▼ Show
Can the Brain Predict the Model of the Brain? Rethinking Brain–Model Alignment
▶ Show abstract
Systems neuroscience increasingly relies on large-scale computational models to understand how neurons give rise to behavior. How good are these models? An intuitive benchmark comes from asking: how well can one primate brain predict the activity of another? A model that is fully brain-aligned should show a similar symmetry with the brain. In practice, however, model–brain comparisons have only followed a one-way approach. Artificial neural networks (ANNs) are typically evaluated by how well their features predict neural responses (forward predictivity), and not whether the ANN's internal responses are equally predictable from primate brain activity. Addressing this gap, we introduced reverse predictivity (Muzellec et al., 2025), quantifying how well neural population activity predicts individual ANN units. Applying this bidirectional framework to macaque inferior temporal cortex we revealed representational mismatches that forward metrics alone fail to detect. Interestingly, factors improving forward predictivity (e.g., increasing model capacity, optimizing single-task performance) reduced reverse predictivity. To avoid such a bottleneck in current model design, we further demonstrated (Ziaee et al., 2025) that ANNs trained with multiple, behaviorally meaningful objectives can achieve improvements in both directions of predictability. Together, these results motivate bidirectional evaluation as a general principle for assessing and developing brain-like ANN models.
Modeling the Role of Early Visual Constraints in the Emergence of Shape-Based Object Recognition
▶ Show abstract
Studies of visual development suggest that early perceptual constraints shape object learning. However, it remains unclear which visual constraints are most influential and how they alter downstream object recognition behaviour. To investigate these questions, we simulated the effects of early visual immaturities during the training of deep neural networks, and examined their downstream performance on object categorization tasks. ResNet-50 models were trained on ImageNet using a self-supervised contrastive loss and a developmental curriculum in which training images were progressively augmented to approximate developmental changes in visual acuity, chromatic sensitivity, or contrast sensitivity from infancy to adulthood. We evaluated the resulting models on cue-conflict stimuli that dissociate shape and texture information. Models trained with contrast sensitivity constraints showed a stronger reliance on shape cues during categorization compared to models trained on the other curricula, producing patterns more closely aligned with human behaviour. However, the same model also exhibited reduced accuracy on standard classification benchmarks. This pattern suggests that the observed increase in shape bias may arise from reduced reliance on texture, a key feature used by neural networks for categorization. In sum, developmental visual constraints systematically shift the features that models rely on, with measurable consequences for downstream object recognition performance.
Convergent neural dynamical systems for task control in artificial networks and human brains
▶ Show abstract
We can flexibly switch between tasks—e.g., switching from speaking English to speaking French—but the neural mechanisms supporting task switching remain poorly understood. For decades, cognitive psychologists have debated the extent to which task switching involves preparation for upcoming tasks versus interference from previous tasks. However, it has been difficult to adjudicate between these theories without concrete formalization or neural predictions. We approached this problem through two key innovations. First, we developed a gated recurrent neural network (RNN) model of task switching, training RNNs to optimize task performance under theoretically motivated curricula. Second, we quantified and compared task dynamics between theory and data by developing a protocol for fitting high-dimensional linear-Gaussian state space models to both task-optimized RNNs and two human EEG experiments. Our control theoretic analyses of RNN dynamics uncovered a novel control strategy in RNNs trained to actively switch tasks: recovering to a neutral task state between trials. Comparing the latent dynamics between RNNs and EEG revealed that humans appear to implement this same strategy. Through new methods for comparing flexible information processing between artificial and biological neural networks, this work provides a major revision to long-standing cognitive theories of task switching.
Sensory expectations shape neural population dynamics in motor circuits
▶ Show abstract
Movement preparation has been studied extensively for self-initiated actions, but corrective responses to mechanical perturbations—which are often predictable—also benefit from preparatory neural activity. Michaels et al. show that when humans and monkeys receive probabilistic cues about the direction of upcoming perturbations, sensory expectations are incorporated into motor cortical preparatory states, improving corrective responses. To explain the neural geometry underlying these expectations, the authors trained recurrent neural networks coupled to a biomechanical model of the arm (MotorNet) to perform the same perturbation task. Crucially, the model reveals that the observed geometry—where preparatory signals scale linearly with perturbation probability and a condition-independent signal rapidly shifts the neural state at perturbation onset—emerges only when sensory feedback first signals that a perturbation has occurred before resolving its direction. This two-stage feedback structure, in which an initial unsigned detection signal precedes direction-specific information, is the key architectural feature that gives rise to the experimentally observed population dynamics. The model thus provides a normative, mechanistic account: preparation for expected perturbations configures the network so that a rapid, initially expectation-driven response can be launched before full sensory resolution, unifying self-initiated and perturbation-driven motor preparation under a common computational framework.
13:45 – 14:45Session 5 · Cognitive Neuroscience (Face/Emotion) · Visuomotor & Spatial · Lifespan & Health · Language (Semantics) · Symposium V: Neural Aesthetics
Talk SessionS5A · HNE Rm 035 · Cognitive Neuroscience – Face, Emotion & EnsembleRoom 035▼ Show
Neural correlates of experience across movie-watching paradigms
▶ Show abstract
Behaviour while viewing naturalistic stimuli, such as movies, provides a rich model of contextual experience that closely resembles the continuous nature of daily life, and can therefore clarify how brain states change over time. In particular, we have demonstrated that common thought patterns have distinct neural mappings that appear replicable across films. However, recent research shows that brain synchronization is highly variable across movies, challenging their purported utility for establishing generalizable neural associations. Here, we test whether mapping neural correlates onto thought patterns reveals consistencies in the neural mechanisms that are either shared or deployed differently across films. We used multi-dimensional experience sampling (mDES) to measure descriptions of thoughts (e.g. sensory features) in one sample watching eight movie clips and used random-effects modelling to predict brain states in a separate sample watching the same clips in the scanner. Across movies, sensory experiences consistently predict activity in visual and auditory cortices. However, regions in higher-order association cortex, including default mode network, were implicated in two opposing states – one predicted by past-knowledge thoughts that supported improved comprehension, and one predicted by distracting thoughts and worse comprehension. Our findings support the utility of mDES for establishing cognitive-neural mappings in naturalistic contexts.
Deep Brain Ultrasound Augments Human Attention
▶ Show abstract
Deep brain ultrasound offers a novel means of modulating human cognition by noninvasively targeting subcortical structures that were previously accessible only through invasive procedures. While decades of research have mapped cortical circuits of attention, the causal roles of deep hubs such as the basal ganglia and thalamus remain poorly understood in the healthy human brain.To test whether low-intensity transcranial ultrasound stimulation (TUS) of two nodes in the basal ganglia-thalamic network, the globus pallidus internus (GPi) and the pulvinar, causally alters visual attention. We hypothesized that TUS-induced modulations in attentional performance would be site-specific, reflecting distinct circuit functions.Across sessions, focal TUS accelerated reaction time in a visual search task, indicating augmented attention. Reaction time improvements were observed after stimulation relative to baseline. A dissociation emerged across sites: both GPi and pulvinar enhanced reaction times, but pulvinar yielded more robust benefits for target-present trials at peripheral eccentricities, and improved search efficiency in the same trials.These findings provide causal evidence that human attentional control can be steered at deep subcortical sites. TUS offers a practical approach for dissecting circuit-level contributions to cognition and a potential noninvasive avenue for enhancing attention and other cognitive or affective functions.
Facial expression ERPs are related to perceived valence and arousal
▶ Show abstract
Identifying others' emotional expressions is important for our social lives, and the temporal course of this process is often investigated using event-related potentials (ERPs). Previous work has proposed that certain ERP components may index the extraction of affective facial information, such as valence (positive/negative) and arousal (calm/agitated). Durston & Itier (2025) is the only study to date to directly test this idea in a within-subjects design with strong methodological controls and robust statistics. They recorded ERPs during a gender task and gathered post-experiment valence and arousal ratings which they linked back to individual face ERP trials. Since ratings were taken after the ERP study, their findings did not capture participants’ in-the-moment perception of faces. The current study (N=86) addresses this concern by having participants rate the valence and arousal of neutral, happy, fearful, and angry faces after each ERP trial. We replicated the original findings that affect ratings are associated with threat-related differences in N170 amplitudes (e.g., fear-neutral and anger-fear). Additionally, we found new associations between ratings and P2, N1 and EPN components for angry contrasts only. These results highlight the temporal dynamics of facial expression affect processing, which seems partially task dependent.
A Neural Investigation of the Other-Race Effect in Ensemble and Single-face Perception
▶ Show abstract
The other-race effect (ORE) refers to a well-documented disadvantage in recognizing faces of other races compared to one’s own. While the ORE has been studied primarily with single faces, real-world visual experience often involves ensembles (e.g., crowds), where perception relies on summary representations (e.g., face averages). This study combined electroencephalography (EEG), neural decoding, and image reconstruction to examine race-contingent representations for both single faces and multi-face ensembles. EEG data were collected while East Asian participants (n = 22) viewed East Asian and White faces displayed one a time or in groups of six. Behaviorally, participants showed a robust ORE. Neurally, single-face decoding mirrored this effect, with higher same-race (SR) than other-race (OR) decoding and tighter clustering of OR faces in the neural face space. Interestingly, this single-face pattern did not generalize to ensembles, suggesting that summary-based encoding alters how race structures face representations. However, neural-based image reconstructions revealed systematic distortions in age, valence and arousal for both single faces and ensembles. Together, these findings demonstrate that the ORE reflects both reduced neural discriminability and multiple representational distortions in the perception of OR faces.
Talk SessionS5B · HNE Rm 037 · Visuomotor & Spatial CognitionRoom 037▼ Show
Home-Based Immersive VR to Improve Motor Performance in Children with Developmental Coordination Disorder
▶ Show abstract
Developmental coordination disorder (DCD) describes a congenital set of difficulties with motor tasks. Immersive virtual reality (VR) offers the potential for feedback-rich movement practice in this population. This study examined the effects of an immersive VR rhythm game within a home-based setting on motor performance in children and adolescents with DCD. Using a crossover design, 27 participants (21 boys, 6 girls) aged 10–16 years with DCD completed two home-based interventions: (1) VR gameplay using Beat Saber, and (2) tablet-based gameplay using Cut the Rope. Each condition required ≥30 minutes of daily gameplay over five consecutive days, separated by a ≥two-week washout period. Motor performance was assessed pre- and post-intervention using the Movement Assessment Battery for Children – Third Edition (MABC-3) and the Box and Block Test (BBT). All motor performance measures showed a significant improvement following the VR intervention, and this improvement was significantly greater than that observed in the tablet intervention for the BBT. VR gameplay yielded higher enjoyment and children reported higher motivation across VR intervention than tablet intervention. This study provides early evidence that playing Beat Saber in VR may support greater short-term improvements in motor performance than tablet gameplay for young people with DCD.
The Influence of Allocentric Cues on Spatial Coding for Memory-guided Actions: An fMRI Study
▶ Show abstract
In this study, we investigated the cortical mechanisms of allocentric coding for memory-guided actions toward naturalistic objects using functional MRI (fMRI). Participants performed an Experimental and Control task. In both tasks, a kitchen scene was presented with six fruits on a countertop (Encoding; 2s), the positions of which had to be remembered. After Delay 1 (1s), during which the fruits disappeared, only five fruits re-appeared (Test; 1.3s), with a possible manipulation – they appeared in their original (Encoding) locations (No-Shift) or one, three, or all five fruits were shifted (Shift) imperceptibly leftward or rightward. After Delay 2 (1s), participants reached to indicate the missing object position (Experimental) or indicated via button press whether the Encoding phase had a duplicated fruit (Control) (Response; 3s). We found differential engagement of early-to-late occipitotemporal cortex in the Encoding phase for the Experimental vs. Control tasks. In the Test phase, we found posterior parietal activation in right supramarginal gyrus (SMG). Lastly, during the Response phase, we found sensitivity to Shift in left intraparietal sulcus and right posterior parietal regions, as well as early-to-mid occipital cortex. Overall, these findings provide novel insights into allocentric spatial coding and movement planning for memory-guided actions in naturalistic environments.
TMS Reveals Causal Contributions of Egocentric and Allocentric Connectivity Hubs to Reach Behavior
▶ Show abstract
Introduction: Goal-directed reaching relies on egocentric (EGO; body-centered) and allocentric (ALLO; landmark-centered) reference frames. Musa et al. (2025) identified two cortical hubs within the visual peripheral (VP) network: a superior hub, ExstrSup-2 (dorsal stream), linked to both EGO and ALLO processing, and an inferior hub, ExstrInf-2 (ventral stream), selectively associated with ALLO. We used transcranial magnetic stimulation (TMS) to test their causal role in reaching. We hypothesized that TMS to ExstrSup-2 would disrupt EGO reaching, while stimulation of either hub would affect ALLO reaching due to coordinate transformation. We also tested whether VP network modularity modulates TMS susceptibility.Methods: Participants (n=16) completed a delayed reach-to-touch task under EGO and ALLO instructions. Resting-state fMRI localized TMS targets and quantified VP modularity. Triple-pulse TMS (1 Hz, 110% motor threshold) was applied during the memory delay to ExstrSup-2, ExstrInf-2, or a control. Accuracy, precision, and reaction time were analyzed.Results: TMS to ExstrSup-2 impaired EGO and ALLO precision and increased reaction time. ExstrInf-2 stimulation selectively worsened ALLO accuracy and precision. Higher VP modularity was linked to smaller TMS-induced performance declines.Conclusion: Dorsal hubs support both reference frames, ventral hubs specialize in ALLO processing, and greater VP modularity confers resilience to disruption.
Active Cognitive Control in Real and Pantomime Grasping
▶ Show abstract
Pantomime grasps are movements performed in empty space to simulate interacting with absent or spatially displaced objects. They are assumed to be more cognitively demanding than real grasps because they must explicitly communicate target features and lack haptic constraints (Whitwell et al., 2020). However, until now, this assumption has not been directly tested. Using a dual-task method, we compared the cognitive control demands of real and pantomime grasps made by healthy young adults (n = 24 per grasping condition). Critically, we used a mirror apparatus that matched visual feedback of the hand and target across the two grasping conditions. Participants performed real or pantomime grasps and a working memory task, both alone (single-task baselines) and concurrently (dual-task). Dual-task interference was observed for both grasping conditions, but was greater for pantomime, as reflected in more working memory errors and greater disruption to grasp kinematics. These results establish that pantomime grasps place greater demands on cognitive control than real grasps. They also have important implications for understanding how actions are compromised in limb apraxia and for designing human-machine interfaces that use natural hand gestures as an input modality.
Talk SessionS5C · SC Room B · Lifespan, Health & Individual DifferencesRoom B▼ Show
Restricted cortical diffusivity with age and obesity: HCP cohort (N=1,682)
▶ Show abstract
Brain microstructural abnormalities as markers of neurodegeneration have been of interest for improving disease prediction. A novel neuroimaging measure, cortical diffusivity, maps cerebral cortex microstructure using diffusion MRI. Cortical diffusivity has been associated with widespread changes in Alzheimer’s disease and prediction of dementia. Obesity is a risk factor for dementia; thus, it is important to investigate cortical diffusivity and metabolism. In this study, we obtained body mass index (BMI), metabolic markers, and cortical mean, axial, and radial diffusivity (cMD, cAD, cRD) from 1064 individuals from HCP Young Adult and 618 individuals from HCP Aging cohorts. We used principal component analysis to detect patterns of covariance across 31 cortical brain regions to investigate associations between age, BMI, and metabolic markers using linear mixed modeling. In both cohorts, older age and higher BMI were associated with lower first principal component (PC1) cMD, cRD and older age with higher cAD. The highest loadings were in temporal and frontal regions. In the young cohort, men had lower cortical diffusivity than women, where sex-related effects were comparable in the aging cohort. These results reflect restricted water movement across the cerebral cortex with age and obesity, which may indicate neuroinflammation or edema in the cortical ribbon.
Examining the Relation Between Math Anxiety and Math Performance: The Mediating Role of Intrusive Thoughts Management Strategies
▶ Show abstract
Higher math-anxious individuals often perform worse on math tasks than lower math-anxious individuals. The cognitive interference model posits that in higher math-anxious individuals, worrisome intrusive thoughts consume working memory resources needed for task completion. However, little research has examined whether intrusive thought management strategies may explain poorer math performance in higher math-anxious individuals, despite their potential influence on cognitive load. The present study tested whether intrusive thought management strategies mediate the relation between math anxiety and math performance. Participants (n = 309 undergraduate students; 53.7% women) completed measures assessing math anxiety, math performance and intrusive thought management strategies. Factor analysis grouped seven intrusive thought management strategies into three categories: Cognitive Regulation (cognitive strategies to reduce thought impact), Attention-Diversion (shifting attention to task-irrelevant distractors), and Spontaneous Resolution (thoughts subsiding without deliberate regulation). Results demonstrated that math anxiety negatively predicted performance (B = -.060, p <.001). A competing mediation using the strategies demonstrated that Cognitive Regulation (B = .007, 95% CI [.0007, .0143]) and Attention-Diversion (B = -.012, 95% CI [−.0207, −.0046]) mediated the relation between math anxiety and performance. These findings suggest that the intrusive thought management strategy used by higher math-anxious individuals plays an important role in their math performance.
Childhood Relationships and Cognition in Older Adults
▶ Show abstract
Normal aging leads to changes in cognitive functioning. Certain protective factors, such as the social environment during childhood, particularly friendships, can decrease the effects of aging on cognition. The current study aimed to investigate the link between childhood social relationships and cognitive functioning in Canadians 55 and older. Participants (N = 115) completed measures of episodic memory, verbal fluency and working memory. They also completed a questionnaire regarding childhood friendship experiences, described their childhood social network using hierarchical mapping technique and completed a loneliness questionnaire. The results from multiple regression analysis showed that when age, gender and present-day loneliness were controlled for, there is a significant link between global cognitive functioning and all types of childhood second circle relationships, termed closer relationships. The same link was also observed for verbal fluency. Particularly, closer childhood friendships, romantic relationships and other types of relationships are associated with verbal fluency in older adults. However, no significant association was found between episodic memory and social relationship variables during childhood. These results suggest that social relationships during childhood explain a part of the variance in older adult cognitive functioning.
Imagery Vividness and Mental Health Through Spontaneous Cognition
▶ Show abstract
Research on the link between mental imagery vividness and mental health has yielded inconsistent results, with prior studies reporting weak positive, negative, and null associations. One possible explanation for this is that imagery vividness relates to mental health through broader differences in internally generated cognition, particularly the nature and frequency of spontaneous mental experiences such as mind wandering. Such experiences are thought to be regulated through attentional control. Additionally, spontaneous mind wandering has been associated with symptoms of anxiety and depression, and while related to attentional control, is not fully explained by it. The role of spontaneous thought in the relationship between imagery vividness and mental health has not been examined, leaving a key explanatory gap in the literature. In the present study, we used structural equation modeling to examine how imagery vividness, attentional control, and spontaneous mind wandering relate to mental health outcomes including anxiety, depression, and emotion regulation. Attentional control mediated the association between imagery vividness and more adaptive outcomes, and spontaneous mind wandering accounted for the link between vividness and negative mental health outcomes. These findings identify spontaneous mind wandering as a potential mechanism connecting imagery vividness to mental health.
Talk SessionS5D · SC Room C · Language – Words, Semantics & Lexical RetrievalRoom C▼ Show
Embedding Culture in Language: Investigating Cross-Dialectical Semantic Alignment
▶ Show abstract
A central question in the language sciences is the extent to which word meanings vary across languages. Under the universalist view, languages are proposed to be based upon a pre-existing conceptual space, while the relativist perspective hypothesizes that language variation reflects solutions to different cultural and social communicative needs. However, it is unknown how the dialects of a language shape the meaning of that language. Given the rise of English globally, English co-exists and interacts with local languages across different cultural settings. Many studies have investigated the distinct phonological, lexical, and grammatical features of these English varieties but there has not been any systematic study of word meaning variation. To investigate this, we trained word2vec models on 14 dialects of English spanning the globe. Using multiple semantic alignment methods (which measures the coherence of word meanings), we found that factors such as geographic distance, history of colonization, and English status all correlate with across-dialect alignment. Moreover, we found that alignment was domain-specific, such that domains anchored in the natural world (i.e., time, numbers) had higher alignment than region-specific domains (i.e., animals, agriculture). Collectively these results highlight how the linguistic environment and cultural embeddedness shapes the semantic structure of a language.
Discovering regularity and mechanisms of word sense acquisition in early childhood
▶ Show abstract
How does language use inform the emergence of word meanings in early life? Prior work in developmental psychology and cognitive science has typically focused on how children acquire words, with much less work studying how different senses within a word emerge over time, especially after the first sense is acquired. Among the available work, most studies rely on manual annotation of word senses. We propose a novel framework based on computational methods and contextual word embeddings to characterize how different senses within a word unfold as children acquire the lexicon. It discovers word senses automatically by forming semantic clusters from natural language use and analyzes 1,270 words from approximately 4 million utterances produced by children (19–144 months) and their caretakers. We tested three hypotheses motivated by language evolution. First, concrete senses of a word tend to emerge earlier in child language. Second, word senses grow incrementally in semantic space across development. Third, algorithms of semantic chaining—how words spawn new senses by extending from existing senses—predict the order of word senses in development. We find support for all three hypotheses. Our work uncovers regularities and mechanisms of word sense acquisition and its similarities to language evolution.
Does AI Know Bouba is Rounder than Kiki?
▶ Show abstract
The Bouba/Kiki effect describes associations between sounds like /l/, /m/, /n/ with roundness, and sounds like /k/, /p/, /t/ with spikiness. Recent work has shown large language models can produce ratings closely aligned with humans on a variety of lexical semantic variables (e.g., valence, arousal, familiarity; Conde et al., 2026). However, it remains unclear whether this extends to sound-symbolic associations. We generated GPT-5.4 shape ratings (ranging from 1 [spiky] to 7 [round]) of 320 nonwords through OpenAI’s API. Model responses were sampled in varied temperature levels, where lower values produce more deterministic outputs and higher values yield greater variability. We also collected human ratings as a comparison, using visual vs. auditory presentation. Participants either saw or heard random 50 out of 320 nonwords and rated each on shape using a seven-point Likert scale. Results showed that nonwords received similar ratings across modalities, though some phoneme-specific effects varied by modality. GPT-5.4 ratings closely resembled human judgements across temperatures. However, nonwords with the most variability differed between humans and GPT-5.4. Additionally, GPT-5.4 rated vowels rounder and consonants spikier than humans. These findings suggest that GPT-5.4 approximates human sound-shape ratings. The internal representations and phonetic processing underlying these judgements will also be discussed.
Exploring guessing behaviour in tip-of-the-tongue states
▶ Show abstract
Tip-of-the-tongue (TOT) states are an interesting phenomenon that occur at the intersection of linguistic, memorial, and metacognitive processing to interfere with our ability to express ideas through language. Here, we sought to explore patterns of guessing behaviour that occur during TOTs while manipulating the guessing context (free report and four-alternative forced choice) and the lexical information available (semantic, phonological, and orthographic) in a series of experiments. Cued recall and paired-associate recall tasks were used to prospect for TOTs towards previously studied target words. When a target was not recalled, participants provided a metacognitive judgement on a scale from complete forgetting to TOT. Finally, the guessing tasks involved either free reporting of words that “feel close” to the unrecalled target, or 4AFC which provided a more structured context and allowed experimental control over the relationships between unrecalled targets and 4AFC options. Comparing results across experiments reveals some consistent patterns and crucially highlights the influence of context on how participants judged their recall progress and directed their subsequent guessing behaviour.
SymposiumSymposium V · SC Room A · The Neural Aesthetics of Imagination and BeautyRoom A▼ Show
Two mechanisms of beauty
▶ Show abstract
People often make rapid judgments about their surroundings to guide subsequent behaviour. Certain visual features shape how one evaluates their environment as beautiful or threatening. Using images of real-world natural scenes, we examined how visual complexity and symmetry influence aesthetic and threat judgements. In Experiment 1, participants rated scenes on beauty and threat, while additional tasks assessed the predictability and memorability of the stimuli. Results showed that both beauty and threat ratings were associated with visual complexity, but not with symmetry, memorability, or expectancy. A subsequent principal components analysis revealed two latent dimensions—'complex beauty' and 'safe beauty'—that accounted for patterns in aesthetic evaluations. These findings suggest that beauty judgments may arise from two distinct mechanisms: one linked to visual complexity and another related to perceived safety. A second study using eye tracking provided converging evidence for this dual-process account. Visual exploration patterns differed depending on whether participants’ judgements were driven by 'complex beauty' or 'safe beauty' when evaluating scenes for beauty and threat. Together, these findings suggest that aesthetic judgements emerge from an interaction between visual complexity and perceived environmental safety, with distinct viewing strategies supporting each evaluative process.
Exploring the varieties of imagination in the visual domain with fMRI
▶ Show abstract
Recently, several theoretical approaches have proposed that rather than being a single faculty, imagination consists of a constellation of processes characterized by their respective properties and contexts of application. One approach has distinguished between instrumental vs. expressive imagination: The former is defined as the detection of patterns, relations, and analogies that foster solutions to specific problems, whereas the latter is characterized by the active creation of vivid, novel constructions based on episodic memories. We sought to explore the neural bases of instrumental vs. expressive imagination by instructing participants in the fMRI scanner to generate stories, situated within specific scenes using four self-selected objects. Critically, they were instructed to be logical and action-oriented on instrumental trials, but attend to feelings and imagery on expressive trials. We examined changes in temporal dynamic functional connectivity in brain states involving six networks (default mode, dorsal attention, salience, left and right frontoparietal, and visual). Our key finding was that when generating stories, there was reduced functional connectivity between the dorsal attention and default mode networks in the expressive compared to instrumental condition. We interpret this to mean that reduced frontal control over the default mode network facilitates spontaneous thought in the service of expressive imagination.
Less is more: Aesthetic liking is inversely related to metabolic expense by the visual system
▶ Show abstract
Energy efficiency is a major driving force in the evolution of organisms, and previous research implies that humans may have evolved pleasure-based signals to guide optimal actions. But could this energy-saving heuristic also apply to aesthetic pleasure? We test this hypothesis using both an in silico model of the visual system (VGG19) and human observers, finding strong evidence in both. First, we measure the proxy for metabolic cost incurred by VGG19—either pretrained for object and scene categorization or randomly initialized—as it processes 4,914 images of objects and scenes, revealing an inverse relationship between aesthetic preferences and metabolic cost, and only in the pretrained model. Next, we compare aesthetic ratings of visual stimuli to metabolic activity in the human visual system, measured by fMRI. We observe the same inverse relationship between BOLD signals and aesthetic preferences in both early visual regions (V1, V2, and V4) and higher-level regions (FFA, OPA, PPA). These findings suggest that aesthetic preferences may at least partially arise from an affective heuristic favoring low-energy states, and they offer a unified framework linking empirical evidence on visual discomfort with theories of processing fluency, image complexity, and prototypicality, providing a straightforward model for understanding aesthetic judgments.
Reimagining aesthetic experiences in colour and in motion
▶ Show abstract
Our perceptual—and aesthetic—experiences are animated by colour and motion. Here I present two studies that reimagine the roles these features play in our aesthetic experiences. First, who imagines in colour? We have been developing an objective property-specific measure of colour imagery via a source memory confusion paradigm. Observers either studied a shape or imagined filling it in with a colour—and we later probed memory for the shape as well as whether its colour had been seen or imagined. Shape memory was robust—but critically, filled-in shapes were often confused as having been seen in colour, with confusion magnitude being reliably predicted only by responses to colour-specific items from the Vividness of Visual Imagery Questionnaire. Second, I will explore the correlates of aesthetic preferences for things in motion. Observers adjusted playback speeds of videos to what was maximally pleasing to them. Speed adjustments were domain-specific (with separate preferences for human vs. non-human motion); converged at a “sweet spot” of motion (with faster dances consistently slowed down and slower dances sped up); and were best explained by a “natural pace” at which we encounter people move. Together, these studies capture new dimensions of aesthetic experiences, within our inner lives and the wild.
15:00 – 16:00Session 6 · CogNeuro (Language) · Language (Reading) · Scene Perception ·Symposium VI: Spatial Vision Renaissance‖Symposium VII: The Empathic Brain
Talk SessionS6A · HNE Rm 035 · Cognitive Neuroscience – Language, Code & Brain OrganisationRoom 035▼ Show
Successful Language Localization in Two Individuals with Non-Speaking Autism
▶ Show abstract
Autism spectrum disorder is a neurodevelopmental condition characterized by challenges across multiple domains, including motor skills, communication, and social interactions. Individuals with non-speaking autism (NSASD) have minimal/absent language production, and some have argued that receptive language may also be limited in this group. However, the absence of spoken language and reliable motor output makes it difficult to obtain valid and reliable assessments of their cognitive abilities. Resultingly, there is limited research to support claims about receptive language. Functional magnetic resonance imaging (fMRI) offers great potential to explore language processing in NSASD because neural responses can be assessed independent of overt output. In this study, we collected fMRI data from two individuals with NSASD (n = 1 left-handed) and 16 healthy controls (n = 8 left-handed). Participants did a modified version of a well-validated auditory language localizer that contrasted 18 s clips of intact and reverse naturalistic speech. We successfully localized language networks in both individuals with NSASD, yielding some of the first neural evidence of language processing in this population. These findings demonstrate that language-selective neural responses can be detected in NSASD in the absence of overt output, enabling future work to examine the nature of their language
Meaning, Action, and the Brain: Evidence from Cerebral Palsy
▶ Show abstract
Cerebral palsy (CP) is a common cause of motor disability in children, resulting from a non-progressive perinatal brain injury, impairing brain development. Although motor and perceptual deficits are fundamental aspects of the disorder, many with CP experience speech and language impairments in two main areas: phonology and lexico-semantic representations. While phonological deficits are reported and often attributed to impairments in gross and fine motor skills, lexico-semantic deficits are reported less frequently and are documented in children, contributing to a lack of understanding of their underlying mechanisms and whether they persist into adulthood.The N400 event-related potential was measured to explore the mechanisms underlying lexico-semantic deficits from the perspectives of the Social Network Hypothesis and the Embodied Cognition Hypothesis. Participants completed a sentence-reading task with sentences that either made sense or did not, followed by an identity paradigm using a semantic decision task comparing lower-limb action verbs (e.g., kick) and psychological verbs (e.g., believe).No evidence was found that CP individuals process domain-general semantic incongruence differently from neurotypicals. There was also no evidence that CP individuals have impairments in action verbs, suggesting that while the motor system contributes to action semantics, it is not necessary for it.
Brains Decoding Code: Neural Signatures of Syntax and Meaning in Programming and Language
▶ Show abstract
This study investigates the neural correlates of syntactic and semantic processing in both natural and programming languages, using event-related potentials. Specifically, we examine the N400 effect, an index of semantic processing difficulty, and the P600 effect, associated with syntactic reanalysis and repair .Participants, all native speakers of English and proficient in Python, read English sentences and lines of Python code. In the natural language condition, sentences were semantically plausible vs. implausible or syntactically correct vs. incorrect. In the programming language condition, participants processed correct and incorrect lines of Python code. Incorrect code contained either semantic anomalies or syntactic errors. Participants performed acceptability judgments to assess real-time brain activity in response to these violations while their ERPs were recorded.Semantic violations in natural languages elicited increased N400 amplitudes, reflecting greater cognitive effort in meaning integration. Syntactic violations, on the other hand, generated P600 effects, indicative of reanalysis and syntactic repair processes. Differences in magnitudes of the ERP effects for the programming language are likely due to (a) the more rigid and less ambiguous nature of programming syntax and (b) due to the fact that variable names and identifiers do not inherently carry meaning in the same way as natural language.
The Role of Individual Differences in Hemispheric Lateralization of Faces and Words in Developmental Dyslexia
▶ Show abstract
This study investigates the relationship between dyslexia severity, cognitive functioning, handedness, and lateralization of faces and word processing in dyslexia. During typical development, face processing tends to lateralize to the right hemisphere and word processing tends to lateralize to the left hemisphere. Some evidence suggests that individuals with developmental dyslexia exhibit atypical lateralization for both word and face processing. Although dyslexia is characterized as heterogenous, with varying severity and cognitive profiles, these differences have yet to be addressed in lateralization studies. This study investigates whether lateralization patterns for faces and words are associated with cognitive performance, handedness, and dyslexia severity of participants with developmental dyslexia. Thirty adults diagnosed with developmental dyslexia and thirty without will complete a dyslexia inventory, handedness questionnaire, a laterality task and cognitive tasks measuring phonological awareness, rapid automatized naming, visual attention span, and visual motion detection. Left-handedness, poor reading skills, and low scores on cognitive tasks are predicted to correlate with lateralization patterns in individuals with dyslexia as similar correlations have been found in research on individuals without dyslexia. The current study will improve the understanding of cognitive and neural mechanisms underlying dyslexia. These mechanisms remain unexplored in adulthood as prior studies have focused on childhood.
Talk SessionS6B · HNE Rm 037 · Language, Reading & Communication across the LifespanRoom 037▼ Show
Conversation Dynamics Among Older Adults with Age-Related Hearing Loss
▶ Show abstract
Background. Age related hearing loss (ARHL) affects 1⁄3 of adults above age 65 and can contribute to social isolation, cognitive problems, and worse life quality. Speech perception difficulty underlies these declines, even with hearing aids. However, traditional, lab-based speech tests fail to capture the interactive, goal-oriented nature of real-world conversations. Conversations are multimodal and dynamic and require integration of linguistic features, paralinguistic cues, turn-taking, spatial orientation, and emotion. These demands are exacerbated by background noise, which forces older adults to leverage more cognitive resources such as attention and working memory. Despite these challenges, few studies have examined speech-based social interactions in older adults with ARHL within ecologically valid contexts. This study addresses this gap by investigating how background noise and hearing loss jointly affect task performance, cognitive load, and conversation satisfaction. Methods. Older adults with ARHL completed a conversation-based task in which they were verbally guided through a map. The task was done in quiet or in 73 dBA background noise. Dependent variables included the NASA Task Load Index (cognitive workload), Positivity Resonance Scale (social connectedness), and conversation satisfaction ratings. Results. Preliminary data show that ARHL predicts lower positivity resonance, conversation satisfaction, and higher effort in background noise.
Not all native readers are the same: How eye movements vary within language groups
▶ Show abstract
Patterns of eye movements change as a function of a reader’s language background. For example, readers of languages with longer words tend to make longer saccades and land further into a word than readers of languages with shorter words. However, the differences in reading behaviours between speakers of the same first language (L1) have often been overlooked. This study uses data from the Multilingual Eye Movement Corpus (MECO) to examine the eye movements of 29 samples of participants speaking 20 different languages while they read in their respective L1s (Experiment 1) and their shared L2, English (Experiment 2). Five of the languages had more than one sample available. We scaled the data across multiple dimensions, including eye movements, comprehension accuracy, and reading rate, and analyzed distances between samples. In both experiments, the distance between replicated pairs (e.g., two samples of Spanish) was smaller than the distance between non-replicated pairs (e.g., an Estonian and a Dutch sample). However, for most samples, their nearest neighbors were samples from a different L1 background, rather than samples with the same L1. This intersample variance suggests that conclusions from a single group of speakers may not be generalizable, even to other speakers of that language.
Redundant Modifiers Enhance Memory Across the Adult Lifespan in Virtual Reality
▶ Show abstract
Speakers frequently produce redundant modifiers, particularly in older adulthood. Traditionally interpreted as inefficiency, redundancy is now considered an adaptive strategy that supports comprehension and memory. Most prior research compared younger and older adults using simple 2D displays, leaving it unclear whether these patterns extend across the adult lifespan (including middle age) or to more naturalistic environments. The present study examined the production of modifiers and their influence on memory across the adult lifespan using an immersive virtual reality (VR) paradigm. Eighty-one adults (18 to 84) wore a head-mounted display and described everyday objects in a virtual environment. On each trial, the target object appeared alongside a competitor of the same category (varying in colour or state), or an unrelated object. Following a 20-minute delay, participants completed unexpected recall and recognition memory tests for the previously described objects. The results reveal high rates of redundant modifier use across our sample, indicating that overspecification persists in realistic virtual environments. Although recall and recognition accuracy declined with age, modifier use improved memory of past referents across age groups. These findings suggest that redundant modifiers support memory across adulthood and may have important implications for communication strategies and assistive technologies for the aging population.
Referential processing draws on multiple perspective cues in parallel
▶ Show abstract
Work in psycholinguistics continues to demonstrate new ways in which active perspective-taking guides language processing. For example, recent work shows that, in sentences like “Sophie [told/asked] Amanda [that/if] she likes learning new languages”, readers draw on perspective reasoning to judge the ambiguous pronoun as near-categorically referring to the subject antecedent with TELL sentences (because Sophie possesses the relevant knowledge) and the object antecedent for ASK (Amanda possesses relevant knowledge). Although these patterns demonstrate a robust perspective effect, could they instead arise from lexical constraints encoded in the verbs TELL/ASK? Experiment 1 manipulated subordinate clause content to introduce modal verbs of possibility/permission (SHOULD/COULD, e.g., “Stella [told/asked] Emily [that/if] she could have another slice of cake”). Participants strongly reversed their preferences from the previously-reported antecedent preferences for TELL/ASK sentences, demonstrating that interpretation is not solely driven by the main verb. Experiment 2 further ruled out surface-cue explanations by showing that preceding context sentences can compel readers to strongly reverse the default TELL/ASK patterns. Experiment 3 demonstrated that simply varying character properties strongly reverses antecedent preferences, e.g., “Max asked his [son/tutor] Gerald if he understood the assignment correctly”, reflecting how role information shifts readers’ understanding of which character holds relevant knowledge.
Talk SessionS6C · SC Room C · Scene Perception, Affect & Social CognitionRoom C▼ Show
Exploring the Influence of Self-Generation during Episodic Simulation on Prosocial Outcomes
▶ Show abstract
Episodic simulation, the ability to construct and vividly imagine future events, has been found to promote prosocial outcomes across a variety of populations (e.g., seniors, temporal lobe epilepsy, etc.). The present study explored the impact of self-generation and narrative transportation on episodic prosociality. Our sample included 150 younger adults aged 18-35. Participants read eight vignettes describing scenarios of someone in need of help. In response, participants either imagined a self-generated or experimenter provided event in which they helped the person in need, or completed a control task (i.e., imagined possible written responses/comments for ways in which the individual could be helped). After each vignette, participants rated the vividness of the imagined helping event, degree of narrative transportation, and prosocial outcomes (i.e., willingness to help the person in need and empathic concern). Individuals in the self-generation and provided narrative conditions reported similar levels of willingness to help and empathic concern, which significantly exceed prosocial outcomes in the control condition. Interestingly, narrative transportation mediated the relationship between vividness and prosociality in both the self-generation and provided narrative conditions. These findings suggest prosocial effects of episodic simulation established for self-generated helping events extend to provided narratives and may rely on similar underlying mechanisms.
Pain as Information: Computational Mechanisms of Learning-Dependent Pain Modulation
▶ Show abstract
In this presentation, I propose that pain functions not merely as a defensive alarm, but as a sophisticated teaching signal that guides long-term behavioral adaptation. Combining behavioral and electrophysiological experiments with computational modeling of learning, I demonstrate that pain intensity is selectively facilitated when it carries high informational value.First, I show that pain is amplified when unexpected, consistent with the principle that prediction errors serve as the foundation of learning. I then demonstrate that pain perception is further modulated by environmental volatility, with elevated uncertainty estimates slowing response times and reducing the influence of prior expectations on subjective ratings.Beyond expectation and uncertainty, I present evidence that pain is significantly enhanced during active exploration compared to exploitation, suggesting the brain prioritizes the teaching signal when we are engaged in learning about our environment. Crucially, while subjective intensity is modulated by these factors, the ability to discriminate between pain levels remains preserved, ensuring that the fidelity of sensory information is maintained.Ultimately, these findings position learning as a core function of pain and suggest that characterizing altered learning parameters may offer vital insights into nociplastic pain conditions.
The forest, the trees, and the whale?
▶ Show abstract
The gist of a scene is computed separately from the objects in it, but it is unclear whether scenes and objects are processed simultaneously or sequentially. Previous studies suggest our ability to categorize a scene as natural vs. artificial is impaired when an incongruent object is present in the scene, suggesting the object’s category is retrieved prior to or in parallel with scene classification. However, it is unknown whether the object is processed to the level of identification because the congruency of the pairing was confounded with its plausibility. To address this, we manipulated both the congruency and plausibility of scene-object pairings. Subjects either classified scenes or objects (task type). For congruent pairings, accuracy did not differ with plausibility or task type. However, for incongruent pairings, implausible objects improved object classification but worsened scene classification. These findings indicate that objects are processed to the level of identification irrespective of task type, suggesting that the visual system extracts high-level information from scene-object pairings. This interpretation supports the simultaneous processing account, and suggests that conflicting findings may be due to the confounding of plausibility and congruency, and the inverse effects of implausibility on scene and object recognition.
Visual Occlusion Disrupts Neural Processing of Facial Emotions and Empathy
▶ Show abstract
Visual face occlusion impairs our ability to infer and share emotions. The neuralmechanisms underlying these social impairments remain poorly understood. Here, in apreregistered study, we investigated the neural processing of occluded and unoccluded faces.Participants viewed images of neutral and happy faces wearing an opaque mask, a transparentmask, or no mask. EEG was recorded from 64 electrodes with analysis time locked to thepresentation of the face. Participants rated emotional intensity and empathy for each face.Behaviourally, the data indicated higher ratings of emotional intensity and empathy for happyfaces, which were reduced when the faces were occluded by an opaque mask. The N170, a classicmarker of early face processing, was more negative when faces were occluded. The N170 was alsomore negative for happy faces, with this effect reduced for faces occluded by opaque masks. Twoneural indices of emotional processing, the Late Positive Potential (LPP) and the Early PosteriorNegativity (EPN), also showed a reduced emotion effect for faces with opaque masks. Thus, visualocclusion of faces disrupts both early face processing and dampens its socio-emotional evaluation.
SymposiumSymposium VI · SC Room A · Spatial Vision RenaissanceRoom A▼ Show
Alternative perspectives on V1 encoding
▶ Show abstract
The receptive-field model has been an effective framework for describing neural and perceptual visual encoding.Classical perceptual and neurophysiological studies have highlighted the importance of a Gabor-like basis in primary visual cortex (V1).While this basis provides an efficient representation of natural two-dimensional images, its limitations for representing three-dimensional scenes remain unclear.In this presentation, I examine a potentially overlooked but ubiquitous constraint on retinal images: perspective projection under natural viewing conditions.Building on these constraints, I propose an alternative prescription for what the visual system should primarily encode.This prescription is defined by a set of transformations describing how retinal images are expected to change with shifts in perspective.I introduce several encoding models that capture these transformation properties while remaining consistent with known characteristics of V1 receptive fields.Based upon this formulation, I discuss whether the standard description of V1 ought to be expanded.
Spatial Profile Analysis in Human Vision
▶ Show abstract
Traditional explanations of spatial pattern discrimination have relied on the concept of local sign: discrimination is based on representations in which the positions of pattern elements are made explicit, and discrimination is limited by the accuracy with which position is represented. An alternative view is that discrimination depends on changes in the response distribution of frequency- and orientation-selective channels, and it is unnecessary to postulate a stage that uses channel responses to construct explicitly spatial representations. Here we examined whether spatial cues are used in a contrast discrimination task. Stimuli were f+2f compound gratings, and a 3-IFC task was used to measure thresholds for a contrast increment added to 2f in conditions in which the contrast, frequency, and orientation of one or both frequency components were randomized across stimulus intervals. When one component was changed, and the spatial luminance profile of the compound grating varied randomly across intervals, threshold increased significantly relative to a fixed baseline. When both components were changed, and the stimulus spatial profile was constant across intervals, threshold was unaffected. These results suggest that observers relied on changes in the spatial profile of the compound grating to discriminate the contrast of 2f.
Peripheral material perception
▶ Show abstract
Humans can rapidly identify materials, such as wood or leather, even within a complex visual scene. Given a single image, one can easily identify the underlying “stuff,” even though a given material can have highly variable appearance; fabric comes in unlimited variations of shape, pattern, color, and smoothness, yet we have little trouble categorizing it as fabric. What visual cues do we use to determine material identity? Prior research suggests that simple “texture” features of an image, such as the power spectrum, capture information about material properties and identity. Few studies, however, have tested richer and biologically motivated models of texture. We compared baseline material classification performance to performance with synthetic textures generated from the Portilla-Simoncelli model and several common image degradations. The textures retain statistical information but are otherwise random. We found that performance with textures and most degradations was well below baseline, suggesting insufficient information to support foveal material perception. Interestingly, modern research suggests that peripheral vision might use a statistical, texture-like representation. In a second set of experiments, we found that peripheral performance is more closely predicted by texture and other image degradations. These findings delineate the nature of peripheral material classification.
SymposiumSymposium VII · SC Room B · The Empathic BrainRoom B▼ Show
Re-Evaluating the Neural Dynamics of Empathy: Task Effects, Emotional Context, and Relationships with Trait Measures
▶ Show abstract
ERP research claims affective empathy elicits early automatic neural processing not seen in cognitive empathy, but whether this effect varies across empathy task types, with priming, and across facial emotions, is unknown. In two experiments, we investigated whether task demands tapping into affective and cognitive empathy constructs differentially modulated ERPs. Participants viewed contextual priming images depicting body parts in neutral and painful scenarios, followed by faces making congruent neutral, pained (Exp. 1), or fearful (Exp. 2) expressions. ERPs to primes and faces were compared when participants rated their own discomfort (affective empathy), their concern for the person (empathic concern, cognitive empathy), how much pain/fear they thought the person was in (perspective taking, cognitive empathy), or how masculine/feminine the face looked (control). Although behavioural results differed between affective and cognitive empathy, no such distinction occurred on early ERPs. Priming occurred across emotional contexts and task types; emotion and task effects interacted differently depending on the facial emotion (pain vs fear), with perspective-taking leading to the shortest emotion effects. None of these neural modulations were correlated with trait empathy. We discuss implications for theories of empathy and our understanding of how social cognition is impacted by task-related attentional processes.
When Strangers Benefit More: Episodic Simulation, Familiarity, and Prosocial Behaviour
▶ Show abstract
Our prior research has linked episodic simulation, the mental construction of imagined scenarios, to empathy and prosocial intentions. In parallel, social psychology has demonstrated that empathy varies with the familiarity with a person. Integrating these lines of work, we examined whether familiarity with a person impacts the extent to which episodic simulation shapes empathic responses and prosociality. In a behavioural experiment, participants were first familiarized with a series of individuals (familiar individuals) or just introduced to individuals by name (unfamiliar individuals). Then, the participants were asked to either read or engage in episodic simulation of these individuals experiencing distressing scenarios, provoking empathy. Following this manipulation, all participants provided ratings of personal distress, empathic concern, and willingness to help the individuals. We found that episodic simulation significantly increased personal distress but not empathic concern for the unfamiliar individuals. No such effect was observed for familiar individuals. Moreover, the increased personal distress for unfamiliar individuals mediated the relationship between episodic simulation and willingness to help the person experiencing distress. Together, these findings suggest that episodic simulation may be particularly effective in motivating prosocial behaviour toward strangers, possibly by amplifying affective self-oriented responses.
Empathy for Pain and Sensorimotor Activity
▶ Show abstract
Previous TMS studies of empathy for pain have consistently reported decreased corticospinal excitability during pain observation, a finding commonly referred to as ""empathic sensorimotor resonance"". In contrast, previous behavioural work has shown faster reaction times following pain observation, suggesting motor facilitation rather than suppression. One possible explanation for this discrepancy is that neurophysiological studies typically constrain movement (to minimize artifacts in measurement), whereas behavioural tasks place the motor system in a state of action readiness. To test this, we used TMS to measure motor evoked potentials while participants observed painful and non-painful actions under active (response preparation) and passive (relaxation) conditions. Contrary to state-dependent predictions, pain observation produced a general increase in corticospinal excitability across both conditions. Parallel EEG work also revealed convergent results, indexing sensorimotor activity via Mu and Beta desynchronization, that were likewise insensitive to motor preparation. Together, these findings challenge the classic inhibitory accounts of empathic sensorimotor resonance, and instead suggests that empathic sensorimotor resonance may be better understood as a type of generalized sensorimotor engagement (perhaps supporting action readiness and bodily defense) rather than a mirror-like suppression of motor output.
Is damage to the ventromedial prefrontal cortex associated with impaired empathy? A neuropsychological study using an empathy motivation task
▶ Show abstract
Patients with damage to the ventromedial prefrontal cortex (vmPFC) classically develop personality disturbances, emotion dysregulation, and social decision-making deficits that have been likened to psychopathic tendencies (“pseudopsychopathy”). This array of acquired deficits is thought to typically entail impairments in empathy. However, these impairments may be due to either motivation to empathize or ability to accurately empathize. We examined whether patients with vmPFC lesions (who have been shown previously to have major deficits in empathy accuracy) would exhibit lower empathy motivation. We conducted a lesion group comparison study and used a free-choice lab paradigm – the Empathy Selection Task – to assess motivation to empathize. The results showed that damage to the vmPFC was not associated with abnormal avoidance of affective or cognitive forms of empathy. Specifically, patients with vmPFC lesions did not avoid empathy any more so than comparison groups (participants with damage outside of the vmPFC or with no brain damage). Our findings suggest that patients with vmPFC damage may have preserved empathy motivation, despite their real-world impairments with deploying empathy. More research is needed to examine whether patients with vmPFC lesions would exhibit normal empathy motivation in
A01
Alert, Aware, or Overwhelmed? How Stress Shapes Human Responses to Takeover Requests in Semi-Autonomous Vehicles
Chloé Lachance-Soulard
chlolachancesoulard@cmail.carleton.ca
▶ Show abstract
Increased vehicle automation is leading drivers to be immersed in another task while the car is driving itself, making it vital for takeover alerts to successfully redirect attention to the driving task when autonomous features fail. The present study investigated how alert salience and modality influence the stress invoked during takeovers in multitasking drivers using semi-autonomous vehicles, and how stress affects takeover success. Using a virtual reality driving simulator, participants completed nine semi-autonomous test drives while performing a secondary task. At a predetermined point during the drive, a takeover alert prompted participants to take over. Preliminary results suggest auditory alerts influence physiological arousal, with significant main effects of time observed for heart rate, F(1, 614) = 37.48, p < .001, along with other variables. For perceived stress, all participants identified high-salience auditory alerts as most stressful; visual salience selections differed from uniform, χ²(2) = 57.95, p < .001, with high salience most frequently chosen. For effectiveness, auditory, χ²(2) = 25.55, p < .001, and visual, χ²(2) = 13.55, p = .001, distributions favoured high salience. This work advances our understanding of how stress, induced by alerts, impacts how effectively attention is redirected to the driving task in distracted, multitasking drivers.
A02
An Association Between Mindfulness and Memory Representations in Undergraduate Students
Julianne Levesque
jclevesque@mta.ca
▶ Show abstract
Mindfulness has been associated with differences in attentional performance. If increased mindfulness is associated with better attention, it is therefore possible that it would also be associated with richer representations in memory. In an online study, we asked undergraduate students to list everything they knew about common objects and to complete scales evaluating trait mindfulness and state mindfulness. We split participants into ‘high mindfulness’ and ‘low mindfulness’ and evaluated the number of sensory and non-sensory attributes they listed, as well as the distinctiveness of these attributes. The items used varied on visual similarity and semantic similarity. We observed a number of effects of visual and semantic similarity as well as associations with both trait and state mindfulness. Surprisingly, individuals high on trait mindfulness listed more non-sensory features for the objects, but not more sensory features. Also, the attributes provided by individuals high on state mindfulness were more distinctive than those of individuals low on state mindfulness, especially when describing what they knew about items that were both visually similar and semantically similar. Our results suggest that mindfulness may be associated with richer and more distinctive representations in memory.
A03
Attentional lapses at encoding have opposite effects on memory in younger and older adults
Emily E. Davis
emily.davis@utoronto.ca
▶ Show abstract
Autism spectrum disorder is a neurodevelopmental condition characterized by challenges across multiple domains, including motor skills, communication, and social interactions. Individuals with non-speaking autism (NSASD) have minimal/absent language production, and some have argued that receptive language may also be limited in this group. However, the absence of spoken language and reliable motor output makes it difficult to obtain valid and reliable assessments of their cognitive abilities. Resultingly, there is limited research to support claims about receptive language. Functional magnetic resonance imaging (fMRI) offers great potential to explore language processing in NSASD because neural responses can be assessed independent of overt output. In this study, we collected fMRI data from two individuals with NSASD (n = 1 left-handed) and 16 healthy controls (n = 8 left-handed). Participants did a modified version of a well-validated auditory language localizer that contrasted 18 s clips of intact and reverse naturalistic speech. We successfully localized language networks in both individuals with NSASD, yielding some of the first neural evidence of language processing in this population. These findings demonstrate that language-selective neural responses can be detected in NSASD in the absence of overt output, enabling future work to examine the nature of their language
A04
Beyond Motor Benefits: Longitudinal Cognitive and Sleep Outcomes in Parkinson’s Disease Following Six-Years of Community Dance
Simran Rooprai
srooprai@yorku.ca
▶ Show abstract
Cognitive and sleep deficits are commonly observed features of Parkinson’s disease (PD). While dance has been increasingly explored as a novel approach for alleviating symptom burden, its potential long-term effects on cognition and sleep remain less understood. The present work examined longitudinal changes in global and domain-specific cognition (memory, attention/working memory, and visuospatial), as well as sleep burden (insomnia and daytime sleepiness), among a PD-Dance group participating in community-based dance compared to a physically inactive PD-Reference group across a six-year period (2014-2019). Generalized estimating equations models were used to evaluate both within- and between-group changes over time. The Dance group demonstrated significant improvements in global (2016-2018, all p < 0.001) and domain-specific cognition (memory = 2016-2017, p < 0.05; attention/working memory = 2016-2019, all p < 0.05), while no significant changes were observed in the Reference group. Insomnia worsened significantly in the Reference group (2018-2019, p = 0.003), while improving in the Dance group over time (p = 0.005). Daytime sleepiness remained stable across both groups. Overall, these findings suggest that engagement in community dance may support cognitive and sleep health in PD, further highlighting the importance of alternative forms of care to alleviate symptom burden.
A06
High Working Memory Load Reveals Hyper-binding in Young Adults
Janessa Lauzon
jl17ae@brocku.ca
▶ Show abstract
Attention filters irrelevant information, allowing individuals to navigate their environments efficiently. However, when cognitive resources are taxed, goal-directed behaviour and distractor suppression become less reliable. Hyper-binding (the binding of irrelevant and relevant information) reflects a failure of distractor suppression. Traditionally, hyper-binding has been found in older adults; however, recent findings suggest that younger adults show hyper-binding under certain conditions. The present study examined whether working memory load modulates hyper-binding in young adults. Participants completed a standard hyper-binding task where they viewed picture-word pairs while performing a low-load (0-back) or high-load (2-back) working memory task with the pictures. Participants then completed a memory test to assess recognition of picture-word pairs that were intact, rearranged into novel combinations, or entirely new. Results revealed significant hyper-binding in the high-load condition, such that participants showed greater memory for intact pairs compared to rearranged or new pairs. No hyper-binding was observed under low-load, as memory accuracy did not differ between pair types. These findings indicate that hyper-binding is not limited to older adults and is shown in younger adults when cognitive resources are taxed. Overall, the results suggest that high cognitive load disrupts the attentional filtering mechanisms relied upon by young adults.
A07
Intuitive Theories of Attention Shape Perceptions of Others’ Attentiveness
yuanze huang
y292huan@uwaterloo.ca
▶ Show abstract
When we look at other people, we extract much information about them, from demographic attributes to emotional states. Social interactions, however, also require that we infer others’ cognitive states, such as whether they seem attentive or distracted. Here we examine how perceptions of others’ attentiveness depend on their behaviour (e.g., how quickly they respond to external stimuli), but also on the context in which that behaviour occurs (e.g., how distracting the environment is). In particular, we hypothesized that judgments of others’ attentiveness in a cueing task would be higher when others have been cued to attend to incorrect locations (‘invalid’ cues) than to correct ones (‘valid’ cues). Participants (preregistered N=50) viewed actors perform a spatial cueing task and rated their attentiveness on each trial. Actors were seen as more attentive when they responded faster, but also following invalid (vs. valid) cues, indicating that observers accounted for the greater attentional demands of detecting targets after an invalid cue. These results suggest that observers draw on multiple sources of information when judging others’ attentional states, evaluating both observed behaviour and contextual factors. In this way, evaluations of attentiveness can reveal folk expectations about how attention operates—their intuitive theories of attention.
A08
Investigating Automatic Attentional Biases Underlying PTSD-CUD Co-Occurrence
Nicholas Murray
n.murray@dal.ca
▶ Show abstract
Cannabis use disorder (CUD) and posttraumatic stress disorder (PTSD) commonly co-occur. An integrated self-medication/attentional bias (AB) model proposes that cannabis use to cope negatively reinforces AB to cannabis information following subsequent trauma cue exposure. This study tests this model using a personalized cue reactivity paradigm (CRP) and an Addiction Stroop Task. Recruitment is ongoing (target N = 50). Eligible participants are 19–65-year-old cannabis users (≥1g/week in the past month; C-TLFB) who report ≥1 lifetime trauma (LEC-5). In Session 1, participants complete measures of PTSD severity, cannabis use frequency/problems/motives, and a semi-structured interview detailing their most severe trauma and a neutral life event. In Session 2, participants complete a CRP involving counterbalanced 2–3-minute audio scripts (trauma vs. neutral). Immediately following each CRP, participants complete an Addiction Stroop Task identifying the ink colour of cannabis and neutral words. Preliminary analyses show CUD+ (vs. CUD−) participants report higher cannabis coping [t(14.80)=4.29, p<.001] and conformity motives [t(12)=2.69, p<.05]. Participants responded less accurately to cannabis (vs. neutral) words, t(33)=-2.41, p<.05, and PTSD+ (vs. PTSD−) participants responded more accurately to cannabis words, t(14.87)=-2.27, p<.05. If trauma reminders activate cannabis-related AB, AB modification with trauma cue exposure may work as a PTSD-CUD intervention.
A09
Investigating the temporal specificity and structure of retrospective reports of engagement following a 0hr versus a 24hr delay
Emily Mashaal
sayersglassey@uwaterloo.ca
▶ Show abstract
In prior work, we established that individuals can accurately report on moment-to-moment fluctuations in attentional engagement immediately following video-lecture viewing. However, given that episodic memory declines over time, it is unclear how accurate retrospective judgments might be following a longer delay. Participants watched an online video lecture and intermittently provided their in-the-moment levels of engagement (immediate reports). Then, either directly following the video lecture (0hr retrospective reports; n = 100) or the following day (24hr retrospective reports; n = 99), participants were presented with short excerpts as cues, and they reported their level of engagement from when they initially watched those excerpts. The results show that the temporal overlap between immediate and retrospective reports of engagement (cross-correlation scores) are lower after 24hr versus 0hr, indicating some loss of temporal specificity. However, the structure of the reports remains similar to the 0hr condition in terms of mean level of reported engagement, mean temporal variability (mean squared successive difference), and the correlation between the temporal variability of immediate and retrospective reports. Overall, this is consistent with the notion that delayed retrospective reports seem to lose some temporal specificity but retain their overall levels of recalled engagement and variability in responses.
A10
Lost in Time: The Role of Time Perception in Mind Wandering
Liam Yeo
YeoL@myumanitoba.ca
▶ Show abstract
Attention plays a very important role in our sense of time perception. Both laboratory takes and anecdotal evidence have shown that increased attention to time causes time to feel like it is slowing down and decreased attention makes time feel like it’s flying. This research has expanded to the mind wandering literature, where off-task thoughts have been found to cause time to feel quicker. However, not only was this found for very short timeframes, but no other study has looked at the reversed relationship. Therefore, we adapted a time perception task designed to manipulate subjective time perception to be more similar to common sustained attention tasks to measure time perception and used thought probes to measure and classify off-task thoughts. We found that off-task thoughts were at their peak when time perception wasn’t faster or slower, but at a more medium pace. We interpreted this as fast or slow time perceptions might represent situations that demand attentional resources and that medium time perception represent more boring moments that allow for off-task thoughts to occur.
A11
Metacognitive Time Investment and Effort Regulation in Adult ADHD: Applying the Diminishing Criterion Model
Adrian Torres
apt98@yorku.ca
▶ Show abstract
Difficulties engaging mental effort during cognitive tasks are well documented in ADHD and reflected in DSM-5-TR inattention criteria. The Diminishing Criterion Model (DCM) is a metacognitive framework that integrates objective performance and subjective confidence to predict time investment. Although validated in neurotypical samples, the DCM has not been applied to neurodevelopmental populations. This study applied DCM principles to an adult ADHD group (n = 69) and a control group (n = 70) recruited via Prolific Academic. Participants completed a 40-item word-analogy task, recording item-level response times (RTs) and confidence ratings. No group differences emerged in accuracy, mean confidence, resolution, or calibration. However, controls invested significantly more time overall than the ADHD group. A significant curvilinear confidence–RT relationship was observed in both groups, consistent with DCM predictions. Critically, neither the linear nor curvilinear components of this relationship differed between groups, suggesting similar patterns of strategic metacognitive effort regulation despite differences in overall time investment. These findings extend ADHD metacognition research beyond one-shot monitoring accuracy measures, indicating that item-level effort allocation processes may remain intact in adult ADHD.
A12
Mind Wandering and Social Conformity
Maryam Hodaie
mhodaie@uwaterloo.ca
▶ Show abstract
Mind-wandering (MW) is a shift of task-related processing to internally generated thought that is known to disrupt sustained attention to a task and impair goal-directed performance. We examined whether levels of MW can be influenced by social context. Participants completed the Metronome Response Task (MRT), intermittently responded to MW thought probes while being shown experimentally manipulated MW "group-averages" of peers (normative feedback), indicating the peers were either highly attentive or inattentive. Rhythmic response time variability (RRTv) served as a behavioral index of attention to the MRT. 176 undergraduate participants were randomly assigned to either an attentive or inattentive normative feedback conditions. After the normative feedback was presented, participants exposed to inattentive norms reported significantly higher MW compared to the attentive condition. While no main effect of condition was observed in RRTv, a significant condition × block interaction pointed to a steeper RRTv increase with time in the attentive group. RRTv predicted MW reports across both conditions, and this relationship did not differ between groups. Thus, MW is sensitive to normative social information, shifting with perceived standards of attentiveness. These findings may reflect heightened meta-cognitive monitoring under attentive norms, where self-evaluation of one’s performance disrupts task focus, resulting in greater response variability.
A13
Personalized Music with Auditory Beat Stimulation Reduces Physiological Markers of Anxiety: Evidence from Cortisol and HRV
Yoseph Lahmany
yoseph.lahmany@torontomu.ca
▶ Show abstract
Music-based digital therapeutics have recently gained attention as scalable approaches to anxiety regulation. This is motivated by the increased prevalence of anxiety in young adults and the capacity for music to induce emotions. Prior research has shown that personalizing music by manipulating its affective components in accordance with the Iso Principle, combined with auditory beat stimulation (ABS), reduced subjective anxiety ratings to a greater extent than a pink noise active control (Mallik & Russo, 2022). Building on this, the present study is the first to examine this intervention in a biopsychological context, using biomarkers of stress regulation. Heart rate variability (HRV) and salivary cortisol are implicated in two distinct stress-response systems: the autonomic nervous system and the hypothalamic–pituitary–adrenal (HPA) axis. These biomarkers were assessed throughout the passive listening intervention, along with subjective measures of anxiety. Pre–post comparisons revealed a significant reduction in cortisol and subjective anxiety, with a larger decrease for personalized music + ABS than pink noise. While pre–post changes in HRV did not differ between conditions, both conditions exhibited significant positive trends across the listening period. These results support the efficacy of music-based interventions incorporating the Iso Principle and ABS in reducing somatic symptoms of state anxiety.
A14
State boredom as a physiological stressor
Chantal Trudel
► Show abstract
Boredom is the feeling of wanting but failing to engage the mind. Commonly described as a restless, agitated state, research has demonstrated conflicting results when exploring the association between boredom and arousal. We induced boredom and contrasted it with interest, while measuring heart rate variability as a measure of autonomic nervous system function.
One hundred and twenty participants rated their boredom and interest levels before and after watching boredom- or interest-inducing videos. Results showed higher heart rate and lower RMSSD during boredom, supporting the idea that boredom is a negatively valenced emotion (Danckert & Elpidorou, 2023). Participants' subjective boredom and interest levels aligned with the induced emotions (boredom and interest) and physiological measures (lower stress response when watching an interesting video). These findings characterize state boredom as a physiological stressor that prompts the organism to seek more optimal cognitive engagement, aligning with our proposed theory that boredom signals suboptimal cognitive resource utilization, akin to a cognitive homeostatic control loop (Trudel et al., 2025). The activation of the sympathetic nervous system in the boredom condition supports this goal, encouraging the switch to a more positively valenced activity to restore cognitive balance.
A15
The Affective Consequences of Oculomotor Selection and Inhibition
Niyatee Narkar
nnarkar@uoguelph.ca
▶ Show abstract
Ignoring task-irrelevant stimuli in our environment can negatively impact their perceived value, possibly reflecting an affective consequence of attentional inhibition. Prior research from our lab suggests this ‘distractor devaluation’ effect may extend to the oculomotor domain, with looked-away-from images subsequently receiving more negative ratings than looked-at images. Whether this difference in affective ratings is really due to inhibition-related devaluation of looked-away-from items or instead due to fluency-related enhancement of looked-at items has remained unclear. We therefore sought to replicate prior work while also obtaining a novel-stimulus affective baseline against which other conditions could be compared. A comparison of the affective impact of selective-looking versus antisaccade tasks allowed us to further explore whether increased demands on oculomotor inhibition would produce greater distractor devaluation. Taken together, the results of four experiments make clear that stimulus value is both enhanced for looked-at targets and reduced for salient distractors to which it is difficult to suppress eye-movements. However, introducing a delay between eye-movement trials and affective-evaluation trials suggests that such effects may be short-lived. These findings further our understanding of the links between oculomotor control and affective preference formation, while highlighting important boundary conditions of the affective impact of controlling what you look at.
A16
The effect of rumination on monitoring and accuracy during a prospective memory task using eye-tracking
Iulia Niculescu
nicules@uwindsor.ca
▶ Show abstract
Background. Rumination may hinder prospective memory (PM) through depleting attentional control capacity. Specifically, rumination may increase negative thoughts in conscious awareness, limiting PM cue detection or PM intention recall. However, there have been no studies clarifying how rumination impacts PM using eye-tracking. Objective. To examine the effect of rumination on PM performance during an ongoing task with emotionally valanced stimuli. using eye-tracking measures. Procedure. Participants (n=90) completed rumination and mood questionnaires. Eye movements were recorded while participants completed a 2-back task with positive, neutral, or negative words, with an embedded PM task. Mixed-effects models were used to examine the effects of rumination, valence, and mood on PM accuracy, dwell time, and fixation counts. Results. Positive valence was associated with marginally higher PM accuracy, while rumination moderated the effect of positive valence on monitoring. For positive words, higher rumination was associated with reduced dwell times and fewer fixations on peripheral PM cues, whereas lower rumination was associated with increased monitoring for cues. Conclusions. Although positive valence was associated with enhanced monitoring and accuracy among low ruminators, this benefit diminished with higher rumination. Rumination may thus reduce PM by interfering with the attentional broadening typically elicited by positive mood.
A17
The impact of brief meditation training on divided attention and recognition memory
Bailey Stokes
bestokes@mta.ca
▶ Show abstract
Attention is known to operate with limited resources. Yet, researchers have shown that when simultaneously memorizing words and engaging in a detection task, recognition memory is best for words paired with target signals—a phenomenon known as the attention boost effect (ABE). When the word list is comprised of low and high frequency words, the ABE is larger for high than low frequency words, which suggests the mechanism that typically produces better recognition memory for low than high frequency words is redundant with the mechanism underlying the ABE. The current study extends these findings by introducing a meditation manipulation, with half of participants exposed to brief meditation training and half exposed to a control condition. Participants then completed a recognition memory task for high and low frequency words that also included a detection component. The results show that in this divided attention task, recognition memory was best on trials in which a target signal was detected, low frequency words were better remembered than high frequency words, and the meditation group showed better memory than the control group. However, these factors did not interact, suggesting attentional enhancements due to detecting targets, encountering low frequency words, and meditating involve separate mechanisms.
A18
When Attention Turns to You: How Eye Contact Alters the Perception of Time
Layla Hussain
► Show abstract
When someone looks at you, time feels different: Past work has shown that people underestimate the duration of faces looking toward (vs. away from) them. Here we asked whether this effect may be due to the perception of others directing their attention towards us, or the perception of directed attention more generally. These possibilities are typically confounded in social perception research, where faces with averted gaze appear to be looking towards empty space, and the object of attention is thus clear for directed but not averted gaze. In a preregistered experiment, participants (N=120) assessed the duration of faces looking at the observer (self-directed), looking at another person (other-directed), and looking at nothing (non-directed). Self-directed gaze elicited time overestimation, whereas other- and non-directed gaze were underestimated. These findings shed new light on how others' attention impacts time perception and underscore the unique salience of being the object of another person's attention.
A19
A neurocognitive investigation of affect and syntax
Borbala Dobos
xk20oe@brocku.ca
▶ Show abstract
Affective state is known to influence cognitive processing. However, its role in the perception of grammatical structure in neutral sentences remains understudied. In an event-related potential (ERP) study, we examined the interaction of dispositional affect with the P600 effect, a marker of syntactic anomaly. To this end, 28 participants read sentence stimuli (from Osterhout & Holcomb, 1992; critical words _underlined_) such as (i) The broker planned _to_ conceal the transaction *_was_ sent to jail vs. (ii) The broker persuaded *_to_ conceal the transaction _was_ sent to jail. We replicated previous P600 findings at _to_ and saw a marginally significant P600 effect at _was_. Furthermore, based on previous findings in our lab, we hypothesized that individuals who scored higher in positive affect would show larger P600 effects at these critical words. Results showed a significant positive correlation between positive affect scores and P600 amplitude differences at _to_, as well as suggestive evidence for this same association at _was_. Data collection is still on-going; however, these important preliminary findings suggest that the underlying neural mechanism for positive affect allows for better perception and increased sensitivity to grammatical structure.
A20
Accessing the meaning of a metaphor in real time: The effects of aptness, conventionality, and familiarity on the time course of meaning access
Cedric Le Bouar
cedric.le-bouar@mail.concordia.ca
▶ Show abstract
In metaphors like my lawyers are sharks, must the literal meaning of the vehicle (i.e., sharks) be accessed before the figurative interpretation is derived? The time course of meaning access during metaphor interpretation was investigated using a crossmodal primed lexical decision experiment. The effects of aptness (i.e., how well properties of the vehicle apply to the topic), conventionality (i.e., association between a vehicle and a figurative interpretation), and familiarity (i.e., experience with a metaphor) on priming was also assessed. Passages containing a metaphor or a simile (e.g., lawyers are like sharks) were aurally presented to participants while they performed lexical decisions to visually presented letter strings. Word stimuli could be related to the metaphor’s figurative interpretation (shark – CRUEL), to the vehicle’s literal meaning (BLOOD), or be matched unrelated controls. Visual targets appeared at the recognition point of the vehicle (determined by gating task) or 500ms later. Response times did not reveal priming at the early timepoint but showed that both meanings were primed at the late timepoint. Only aptness improved model fit at the late timepoint. Results suggest that the literal meaning of the vehicle is accessed during metaphor interpretation, and that aptness influences the speed of meaning access.
A21
Action-Dependent Attention Selects Behaviourally Relevant Object Segments
Jiyoon Jeong
jy.jeong@mail.utoronto.ca
▶ Show abstract
It is often assumed that during grasping, attention splits into multiple foci near the intended grasp-points, whereas during reaching, attention is directed toward the object's centre of gravity. Here we revisited this distinction by asking participants to grasp or reach for objects that differed in local geometry near grasp-points, coloured surface patches, and object orientation. To decode feature representations underlying these two action types, we used electroencephalography (EEG) combined with Support Vector Machine (SVM) classifiers applied to event-related potentials (ERPs). Classification accuracy for grasp-point geometry was significantly higher during grasp planning compared to reach planning (234–332 ms), as was accuracy for object orientation (308–371 ms), consistent with previous studies. However, contrary to the classical split-attention account, classification accuracy for central colour was significant regardless of action intention, suggesting that central colour is processed independently of the planned action. Taken together, these findings suggest that action-dependent attention does not split into isolated foci at the contact points, but instead parses objects into behaviourally relevant units that are defined by the motor demands of the task at hand.
A22
Acute Auditory Degradation Enhances Auditory–Frontal Connectivity During Emotional Speech Perception
Katya Tikhostoup
ctikhostoup@torontomu.ca
▶ Show abstract
Previous research has shown that older adults with hearing loss are less accurate in the recognition of affective auditory stimuli than those with normal hearing. Some studies have also found that these drops in accuracy are accompanied by task-based increases in fronto-temporal connectivity. However, it is unclear whether this increased connectivity only arises following years of hearing loss or whether it may arise suddenly following signal degradation. Using functional Near-Infrared Spectroscopy (fNIRS), the present study investigated the effect of sudden simulated hearing loss in normal hearing young adults. Forty-five participants listened to emotional speech in multitalker babble and judged the expressed emotion. Half of the participants heard the original, unaltered stimuli (NH), while the other half heard stimuli that were modified to simulate hearing loss (SHL). Compared to NH, SHL had worse task performance and increased connectivity between superior temporal gyrus and inferior frontal gyrus (IFG). One interpretation of this difference in connectivity is that the IFG of participants with SHL may have been differentially recruited to support the “filling-in” of missing detail that would support emotion recognition. These findings indicate that neural signatures commonly attributed to long-term hearing loss can emerge immediately following signal degradation.
A23
Acute Stress Increases in Delta-Beta Coupling Are Partially Moderated by Sympathetic Activity
Mohammad Soleyman Nejad
soleymam@myumanitoba.ca
▶ Show abstract
Delta-beta coupling (DBC) has been proposed as a neural marker of cortico-subcortical coordination relevant to stress regulation, though prior findings are mixed. Little is known about how DBC might respond to laboratory manipulations of acute stress, such as the Trier Social Stress Test (TSST), a gold-standard paradigm for inducing acute psychosocial stress, reliably increasing social-evaluative threat and feelings of uncontrollability. We hypothesized that acute stress would increase resting-state DBC after the TSST relative to a pre-TSST resting measure. We also examined whether autonomic indices of stress reactivity, specifically high-frequency heart rate variability (HF-HRV) and pre-ejection period (PEP), would operate as potential moderators or mediators of stress-related neural changes. Eighty-seven undergraduate students were randomly assigned to complete the TSST or a placebo condition. Resting-state DBC was assessed before and after the manipulation. Results indicated that DBC increased following the stressor but remained stable in the control condition. Exploratory analyses indicated that HF-HRV neither moderated nor mediated stress-related changes in DBC, whereas PEP showed a significant condition × time interaction, with stress eliciting decreases in PEP. Further, changes in PEP were associated with changes in DBC, suggesting a partial indirect pathway linking stress exposure to altered DBC.
A24
Are different bilingualism factors associated with resting-state brain oscillations that underly cognition?
Thamayanthi Giritharan
tgiri012@uottawa.ca
▶ Show abstract
Bilingualism has been associated with superior executive function, and, in aging, it has been found to contribute to cognitive reserve (i.e., spared cognitive functioning despite neuropathology). However, the bilingualism factors that contribute to this effect remain underexplored. Little research to date has examined the impact of bilingual language experiences on resting-state electroencephalography (EEG). Given that different frequency bands are associated with different cognitive and executive processes, we examined the strength of relationships between neural oscillations in the theta, alpha, and beta EEG frequency bands with second language (L2) age of acquisition (AoA), entropy (a measure of contextual language use), and L2 proficiency. Resting-state EEG was recorded from French/English bilingual young adults (n=74) while they fixated on a central point and were instructed to clear their mind and not think of anything. Data analysis is ongoing; it is hypothesized that power in the theta, alpha, and beta frequency bands (i.e., indices of memory, attention, executive function, and cognitive processing) will be positively correlated with entropy and proficiency, while AoA will be inversely correlated. This work will illuminate the impact of different bilingualism factors on the neural responses that support cognition in young adults. Future research will examine this in older adults.
A25
Associations between early life stress and language outcomes; an Event Related Potential (ERP) study investigating phoneme discrimination in 6- and 10-month-old infants
Aaliyah Daruwala
aali786@yorku.ca
▶ Show abstract
Language develops such that early established foundations scaffold development of more complex language abilities. Processes like phonemic discrimination and perceptual narrowing lay the foundation for vocabulary development and even later literacy. The current study tests associations between early life stress, event-related-potential (ERP) amplitude during a phoneme discrimination task, and language outcomes in a longitudinal sample of 6- and 10-month-old infants (n= 11), recruited as part of a larger ongoing study. EEG was recorded while infants completed an auditory oddball task in which they listened to a stream of standard (presented 80% of the time) and deviant (presented 20% of the time) phonemes. In this preliminary sample, early life stress (defined using a cumulative risk index) did not significantly predict the amplitude of the P150 difference wave (p= 0.5). P150 amplitude, similarly, did not predict scores on a standard language assessment (PLS-5) (Auditory comprehension: p= 0.6; Expressive communication: p= 0.9). Ongoing analyses will be updated (anticipated n by conference = 25) and will explore additional associations between early life stress, early language outcomes, and the late slow wave (LSW) component, where visual condition differences are currently observed.
A26
Associations Between Joint Attention and Event-Related Potential (ERP) Response During Auditory Statistical Learning
Prabhleen Kaur
prabhleenkaur4060@gmail.com
▶ Show abstract
Joint attention (JA)—the shared focus of a child and caregiver over an object or event—is a critical early social-cognitive skill linked to language acquisition, yet the neural pathways connecting JA to language outcomes remain unclear. This study examined whether quantity and quality of JA engagement at 24 months predicts neural responses during statistical learning, a foundational skill that supports early language acquisition. A sample of mother-infant dyads (n = 28) completed a semi-structured free-play task to assess JA when infants were 24 months old. Event-related potentials (ERPs) were recorded from infants during a passive auditory statistical learning task, during which infants heard a continuous stream of three syllable “tone-words” presented in random order. Linear regression demonstrated a significant association between JA quantity and P2 amplitude, such that greater JA quantity predicted a larger difference in P2 amplitude between predictable and unpredictable tones. Results suggest that frequent JA engagement may contribute to development of early neural mechanisms that support statistical learning. These results highlight the importance of early social interactions in supporting neural mechanisms for language learning and point to the need for longitudinal studies that can test mediating pathways connecting early social engagement with language development.
A27
Detecting differential modulations during cognitive tasks
Benjamin Corrigan
bcorrig@yorku.ca
▶ Show abstract
Cognitive tasks frequently generate neural responses that are affected by different features and at different times as a trial progresses. In a stop/change-signal task, where there is an instruction for one movement, and then sometimes a second instruction, neurons in medial frontal cortex can be modulated by either instruction or by participating in performance monitoring, or a mixture of any of these. To identify modulations error signals in the presence of visual responses, we developed a permutation-based method using the difference between the spike density functions of aligned trial types. We identified modulation periods in three areas of the medial frontal cortex of macaque monkeys. We found differences between the two banks of cingulate cortex and supplementary eye fields in terms of the modulation onsets, durations and reliability of firing rates during these periods in macaques during a stop-signal task. We also show that these methods work for neurons recorded in humans in a change signal task in cingulate cortex and pre-supplementary motor area.
A28
Do Physical Constraints Shape Feature Integration? Human EEG Validation of a Reinforcement Learning Model in Memory-Guided Reaching
Melina Halkias
melinahalkias@icloud.com
▶ Show abstract
Goal-directed reaching requires the brain to integrate visual, spatial, and motor information across perception, memory, and action. While prior research shows that feature integration evolves over the course of movement planning, it remains unclear whether this integration is shaped by the physical constraints of the environment. To address this, we combined computational modeling with human neurophysiology to test whether real-world physics drive feature integration during memory-guided reaching. A reinforcement learning agent controlling a simulated arm was trained under two conditions: one approximating real-world joint dynamics and another with altered physical constraints. Representational analyses revealed greater feature integration in the real-physics model, whereas altered physics led to more disentangled and localized representations, suggesting that integration emerges only when required for efficient action. To validate these findings, ongoing human EEG and EMG data are being collected during a memory-guided reaching task closely matched to the model paradigm. Representational similarity and regression analyses will assess whether neural activity reflects physics-dependent integration patterns. Together, this work provides a novel framework linking environmental constraints, computational learning, and neural dynamics, offering insight into how the brain optimizes action through adaptive feature integration.
A29
Effects of Anthropogenic Noise on Song-Evoked Neural Activity in the Zebra Finch Auditory Forebrain
Patrick Curiston
curiston@ualberta.ca
▶ Show abstract
Anthropogenic noise is an increasingly pervasive feature of modern environments and poses a substantial challenge to animals that rely on acoustic communication. In songbirds, conspecific vocalizations are essential for social interaction and reproduction, yet the neural consequences of hearing these signals in noisy conditions remains poorly understood. While behavioural adaptations to noise pollution are well documented, far less is known about how background noise alters activity within auditory processing circuits. Here, we examine how urban-like anthropogenic noise influences neural activation in higher-order auditory forebrain regions of zebra finches (Taeniopygia guttata).Auditory information in songbirds is processed through a hierarchical forebrain pathway, culminating in the caudomedial nidopallium (NCM) and caudomedial mesopallium (CMM), regions implicated in auditory memory, stimulus evaluation, and song recognition. Following exposure to song playback embedded in anthropogenic noise we assessed neural activity within NCM and CMM using immunohistochemical staining for ZENK (Egr-1), an immediate early gene rapidly transcribed following neuronal depolarization and widely used as a marker of auditory-driven neural activation. Together, our findings aim to clarify how real-world noise affects neural mechanisms of auditory perception and memory, providing insight into the resilience and vulnerability of songbird auditory systems in increasingly noisy environments.
A30
Effects of Optic Flow Perturbations on Postural Control in Working-Aged Adults Post Concussion
Sonia Vovan
svovan2@yorku.ca
▶ Show abstract
Background: Individuals with concussion often report dizziness and instability in visually complex environments, yet the sensory integration mechanisms underlying these symptoms remain poorly understood, particularly in working-aged adults where such deficits may affect occupational safety. Methods: Ninety adults (25–65 years; 50% female) participated: Controls (n=30), Lower Symptom Burden (LSB; n=30), Higher Symptom Burden (HSB; n=30). Participants stood on a force plate while immersed in a 240° VR grocery store environment. Optic flow perturbations were presented at three speeds (1.5, 3.0, 4.5 m/s) with predictable or unpredictable timing. Centre of pressure (COP) path length and burst response metrics (peak amplitude, latency, balance correction time) were measured. Results: HSB demonstrated longer COP path lengths than Controls, but only at faster optic flow speeds (3.0 and 4.5 m/s). Only HSB showed speed-dependent increases in path length. HSB took longer to complete balance corrections than Controls across all conditions. No group differences in peak amplitude or latency were observed, indicating that group-related differences emerged in balance recovery rather than the initial response. Conclusion: Individuals with higher symptom burden demonstrate impaired efficiency in balance correction during visual perturbations, particularly at faster optic flow speeds.
A31
Exploring Whether Polygenic Risk Moderates Associations Between Maternal Wellbeing and Infant Neurodevelopment in a Low-SES Sample
Zahra Wakif
zwakif11@gmail.com
▶ Show abstract
Early brain development is shaped by the environment in important ways. Early environmental risks to neurodevelopment include low socioeconomic status (SES) and poor caregiver wellbeing. Previous work has used electroencephalography (EEG) to explore associations between early environmental risk and infant neurodevelopment, however the moderating role of polygenic risk is less well understood. The current study tested 116 mother-infant dyads recruited from primary-care clinics serving predominantly low- and mid-SES families, across the first year of life (at 2, 6, 9, and 12 months). Infant polygenic risk for depression was tested as a moderator of associations between cumulative risk (including SES variables and maternal wellbeing) and infant neurodevelopment, measured as frontal alpha asymmetry (FAA) at each time point. Polygenic risk did not significantly moderate these associations. Ongoing analyses will test whether environmental and genetic risk predict early parietal alpha asymmetry - an EEG marker associated with early emotion regulation that may be associated with environmental risk earlier in infancy than FAA. Understanding how the interaction between genetic and environmental risk contributes to variation in neurodevelopment from early infancy bridges current research gaps and provides an avenue to identify who may be at higher risk for depression later in life.
A32
Eye vs. hand effector-specificity in human parietal and frontal cortices
Kiana Masoudi
kmasoudi@yorku.ca
▶ Show abstract
Humans rely on visually-guided saccades and reaches to interact with the world. Although non-human studies report effector-specific signals for saccades vs. reaches in parietal and frontal cortices, human fMRI studies often show overlapping activation, perhaps due to temporal “blurring” of closely-timed sensory, cognitive, and motor events. Here, we use an fMRI paradigm that separates these sensorimotor events to better characterize saccade versus reach networks, and to determine effector-specific activation. Participants (N = 20) performed cue-separated saccade, reach, and control tasks. Each trial included 3 events: an effector-independent target-encoding phase, and effector-dependent motor-planning and execution phases. Relative to control, encoding of visual target information engaged bilateral temporal cortex. During planning, saccades activated occipito-parietal and frontal regions whereas reaching showed a broader occipito-parietal-frontal activation pattern. During execution, both effectors activated overlapping occipital-parietal regions, with distinct frontal motor regions. Direct comparison of saccade vs. reach activation patterns during planning revealed greater reach activity in parietofrontal regions. During execution, saccades had greater activity in lateral occipito-parietal and frontal eye-field related areas while reaches had greater activity in medial occipito-parietal and left motor-related frontal regions. Together, these findings suggest distinct saccade- and reach-related activation across occipito-parieto-frontal regions during movement execution.
A33
Face size modulates face perception: Evidence from mass univariate analyses of ERPs in a parametric design
Lily M. R. Laevens
lmrlaeve@uwaterloo.ca
▶ Show abstract
In everyday life, we encounter people at various distances, causing their faces to vary in size on the retina. Face size is thus an important factor that may modulate face-related processes, yet its impact at the neural level remains understudied. Importantly, face sizes larger than 6° of horizontal visual angle are rarely tested and methodological concerns such as lack of enforced fixation location and small sample sizes are likely contributors to result inconsistencies. The present study used a fixation-contingent approach with an eye tracker to investigate the effects of 7 face sizes (2°, 4°, 6°, 8°, 10°, 12°, 14° horizontally) on gender discrimination. Data were analyzed using data-driven whole scalp mass-univariate statistics. Results (N = 56) indicate main effects of gender and size, but no interaction. Face size impacted neural activity from 50 to 350ms across the scalp. Paired contrasts showed the strongest N170 differences between 2-4°, with some differences also observed between 4-6°. Differences between sizes 6-8°, 8-10° and 10-12° were non-significant, while those between sizes 12-14° occurred between 200-250ms only at right posterior sites. Results suggest that face-related neural activity is mostly impacted by small and large face sizes but remains similar for faces at conversational distances.
A34
Factorial Analysis of Ensemble Representation of Attention and Decision-making in Prefrontal Cortex of Macaque Monkeys
Wanyi Lyu
wanyilyu@yorku.ca
▶ Show abstract
Macaque monkeys were trained with positive reinforcement to perform a visual search task with separate interleaved 2x2 factorial manipulations of attention and of response decision operations. Attention was manipulated by varying the similarity between a singleton and distractors. Response decision was manipulated by varying the discriminability of a go/nogo cue. The organization and termination rule of the two operations were previously determined using System Factorial Technology (SFT; Lowe et al, 2019). Single-units were sampled in the frontal eye field (FEF) with linear electrode arrays. Here, results are reported from an ensemble level analysis using demixed Principal Component Analysis (dPCA). The ensemble of neurons in FEF embodied reliable representations of singleton location and response decision. The separate manipulations of attention and decision operations influenced the timing of these representations. The dPC for attention and for decision were not entirely orthogonal. The results indicate that attention and decision processes are embodied in overlapping neuronal subpopulations rather than in fully distinct groups. Future work will determine whether variation of the dPC accounts for systematic variation of response times and how single-neuron properties relate to the ensemble representations. We will also compare these representations across different processing architectures characterized by SFT.
A35
Fingerprinting The Social Brain: A Recently Developed Mentalizing Signature Tracks Age and Autism-Related Differences.
Natalia Castro Gonzalez
natalia.castrogonzalez@queensu.ca
▶ Show abstract
Like maps represent landmarks, stable whole-brain patterns of neural features can capture specific mental processes. Known as brain signatures, these multivariate patterns can track subjective experiences (pain, craving) and treatment-related changes. Here, we tested the sensitivity of a recently developed brain signature of mentalizing - the ability to infer others’ mental states. Specifically, we assessed whether neural signature expression (i) tracks developmental differences in mentalizing across adulthood, (ii) shows translational relevance in autism spectrum disorder (ASD), and (iii) captures variability in mentalizing ability. We analyzed fMRI data from two datasets from participants completing mentalizing tasks: dataset 1 (265 young adults aged 18–30; 42 older adults aged 65–77), and dataset 2 (232 neurotypical individuals; 23 with ASD). Signature expression was quantified as the dot product between task-evoked whole-brain activation patterns and the mentalizing signature. The neural signature reliably distinguished mentalizing from other processes, such as factual reasoning, and relative signature expression reflects individual differences in mentalizing ability. Signature expression was reduced in older adults, mirroring age-related declines in mentalizing, and in ASD compared with neurotypical participants. These findings demonstrate sensitivity to developmental and clinical variation and highlight the potential of neural signatures for studying socio-cognitive abilities across populations.
A36
Following the Story: Brain and Behaviour in Children’s Engagement
Rafaela Platkin
rplatkin@uwo.ca
▶ Show abstract
In complex listening environments, such as classrooms, successful auditory perception is related to listening effort – the dynamic interplay between the processing demands of auditory stimuli and an individual's available cognitive resources. While previous research has identified factors that influence listening effort, much of this work has relied on isolated, unnatural stimuli leaving a gap in our understanding of how listening effort is exerted during naturalistic, real-world listening. Thus, in the current study children listened to a series of context-rich stories while neural activity was recorded. EEG power spectral density analysis was used to evaluate changes in alpha and theta band activity over the course of each story, and intersubject correlation was used to assess neural synchrony across participants. Following each story, participants completed an adapted version of the Story World Absorption Scale measuring subjective engagement. Across listeners, attention, enjoyment, and mental imagery were found to be the primary drivers of subjective engagement. EEG analyses revealed greater neural synchrony during more engaging stories relative to a less engaging control story. These findings demonstrate that child-oriented narratives can serve as valid stimuli for investigating listening and engagement under naturalistic conditions, with implications for classrooms and the design of accessible auditory materials.
A37
Freezing of Upper Limbs Parkinson's Disease: A Reaching Study
Jennifer Stevenson
jenniferstevenson@trentu.ca
▶ Show abstract
Freezing of gait (FOG) affects approximately 50% of people with Parkinson's disease (PwPD) and may reflect a global motor control deficit extending beyond gait. We investigated whether freezing-like episodes could be elicited in the upper limbs during reaching tasks designed to mimic established FOG triggers. Because PwPD exhibit altered processing of visual spatial-frequency information, we also examined whether manipulating visual backgrounds would influence motor performance. PwPD and similarly-aged controls completed three reaching tasks: reversals, corners, and navigating through a virtual tunnel, analogous to walking through a doorway. Targets were presented against low-frequency, high-frequency, and stippled control backgrounds.PwPD exhibited significantly longer movement times and lower mean speeds than controls, and the tunnel condition was disproportionately challenging for PwPD. A movement-phase analysis revealed that PwPD allocated reach distance differently than controls in both the reversal and tunnel conditions. Ongoing analyses examining velocity at specific time and distance points along the reach aim to identify discrete signatures of freezing-like behaviour. Spatial frequency manipulations did not produce consistent effects. These findings suggest that reaching tasks involving transitions elicit kinematic disruptions in PwPD, supporting the view that freezing reflects a broader motor control deficit.
A38
Hearing Aids Hinder Speech Emotion Perception in Noise: Inefficient Sensorimotor Integration
Carmen Dang
c1dang@torontomu.ca
▶ Show abstract
Hearing aids are the primary treatment for age-related hearing loss, but fail to alleviate the difficulties experienced by users when perceiving emotion in speech. Although fMRI has found that even mild hearing loss alters neural processing of affective sounds, research in hearing-aid users is limited because fMRI is incompatible with hearing-aid metal components. In contrast, fNIRS is an optical neuroimaging technology that is silent, portable and hearing-aid compatible. By leveraging fNIRS, we investigated how hearing-aid users neurally process emotion in speech, particularly in an acoustically noisy environment. We found that hearing-aid use can, under certain conditions, impair speech emotion perception. Furthermore, we identified three graded levels of efficiency in sensorimotor integration (inferior frontal gyrus and inferior parietal lobe task-based connectivity). Normal-hearing older adults had high task performance with low sensorimotor integration. Followed by unaided listening, with comparable performance but marginally elevated sensorimotor integration. Lastly, aided listening exhibited the worst performance, despite sensorimotor integration greater than normal hearing and comparable to unaided listening. This ordering suggests that signal-processing strategies, designed primarily to prioritize speech intelligibility, may inadvertently alter acoustic emotion cues and increase emotional ambiguity; highlighting the importance of affective and social dimensions of speech when designing and evaluating hearing aids.
A39
How is the BRIEF Used to Assess Executive Function in Pediatric Epilepsy: What is Being Reported?
McCaley Campbell
mccaley.campbell21@gmail.com
▶ Show abstract
Epilepsy is a chronic neurological disorder characterized by recurrent seizures. Executive function (EF) difficulties are common in children with epilepsy and are often assessed using the Behavior Rating Inventory of Executive Function (BRIEF). This review synthesizes 49 articles, examining which BRIEF subscales and informant types have been used in the pediatric epilepsy literature. We found that all nine subscales (e.g., inhibition, working memory, shifting, emotional regulation) and composite scores (behavior regulation index, metacognition index, global executive composite) have been used to describe EF in children with epilepsy however working memory, inhibition, and composite scores were reported the most often. Informant type is also important as parents, teachers, and youth themselves provide unique perspectives regarding EF in daily life. The review indicated that parent reports were most often used, whereas teacher and self-reports were used the least. The findings show an uneven focus across executive function aspects and informant types, with some being overrepresented while others are rarely assessed. Understanding these patterns is essential for determining which executive function aspects and informant perspective is over- or under-represented in the pediatric epilepsy literature and for improving the design of future studies to ensure more comprehensive and appropriate assessment and reporting.
A40
How Meaning Becomes Memory: ERP Correlates of Long-Term Recognition for Concrete and Abstract Words
Alexandra Doiron
alexandradoiron@cmail.carleton.ca
▶ Show abstract
Background: Concrete words are remembered better than abstract words, known as the concreteness effect. However, the neural mechanisms underlying their encoding and retrieval remain unclear. This study examined behavior and neural responses for concrete and abstract words. Methods: Forty-six younger adults completed a long-term memory task while behavior and electroencephalography were recorded. During encoding, participants categorized 100 words as concrete or abstract. Following a 20-minute distractor task, participants completed a recognition test with 200 words (100 foils, 100 targets) and indicated whether each word appeared during encoding. Results: Participants were more accurate when categorizing concrete vs. abstract words during encoding. During recognition, concrete words produced more hits and fewer false alarms, resulting in higher d’ scores and better memory performance than abstract words. ERP analyses revealed larger N400s for concrete than abstract words during encoding and recognition. LPC amplitude was larger for abstract than concrete words during encoding, while only concrete words showed an LPC increase from encoding to retrieval.Discussion: Greater accuracy and larger N400s for concrete vs. abstract words reflect richer semantic representations supporting the memory advantage for concrete items. Larger LPC for abstract words during encoding reflects greater elaborative semantic processing to categorize abstract vs. concrete words.
A41
Hungry Brains Choose Foods Differently: State-Dependent EEG Microstates Predict Health Evidence Processing
Tavneen Sandhu
tavneen.ksandhu@queensu.ca
▶ Show abstract
Dietary decisions often occur under varying physiological states (i.e., hunger), yet most neurocomputational models assume stable decision processes and rarely consider how these states influence the accumulation and integration of health and taste information during food choices. Moreover, intrinsic brain dynamics measured during rest—before the decision process unfolds—may shape how individuals integrate these competing attributes when evaluating foods. Here, we combined resting-state EEG with computational modeling to examine how hunger influences dietary decision-making. In a within-subject design (n=110), participants completed an established food choice paradigm across two sessions—fasted and satiated. Choices were modeled using an attribute-based drift diffusion model that characterizes how health and taste information accumulate during decision-making. Hunger significantly reduced the likelihood of choosing healthy foods and shifted decision processes toward prioritizing taste over health considerations. Resting-state EEG microstates—brief patterns of large-scale brain activity measured before the task—predicted subjective valuation and selectively predicted the accumulation of health-related evidence, but not other decision parameters. These findings suggest that intrinsic brain dynamics prior to decision-making may shape the degree to which individuals incorporate health information during dietary decisions, revealing a neural–computational pathway linking physiological state, intrinsic brain dynamics, and mechanisms underlying dietary decisions.
A42
Investigating spatial consciousness across the visual blind spot
Bernard 't Hart
thartbm@gmail.com
▶ Show abstract
Compared to the rest of our visual field, the cortical representation of the physiological blind spot lacks feedforward input from one eye. Within TWCF-Intrepid (https://arc-intrepid.com) we conducted a series of psychophysical experiments investigating how this alteration of retinotopic structure may affect spatial consciousness. We gathered adversarial predictions from three theories of consciousness. According to Integrated Information Theory, the blind spot’s structural alterations give rise to a cause-effect structure different from the one of eccentrically comparable retinotopic regions, causing space to be perceived as smaller when spatial judgements include the blind spot. Predictive Processing accounts, in contrast, posit that internal models will accommodate structural deviations, with Neurorepresentationalism predicting small disruptions at most, and Active Inference predicting that judgements across the blind spot may be less precise while remaining unbiased. We are testing these predictions using three paradigms where participants estimate distances, area and curvature of motion. Our stimulation is presented dichoptically to the ipsilateral or contralateral eye (relative to mapped blind spot) at locations that either span or do not span the mapped blind spot region. We fit psychometric functions for all experimental conditions to test bias and precision of spatial perception at blind spot and comparable locations.
A43
Is bilingualism a predictor of the brain age gap in older adults?
Kareem Mukbil
kmukb011@uottawa.ca
▶ Show abstract
Brain age can be estimated from neuroimaging data and is increasingly used as a biomarker of biological brain aging. A difference between brain age and chronological age (i.e., a brain age gap) is taken as an indicator of vulnerability to neurodegenerative disease when brain age is greater than chronological age. However, some lifestyle factors may promote resilience to age-related neural decline and thus be associated with a smaller brain age gap. Bilingualism may be one such factor due to the sustained cognitive control demands required to manage multiple languages. In this exploratory study, we used neuroimaging data from 53 French/English bilingual cognitively healthy older adults to estimate brain age using the brainageR package. Brain age was significantly correlated with chronological age (r=0.60, p < .001). The mean brain age gap in the sample was -7.11 years (SD = 6.41; range = -20.79 to 6.63), with eight participants showing a positive brain age gap (i.e., brain age was greater than chronological age). Linear regression analyses showed a limited relation between self-reported bilingual language experience and the brain age gap. Future analyses will include additional measures of bilingual language experience and account for potential biases in age estimation with brainageR.
A44
Keeping the Mind Moving During Aging
Leah Durham
leahdurham@cmail.carleton.ca
▶ Show abstract
Introduction: Physical activity (PA) is linked to cognition, but how cardiovascular regulation (heart rate) relates to activation and connectivity remains poorly understood. fNIRS will be used to measure prefrontal activation and connectivity at rest and during a working-memory task. We predict that physically active older adults will outperform less active older adults and will exhibit brain activation and connectivity patterns consistent with those of younger adults during the n-back task. Methods: 68 younger (18–25) and 65 older (60+) adults completed MoCA screening, then underwent fNIRS during a 5-minute resting-state scan and an n-back (1 & 2-back) task. HbO and HbR activation will be estimated from fNIRS time series using task (task > baseline) and load (2-back > 1-back) contrasts across montage channels.Results: Multivariate partial least squares analyses are ongoing to test whether PA levels are associated with load-dependent fNIRS activation patterns and n-back performance across age groups.Significance: This work will clarify how cardiovascular health and PA relate to cognitive reserve and functional connectivity in aging.
A45
Lateral prefrontal ‘gaze’ signals encode future head and hand motion during visually guided reach.
Veronica Nacher
► Show abstract
Lateral prefrontal cortex (LPFC) is associated with executive function, working memory and response selection. There is evidence that posterior LPFC (pLPFC) is concerned with the high-level selection of specific motor repertoires, but it is not known if it is involved in the planning and coordination of the motion of specific effectors. We investigated this question by recording from pLPFC (spanning Brodman areas 45, 46, & 8a) while two Rhesus monkeys performed head-unrestrained reaches toward visual targets. Task-related neurons (208 / 499) showed time-locked gaze- (and later reach) related responses, but surprisingly these ‘gaze’ responses disappeared (35 / 84 neurons) or diminished during gaze shifts toward the same targets without reach. Further, in-depth spatial analysis confirmed that these ‘gaze’ responses were not what they appeared to be. In direct contrast to the saccade system, gaze displacement models provided the worst fits to the data. Instead, pLPFC ‘gaze’ neurons preferentially coded skeletomotor motion, either future head (49%) or hand (33%) motion, with reach codes predominating later in the task. This is an important demonstration that signal timing does not always reflect spatial tuning in the same neurons. We conclude that many pLPFC ‘gaze’ responses reflect gaze inputs that trigger complex head-hand repertoires.
A46
Lower socioeconomic status is associated with accelerated development of memory integration in children
Megan Kelly
mo.kelly@utoronto.ca
▶ Show abstract
Integrating memories of multiple events is vital for learning and problem-solving. Although evidence suggests that memory integration emerges across childhood and adolescence, little is known about how this process becomes the flexible ability seen in adulthood. One factor that may shape its development is socioeconomic status (SES). Lower SES is oft associated with poorer performance on various cognitive measures such as working memory. However, nonhuman animal models suggest that early stress exposure can lead to precocious memory development, consistent with an accelerated memory maturation hypothesis. The present work tests this hypothesis by investigating whether early stress (indexed via SES) relates to adult-like memory integration processes in children. Results provide initial support for adult-like memory integration, in line with accelerated maturation of memory function: lower-SES children but not higher-SES children benefitted from an instruction to engage in a mature memory integration strategy, achieving performance not significantly different from that of adults. Findings align with the perspective that rather than being objectively detrimental to development, early stress shifts the developmental trajectory of memory toward functional strategies given the developmental environment. Understanding how memory integration develops in tandem with children’s environments is critical for explaining the emergence of flexible learning and problem-solving abilities.
A47
Math Anxiety Leads to Objective and Perceived Cognitive Interference through the Emotional Impact of Intrusive Thoughts
Fraulein Retanal
freta059@uottawa.ca
▶ Show abstract
Math anxiety is a well-documented predictor of lower math performance. A dominant account for this relation posits that highly math anxious individuals underperform because intrusive thoughts consume working memory resources necessary for math performance. However, direct evidence linking intrusive thoughts to math performance deficits of highly math anxious individuals remains limited. The present research addresses this gap by examining how the frequency and emotional impact of intrusive thoughts relate to math anxiety, math performance, and perceived interference of these thoughts on math performance. In Study 1 (N = 309) structural equation modelling revealed that the emotional impact, rather than the frequency, of intrusive thoughts, mediated the relation between math anxiety and both poorer math performance and greater perceived interference on performance. In Study 2 (N = 437), where working memory load was manipulated, the emotional impact of intrusive thoughts mediated the relation between math anxiety and math performance across both working memory conditions. Notably, the emotional impact mediated the relation between math anxiety and perceived interference only for the high working memory demanding task. Our findings demonstrate that emotional dimension of intrusive thoughts should be incorporated into cognitive models of how math anxiety undermines performance.
A48
Measuring neural mechanisms underlying infant joint attention: live-interaction electroencephalography (EEG) with caregiver-infant dyads.
Ana Badal
anabadal@yorku.ca
▶ Show abstract
Early caregiver-infant joint attention (JA), defined as the shared attention between caregivers and infants over an object, play a foundational role in social and cognitive development. Our understanding of JA has relied primarily on structured paradigms that isolate responding to joint attention (RJA) and initiating joint attention (IJA), limiting insight into neural activity during natural caregiver–infant interactions. We introduce a novel live-interaction electroencephalography (EEG) paradigm to examine neural activity during ecologically valid caregiver–infant JA in a longitudinal sample of 6- to 12-month-old infants and compare it to a previously validated structured JA paradigms. EEG was recorded at baseline and during three JA contexts: naturalistic book reading, caregiver-elicited responding to JA (RJA), and infant-initiated JA (IJA). Preliminary data at 6 months (n = 8) show greater theta power during IJA relative to baseline, with similar trends for RJA and book reading. By 12 months, preliminary observations suggest reduced alpha power during JA compared to baseline. Planned analyses will test condition and region effects on alpha and theta power and evaluate whether naturalistic book reading elicits neural responses comparable to structured RJA.
A49
Multidimensional Temporal Dynamics of Distraction in the Context of Visuospatial Attention
Nicolas Joyal
Nicolas.Joyal@gmail.com
▶ Show abstract
In the context of attentional capture, salient stimuli can automatically and involuntarily orient attentional resources, often to the detriment of the ongoing task, a phenomenon related to visual distraction. The present study aimed to investigate the temporal dynamics of distraction effects in the context of a voluntary attention task. To this end, participants (N = 32) completed three experimental tasks while brain activity was recorded using electroencephalography (EEG). Using a visual discrimination task, we identified three components associated with target-evoked responses that accounted for more than 90% of the total variance. We then examined the effect of a visual distractor (absent vs. present) on voluntary attention across these EEG components and linked their modulation to the distractor’s impact on discrimination performance. The results revealed that attentional capture influenced the temporal dynamics of information processing through at least three distinct effects: an early effect involving a single component, a more sustained effect involving two visual components, and a later effect involving an ipsilateral visual component. Mediation analyses further indicated that early effects of distraction shaped later effects. Taken together, these findings highlight the multidimensional nature of distraction in the control of voluntary attention.
A50
Neural Signatures of Event Segmentation During Narrative Listening in Background Noise
Ryan Panela
ryan.panela@utoronto.ca
▶ Show abstract
Investigations into speech-comprehension difficulties often involve intelligibility of short, disconnected sentences, limiting generalization to real-life listening. Novel research approaches are critical for understanding impaired speech processing. Research has demonstrated that humans perceive, encode, and recall experience as discrete events, a process known as event segmentation; however, few studies have leveraged this framework to examine speech-comprehension. In this experiment, participants (N=40, 17–24 years) listened to three stories from The Moth during fMRI scanning. Stories were overlaid with twelve-talker babble across five intelligibility conditions (clear, +14, +9, +4, and –1 dB SNR). Event boundaries were identified using large language models and aligned to the audio. BOLD responses were time-locked to boundaries and compared to event centres. Boundaries elicited transient activity across frontoparietal, attention, default-mode, somatomotor, and visual networks, peaking ~5 seconds post-boundary. Activation at boundaries exceeded event centres, confirming listeners parsed narratives into meaningful events. High-clarity speech engaged broader regions than low-clarity speech, though overall response magnitude did not differ significantly. Subtle right-temporal deactivations suggest anticipatory processing preceding boundaries. These findings indicate that neural signatures of event segmentation persist despite background noise. Even when intelligibility declines, listeners maintain coherent mental models, highlighting event segmentation as a stable mechanism supporting real-world comprehension.
A51
No evidence of a general athlete advantage across the networks of attention
Ralph Redden
x2021avz@stfx.ca
▶ Show abstract
Effective engagement with the world depends on our ability to detect, interpret, and prioritize meaningful information while resisting interference from competing stimuli. In sport, the demands on cognition are amplified: athletes must rapidly detect and interpret cues, recall strategic plans, make split-second decisions, and adapt to continuously evolving situations. Consequently, exceptional athletic performance reflects not only physical prowess but also the integration of advanced perceptual-cognitive skills that enable precise, adaptive responses under pressure. Given the complex attentional demands of sport, the present study aimed to investigate whether a general varsity athlete sample would demonstrate enhanced efficiency across the three attentional networks (alerting, orienting, and executive control) compared to normative undergraduate controls. Using a gamified version of the Attention Network Test (called “AttentionTrip”), we found faster but more error prone responses to targets for the athlete group relative to the control group. Pertaining to our primary research question, though, there was no difference in efficiency for athletes as compared to controls on any of the three networks, suggesting that athletes do not differ in general attentional processing. Rather, previously reported differences between high-performance athletes and non-athletes may be attributed to domain-specific factors associated with the nuanced cognitive dynamics of their sport.
A52
Posterior Intraparietal Sulcus activity during a head unrestrained, memory guided reach task
Brando Sheldrick
brandocatt@hotmail.com
▶ Show abstract
The purpose of the current study is to investigate how local field potential (LFP) activity along the intraparietal sulcus (IPs) is modulated by visual landmarks before and during reaches to remembered visual targets. We recorded LFPs using a 32-channel neural probe in one female Rhesus monkey. A landmark (four identical dots positioned at the vertices of a virtual square) was displayed at one of fifteen locations on a touch screen. A visual target then appeared, either within or outside of the landmark square, followed by a visual mask. After the mask disappeared, the landmark reappeared either at the same location (stable landmark condition) or shifted by 8 degrees in one of eight directions (landmark s22hift condition). Gaze and head position were allowed to move freely, and the animal was rewarded for reaching within 4.7 cm of the target. In the 'no-landmark' control trials, the procedure was the same, but the landmark was not presented. Preliminary analysis of the current IPS LFP dataset suggests a decrease in beta band power that is time-locked to the reaching movement. Additionally, there appears to be an increase in delta and theta band power before reward in the landmark task conditions compared to no-landmark controls.
A53
Predicting Human Viewpoint Preference with Deep Neural Network Embeddings
Athanasios Bourganos
Athan.bourganos@mail.utoronto.ca
▶ Show abstract
Humans prefer some object views over others, but the representational basis of these judgments remains unclear. We collected human ratings of viewpoint canonicity across multiple perspectives of the same objects to obtain average canonicity measurements around the objects. We show that deep image models accurately predict human canonicity judgments when anchored to the most canonical perspective (the view with the highest average canonicity rating). Specifically, latent space distances between the preferred view and other views closely track human ratings. These results suggest that the statistical distributions of object views in large image datasets may reflect human canonicity bias.
A54
Randomly Titled but Better Remembered: How Arbitrary Titles Shapes Art Memorability
Lauren Nordstrom
lauren.nordstrom@mail.mcgill.ca
▶ Show abstract
Art is often considered personal and highly subjective. Yet, research has shown that external context can shape the way art is perceived, appreciated and later remembered. One external factor thought to affect art appreciation and memorability is the presence of a title, which can alter the way a viewer attends to and interprets an art piece. In the present study, we tested whether the presence of an arbitrarily assigned title could impact the encoding and later retention of visual art pieces. In two experiments, participants encoded art pieces present with or without a random title and then completed an old/new recognition test that included old and new art pieces. Recognition was significantly better for the art pieces with a title than without a title, but appreciation was unaffected. A third experiment included a rating of title-art congruency during encoding to illustrate that higher congruency led to stronger recognition memory. These findings illustrate how random external factors like title can alter the memorability and retention of art.
A55
Shape Selectivity for Tangible Objects and Images in Ventral and Dorsal Cortex
Jacqueline Walsh-Snow
u6071610@utah.edu
▶ Show abstract
Although the neural basis of object perception has been studied extensively, mostresearch has relied on artificial stimuli in the form of pictures or computer-rendered images, rather than real-world, tangible objects. This approach may overlook the possibility that the brain’s processes and organization are shaped by behaviorally relevant features of naturalistic objects and environments, including their potential for action. Here, we used fMRI to compare shape selectivity for tangible objects versus matched planar photographs of objects across the ventral and dorsal visual cortex. Using a parametric shape-scrambling paradigm, participants directly viewed 3D-printed everyday objects and matched photographs presented at varying levels of scrambling. We examined whether dorsal object-selective regions, given their established role in spatial processing and action guidance, show differential sensitivity to object format relative to ventral regions. Preliminary analyses suggest format-dependent variation in shape selectivity across ventral and dorsal regions. Ongoing analyses are refining these estimates. These findings underscore the importance of testing object representations using both tangible objects and images when characterizing shape selectivity in human cortex.
A56
Tempo-Evoked Arousal: Distortions of Time Perception in Event Memory
Mikaila Tombe
mikaila.tombe@mail.mcgill.ca
▶ Show abstract
Arousing events hold privilege within our memory; however, the precise nature of this is unclear.Two key components of temporal memory include the ability to successfully recall the sequence and duration of events, leading researchers to investigate how arousal impacts temporal memory during arousing and non-arousing events. In the current study we aimed to investigate arousal in relation to neural entrainment and internal clock theories within temporal memory. If arousal is a mechanism of neural entrainment, we should see distortions of temporal duration during high arousing events. We conducted a within-subjects behavioural experiment in which participants encoded four videos (neutral events) paired with highly arousing or non-arousing musical excerpts. After a short delay, participants were shown pairs of still images and asked to indicate the sequence (which image came first?) and duration (how much time between?). Temporal sequence did not differ between music tempo, although videos paired with fast music experienced a significant dilation of time. This maps onto neural entrainment theories as fast music increases physiological arousal, speeding up one's internal clock. We interpret these results as indicating tempo evoked arousal may not influence the sequence but rather the duration of temporal memory, as a function of neural entrainment.
A57
The effect of size and distance on neural responses to realistic 3D faces
Eva Deligiannis
edeligia@uwo.ca
▶ Show abstract
Binocular vision provides rich three-dimensional (3D) information about the visuospatial geometry of faces, along with their physical size and egocentric distance. Despite this, most neuroimaging studies present two-dimensional faces at fixed distances. Some face-selective neurons in macaque face patch AF encode physical face size, in contrast to the typical retinotopy found in early visual cortex; however, it remains unclear whether 3D visuospatial geometry (i.e., size and distance) impacts face processing in the human brain. We used functional MRI to measure brain activation while participants viewed realistic human avatars in stereoscopic 3D. Faces were rendered at three sizes and three distances, with some size–distance combinations matched for retinal angle, allowing us to dissociate neural responses to physical versus retinal size. Activation in posterior superior temporal sulcus (pSTS) was highest for faces that appeared larger or smaller than a typical face, even after contributions from retinal angle were partialled out. This response profile resembled macaque face patch AF, which is thought to be functionally similar to human pSTS. Surprisingly, higher-order visual depth areas (including V3A/B) showed limited sensitivity to size and distance. These results suggest that higher-order face areas may use depth information to encode a typical or expected face size.
A58
This is your brain on poetry: EEG patterns dissociate across frequency bands during poetry reading
Lina Khayyat
lina.khayyat@mcgill.ca
▶ Show abstract
Despite increasing interest in how attention is engaged by aesthetic experiences, its correlates and relationship to engagement with poetry remain understudied. Here we investigated the neural correlates of ‘deep attention,’ which we operationalized as an absorptive mental state elicited by engagement with poetry. Brain activity was recorded using a 32-channel EEG from 54 participants while they read four poems and four matching prose texts. Participants also rated each piece for liking, interest, and engagement, with poems rated significantly higher than prose on all metrics. Brain activity in the Alpha (8-12Hz), Beta (13-30Hz), and Theta (4-8Hz) frequency bands during each poem and prose epoch was examined using a sliding window FFT-based event-related spectral perturbation (ERSP) analysis. The results showed an anterior-posterior dissociation in Alpha event-related desynchronization (ERD) with frontal regions showing greater Alpha band ERD during the prose epochs, and posterior parietal and occipital regions showing greater Alpha band ERD during the poem epochs. For the Beta band, poetry epochs were marked with significantly greater occipital Beta ERD. There were no differences in the Theta band activity. These findings provide evidence in support of a specific pattern of whole-brain neural dynamics for engagement with poetry.
A59
Transdiagnostic Examination of Amygdala Structure and Anxiety Across Attention-Deficit/Hyperactivity Disorder, Autism Spectrum Disorder, Obsessive-Compulsive Disorder, and Typically Developing Children
Aditi Ilangovan
aditi.ilangovan@ucalgary.ca
▶ Show abstract
Comorbid anxiety is highly prevalent among youth with neurodevelopmental disorders (NDDs), including attention-deficit/hyperactivity disorder (ADHD), autism spectrum disorder (ASD), and obsessive-compulsive disorder (OCD), and is associated with poorer clinical outcomes. Despite this, neural markers of anxiety in NDDs remain unclear. Building on evidence implicating amygdala subnuclei in socioemotional processing, this study examined the right basolateral complex, central nucleus, and paralaminar nucleus as structural correlates of anxiety across ADHD, ASD, OCD, and typically developing (TD) youth. Magnetic resonance imaging (MRI) and parent-reported anxiety scores from the Revised Child Anxiety and Depression Scale (RCADS) were obtained from 330 participants (92 ADHD, 149 ASD, 56 OCD, 33 TD youth; M = 11.61 years, SD = 2.70) in the Province of Ontario Neurodevelopmental Disorders Network. General linear models assessed group and transdiagnostic effects, adjusting for age, sex, total cerebral volume, and MRI site. Both generalized and total anxiety scores differed significantly across groups (p < .001). A nominal group effect was observed for right BLA volume (p = .040). Larger right paralaminar volume was nominally associated with lower generalized anxiety (p = .024) but did not differ across groups. These findings suggest potential relationships between distinct amygdala subnuclei and anxiety severity across NDDs.
A60
The Lasagna Model: a layered approach to sustainable teaching methods
Lindsay Richardson
LindsayRichardson@cunet.carleton.ca
▶ Show abstract
Generative artificial intelligence (GenAI) tools such as ChatGPT are rapidly becoming integrated into how students study and complete academic work. While some evidence suggests AI can support learning, other findings indicate that reliance on AI-generated answers may promote superficial engagement with course material. The present study examined whether the manner in which students engage with GenAI tools influences learning outcomes. Undergraduate participants were randomly assigned to one of three conditions: No AI, AI Instructed, or Unrestricted AI. Participants studied a neuropsychology passage and completed an initial comprehension test under their assigned condition. They then completed a surprise retention test without AI access. The dependent measure was percentage of correct responses on each test. This mixed-factorial design allowed direct comparison between AI-assisted performance and unaided retention. Unrestricted AI use is expected to produce the highest comprehension scores during AI-assisted testing but the lowest retention once AI access is removed. In contrast, scaffolded AI use is expected to produce stronger retention outcomes. These findings will help clarify when AI supports learning and when it may instead encourage unwanted cognitive offloading.
A61
Adaptive norm-based encoding of within-person variability
Yaren Koca
yarennkoca@gmail.com
▶ Show abstract
According to the face space theory, faces are encoded in a multidimensional space in memory based on their relative similarity. The exact dimensions of the face space are unknown, but it is assumed that they serve to differentiate faces. Recent research, however, showed that each person’s face varies uniquely under different viewing conditions such as lighting, angle, or emotional expression. Therefore, it is likely that each face is represented in the face space based on idiosyncratic set of dimensions that best serve to recognize them. To understand the architecture of such identity-specific face spaces, we created anti-faces of familiar faces by distorting either their identity-specific average or a gender-matched average of many identities. Observers (N = 42) were asked to identify different images of these familiar faces before and after adapting to the anti-faces. We found that the magnitude of the adaptation effect was larger for anti-faces created through distorting the identity-specific average. These findings show that different images of familiar faces are encoded in reference to an identity-specific norm in the face space.
A62
Affective and Behavioural Factors that Influence Conversational Memory in a Bilingual Context
Colin Hoare
choar057@uottawa.ca
▶ Show abstract
In bilingual environments across Canada, individuals often engage in dynamic social interactions in their second language (L2). Little research to date has examined the association between memory for the information conveyed in bilingual or L2 conversations and the characteristics of the individual (e.g., linguistic anxiety) and of the conversation (e.g., the use of gesture to support communication). The current research examines the link between memory for conversations and various cognitive and behavioural features using an ecologically valid design that simulates daily interactions. Sixty-seven English/French bilingual young adults (17-32 years old) engaged in 15 minutes of conversation (five minutes per language condition, including English, French, Bilingual). Participants then completed various measures of language history and anxiety, and conversational recall. Preliminary analyses showed similar state anxiety across language conditions. However, trait linguistic anxiety was higher in L2 than L1 and was positively associated with post-conversation state anxiety. Subsequent analyses will explore factors that may impair or enhance recall in rich communicative contexts, including linguistic anxiety and gestures. Whereas anxiety may impair the recall of L2 or bilingual conversations, we expect gestures conveying meaningful information (i.e., iconic gestures) to favour retention.
A63
CBM-DF: Directed forgetting of negative information reduces state levels of depressive symptoms
Evan Curtis
evan.curtis@boothuc.ca
▶ Show abstract
It is well established that people experiencing depression are prone to a variety of fundamental cognitive biases towards negative information that are reliably observable even in basic laboratory tasks. In response to this empirical pattern, researchers have explored Cognitive Bias Modification (CBM) as a potential tool for alleviating depressive symptoms. CBM, at a general level, involves adapting an experimental design such that cognitive processing is forced to shift either towards positive information, away from negative information, or both. Findings have been promising in cognitive processes such as attention and interpretation, but progress in CBM for memory processing has been both lacking and generally unsuccessful. We first suggest potential reasons for the low success rates in recent attempts to modify memory biases in depression and then report a pilot study examining the efficacy of directed forgetting of negative information in reducing state levels of negative mood and depressive symptoms. The results show a small but reliable decrease in state levels of negative mood. We argue that the results warrant further research into whether directed forgetting can be effectively leveraged as an additional tool in the alleviation of depressive symptoms.
A64
Choosing to forget emotional words.
Brendan Redmond
bmr785@mun.ca
▶ Show abstract
The production effect refers to the finding that produced items (e.g., those read aloud) are better remembered than items that were not produced (e.g., those read silently), although this difference is smaller for between- compared to within-subject designs. One perspective is that participants use memory of having produced a given item to guide recognition judgements at test (the distinctiveness heuristic), giving rise to the performance benefit. It is theorized that when participants only read words e aloud (as in between-subjects studies), they are less likely to use this strategy. The current study examined whether providing participants with instructions to use a distinctiveness heuristic before completing a recognition test increased the production effect in a between-subjects design. Participants were assigned to one of three conditions: (a) Reading silently, (b) reading aloud , or (c) reading aloud with the addition of specific instructions to think about whether each item at test had been read aloud as a useful strategy to differentiate between “old” and “new”. Participants then completed a recognition test. Although we observed higher memory accuracy for produced items compared to non-produced item, our current findings suggest minimal to no additional effect when participants receive instructions to use a distinctiveness heuristic.
A65
Chronotypes and intentional forgetting: Evidence from evening types
Jayden Roberts
jayden.roberts@unb.ca
▶ Show abstract
Intentional forgetting refers to the active process of memory suppression; sleep affects this ability. Chronotype reflects an individual’s inclination for alertness at certain times of day (e.g., morning or evening types). While a "synchrony effect” has been found for an array of cognitive tasks—where superior performance is observed when task completion time aligns with chronotype (e.g., evening type tested in the evening)—it is unknown if this effect impacts the ability to intentionally forget. The current work sought to investigate how the ability to actively suppress unwanted information is influenced by time of day. Participants completed the Morning-Eveningness Questionnaire to determine chronotype. Individuals classified as evening chronotypes then completed a Think/No-Think (TNT) task. In the TNT task, a list of word pairs (e.g., HAPPY-MEMORY) is learned. Some items are then repeatedly recalled, leading to increased memory, while others are actively suppressed, leading to the intentional forgetting of those words. Participants completed the TNT task in either their optimal (evening) or suboptimal (morning) arousal period of the day. Of interest is how chronotype and time of testing relate to memory suppression. Results are discussed in terms of the impact of synchrony effects and their implications for inhibition and intentional forgetting.
A66
Congruent Interleaved Words Generate Substantial False Memory Effect
Harris Miller
milleh11@mcmaster.ca
▶ Show abstract
In studies of recognition memory, fluent processing of ‘new’ test items can lead to false recognition of those items as ‘old’. The Jacoby-Whitehouse effect used subliminal presentation of a context word prior to a matching test word to produce this effect. A key property of the Jacoby-Whitehouse effect is that it does not occur when the context word is presented for a longer duration, presumably because participants can then identify the correct source of increased processing fluency. We examined whether a similar effect might occur with two simultaneous and interleaved test word stimuli. Indeed, false alarms were higher when the two test words were the same than when they were different. In a following experiment, participants judged whether the two test words were same or different prior to producing a recognition judgement. We predicted that directing attention to the source of fluency would eliminate the false recognition effect. However, the false recognition effect was equally robust when participants did and did not perform the additional same/different discrimination task. These results demonstrate that processing ease is a powerful cue to prior experience in this task context, biasing recognition judgments even when participants were directed to the true source of the fluency.
A67
Dissociating Valence and Arousal in Autobiographical Memory
Melissa Meade
stielker@uwo.ca
▶ Show abstract
Emotional valence (negative-positive emotion) and arousal (emotional intensity) are frequently conflated in the autobiographical memory literature, limiting understanding of how these dimensions interact in autobiographical memory. This is important in the context of emotional disorders like depression, with reduced physiological responsiveness to arousal and bias towards negative information. In an online survey, participants completed the Minimal Instruction Autobiographical Memory test (MIN-AMT): 12 cue words (e.g., ‘excited’, ‘relaxed’, ‘nervous’, ‘sad’) were used to prompt associated autobiographical memories. Cue words were selected to reflect dimensions of high- and low-arousal, and negative- and positive-valence, with 3 cues per valence-arousal combination. Effects of arousal and valence on emotional sentiment in recall and memory specificity were examined. Sentiment analysis using VADER (Hutto et al., 2014) confirmed that recall elicited by negative cues had more negative sentiment, and positive cues led to more positive sentiment. Participants with high Beck Depression Inventory (BDI-II) scores showed more negative sentiment for negative cues than low BDI-II participants, whereas positive-sentiment memories were similar across groups. Memory specificity was greater for high- than low-arousal cues, and for positive- than negative-valence. Memory specificity was lowest for low-arousal, positive-valence cues, regardless of BDI-II score, suggesting reduced episodic richness in memory for peaceful experiences.
A68
Drawing in class: Does drawing benefit memory for lecture content?
Katarina Jovanovic
kjovano4@uwo.ca
▶ Show abstract
Drawing pictures of to-be-remembered words and definitions has been shown to benefit later retrieval, more than other mnemonic strategies such as verbatim writing. To date, work examining the drawing effect in an educational context has involved relatively controlled presentation of stimuli where participants memorized isolated words and their definitions. Further work is needed to determine the utility of drawing in enhancing memory for lecture content in the classroom. The goal of the present study was to investigate whether the drawing effect persists when participants encode more complex, ecologically valid stimuli, similar to typical undergraduate lectures. In a within-participants design, 50 participants saw two 15-minute lectures for which they took notes by either writing or drawing. After each lecture, participants completed free recall and cued recognition measures. In both cases, participants showed better memory performance when taking notes by writing, relative to drawing. These findings suggest that while drawing may provide benefits for remembering individual terms and definitions, the drawing effect may not extend to a broader educational context, such as note-taking during a lecture.
A69
Effects of Event Tagging on Memory and Eye-Movements
Joseph Merante
ff22ge@brocku.ca
▶ Show abstract
Event segmentation theory states that we take a continuous stream of activity and parse it into multiple events, separated by event boundaries. As we age, our ability to segment events seems to decline, as older adults sometimes show less agreement about when boundaries occur and less distinction between successive events in long-term memory. To remedy this, we propose a novel intervention, “Event Tagging”, which entails pausing a video at event boundaries and asking participants to provide key words related to the scene they watched. We had younger (N = 37) and older (N = 36) adults watch an edited episode of BBC’s Sherlock continuously or with occasional pauses between events to allow for event tagging while eye movements were recorded. Contrary to our previous work, tagging did not improve older adults’ cued recall relative to continuous viewing, but there was a trend towards improvement in younger adults. In terms of eye movement synchrony, we did not observe any age differences in gaze similarity. However, gaze similarity did differ across encoding conditions, with participants showing slightly more synchrony in the tagging condition, but this differed by block. This suggests that under certain conditions tagging can impact memory performance and our eye-movements.
A70
Emotion modulates directed forgetting but not attentional withdrawal
Tracy Taylor
► Show abstract
Across six experiments, we determined whether the emotional study words modulate the attentional changes that occur after instructions to Remember and Forget. Using rapid serial visual presentation (RSVP), neutral and emotional study words were embedded in a stream of nonsense strings, each followed immediately by an instruction to Remember or Forget, and then by a probe string at one of 7 post-instruction positions. Following all study trials, participants completed a yes-no recognition test. Probe identification revealed an attentional blink that recovered sooner following Forget instructions than following Remember instructions, with no discernible influence of negative or positive emotion. Recognition memory, by contrast, showed that directed forgetting effects were modulated by emotion. Although there was no consistent pattern for positive words, negative words showed a smaller directed forgetting effect than neutral words. No-memory and no-probe controls confirmed that the memory effects required a memory intention and were not due to dual-task interference. The results indicate that the intention to forget withdraws limited-capacity attentional resources from unwanted word processing, without regard to emotional valence; effects of emotion on memory outcomes operate independently of this withdrawal.
A71
Episodic Memory at Midlife: The Influence of Perimenopause
Millie R. Bhaskara
millie.bhaskara@torontomu.ca
▶ Show abstract
Episodic memory begins to decline at midlife (age range: 40-65), the period when most female adults transition from premenopause to perimenopause and postmenopause. Past studies have shown reduced spatial context episodic memory performance in postmenopausal compared to premenopausal adults. However, there is little research examining differences in memory performance between premenopause and the transitional perimenopause phase. The current study addressed this literature gap and examined whether age within each menopause phase influences memory performance. Middle-aged, premenopausal (n = 42) and perimenopausal (n = 33) adults completed four easy and four hard runs of a spatial context memory task, where they were asked to remember the spatial locations of face stimuli. Participants were later asked to recall the spatial locations of previously seen (source memory) and novel stimuli (novelty detection). Outcome variables included memory accuracy, reaction time, and variability in these measures across task runs. Linear mixed-effects analyses demonstrated phase differences in source memory performance, such that perimenopausal participants made less accurate responses on easy runs and slower responses across easy and hard runs. There was no main effect of age on memory performance in perimenopausal participants, suggesting that there may be other factors that influence episodic memory during this phase.
A72
Executive Functioning and Sleep Quality Across Cognitive Aging Profiles
Narissa Byers
nbyers1@unb.ca
▶ Show abstract
Both declines in executive functioning (e.g., inhibition, working memory) and sleep quality (e.g., awakenings, night-time wakefulness) are common in normative aging, with more pronounced impairments observed in non-normative aging, including mild cognitive impairment (MCI), a pre-dementia stage (Hof & Mobbs, 2009). However, despite the close relationship between executive functioning and sleep quality, and evidence that poor sleep quality is a risk factor for MCI, scant research has examined the relationship between these two factors in MCI. Of the limited extant work, findings have been inconsistent (Overton et al., 2025; Palmer et al., 2018). Thus, the current study investigated executive functioning and sleep quality in 32 older adults (60-80 years) with different cognitive aging profiles: healthy and probable MCI, determined via Montreal Cognitive Assessment (MoCA) scores. Executive functioning was assessed with an antisaccade task (inhibition, working memory) and Wisconsin Card Sorting Test (cognitive flexibility, set shifting). Sleep quality was assessed with the shortened Pittsburgh Sleep Quality Index. Regression analyses revealed that participants with higher MoCA scores exhibited better executive functioning, regardless of sleep quality. However, this relationship was stronger among those with better sleep quality, suggesting that sleep quality may play a critical role in maintaining cognitive health in older adults.
A73
Influence of Emotional Episodic Future Thinking on Delay Discounting of Rewards
Nardeen Yalda
nyalda@yorku.ca
▶ Show abstract
Decision-making often involves choosing between smaller, immediate outcomes and larger, delayed ones. Delay discounting (DD) reflects the tendency to devalue delayed outcomes and is associated with impulsivity and addiction. Episodic future thinking (EFT) of specific, personal events is shown to reduce DD for rewards in both younger and older adults. However, the role of imagined emotional content remains unclear, especially with aging, where the influence of EFT on DD is weaker. This weakened DD effect in aging is also paired with shifting affective and motivational processes, which may further influence the effect of emotional EFT on DD relative to younger adults. In the present study, 217 older and 220 younger adult participants imagined either positive or negative future events or completed a control task before making reward choices that varied by amount and delay. DD did not significantly differ between older and younger adults. Both positive and negative EFT significantly reduced DD relative to the control condition, and this effect was comparable across age groups. These findings indicate that EFT is effective at reducing DD in both younger and older adults, irrespective of emotional valence. These findings may be leveraged to support individuals who are at risk for suboptimal decision-making.
A74
Playing it Safe: The Role of Judgments of Learning in Prototype Abstraction
Anthony Cruz
acruz27@uwo.ca
▶ Show abstract
It is well-established that making judgments of learning (JOLs) while studying can influence word list and word pair memory, a phenomenon called JOL reactivity. However, the role of JOLs in category learning, which requires abstraction to novel stimuli, remains less well-explored. We examined how making JOLs during study impacts prototype abstraction, a task typically thought to rely on implicit learning. Undergraduates (N = 146) studied distorted exemplars of a novel category (“Blargs”) under three conditions: passive study (control), concurrent random number generation (RNG; dual-task control), or concurrent JOLs. They were then tested on their ability to discriminate prototypical Blargs and distortions from foils. The JOL group responded more slowly than the RNG group and was less likely than the RNG group to correctly identify the prototypical Blargs, suggesting increased evidence accumulation thresholds. Signal detection analyses indicated that both JOL and RNG groups had higher sensitivity (d’) than the passive group. These findings suggest that JOLs shift learners toward more deliberate, criterion-based categorization strategies. These results are contextualized within multiple-system theories of category learning, and connections to existing JOL reactivity literature are established.
A75
Rewriting the Past: The Role of Plausibility in Updating Self-Beliefs in Social Anxiety
Sara Toca
sara.toca@mail.mcgill.ca
▶ Show abstract
Maladaptive self-beliefs are central to many mental health conditions, including social anxiety. Autobiographical memory plays an important role in constructing and maintaining self-beliefs with recent work suggesting that modifying the content of specific memories can alter associated self-beliefs. However, the mechanisms through which modifying memory content can alter and update self-beliefs remain unclear, particularly in the context of mental health conditions. To address this gap, we tested whether modifying (i.e., rescripting) negative autobiographical memories with either plausible or implausible positive outcomes differentially updated negative self-beliefs in individuals with social anxiety. In a between-subjects experiment, participants described negative autobiographical memories, completed a series of self-belief measures, after which they were randomly assigned to rescript their memories with plausible or implausible positive outcomes. They provided plausibility ratings for the rescripted memories and completed the same self-belief measures collected prior to rescripting. While rescripting negative memories significantly reduced negative self-belief endorsement for both groups, this effect interacted with post-rescripting plausibility outcome ratings. Higher plausibility ratings predicted greater belief change in the plausible condition, but smaller belief change in the implausible condition. These findings show that modifying memories with plausible and implausible rescripting techniques support belief change through multiple mechanisms.
A76
Sex- and Menopause-Specific Hippocampal Subfield Volumes and Memory at Midlife: A Period of Change
Sara Ahmed
sara2.ahmed@torontomu.ca
▶ Show abstract
Midlife is when age-related episodic memory decline is first detectable. Midlife also coincides with when most females experience spontaneous menopause. Age-related episodic memory decline is associated with hippocampal volume reduction. Yet, it remains unclear if age, sex, and menopause affect hippocampal subfield volumes, and their associations with episodic memory, at midlife. Using high-resolution T2-weighted MRI, 132 cognitively unimpaired middle-aged adults (38 males, 49 premenopausal females, 45 postmenopausal females; aged 40–65 years) completed a face-location episodic memory task. Hippocampal subfields were segmented using a customized-automated pipeline and volumes were adjusted for intracranial volume. Regression analyses were used to examine sex- and menopause-specific effects on age-related differences in memory and subfield volumes, respectively, and whether subfields contributed to memory. Preliminary results indicated sex differences in episodic memory wherein females performed worse with age compared to males, which was driven by postmenopausal females. In postmenopausal females, age was negatively associated with both CA1 and subiculum volumes, though these age-related differences in volume did not contribute to their memory performance. Our findings may indicate that the relationship between hippocampal subfields and memory is not linear during midlife, such that differences in volume do not translate directly into differences in memory.
A77
Sounds of the Past: Music-Cued Autobiographical Recall in Older Adults
Nadya Drury
► Show abstract
Autobiographical memory declines with age and is further reduced in individuals with mild cognitive impairment. Music can cue vivid meaningful memories, making it a promising way to support retrieval in older adults. We examined whether there exists a relationship between self-efficacy—belief in one's ability to succeed—and the number of music-evoked autobiographical memories in older adults with varying cognitive status. Thirty participants listened to 15-second clips of popular late-1960s songs, presented as either unaltered songs or in modified format (lyric-only). Following each cue, participants were given one minute to verbally describe any autobiographical memory that was evoked. Self-efficacy was assessed using the General Self-Efficacy Scale. Unaltered song cues led to a greater number of reported autobiographical memories than lyric-only cues, with total number of memories recalled increasing with self-efficacy. Importantly, self-efficacy differentially increased the ability of song versus lyric cues to elicit memories. Although individuals with poorer cognitive status, measured using the Montreal Cognitive Assessment, recalled fewer memories overall, self-efficacy remained a significant predictor of memory. These findings indicate that music reliably supports autobiographical recall and that self-efficacy enhances its effectiveness.
A78
The effect of context on speech intelligibility and gist recall when listening to stories amidst competing speakers
Tara Raessi
traessi@research.baycrest.org
▶ Show abstract
Background: Context is observed to facilitate speech intelligibility performance, yet there is limited evidence on whether this applies within naturalistic listening environments. Further, it is unclear whether the facilitation extends to gist recall. This study investigated the facilitation and contrast of context in speech intelligibility and gistrecall against informational masking (i.e., amidst competing speakers).Methods: Young adults (N = 19) listened to ~45-second stories. Half of the stories were played originally (intact), while the other half had reduced context by shuffling the sentences within the story, excluding the final sentence. Speech was presented in quiet or amidst two competing speakers at a signal-to-noise ratio (SNR) of 0 and -6 dB. After each story, participants repeated the final sentence (intelligibility) and provided a full, detailed recall of the story (gist).Results: Across all SNR levels, context facilitated gist recall (p < .001) but not speech intelligibility (p = .876). The temporal order of gist recall for shuffled stories did not significantly align with either the original story sequence or the shuffled order, suggesting participants mentally realigned the story content.Conclusion: Context appeared to only facilitate gist recall, recognizing the potential limitations of narrative context for speech intelligibility.
A79
The effect of video presence on associative and source memory in a virtual socialization event
Sophie Bhaskara
sbhaskara@torontomu.ca
▶ Show abstract
During virtual socialization, video presence has been shown to support associative memory for name-occupation pairs in older and younger adults, and enhance source memory in older adults. Previous work from our lab tested whether video presence may also benefit memory for a video-independent source (i.e., font colour). Participants watched a pre-recorded Zoom meeting where speakers introduced themselves, half with video on and half with video off. Each name-occupation pair appeared next to the video in white or black font, signaling good vs. evil speaker characteristics. Associative memory was measured with an old-new recognition test and source memory was tested by identifying the font colour of pairs recognized as “OLD”. Video presence improved associative memory, but not source memory. To enhance encoding of to-be-remembered information, the current study overlapped the name-occupation pairs with the video and tested source memory for socially salient good vs. evil characteristics (represented by coloured frames around the video). We expect to replicate the benefit of video presence on associative memory, and the hypothesis for source memory is open. The findings will clarify whether video presence may support older and younger adults’ source memory for video-independent and socially important information.
A80
The influence of emotional irrelevant speech on serial order memory
Dana Murphy
pearmstrong632@my.nipissingu.ca
▶ Show abstract
In this study, participants attempted to remember the order of presentation of seven words while ignoring single words that were spoken in either a happy or angry emotional prosody. After the ordered presentation, words were presented in alphabetical order and participants put them back in their original order. In the changing state (CS) conditions, different words spoken in either a happy or angry prosody were presented as irrelevant speech (IS). In the steady state (SS) condition, a single word spoken in happy or angry prosody was presented as IS. In an oddball condition, a single happy/angry word was repeated while a word in the other prosody was presented approximately 1/3 through the list. While all CS and SS conditions resulted in poorer serial order memory, the happy CS condition appeared to have the largest influence on memory. Angry speech with a happy oddball did not affect memory, but an angry oddball proved to be disruptive to memory. Despite expectation, with many different distracting words, angry speech was not more distracting than happy speech. Where angry speech may be more distracting is when an angry oddball is mixed in with multiple repetitions of a single happy word.
A81
The Role of Cognition and Motivation in Lexical-Semantic Processing and Recognition Memory
Ella Goldman
egoldma5@uwo.ca
▶ Show abstract
Cognition and motivation are two newly proposed dimensions of abstract word meaning. Initial studies of these semantic dimensions revealed an uncommon pattern: words with high cognition ratings and words with high motivation ratings show lexical-semantic processing advantages but memory accuracy disadvantages. The purpose of the present study was to further understand the role of these variables in the retrieval of word meanings. Cognition and motivation are correlated, so we first examined whether they have independent behavioural effects, above and beyond the effects of other variables (frequency, length, age of acquisition, socialness, imageability, semantic diversity, valence). When cognition and motivation were entered in models simultaneously, results showed that both were related to faster lexical processing but only motivation was related to less accurate recognition memory. Second, we tested whether meaning similarity explains the lexical processing/memory dissociation observed for high cognition and motivation words. Meaning similarity was a significant predictor of recognition memory accuracy but not of lexical processing and did not change the pattern of effects for cognition and motivation. These results suggest that cognition and motivation are both likely to be underlying dimensions of word meaning and that motivation best captures the memory disadvantage observed for internal meanings.
A82
The roles of biological sex and cognitive reserve in shaping sleep-memory dynamics
Alana Brown
abrown@research.baycrest.org
▶ Show abstract
Background: Hippocampal structural integrity is critical for memory, yet it remains unclear whether sleep quality determines when and for whom hippocampal structure supports memory. Compared to males, females are at higher risk for Alzheimer's disease (AD) and sleep disturbances. We examined whether sleep mediated or moderated hippocampal volume-memory associations and whether biological sex and cognitive reserve proxies influenced these pathways. Methods: Data from COMPASS-ND and CIMA-Q were harmonized across cognitively unimpaired (CU, n=169), subjective cognitive impairment (SCI, n=222), and amnestic mild cognitive impairment (aMCI, n=380) groups. Principal component analysis of self-reported sleep variables yielded a principal component (PC1) reflecting overall sleep quality. Structural equation models tested whether PC1 mediated or moderated hippocampal volume-memory relationships, including sex as a moderator. When sex moderated relationships, linear models were used to assess whether cognitive reserve proxies moderated sleep-memory relationships. Results: Larger hippocampal volume was associated with better memory, but sleep did not mediate this relationship. Good sleep strengthened volume-memory associations, with a stronger effect in females compared to males. In females, better sleep was associated with poorer memory, whereas cognitive reserve buffered the negative effect of poor sleep on memory. Conclusions: Findings highlight sex- and cognitive reserve-dependent pathways through which sleep influences cognition.
A83
Uncalibrated: Strategy use, beliefs, and confidence when learning a new face
Rebekah L. Corpuz
rebekah.corpuz@uregina.ca
▶ Show abstract
Growing evidence suggests that when learning a new face, unsystematic variability (i.e., differences in expression, make-up, hairstyle) leads to more accurate recognition than both systematic variability (i.e., differences in viewpoint or lighting) and no variability (i.e., learning from a single photo). In the present study, we investigated whether subjects can intuit this benefit. Subjects learned three unfamiliar target identities with unsystematic, systematic, or no variability, then sorted new photos of the targets from photos of similar-looking distractors. Subjects rated how confident they were that they would recognize the target at pre-training, post-training, and post-test. Interestingly, face recognition accuracy did not differ across variability conditions. However, results suggest that subjects’ confidence judgements, both prospective and retrospective, differ across variability conditions, self-reported strategy use, and beliefs about their general memory ability. The results provide evidence that subjects’ beliefs about their face recognition abilities do not align with their performance.Keywords: face recognition, within-person variability, confidence, calibration
A84
Unsteady Lines, Unshaken Recall: Why Handedness and Production Quality Does Not Affect the Drawing Benefit
Karen Nayiga
kmnayiga@uwaterloo.ca
▶ Show abstract
Drawing a sketch, as a way to encode information, boosts memory compared to other techniques, such as reading or writing to-be-remembered information. Prior research suggests three key components are responsible for the drawing boost to memory: semantic elaboration, pictorial, and motoric processes. We examined whether fine motor control of hand movements, manipulated by using the dominant or non-dominant hand to draw or write words during encoding, influenced the magnitude of the memory benefit. Participants (n=65) were visually presented with a list of 40 concrete nouns, one at a time, and were asked either to write or to draw a picture representing each word, with cue-type intermixed. In separate blocks, writing and drawing were performed using either the dominant or non-dominant hand (within-subject). Drawing compared to writing at encoding led to higher recall, as did using the dominant compared to the non-dominant hand. Importantly, encoding strategy and hand type did not interact: the typical memory gain from drawing at encoding was preserved regardless of the hand used. Results suggest that the quality of productions does not contribute to the magnitude of the drawing benefit. Instead, simply planning the motor actions required to draw an image is sufficient to boost memory.
A85
Boundary Conditions of Imagery Rescripting: Dissociating Episodic Self-Images from Semantic Self-Schemas
Rotem Paz
► Show abstract
Negative autobiographical images derived from painful episodic memories are theorized to reinforce the maladaptive self-schemas that lie at the heart of many clinical disorders. In a preregistered nonclinical study, we examined whether memory updating provides a pathway to schema change in 311 participants who recalled a negative autobiographical memory and were randomized to one of two active memory rescripting conditions or a reflective control condition. Both rescripting conditions lowered image negativity (b = -18.51, p < .001; -20.99, p<.001) and increased image positivity (b= 25.21, p<0.001; b= 29.60, p<.001) relative to control, but – contrary to hypotheses – neither condition produced immediate changes in generalized self-schemas (p's > .05). Several interpretations may account for this discrepancy, including that updated images are encoded as event-specific "exceptions" rather than prediction errors that signal a revision rule, or that downstream schema change unfolds over a longer timeframe than measured in this study. While rescripted images might signal successful encoding of schema-incongruent representations, broader self-knowledge revision may require more time and repeated rehearsal to fully crystallize. These findings highlight that the neurocognitive mechanisms through which schema updating occurs remain poorly understood and suggest key boundary conditions for memory-based interventions warranting future investigation in clinical populations.
A86
A Model-Based Source Monitoring Explanation for Increased Unconscious Plagiarism in Response to Generative Elaboration
Steve Lindsay
slindsay@uvic.ca
▶ Show abstract
Unconscious plagiarism is a memory failure in which individuals misattribute the source of external information to themselves. Previous research suggests that the likelihood of unconscious plagiarism increases after generative elaboration (e.g., generating ways to improve an idea). Separating source memory and guessing with a cognitive modeling approach, the present research offers a process-oriented explanation for this phenomenon. We explored these issues in the context of advice-taking interactions, a high-risk setting for source misattribution. Participants (N = 250) were tested individually online. They generated and read health advice in turns. Half of the participants rated the advice after generation/reading, whereas others elaborated the advice. The behavioral results indicated the expected generation effect in recognition memory, which was confirmed by the modeling results. Whereas the behavioral results suggested also a generation effect in source memory, the modeling results revealed that the self-generation advantage in source identification performance originated from source guessing. Crucially, after elaboration source memory was worse for self-generated ideas than for read ideas, and participants exhibited compensatory source-guessing, leading to substantially more source guessing toward the self. Our results suggest that generative elaboration poses a risk to accurate credit assignment in advice taking.
A87
Interest and Mind Wandering
Om Patel
opatel2005@gmail.com
▶ Show abstract
Intuitively, readers should be more focused on a story if they are interested in it. In the present study, we investigated how this effect interacts with other factors that affect mind wandering. Participants read two different stories presented one word at a time at rates of 150, 200, or 300 words per minute. Afterwards, they completed a questionnaire on their level of interest. Generally, more mind wandering was reported as the presentation speed decreased, presumably because readers had more time for task-unrelated thoughts. However, this effect was much smaller when readers were interested in the story. In addition, mind wandering increased over the course of the text, presumably because readers became more distracted over time. However, again, readers who were interested in the story showed a relatively small decrease in task focus. Our conclusion is that text processing is mediated by interest, so that additional textual processing occurs with slower presentation and over time when readers are interested. This additional processing curtails mind wandering.
A88
Knowing What You Know and What You Don’t Know: How Online Health Information Shapes Perceived and Objective Knowledge of Polycystic Ovary Syndrome
Eleni Dubé-Zinatelli
edube024@uottawa.ca
▶ Show abstract
Health literacy encompasses four key processes involved in health-related decision making: acquiring, understanding, evaluating, and applying health information. In chronic conditions such as polycystic ovary syndrome (PCOS), these competencies shape how individuals interpret medical information, navigate healthcare systems, and engage in health behaviours, making health literacy an important determinant of health outcomes. The present study examined the role of PCOS literacy and its underlying competencies in predicting quality of life (QoL) among women with PCOS. A sample of 263 women with PCOS was recruited through online platforms, PCOS support groups, and personal networks. Participants completed an online survey assessing demographic and clinical characteristics, PCOS literacy, and PCOS-related QoL. Greater PCOS literacy was associated with better QoL (β = .385, t = 5.63, p < .001). Among the four literacy dimensions, acquisition (β = .279, t = 3.39, p < .001) and application (β = .178, t = 2.32, p = .021) were the strongest predictors of QoL, suggesting that the ability to access relevant information and apply it to self-management may be particularly important for well-being. Overall, these findings identify PCOS literacy as a potentially modifiable target for future interventions aimed at improving QoL in this population.
A89
Determining the threshold of curvature in architectural spaces
Chris Arrabito
chris.arrabito@mail.utoronto.ca
▶ Show abstract
People can judge whether something appears curvy or angular. However, computational algorithms that measure curvature do not always align with human judgements. We conducted a psychophysics study using images of 3D architectural spaces where foreground and/or background curvature was selectively manipulated, allowing us to examine independent and combined influences on perceived curvature. Using the method of constant stimuli, participants classified randomly selected images as "curvy" or "angular". Fitting psychometric functions determined thresholds at which a curved element was perceived as curvy. Psychometric functions revealed that when only foreground was manipulated, participants were sensitive to curvature manipulations. When only background was manipulated, participants were less sensitive to curvature manipulations. We found 80% of curvature perception depended on foreground curvature, whereas background drove the other 20%. These contributions were used to predict thresholds when both foreground and background are manipulated. Compared to human judgements, these predictions were accurate. Our results suggest foreground curvature plays a larger role in curvature perception of architectural spaces. These findings clarify how people process curvature in complex scenes and contribute to an understanding of how aesthetic properties shape experiences of architectural spaces. Additionally, our results will help align computational curvature models with human judgements.
A90
Education outperforms income as a predictor of public-health-related outcomes
Abhishek Dedhe
adityasingla122@gmail.com
▶ Show abstract
Resilient data infrastructures are critical for tackling global public health crises. Whereas medical performance (clinical outcomes, mortality rates, etc.) has been extensively studied in previous research, statistical performance (data reliability, health surveillance, etc.) remains relatively understudied. In this study, we investigated both a medical performance metric (COVID-19 death rate) and a statistical performance metric ("undercount factor" i.e., the ratio of excess to reported deaths) of 190+ countries across the COVID-19 pandemic (2020-21). We tested whether medical performance and statistical performance are correlated with Human Development Index (HDI) and its constituent components of income, education, and life expectancy. Our analyses showed that medical performance was inversely correlated with HDI, with more developed countries showing higher COVID-19 death rates (r = .50). However, statistical performance was more significantly correlated with HDI, with more developed countries showing lower undercount factors (r = -.71). Notably, education was a better predictor of both medical performance and statistical performance (r = .51 and r = -.59) compared to GDP per capita (r = .16 and r = -.39). These findings suggest that investment in human capital such as education rather than economic wealth alone may have better outcomes for accurate global health surveillance and pandemic reporting.
A91
Expressions: Real-time face and head tracking using iPhones and iPads
Jesse Pazdera
pazderaj@mcmaster.ca
▶ Show abstract
We developed Expressions, an iOS application and data processing pipeline that leverages the TrueDepth camera systems on iPhones and iPads to collect 60 Hz data on 51 low-level features of facial expressions, 3D head movements, and eye gaze direction of up to three individuals simultaneously. To test the feasibility and flexibility of this approach for studying both children and adults during screen-based tasks and naturalistic interactions, we deployed Expressions in two developmental studies. In Study 1, toddlers watched emotional videos on a computer that wirelessly streamed event markers to an iPhone running Expressions. This timestamping allowed us to time-lock and analyze the toddlers' micro-expressive responses to positive versus negative stimuli. In the second study, two wirelessly networked iPads captured simultaneous data from caregiver-infant dyads during naturalistic interactions. This dual-device setup allowed assessment of complex interaction dynamics without disrupting the interaction through the presence of attached motion-tracking markers and sensors. These studies demonstrate that consumer mobile devices can be a feasible and flexible platform for studying facial expressions. Because the tracking in Expressions runs entirely on-device without requiring internet access, we hope to empower accessible facial expression research, not only in labs and urban homes but also remote and rural communities.
A93
iTemplate2: An Open-Source Python Toolbox for Standardized Eye-Movement Analysis of Dynamic Face Stimuli Using Automatic Landmark Detection
Gabriel (Naiqi) Xiao
xiaon8@mcmaster.ca
▶ Show abstract
Research in perception, cognition, and communication increasingly relies on dynamic face stimuli, such as videos or live interactions, to improve ecological validity. Eye-tracking data collected during these naturalistic tasks helps researchers understand cognitive processes in real-world scenarios. However, the lack of accurate analysis pipelines for dynamic gaze data limits this work.Traditional workflows rely on static, manually drawn Areas of Interest (AOIs). When applied to moving faces, this approach requires extensive manual coding and often leads to spatial misclassification.We introduce iTemplate2, an automated, open-source toolbox with a graphic user interface that addresses this limitation. The software uses deep learning to detect facial landmarks and co-register gaze coordinates from unconstrained dynamic stimuli onto a standardized face template.This approach converts varied visual data into a unified analytical format. We demonstrate the toolbox’s utility across three studies: infant face-scanning, cross-cultural perception tasks, and live caregiver-infant interactions. iTemplate2 allows researchers to use naturalistic stimuli while maintaining both automation and measurement precision.
A94
Sensitivity of infrared optical technologies vary with skin tone
Katherine O. Mordi
katmordi@my.yorku.ca
▶ Show abstract
Melanin pigment determines skin tone, where higher levels of melanin are found in darker skin tones. Because melanin absorbs light, differences in skin pigment, particularly darker skin tones, may adversely affect the accuracy and sensitivity of infrared optical technologies. The objective of this study was to investigate how different skin tones influence performance of infrared devices, namely pulse oximeters. Participants’ skin tones were objectively measured using a reflectance spectrophotometer and skin tones were categorized from Type 1 (Very Light) to Type 6 (Dark) based on the Individual Typology Angle. Blood oxygen saturation (SpO2) was then measured non-invasively using five commercial infrared devices simultaneously (Masimo, ToronTek, Blue Echo, Apple Watch Series 10, and NIRSport2 fNIRS cerebral cap) at rest and during physical activity. Preliminary findings indicate that skin tone influences the sensitivity and accuracy of infrared optical measurements, with greater variability observed with darker skin tones. Measurements with the Masimo were most stable at rest and during decreased SpO2 levels with physical activity. These findings highlight the potential for skin tone bias in the performance of devices that rely on near-infrared light and underscore the need for improved design and calibration to ensure accurate measurements across diverse populations.
A95
The Generative Repository of Anagrams and Metrics from SUBTLEX
Ian Newman
ian.newman@usask.ca
▶ Show abstract
Solving anagrams involves rearranging a string of letters (e.g., “cheab”) to identify the solution word (e.g., “beach”). Structural features of the solution, like number of syllables, can influence difficulty. Anagram surface cues like pronounceability (e.g., “cheab” vs “hcaeb”) can impact both actual and perceived difficulty. Meta-reasoning is concerned with how cue-based judgments influence problem-solving and effort regulation (Ackerman & Thompson, 2017), making anagrams useful experimental stimuli. Myriad other relevant meta-reasoning cues are computable, such as orthographic neighbourhoods and letter frequencies. However, computation of these metrics is often non-trivial or time consuming. We present an anagram database, the Generative Repository of Anagrams and Metrics from SUBTLEX (GRAMS), and the associated grams R package (https://github.com/irnewman/grams) used to create it. The package simplifies computation of various metrics for use as model predictors and facilitates generating matched stimuli lists by manipulating or controlling for these metrics. By default, GRAMS is based on the SUBTLEX-UK corpus (van Heuven et al., 2014), well-established in psycholinguistic research. Notably, the package allows for alternative corpora and resources in its computations. Our analyses demonstrate that several metrics computed by the package are important to consider, predictive of anagram solving performance and of pronounceability judgments.
A96
Straining the mind or moving the body? Individual differences in effortful exertion while listening
Carson Rumble-Tricker
crumblet@uoguelph.ca
▶ Show abstract
Models of auditory functioning posit that physical behaviours (e.g., moving closer to a talker) and the effortful allocation of cognitive resources (e.g., listening effort) are adaptive responses to adverse auditory conditions. While each have costs, they can be used as needed to maximize hearing success. Here we examined how individual-difference factors in cognitive-affective engagement predict these different forms of effortful exertion. Prior work suggests that individuals more prone to boredom and spontaneous mind-wandering have greater subjective hearing difficulties (Crawford et al., 2023), potentially leading to a stronger motivation to compensate for degraded listening conditions in our study. While listening to stories with increasing background noise, participants (N = 107) could exert physical effort (repeated key-pressing) to reduce the noise in real time. After each story, they reported subjective effort and memory for the story content. Individual differences predicted both key-pressing and ratings of subjective effort: boredom proneness was associated with both higher key-pressing and greater subjective effort, whereas the propensity for spontaneous mind-wandering predicted less key-pressing but greater subjective effort. These results underscore how cognitive-affective traits shape effortful exertion during listening and suggest that considering these individual differences may inform strategies or interventions to support listeners in challenging auditory environments.
A97
Oral contraceptives and physical and cognitive effort while listening: An exploratory analysis
Alyssa C. Smith
asmit114@uoguelph.ca
▶ Show abstract
There is evidence that the hormonal shift associated with oral contraceptive (OC) use can involve neurological changes that may affect hearing and cognition. As such, we aimed to explore whether OC users and non-users differed in terms of the physical and cognitive effort exerted across two listening tasks. In Study 1, female participants (OC N = 50, non-OC N = 135) listened to six stories with increasing background noise. In Study 2, female participants (OC N = 10, non-OC N = 27) completed the Revised-Speech Perception in Noise (R-SPIN) task. Background noise could be reduced via key-pressing in Study 1 or by squeezing a hand dynamometer in Study 2. During both tasks, participants responded to thought probes assessing listening-effort and boredom. Our findings indicate that across both tasks and compared to non-users, OC users exert nominally greater physical effort (via key-pressing in Study 1 and via the force exerted on the dynamometer in Study 2) to reduce background noise, and report greater boredom and effort during the task. This work suggests that OC use may affect the physical and cognitive effort exerted during listening tasks and indicates more work is needed to better understand the role of OCs in listening-effort.
A98
The Temporal Organization of Free Recall in the Real-World: The Temporal Contiguity Effect and Time–Space Interactions
Casey Aurin
▶ Show abstract
Past research reveals consistent patterns in how memories are freely recalled (Kahana, 2020; Healey et al., 2019). For instance, the temporal contiguity effect is the tendency for events experienced close together in time to be recalled consecutively. However, most studies utilize lab-based word lists, which do not capture the spatiotemporal context in which real-world events are experienced. The present study adds to the small number of studies that have examined the temporal contiguity effect in a naturalistic setting (Diamond & Levine, 2020) by examining how memories encoded during a 1-hour guided tour of a living history museum are recalled. Young adults (n = 85; age: M = 19, SD = 0.99) walked through a historic town, took photographs, and learned information from the tour guide, as they explored various historical buildings. After a five-day delay, participants were asked to recall all the locations they visited. Results revealed temporal clustering: participants were more likely to consecutively recall locations experienced close together. We also observed a forward asymmetry effect: locations were more often recalled in the forward order than in reverse. These findings advance our understanding of how real-world memories are organized and provide insight into the processes supporting spatiotemporal memory structure.
B01
Anticipated emotions and loss aversion in intertemporal risky choice
Avery Bernardin
avery.bernardin@uwaterloo.ca
▶ Show abstract
Economic choices often involve both delay and risk. While people typically prefer to expedite gains and postpone losses, the opposite pattern—delaying gains and hastening losses—has sometimes been observed. This preference reversal has been linked to anticipated emotions, with dread (negative affect) and savouring (positive affect) influencing the valuation of delayed outcomes. In this experiment, we investigate the effect of anticipated emotions on intertemporal risky choice. Specifically, we test whether dread, savouring, and vividness predict changes in loss aversion over time. While choosing between hypothetical 50/50 mixed lotteries, participants displayed significantly greater loss aversion when the resolution of uncertainty was delayed by six months, compared to with no delay. This difference was observed across lotteries with potential gains of $100, $500, and $1000. Further, dread—but not savouring—predicted larger changes in loss aversion for delayed lotteries, over and above discounting of sure amounts. These findings indicate that people are more loss averse for delayed outcomes in part due to anticipated emotions, with dread exerting more influence than savouring.
B02
Beyond Introspection: Inferring Decision Weights from Others’ Multi-Attribute Choices
Trent N. Cash
tcash@uwaterloo.ca
▶ Show abstract
Metacognitive judgments are informed by both privileged cues and observable cues. Across three studies implementing a variant of the learner-observer-judge paradigm, we explore the extent to which decision makers and observers can leverage these cues to infer the processes underlying a decision maker’s multi-attribute choice behavior. In Study 1, we demonstrate that decision makers (n = 221) more accurately estimate the weights they applied to each attribute in a multi-attribute choice task than observers (n = 220) who studied their choices. However, observers’ estimates were still weakly correlated with decision makers’ true weights, suggesting that reasonably accurate inferences can be made without access to privileged metacognitive cues. In Study 2, we demonstrate that the relative (but not absolute) accuracy of observers’ (n = 218) inferences decreases when the context of the decision is obscured, suggesting that domain-specific beliefs slightly improve metacognitive inferences. In Study 3 (n = 662), we replicate these findings and investigate differences in subjective experiences (e.g., motivation) across conditions. Together, our findings suggest that privileged metacognitive cues improve the accuracy of decision makers’ metacognitive knowledge, but that observers can make reasonably accurate inferences about decision makers’ choice behavior, particularly when they combine statistical monitoring with domain-specific beliefs.
B03
How We Think We Know Ourselves: Investigating Meta-meta Knowledge in Decision Making
Trent N. Cash
tcash@uwaterloo.ca
▶ Show abstract
Metacognitive judgments are informed by both privileged cues and observable cues. Across three studies implementing a variant of the learner-observer-judge paradigm, we explore the extent to which decision makers and observers can leverage these cues to infer the processes underlying a decision maker’s multi-attribute choice behavior. In Study 1, we demonstrate that decision makers (n = 221) more accurately estimate the weights they applied to each attribute in a multi-attribute choice task than observers (n = 220) who studied their choices. However, observers’ estimates were still weakly correlated with decision makers’ true weights, suggesting that reasonably accurate inferences can be made without access to privileged metacognitive cues. In Study 2, we demonstrate that the relative (but not absolute) accuracy of observers’ (n = 218) inferences decreases when the context of the decision is obscured, suggesting that domain-specific beliefs slightly improve metacognitive inferences. In Study 3 (n = 662), we replicate these findings and investigate differences in subjective experiences (e.g., motivation) across conditions. Together, our findings suggest that privileged metacognitive cues improve the accuracy of decision makers’ metacognitive knowledge, but that observers can make reasonably accurate inferences about decision makers’ choice behavior, particularly when they combine statistical monitoring with domain-specific beliefs.
B04
Pain catastrophizing and executive dysfunction: Preliminary findings from a chronic mTBI sample
Mary-Jo Daher
mary.daher@torontomu.ca
▶ Show abstract
Background: Cognitive impairment and pain are common following mild traumatic brain injury (mTBI), yet whether pain characteristics shape cognitive deficits remains unclear. Catastrophizing, involving rumination and magnification of pain, may compete with executive processes. This study examines whether different pain dimensions differentially predict cognitive performance after mTBI, focusing on catastrophizing as a risk factor. Methods: This retrospective study includes patients with mTBI-related chronic pain (N=131; mean 24.4±20.0 months since accident) who completed neuropsychological assessment. Cognitive composites spanned processing speed, executive functioning, episodic memory, and attention/working memory. Pain dimensions were assessed via pain intensity ratings and the Pain Catastrophizing Scale (PCS), Pain Disability Index, and Hospital Anxiety and Depression Scale. 69% of participants exceeded the clinical catastrophizing threshold (PCS≥30).Results: Cognitive impairment was observed across domains (z=-0.65 to -1.11), with episodic memory most affected. Catastrophizing was related to executive functioning (r=-.50, p=.002); no other pain dimension was related to cognitive outcomes. Regression models were significant for processing speed (R²=.46, p=.016) and attention/working memory (R²=.46, p=.030), driven by pain medication use (β=-.46, p=.006) and time since accident (β=.47, p=.017).Conclusions: Preliminary findings suggest a relationship between catastrophizing and impaired executive functioning, consistent with attentional resource competition.
B05
Curiosity and prior knowledge shape naturalistic information-seeking
Michelle Hirsch
mhirsch@yorku.ca
▶ Show abstract
Curiosity is crucial to well-being and acts as a key driver of information-seeking behaviour. Such behaviour can be more exploratory (i.e., seeking new information) or exploitative (i.e., seeking information within known territory). Older adults possess more robust prior knowledge and are known to rely on it more. Hence, we investigated whether age differences exist in naturalistic information-seeking using an information-foraging task in which participants freely browsed Wikipedia for 15 minutes per day over three consecutive days. After collecting individual browsing data, the text from each URL was transformed into numerical vectors (via tf-idf), and the semantic similarity between URLs was estimated using cosine similarity. For each visited URL, participants indicated their curiosity and prior knowledge, perceived similarity between pages, and search motivations. Analyses characterized moment-to-moment instances of exploitation versus exploration based on their relative semantic similarity and associated ratings. Our work presents a novel approach to understanding manifestations of curiosity in real-world information-seeking.
B06
Delineating the impact of early-life adversity on impulsivity
Anthony Chirila
ykambari@yorku.ca
▶ Show abstract
Early-life adversity (ELA), defined as exposure to adverse environments and experiences during childhood or adolescence, is associated with increased engagement in health risk behaviours. Emerging research suggests that ELA may disrupt the development of neurocognitive systems involved in behavioural regulation, thus contributing to impulsive and maladaptive behaviours. Given the conceptual complexity of impulsivity, however, the specific nature of the relationship between ELA and impulsivity remains to be fully explored. Therefore, this study examines the relationship between ELA and the construct of impulsivity across three empirically delineated categories: impulsive personality, impulsive choice, and impulsive behaviour. Data will be collected from community dwelling adult participants who self-identify as having experienced a stressful childhood. Participants will be asked to complete a brief survey evaluating ELA and impulsive personality traits, as well as standardized behavioural tasks assessing impulsive choice and impulsive behaviour. Using multivariate multiple regression, we will then explore the relationship between the severity of ELA and domains of impulsivity (i.e., personality, choice, and behaviour), with age as a covariate. Results will be discussed in the context of clarifying the role of ELA in the development and maintenance of impulsivity across its multidimensional presentation.
B07
Developmental Changes in the Symbolic Distance Effect During Double-Digit Comparison in Grade 1 and 2 Students
Hiva Bagheri Samghabadi
► Show abstract
Symbolic magnitude processing predicts early arithmetic achievement, and difficulties are associated with poorer long-term academic, financial, and health outcomes. The symbolic distance effect, whereby numerical comparisons are performed more accurately when the numerical distance between values is larger, is well documented in single-digit contexts. However, comparatively little research has examined how they emerge in double-digit comparison during early schooling, when symbolic processing is consolidating. The present study replicated and extended symbolic distance findings by examining cross-sectional differences in a timed double-digit number comparison task in Grade 1 (N = 581) and Grade 2 (N = 503) students. Data were drawn from a large-scale digital numeracy screener administered in classroom settings. Trial-level accuracy was analyzed using linear mixed-effects models, with additional models controlling for response pace to distinguish symbolic processing from general task engagement. Results indicated no symbolic distance effect in Grade 1, where accuracy was primarily predicted by response pace. In contrast, Grade 2 students demonstrated higher accuracy on large-distance trials, and this effect remained significant after accounting for response pace, which positively predicted accuracy across both grades. These findings suggest increasing stability in symbolic comparison performance across early elementary school.
B08
Effects of amnestic mild cognitive impairment on cardiac manipulations of familiarity and interoception
Hannah Del Gatto
hdelgatto@research.baycrest.org
▶ Show abstract
Familiarity can be both mentally experienced and physically embodied and is impaired in amnestic mild cognitive impairment (aMCI). Research has linked familiarity to cardiovascular baroreceptor signaling during systole. The present study uses a frequency judgment task, which includes a cardiac phase manipulation that synchronizes the presentation of stimuli to either systole or diastole, to determine whether familiarity is higher during systole than diastole. Older adults with aMCI show impaired familiarity overall compared to healthy controls, and do not show the predicted higher familiarity during systole than diastole. Interoceptive measures and structural magnetic resonance imaging of the perirhinal cortex are also collected to examine their relationship with autonomic nervous system (ANS) activity in supporting familiarity. Overall, these findings suggest that ANS coupling with familiarity is distinctly different in older adults with aMCI.
B09
I Can’t Look Away: How Attentional Biases Differ Across Neurodiversity and Sensory Environments in Children
Adrianna Molenaar
mole5490@mylaurier.ca
▶ Show abstract
Attentional biases, or the tendency to allocate more attention to emotional (positive and/or negative) compared to neutral stimuli, have been demonstrated across a variety of stimulus types. Additionally, attentional bias often differs in neurodivergent (e.g., Autism, Attention Deficit Hyperactivity Disorder; ADHD) individuals compared to neurotypical. Neurodivergent individuals also often show sensory processing differences, viewing high sensory environments as more overwhelming than their neurotypical peers. It is currently unknown if/how attentional biases differ across sensory environments, and how this relates to neurodiversity. Thirty-two school-age children; six autistic, five ADHD, and twenty-one neurotypicals, completed a dot-probe task in which they viewed pairs of positive-neutral, and negative-neutral images, in virtual reality. One sensory environment was a busy cityscape (high sensory) and the other was an empty (low sensory) environment. Results show significant differences in positive and negative bias scores between diagnoses in the low sensory condition, F(1,25) = 4.94, p = .035, η² = .165, but not the high sensory condition, F(1,25) = 0.30, p = .586, η² = .012. Overall, this research suggests there may be a role of diagnoses only when there is a low level of sensory information, with high sensory environments influencing children’s attentional biases more equally across diagnoses.
B10
Parental Math Self-Efficacy and Child Math Self-Efficacy: The Moderating Role of Homework Behaviors
Diya Kamineni
dkami024@uottawa.ca
▶ Show abstract
Parents’ math attitudes relate to their children's math attitudes and math learning. However, little is known about the mechanisms underlying these relations. We examined: (1) whether parents’ math self-efficacy relates to their child’s math self-efficacy and (2) whether parental behaviours during math homework influence this relation when parents’ math skills are controlled. Parent-child dyads (children aged 7-10 years; N = 104 dyads) independently completed self-reported measures of math self-efficacy and a test of their knowledge of equivalence (i.e., balancing equations). The dyads then participated in a simulated math-homework task that involved solving equivalence problems together. The quality of the parent-child interaction was coded for parental effort–how hard the parent tried to help their child. Analyses showed that parents with lower math self-efficacy exerted less effort during the homework task. Importantly, parental self-efficacy was positively related to child self-efficacy only when parental effort was low. Together, these findings suggest that parents who are less confident in their math skills and put less effort into homework helping have children with lower math self-efficacy. As such, homework-helping behaviours may influence the link between parents’ math attitudes and their children's math attitudes, which has important implications for supporting children's math achievement.
B12
The Price of Hierarchical Reasoning
Abhishek Dedhe
ridhib2422@gmail.com
▶ Show abstract
Despite generative AI’s (GenAI) ubiquity, its "black-box" nature obscures the reasoning demands governing its behavior. We propose a unified setup to compare reasoning across humans (children and adults), animals, and machines using an Artificial Grammar Learning (AGL) task (Ferrigno et al., 2020). This task serves as a "common currency" for evaluating hierarchical sequencing and abstract pattern processing.To tease apart behavior from underlying processes, we implement cognitive algorithms using a Language of Thought (LoT) with primitive functions and formal data structures (e.g., stacks or queues). We treat Chain-of-Thought (CoT) explanations as windows into generative mechanisms, applying program synthesis to map these informal descriptions into formal programs within a domain-specific LoT.By implementing these programs, we analyze them through a suite of complexity measures, including Minimum Description Length (MDL), working space requirements, and Kolmogorov complexity. This approach moves beyond benchmarking to ask what computational assumptions are required for a sequence-producing mechanism that underpins a certain behavior. By mapping observable behavior to latent algorithmic structure, our framework provides a principled basis for measuring cognitive cost and comparing the representational efficiency of radically different intelligent systems.
B13
A feasibility study evaluating a home-based music intervention for emotional regulation in older adults
Nathalie Gosselin
nathalie.gosselin@umontreal.ca
▶ Show abstract
Advancing age is associated with increased vulnerability to mental health problems, including anxiety and depression. Music has been shown to support emotional regulation and represent a tool to improve mental health. However, music intervention research faces methodological challenges (e.g., determining appropriate control conditions). This study assessed the feasibility of a randomized controlled trial examining the effects of a home-based personalized music listening intervention on emotional regulation in older adults, using podcast listening as an active control condition. Healthy older adults (e.g. without neurocognitive disorders) were randomly assigned to either the music or the podcast group. Participants listened to self-selected stimuli for 20 minutes every two days over four weeks. Psychological outcomes (e.g. anxiety and depressive symptoms; sadness and calm ratings) were collected before, during and after the intervention. Feasibility was evaluated by the retention, recruitment and adherence rates. Fifty-two participants were screened; 62% were eligible, and 64% were randomized. Recruitment, retention, and adherence rates were 77%, 91% and 96%, respectively. Findings support the feasibility of a randomized home-based music intervention trial and will inform the design of a future definitive study.
B14
Beyond Intelligence: Actively Open-Minded Thinking as a Unique Predictor of Accuracy on Psychological Misconceptions
Daniel Vinar
vinardaniel@yahoo.com
▶ Show abstract
Psychological misconceptions remain persistent even among educated adults, suggesting that accurate knowledge may depend on more than cognitive ability alone. The present study examined whether actively open-minded thinking (AOT), defined as the tendency to consider alternative viewpoints and revise beliefs in light of evidence, was associated with accuracy on psychological misconceptions beyond cognitive ability. Undergraduate participants (N = 262) completed the 40-item Test of Psychological Knowledge and Misconceptions (TOPKAM), a 13-item AOT scale, and a verbal-numerical reasoning test (VN). For each misconception item, participants also provided confidence ratings, allowing exploratory analyses of metacognitive calibration. Hierarchical regression analyses showed that verbal-numerical reasoning was a significant positive predictor of misconception accuracy. AOT was also positively associated with accuracy, although its incremental contribution beyond VN was modest. Participants were substantially overconfident overall, with confidence exceeding accuracy across the sample. The findings suggest that accuracy on psychological misconceptions may reflect not only reasoning ability, but also thinking style. More broadly, the results highlight persistent overconfidence in psychological knowledge, with open-minded thinking emerging as a potentially relevant factor in understanding why some individuals are better able to resist widely held but inaccurate beliefs.
B16
Can more accurate time estimation facilitate problem solving performance?
Veronica Bodea
veronicabodea@cmail.carleton.ca
▶ Show abstract
Problem solving is one of the most fundamental cognitive abilities. We propose that people allocate effort to solving a problem based on subjectively estimated solving time. As a result, individual differences in subjective time estimation should affect performance in a complex problem solving task. To test for this hypothesis, participants were recruited to perform a target time detection (TTD) task and solve the travelling salesperson problem (TSP). In the TTD, participants were shown a black circle for a target time (12 s) and asked to determine if subsequent circles were presented for the target time (or not). In the TSP, participants were asked to find the shortest path visiting 50 different cities exactly once and coming back to their starting point. The results in the TTD show that participants were generally able to distinguish between target and non-target times, and the deviations in the TSP replicated earlier findings from the literature. Preliminary results show a trending negative correlation between sensitivity (d’) in the TTD and deviations in the TSP. However, no relationship was observed between the decision criterion (beta) and problem solving performance. These results suggest that more accurate time estimation may facilitate optimal problem solving.
B17
Can stress modulation influence motion sickness in virtual reality?
Carina Baldassarra
carina.baldassarra@torontomu.ca
▶ Show abstract
Virtual reality (VR) offers powerful opportunities for research, education, healthcare, and entertainment, yet visually induced motion sickness (VIMS) presents a barrier to its widespread adoption. As VIMS has been previously compared to a stress response, the main goal of the present study was to investigate the relationship between stress and VIMS. We manipulated stress in a between-subjects design before having participants engaged in a 15-minute VR task. Sixty-six younger adults (24 male, 42 female) were randomly assigned to one of three experimental conditions: 1) stress reduction via auditory Mindfulness Based Stress Reduction [MBSR] task), 2) stress induction via auditory Paced Auditory Serial Addition Test [PASAT] task), and 3) control condition listening to a neutral podcast. Self-reported stress was captured by Visual Analog Scale (VAS), and VIMS severity was captured using the SSQ and FMS. While experimental groups did not differ in motion sickness, grouping participants by pre-audio stress (VAS) severity (low, medium, high) revealed significant differences in SSQ and FMS measures (p < .05). Findings showed pre-audio stress is positively correlated with disorientation and average FMS scores. Trait-level stress was correlated with SSQ reported sickness. Overall suggesting baseline stress levels may predict VIMS severity more than the experimental stress modulation.
B18
Conspiracy theory success: Characteristics of good explanations and stories predict belief and engagement with conspiracies
Chelsea Russill
crussill@uwaterloo.ca
▶ Show abstract
There is a pressing need to understand why conspiracy theories achieve widespread influence and popularity relative to official accounts. Past work has primarily focused on identifying the individual differences and motivations underlying belief in epistemically suspect claims, providing insight into who is most receptive to these beliefs, but not what makes the claims themselves compelling. Previous research suggests that the internal content and structure of conspiracy theories may influence attention and engagement, but few studies have examined the characteristics contributing to their appeal. We investigate the internal features of conspiracy theories that may contribute to their success (i.e., belief and engagement), testing the account that successful theories operate as convincing explanations and compelling stories. In two studies (N = 456), participants rated conspiracy theories on features associated with good explanations and narratives and completed measures of belief and engagement. We find that conspiracy theories perceived as better explanations and stories were more likely to be believed and shared, even after controlling for individual differences in conspiratorial ideation. Preliminary results from a third study (n = 489) using an expanded set of theories and a within-subjects design replicate these findings, suggesting that explanatory and narrative characteristics reliably predict conspiracy theory success.
B19
Embodied Cognition Meets Personality Psychology: Introducing the Loss of Grip Scale
Garri Hovhannisyan
garri.hovha@gmail.com
▶ Show abstract
This study introduces the Loss of Grip Scale (LOGS), a novel assessment tool designed to capture breakdowns in person–situation fit. “Loss of grip” refers to disruptions in an individual’s ability to effectively engage with their environment, often experienced as disorientation, anxiety, or detachment . Existing personality measures assess stable traits but do not adequately capture how these traits become dysregulated in context. The LOGS addresses this gap by measuring both the direction (i.e., excess vs. deficit) and magnitude of trait expression across situations. Items were developed to reflect Big Five aspects across domains such as relationships and work, with parallel indicators of “too much” and “too little.” Exploratory factor analysis in a university sample (N = 191) yielded a 12-factor structure, including Hypervigilance, Impulsivity, Perfectionism, and Interpersonal Detachment. Higher-order factors aligned with Internalizing, Disinhibition, and Detachment dimensions, consistent with contemporary models of psychopathology. Correlations with Big Five Aspect Scales supported convergent and discriminant validity, suggesting that LOGS captures maladaptive extensions of normative traits. Clinically, the LOGS complements trait-based assessment by quantifying when and how traits result in person-environment misfit, offering a structured bridge between personality structure and lived dysfunction.
B21
Testing a Behavioral Measure of Persistence in Problem Solving
Mengxi Liu
mengxii.liu@mail.utoronto.ca
▶ Show abstract
Persistence plays a critical role in problem-solving, motivation, and goal-directed action. It reflects an individual's capacity to sustain effort in the face of obstacles, uncertainty, or delayed outcomes. The purpose of this study is to develop and validate a novel and fine-grained approach to the measurement of persistence in problem solving. We assume that participants invest a set amount of time to a given class of problems and attempt to estimate this latent value by comparing the time spent on problems of the given class to theoretically-established processing time distributions (e.g., via cognitive modeling). To collect some initial evidence regarding the validity of such an approach, we focus in this study on the voluntary time individuals spend on a word fluency task with an experimentally determined stochastic completion time. We examine the relationship between the proposed measure and task motivation, personality traits, and delay discounting for convergent validity.
B22
Does Providing Students with Choice Increase Academic Confidence and Reduce Academic Procrastination and Stress?
Delaney O'Brien-Ristau
obrienristaud@mymacewan.ca
▶ Show abstract
Research on Self-Determination Theory and Universal Design for Learning has shown that autonomy is necessary to foster student motivation and engagement. Motivation and engagement have been linked to increased confidence and reduced procrastination and stress in academic environments. Autonomy can be supported by providing students with many opportunities for choice within their courses (e.g., assignment topics, modality, and deadlines). The current study investigated the relationship between offering students choices in their university coursework and their levels of academic confidence, procrastination, and stress. Three hundred university students were asked to rank the courses they were currently enrolled in based on flexibility. Participants then responded to a series of questionnaires assessing course choices, confidence, procrastination, and stress. Participants first completed the questionnaires thinking about their least flexible course and then completed the same set of questionnaires thinking about their most flexible course. Results showed that students tended to procrastinate more, experience more stress, and feel less confident in lower-flexibility courses compared to higher-flexibility courses. However, students' perceptions of whether the course was flexible had a greater impact on outcome variables than the actual number of choice opportunities offered.
B23
Exploring the Role of Feedback Timing and Response Format in Verbal and Implicit Category Learning
Isaac Withers
iwithers@uwo.ca
▶ Show abstract
The ability to categorize allows us to understand vital conceptual groups such as “edible vs inedible”. The COVIS Model, (COmpetition between Verbal and Implicit Systems), proposes that two systems compete during category learning: a Verbal System, (VS), using verbal rules (“If large, then A; if small, then B”), and an Implicit System, (IS), which supports unconscious categorization necessary when a category is too complex for easy verbal learning. This study investigates COVIS, showing that it is possible to effectively dissociate the VS and IS systems even when confounds and alternate explanations are accounted for; holding task complexity constant and addressing common criticisms from previous literature. In Experiment 1, 200 participants complete a novel delayed feedback task to validate suggestions that variations in feedback timing can dissociate VS and IS. Experiment 2 (n = 200) extends this logic to ask whether altered choice formats (A/B choice or Yes/No verification) can dissociate VS and IS, a technique previously unexplored. By rooting Experiment 1 in the previous literature, we lay a firm conceptual groundwork for the novel categorization technique proposed in Experiment 2. This study will shed light on the COVIS model, as well as provide a new paradigm to explore category learning.
B24
From Meaning to Mechanism: Semantic Constraints on Statistical Learning
Laura Li
20sl101@queensu.ca
▶ Show abstract
Statistical learning allows individuals to efficiently extract regularities from their surrounding environment, but how prior knowledge influences this process remains unclear. In real-world environments, objects are associated with specific spatial locations. We examined whether statistical learning operates independently of such knowledge. Participants searched for a target object in 4-item arrays. Targets appeared in either high- or low-probability locations (80% vs. 20%) and in either semantically consistent (e.g., boots in the lower quadrants) or inconsistent (e.g., boots in the upper quadrants) locations. Participants responded faster to targets appearing in semantically consistent than inconsistent locations. This effect interacted with statistical learning: the high-probability advantage was stronger for semantically consistent targets than for semantically inconsistent targets. These results indicate that prior semantic knowledge does shape statistical learning during visual search, even in the absence of scene context. Together, these findings demonstrate that statistical learning is not purely frequency-driven, but is systematically constrained by prior semantic knowledge.
B25
Integrating open science in the teaching of cognitive research methods 2.0: Meta-analytic thinking & Registered Reports
Ralph Redden
x2021bqn@stfx.ca
▶ Show abstract
Openness, transparency, and reproducibility are widely accepted as fundamental aspects of scientific practice. However, a growing body of evidence suggests these features are not readily adopted in the daily practice of most scientists. The Centre for Open Science has championed efforts for systemic change in the scientific process, endorsing practices such as preregistration and open sharing of data and experimental materials. In an effort to inculcate these practices early in training, we integrated several key components of open science practice into an undergraduate research methods course in the cognitive sciences. A previous iteration of the course with a lab component focused on teams carrying out a preregistered replication experiment related to the topics in the course. This version (non-lab class) had teams perform a meta-analysis of the Attention Network Test literature using the AttentionNetwork.ca database, pertaining to a theme of their choice (e.g., Exercise, Alcohol, Depression, PTSD, and Bilingualism). Teams conceptualized a follow-up experiment of their own choice, and prepared a Stage 1 Registered Report protocol motivated by their meta-analysis. Results from meta-analyses, and critical appraisal of the goals and implementation of the course across formats are discussed.
B26
Investigating rapid adaptation to speech in noise using reverse correlation
Hanna Zhang
zhanghanna20@gmail.com
▶ Show abstract
The auditory system is able to rapidly adapt to noisy conditions, but how and how fast phonetic processing changes still needs to be better understood. Here we ran a categorization task using the syllables /da/ and /ga/ embedded in white noise. In one condition (LogatomeA) the noise-signal onset delay was 250ms while in the other (LogatomeB) it was 50ms. All participants (N = 10) completed 2000 trials per condition, divided by blocks of 200 trials during which the speech-to-noise ratio (SNR) was adaptively adjusted to reach an accuracy of 70%.As expected, we observed a ~2.5 dB improvement in SNR thresholds for LogatomeA. Further, using a reverse correlation approach (correlating the spectrotemporal noise patterns in each trial with participants) we observe that while listeners recruit the same cues under both conditions, LogatomeA results in greater reliance on the lower frequency cue relative to the higher frequency cue. This pattern is consistent with adaptation-to-noise theory. Since white noise produces stronger masking at higher frequencies, LogatomeA, but not LogatomeB gives the auditory system enough time to adapt to the noise and change its frequency-dependent gain to make the most efficient use of the cues.
B27
The Effects of Relaxation Training Interventions on Mitigating Virtual Reality Sickness
Narmada Umatheva
numatheva@torontomu.ca
▶ Show abstract
Virtual Reality (VR) sickness is associated with the activation of the sympathetic nervous system (SNS), producing physiological changes like increased heart rate, respiration and sweating. Auditory beat stimulation (ABS) in the theta range (4–8 Hz) combined with music, and deep diaphragmatic breathing (DDB) shows potential in reducing SNS responses, which may reduce VR sickness. The current study explores whether relaxation trainings of ABS + music or DBB can reduce the stress response and SNS activity seen in VR sickness and alleviate symptoms. Healthy participants (ages 18 – 49) completed pre- and post-training sessions in-lab, separated by two weeks of at-home training (10 minutes/day, 5 days/week) in one of three training conditions: ABS + music, DDB, or control. During the in-lab sessions, participants viewed a VR stimulus while subjective VR sickness ratings and physiological measures (heart rate, respiration rate, and galvanic skin response) were recorded. Preliminary analyses (N =33) using linear mixed models indicate a downward trend in VR sickness scores for participants in the ABS + music and DDB conditions compared to the control. These preliminary findings tentatively suggests that relaxation trainings using ABS + music or DDB may be effective strategies in reducing VR sickness symptoms.
B28
Acquisition and automatization of a complex motor skill: the role of baseline visual and motor abilities
Mallory E. Terry
terryme@mcmaster.ca
▶ Show abstract
Juggling is a complex motor skill requiring the integration of visual and motor processes. Past research has identified distinct juggling learning trajectories, which can be quantified through rates of skill acquisition and automatization. The present study investigated factors that may account for these individual differences, and in particular, the influence of baseline visual and motor abilities. Ten participants completed 14 days of juggling practice (20 minutes per day). Acquisition was assessed daily as the mean number of consecutive catches. Automatization was assessed across four probe sessions (pre-training, Days 4, 9, and 14) as the dual-task cost for alphabet recitation performed concurrently with juggling. Baseline visual and motor abilities were assessed prior to and post training. Consistent with past research, three distinct learning patterns emerged, reflecting differences in the rate of acquisition and degree of automatization throughout training. Baseline aiming and catching abilities were associated with both learning rate and juggling improvement across training, while overall motor score was associated with the degree of automatization at the end of training. These findings suggest that baseline motor abilities contribute to individual differences in both the acquisition and automatization of complex motor skills, with different motor domains playing distinct roles in each process.
B29
Adjustment of Response Bias Requires Task Focus
Carmi Ampo
ampo@ualberta.ca
▶ Show abstract
In visual detection at threshold, response bias has been shown to depend on target frequency: If targets are more likely, “yes” responses are more likely in general. This might be characterized as a form of probability matching. In the present research, we manipulated target frequency across blocks of trials and also asked participants to report their level of task focus. Consistent with previous research, when targets were more frequent, target detection confidence increased. However, this effect was much smaller when participants reported being off task. Our interpretation is that the adaptive adjustment of response bias is not automatic and requires task focus.
B30
Age-Related Effects on Spatial Navigation and Gait During Motor–Cognitive Dual-Tasking
Jordan Fairlie
j.fairlie@mail.utoronto.ca
▶ Show abstract
Locomotor spatial navigation relies on the integration of sensory, cognitive, and motor processes, all of which may display age-related decline. Investigating how these domains interact may improve our understanding of age-related changes to real-world mobility and characterize performance trajectories across the lifespan. The current study examined age-related differences in locomotor spatial navigation in healthy younger (n = 29, 18–35 years) and older (n = 31, 65+ years) adults using a locomotor triangle completion task in virtual reality. Participants walked two guided segments of a triangle and then attempted to return unassisted to their unmarked starting location, either in silence (single-task) or while performing a concurrent auditory task (dual-task). Spatiotemporal gait metrics and navigation performance measures were collected to assess task performance. Older adults showed greater endpoint and rotational errors, and greater dual-task costs to distance error than younger adults. Gait speed decreased from the first to the final segment (segment requiring active navigation) across groups, and dual-tasking further reduced gait speed in the final segment. These findings may help characterize age-related differences in locomotor spatial navigation and provide a foundation for identifying early deviations associated with functional and cognitive decline.
B31
Assessing the Temporal Dynamics of Attentional Control across Tasks
Jocelyne Harling
jocelyne.harling@yahoo.ca
▶ Show abstract
Attentional control is critical for regulating task-focused attentional states, with prior research suggesting that this ability is relatively consistent within individuals across tasks. However, little is known about how this consistency unfolds over time. As such, we recruited 189 participants to complete three well-established experimental tasks, the Sustained Attention Response Task (SART), the Visual Search Task (VST), and the Attentional Network Task (ANT). To assess temporal patterns of attentional control within each task, autocorrelation analyses were conducted, revealing predominantly positive values across tasks [SART: M=0.25, SD=0.22; VST: M=0.17, SD=0.16; ANT: M=0.18, SD=0.21] and indicating low-to-moderate repetitions in temporal patterns within individuals. To assess consistency across tasks, we ran a repeated measures ANOVA, which revealed significant differences across tasks [F(2,278)=8.80, p<.001, ηp²=.06 ], with post-hoc comparisons indicating that patterns of repetition in the SART differ significantly from the VST and ANT. Together, these findings provide insight into the individual variability underlying the temporal dynamics of attentional control by suggesting that attentional control may be driven by the specific demands of each task rather than reflecting a stable, overarching trait that can be applied uniformly across tasks.Keywords: Attentional control, cognitive tendencies, questionnaires, experimental tasks
B32
Associative learning from contextual cueing effect: Interference from previous learning
Sm Imran Faruqui
faruqs4@mcmaster.ca
▶ Show abstract
During visual search, reaction time becomes faster in repeated scenes with repeated target and distractor positions across trials than in novel scenes with changing target-distractor relations. Such learning of the association between the target and the distractor layout in the repeated scene is called the contextual cueing effect (CCE, Chun and Jiang 1998). Prior research has established that, following learning, relocating the target makes relearning the association between the relocated target and the distractor layout more difficult. The present study examines whether the target relocation cost would be smaller if the visual features (e.g., colour) of the search items changed, making the entire scenes appear more dissimilar to those during earlier learning. The results showed that, following successful learning, the CCE during the test phase was generally smaller than during the learning phase. In addition, the CCE for scenes with colour change was larger than for scenes without. These results suggest that prior learning of the association between a specific target location and a given distractor layout can interfere with subsequent learning of a new target location within the same layout. However, introducing a salient non-spatial cue that clearly differentiates the scenes from the original can facilitate this relearning process.
B33
Behind the bias: How attention shapes ensemble perception
Hannah Arabella Gabling
hannah.gabling@mail.utoronto.ca
▶ Show abstract
Many studies investigating ensemble perception (the rapid extraction of summary statistic information from groups of objects) use displays composed of identical items, which does not match our visual reality. Previous work using heterogeneous displays has shown that item variability can bias ensemble perception (e.g., reports of average orientation for displays composed of triangles and ovals were biased in the direction of the triangles). Across two experiments, we examined whether attention explains the source of this “ensemble bias”. In Experiment 1, participants viewed ensembles of ovals and triangles and reported the average orientation. On some trials, letters appeared in the same location as the shapes, and participants reported the letters they recalled. This allowed us to examine where attention was allocated in the display. We found that spatial attention was biased in the direction of the triangles, as participants were more likely to recall letters that were in the triangle locations. Experiment 2 used eye tracking to examine if overt attention causes this bias (i.e., more saccades towards triangles versus ovals). We did not find differences in the number of saccades towards different stimuli, suggesting that covert, rather than overt, attention likely explains the ensemble bias, possibly due to attentional amplification.
B34
Capturing the Real You: Using Gamification to Assess Real-World Attentional Abilities
Nicole Hernandez
21nhm5@queensu.ca
▶ Show abstract
Attentional abilities are traditionally assessed using laboratory tasks (e.g., SART, visual search) that carefully control the study design (e.g., stimuli used, presentation timing) and motivate participants through external means (e.g., monetary compensation, course credit). However, it is unclear if these methods capture ‘true’ functional attentional abilities given that they are often devoid of the intrinsic motivation (e.g., personal interest, curiosity, enjoyment) that we apply to tasks in our everyday lives. One means of capturing this is by applying gamification to these traditional tasks, as prior work has shown that elements of game design (e.g., narrative framing, presenting feedback, current score) can increase participants’ overall task performance and levels of interest. We will present data comparing participants’ objective task performance and subjective affective reports for a traditional versus gamified sustained attention task (i.e., SART). These findings also guide the development and launch of our laboratory’s smartphone application that provides gamified versions of multiple traditional attentional tasks, allowing researchers to capture functional attentional abilities that encompass participants’ intrinsic motivation and to create their own gamified attentional tasks for data collection.
B35
Characterizing mental imagery and its relation to memory using multidimensional experience sampling
Hala Rahman
13har1@queensu.ca
▶ Show abstract
Individuals' ability to engage in visual imagery can vary dramatically in its nature and content, contributing to later cognition. Existing measures make it difficult to capture this variability, limiting the establishment of associations with downstream cognition. Here, borrowing from tools used to measure ongoing thought, we developed a 15-item multidimensional experience sampling (imDES) approach for imagery, to measure and identify the unique experiential features of visual imagery (e.g. sharpness, colorfulness, prototypicality). Participants encoded images of objects by spending 4-8s visualizing them, and then made a judgment about the likelihood they would remember the item later. Periodically, they were asked to rate their imagery using imDES. Their memory was tested using a difficult 4AFC remember/know/guess task. Our results show three distinct principal components along which mental imagery varied: 'vivid detailed,' 'dynamic contextual,' and 'difficult prototypical.' Critically, this meaningfully mapped to anticipated and actual memory performance, where 'vivid detailed' imagery was associated with higher anticipated, and more recollection-based actual memory, while 'difficult prototypical' imagery showed the opposite. Together, these results highlight mDES as a means of identifying within- and across-participant differences in imagery, and establish critical links between imagery subtypes and perceived and veridical memory performance.
B36
Contours Drive the Tilt Aftereffect in Naturalistic Images
Seohee Han
seohee.han@mail.utoronto.ca
▶ Show abstract
Determining orientation is a fundamental visual computation, yet natural scenes contain conflicting cues: contours reflect object boundaries and global structure, whereas filter-based orientation mixes boundaries with surface texture and local gradients. Across five experiments, we used the tilt aftereffect to test which cue the visual system relies on. Adaptation to natural-scene patches consistently followed contour-defined orientation, and explicit judgments showed the same bias. Replication with artist-generated line drawings and a blocked design strengthened these effects. The findings show that contour structure provides a more perceptually meaningful orientation signal in natural scenes.
B37
Driven to Distraction: Effects of visual distraction and eyes-off-road (EOR) warnings on driver gaze behaviour and hazard detection accuracy
Ginnie Wee
ginnie.wee@mail.utoronto.ca
▶ Show abstract
Distracted driving, or dividing attention between tasks while driving, is dangerous, but why? To answer this, we investigated how distraction impacts driver eye movements, and how eyes-off-road (EOR) warnings may mitigate distraction. In Experiment 1, twenty licensed drivers performed two tasks – a hazard detection task where they reported the locations of hazards in dashcam videos while performing a secondary visual task. Tasks were performed either separately under focused attention, or simultaneously under divided attention. Under distraction, hazard detection accuracy was significantly lower, and two-thirds of misses were due to failures to look at the hazard. Experiment 2 investigated whether real-time gaze monitoring with EOR warnings reduced hazard miss rates. Thirty-six drivers completed the divided attention condition from Experiment 1 in an EOR warning condition and a warning-absent condition. In the warning condition, a visual alert was given when drivers looked away from the road video for a set duration (1 or 1.5s, in separate conditions). EOR warnings did not significantly change driver gaze behaviour or hazard detection accuracy. However, warnings helped speed drivers’ responses. These results show how distraction changes driver gaze behaviour, and point to why drivers fail to notice key events on the road.
B38
Examining Light Sensitivity and ADHD Symptomatology
Savannah Mancebo Bodden
smb13@my.yorku.ca
▶ Show abstract
Light sensitivity in Attention Deficit Hyperactivity Disorder (ADHD) represents a crucial symptom that is often disregarded in clinical assessments and treatment plans, despite its prevalence among ADHD populations. To better understand the impact of sensory sensitivity to light, we developed a protocol to assess differences between ADHD participants and controls on a modified Sustained Attention to Response Task (SART) under varying lighting conditions. As a secondary objective, we aim to examine differences in performance across ADHD subtypes. Participants will be assigned to an ADHD or control group based on their performance on the Adult ADHD Self-Report Scale (ASRS). Following this, the modified SART will include four separate blocks of varying lighting conditions: Brightness (Dim, Full) and Light Temperature (Warm, ~3200K; Cool, ~6500K). Data obtained from the SART will be analyzed using a 3-way mixed-model analysis of variance (ANOVA) to test for main effects of ADHD status, brightness and light temperature on performance, as well as interaction effects between conditions. We hypothesize that the ADHD group will demonstrate performance variability, as measured by reaction time, between lighting conditions, compared to stability within the control group. Findings of the proposed study may contribute to a better understanding of light sensitivity in ADHD.
B39
Examining the Developmental Trajectory of Elderspeak in Children
Katia Perry
katiaperry@trentu.ca
▶ Show abstract
Elderspeak is a form of implicit ageism characterized by speech accommodation towards older adults (e.g., semantic elaboration, slower speech rate), which negatively impacts their well-being. Little is known about when elderspeak develops; thus, this study explored the developmental trajectory of elderspeak in 3- to 7-year-old children. In computer tasks involving younger and older characters, we measured production of elderspeak (how children provide instructions to each character), perception of elderspeak (which character children think elderspeak and normal speech is intended for), and children’s ageist stereotypes (e.g., who is better at running? knitting?). Preliminary results from 27 children found little evidence of children producing or perceiving elderspeak. Although older children elaborated more when producing instructions than younger children, neither their speech rate nor degree of elaboration varied according to character age. The perception task showed no evidence that children associated elderspeak with older adults. The stereotype task, however, revealed an age effect with older children holding more ageist stereotypes than younger children. The study, currently in progress, will further explore age-related differences as a function of cognitive (e.g., theory of mind) and social factors (e.g., experiences with older adults). Findings will inform educational interventions to help reduce implicit ageism in early childhood.
B40
How multilingualism affects perceptual sensitivity to non-native speech sound features
Jade HY. Fok
jade.fok@mail.utoronto.ca
▶ Show abstract
It is often claimed that bilinguals acquire foreign languages more easily than monolinguals. The present study investigates whether bilinguals are better at discriminating non-native speech sounds, and how general this “bilingual advantage” is. English monolinguals, bilinguals who speak English and a “prevoicing language” (e.g. Spanish, Tagalog), and bilinguals who speak English and a “tone language” (e.g. Mandarin, Cantonese) were tested on their ability to discriminate two non-native Thai speech sound features. To investigate whether discrimination is better a) for bilinguals in general, or b) only for bilinguals experienced with the non-native speech sound features tested, participants discriminated Thai words that differed only in prevoicing (vibration vs. lack of vibration of the vocal folds during the production of certain sounds), and words that differed only in tone (e.g., ‘ma’ spoken with high vs. low tone). Preliminary findings suggest that the “bilingual advantage” is less general than sometimes claimed. There appeared to be no advantage for either bilingual group in discriminating the prevoicing contrast, and only the bilinguals who speak English and a tone language discriminated the tone contrast better. These findings provide some evidence that the “bilingual advantage” may depend on bilinguals having experience with the non-native speech sounds they encounter.
B41
Lexical Restructuring of Auditory Categories in Chinese–English Bilinguals
Zoe Hu
zhu358@uwo.ca
▶ Show abstract
Lexical categories vary across languages, requiring L2 learners to reorganize how perceptual experiences are mapped onto words. Auditory lexical categories, despite their importance in everyday communication, remain largely understudied in lexical restructuring research. This study addresses this gap by examining how Mandarin–English bilinguals label sounds in English to trace restructuring from Mandarin-influenced lexical patterns toward more English-like ones. Mandarin often lexicalizes sounds through object and action (e.g., bee make sounds), whereas English uses sound-quality labels (e.g., buzz) more, making this contrast a strong test case of bilingual lexical restructuring.Mandarin–English bilinguals and English monolinguals heard clips of everyday sounds and produced their names in English (e.g., pop, beep). Results show that English monolinguals used sound-quality labels more often than bilinguals, whereas bilinguals produced more combined responses, especially sound quality + object and object + action, consistent with Mandarin auditory lexical structure. Further analyses were conducted to examine how individual differences, such as length of immersion in English and current usage of English, impacted bilinguals’ responses. Our findings suggest that bilingual sound labelling in English remains influenced by Mandarin and gradually reorganizes with greater English exposure.
B42
Lightness perception in augmented reality: effects of context, background, motion, and disparity
Richard Murray
rfm@yorku.ca
▶ Show abstract
Lightness constancy is critical for understanding visual scenes, including virtual content in optical see-through augmented reality (OST-AR). Blending of virtual and real-world scenes in OST-AR elevates stimulus luminance, and makes it difficult to create convincing images of low-reflectance surfaces. We investigated how several visual cues affect lightness perception in OST-AR, with the goal of understanding how to expand the range of perceived reflectance. In two experiments, participants viewed virtual test patches in a foreground plane, and rated their lightness relative to real paper patches. We manipulated the composition of the foreground and background planes, motion of the background, and relative disparity between foreground and background. The strongest effects on perceived reflectance came from foreground manipulations: isolating the test patch or eliminating luminance variation substantially reduced the range of perceived reflectance. Even large changes to the background, such introducing motion or eliminating disparity, had much smaller effects. Thus lightness perception in OST-AR depends heavily on foreground structure, and including rich foreground context is important for generating a wide, realistic range of perceived reflectance. These are encouraging findings for applications of AR, since designers have much more control over the virtual foreground of OST-AR displays than over the real-world background.
B43
Link Between Face Detection and Identification: Evidence from Individual Differences
Laurianne Côté
► Show abstract
Face detection and identification are widely assumed to rely on distinct processes. Supporting this view, individual-differences studies typically report little to no association between these abilities, and prosopagnosic individuals often exhibit intact detection despite profound identification deficits. Yet, evidence from a highly sensitive psychophysical paradigm (Xu and Biederman, 2014) challenges this separation, showing that impairments in identification can co-occur with deficits in detection. These findings call into question a strict functional dissociation between detecting a face and knowing who it is. Here, we measured face detection ability in 34 neurotypical adults (22 women) using this paradigm. Participants also completed three established face identification tasks (CFMT+, CFPT, GFMT2), alongside an object identification task (VET) to control for domain-general visual processing. Face detection thresholds were negatively correlated with identification performance (r = −0.54, p < 0.001), such that individuals with superior recognition required less information to detect faces. Critically, this association remained robust after controlling for object recognition (r = −0.49, p = 0.004). These findings point to shared mechanisms underlying face detection and identification, challenging accounts that posit a strict functional separation between seeing a face and knowing who it is.
B44
Locomotor Spatial Navigation Reveals More Cautious Gait Patterns in Older Adults with Age-Related Hearing Loss
Lianna Montanari
lianna.montanari@mail.utoronto.ca
▶ Show abstract
Age-related hearing loss (ARHL) is associated with mobility-related problems and falls; however, the mechanisms underlying this relationship remain poorly understood. The cognitive load hypothesis proposes that degraded auditory input during listening draws cognitive resources away from concurrent tasks, including mobility tasks. Traditional mobility assessments often fail to capture the multitasking and multisensory integration demands required for real-world mobility, highlighting the need for more complex, multi-domain assessment paradigms. This study aimed to address this gap using a locomotor spatial navigation task requiring concurrent navigation, walking, and listening. Older adults with and without ARHL (n=31/group) completed a locomotor triangle completion task in virtual-reality; they walked along two legs of a triangle and attempted to return to their unmarked start location. This was performed alone (single-task) or with a concurrent listening task (dual-task) during which gait speed, cadence, double support bouts, and stationary time were measured, enabling systematic investigation of gait under increased cognitive load. Older adults with ARHL demonstrated greater dual-task costs to all gait parameters than those with normal hearing, reflecting more cautious gait patterns under more cognitively demanding conditions. This suggests that ARHL may meaningfully influence mobility when cognitive demands are high, underscoring the value of complex, multi-domain mobility paradigms.
B45
Mechanisms of Emotional Sentiment in Autobiographical Memory Reactivation
Aidan Steeves
aidan.steeves@mail.utoronto.ca
▶ Show abstract
Autobiographical memory shapes not only how we recall past events but how we feel about them. However, the mechanisms linking episodic richness to emotional sentiment remain unclear. Using HippoCamera, a smartphone application that captures and replays audiovisual memory cues, we examined these mechanisms across several samples of older and younger adults using linear mixed-effects models.Episodic richness was the most robust predictor of emotional sentiment across all datasets. Mediation analyses revealed that internal detail partially accounted for the replay-sentiment relationship, though most remained direct, suggesting reactivation carries affective consequences beyond elaboration alone. This benefit was time-sensitive, strongest for recently encoded memories and diminishing with age, driven primarily by younger adults, who showed weaker episodic-sentiment coupling overall. Older adults showed a tighter episodic-sentiment relationship and more stable emotional tone across memory age. Emotional intensity, rather than positive valence specifically, drove episodic richness: both positive and negative memories were recalled with greater detail than neutral ones, with a positive valence advantage. Longitudinally, replay improved retention of episodic detail over time independent of valence.These findings suggest emotional benefits of memory reactivation emerge through broad preservation of episodic representations, rather than selective amplification of positive memories.
B46
Mental rotation ability: Sexual orientation matters for males and females
Jordyn Heron
heronj@yorku.ca
▶ Show abstract
Biological sex differences have been identified on several cognitive tasks. Cross-sex shifts in the performance of gay and lesbian individuals have also been found in some cognitive tasks that show large sex differences. Previous studies on mental rotation, the ability to visualize and manipulate two- or three-dimensional objects in the mind, report a robust male advantage and cross-sex shifted performance among gay males. However, differences between heterosexual and lesbian females have been reported inconsistently and are often omitted from study designs. The present study addressed this gap by examining mental rotation ability using a balanced design with four groups of participants: heterosexual males, heterosexual females, gay males, and lesbian females. Participants completed a 24-item mental rotation test in which they compared three-dimensional cubed figures rotated at different orientations along the vertical axis. Results revealed cross-sex shifted performance in both sexes: heterosexual males exhibited higher accuracy than gay males, while lesbian females exhibited higher accuracy than heterosexual females. These findings provide a foundation for future neuroimaging studies to further examine how biological sex and sexual orientation relate to neural mechanisms underlying mental rotation and other visuospatial abilities.
B47
Modelling embodied language under active inference
Vitoria de Souza
vsouza@mun.ca
▶ Show abstract
How do we understand sentences like “the ranger saw the eagle in the sky”? Models of grounded and embodied cognition suggest that comprehension is supported by the partial reactivation of sensorimotor states. For instance, to understand "the ranger saw the eagle in the sky" we partially reactivate visual perceptual states of outstretched wings. While verbal theories have been offered to account for sensorimotor contributions to comprehension, mathematical, mechanistic models are scant. In the present study we provide a proof-of-concept mechanistic model of sensorimotor contributions to sentence comprehension using the framework of active inference, a novel computational neuroscientific technique for modelling embodied perception and action. We simulate sensorimotor simulations in a seminal sentence-picture verification task in which participants read a sentence and then are shown a picture of an object whose shape either matches or mismatches the shape implied by the sentence. Behaviourally, participants are faster at deciding that the picture reflects the concepts of the sentence when the shape matches, implicating sensorimotor reactivation. Our model simulations show that mismatching shapes cause increases in prediction error and abrupt changes to sensorimotor simulations during sentence comprehension. This model provides the first mathematical and mechanistic account of embodied concepts, extending verbal theories.
B48
More Than Meets the Eye: Exploring Binocularity in Self-Reported Amblyopia
Fermin Retnavarathan
retnavf@mcmaster.ca
▶ Show abstract
Amblyopia is a neurodevelopmental condition characterized by poor vision primarily in one eye. This condition arises when the two eyes do not receive concordant visual input early in life. Even after corrective surgery, input from the two eyes rarely matches in these individuals. Thus, they do not experience binocular vision and cannot use binocular-supported depth cues to perceive depth. However, Maehara et al. (2019) discovered a subset of amblyopia patients, who failed all clinical tests of binocularity, demonstrated the Pulfrich effect. Perceiving depth through this effect indicates these patients retain intact binocular abilities—hidden binocularity. Additionally, clinical tests may not provide sensitive enough measures of binocular vision. To explore this further, I am part of a team funded by CIHR that has developed a battery of binocular vision tests. In this study, control participants and self-declared amblyopes completed three tasks in our battery for which we have confirmed reliability: a letter dominance, Pulfrich, and plaid motion task. Preliminary findings provide insight into the extent of residual binocularity in this population and hidden binocularity in select cases. Given re-establishing binocular vision has been the primary goal of many rehabilitation paradigms for this population, this represents a meaningful advance in the field.
B49
Passive and Active States in Visual Working Memory
Alex Huynh
huynha37@mcmaster.ca
▶ Show abstract
Emerging research in visual working memory suggests the existence of active and passive memory states. For example, in a blocked design, the memory of array 1 switches from an active to a passive state when the encoding of array 1 is immediately followed by the encoding of array 2. In this study, we study the benefits/costs of state-switching. Participants performed memory tasks in two sessions, with each session containing trials mixed from two conditions. Both sessions included Condition 1 (70% of the trials), in which arrays 1 and 2 were sequentially presented. Following encoding, in Session A, memory retrieval was required in a fixed order (array 2, then 1), encouraging the state switch for array 1. In session B, memory retrieval was required for either array at random, encouraging array 1 to remain active. Condition 2 in both sessions (30%) involved encoding and retrieval of only array 1, which likely involved state-switching in Session A, but not in Session B. The results showed that, for Condition 1, accuracy in Session A was higher than that in Session B, suggesting a benefit from state-switching; for Condition 2, accuracy was comparable across the two sessions, suggesting little cost to state-switching.
B50
Perception of the sound induced flash illusion measured in-person compared to online
Julie Lewczuk
julie111@my.yorku.ca
▶ Show abstract
Conducting psychological studies online has become increasingly popular in recent years, especially following the period of in-person COVID-19 pandemic restrictions. Compared to other disciplines in psychology, perceptual research is typically conducted in a traditional in-person laboratory testing environment. An online environment presents unique challenges that introduces a lack of experimental control for technical set-up and stimulus presentation that may impede collecting high quality data when studying perceptual phenomena. Nonetheless, the opportunity to conduct perceptual research online is valuable, as it may facilitate generalization to the broader population through increased recruitment opportunity and can assist outreach to special populations with rare characteristics. Perceptual integration of audiovisual events is important to our understanding of how people experience the world. The present study investigates the feasibility of conducting a standard audiovisual perception experiment, the sound induced flash illusion (SIFI), online compared to a traditional in-person laboratory environment. Overall, susceptibility to the SIFI was observed in both online and the in-person lab setting with no difference between environments. This research indicates that it is feasible to conduct high quality audiovisual perceptual studies in an online environment with comparable results to a traditional in-person lab setting.
B51
Perceptual grouping reduces perceived speed but does not change uncertainty
Alejandro Gonzalez Garcia
aggarcia@yorku.ca
▶ Show abstract
Four dot pairs rotating about vertices arranged in a square configuration group together to form the illusory percept of two overlapping squares (global percept). The transition from local to global motion percepts leads to a reduction in perceived speed: the global slowdown effect. Prior work has ruled out effects of perceived size, emergent rotation or perceived number of items as explanations for this illusion. Here we test a new possibility: If the global percept is associated with greater sensory uncertainty than the local percept, a Bayesian prior toward slow motion could lead to a reduction in perceived speed. To test this, we ran a 2AFC speed-comparison task in which participants compared global-biased and local-biased configurations, and measured the global slowdown effect as the Point of Subjective Equality. We then estimated the uncertainty of speed judgments for each participant, as the slope of the psychometric functions for trials where the compared configurations were either both global or both local. Our results replicated the global slowdown effect, but we found no systematic difference in uncertainty between local and global, and no relationship between uncertainty differences and illusion strength. We conclude that a Bayesian prior for slow motion cannot explain the effect.
B52
Predicting Road Hazards from Brief Dynamic Previews: Accuracy and agreement among young adult drivers
Jiali Song
jiali.song@utoronto.ca
▶ Show abstract
Prediction is a fundamental visual process and is essential for safe driving. We examined whether drivers could correctly predict the hazards in road scenes. Twenty-four licensed drivers viewed 340ms excerpts from 174 dashcam videos that each contained a collision or near-collision. Video excerpts were created from each video before the moment of collision: 2000ms, 500ms, 0ms, and just before the driver in the video responded. After viewing each excerpt, participants were asked to report the location of the most likely hazard in the scene by clicking on it with a computer mouse. Prediction accuracy was defined as the proportion of clicks that corresponded with the location of the collision. Prediction accuracy was above chance across all excerpts, and increased monotonically with time. We indexed agreement using average pair-wise distance between all predictions, which also decreased monotonically with time, indicating increasing agreement. Click locations were also moderately correlated with fixation locations collected previously in a separate sample of 30 drivers who localized hazards in the same stimuli (mean Spearman’s rho = 0.52), suggesting that drivers look at predicted hazard locations. These results indicate that on-road predictions are guided by scene context, are moderately consistent among individuals, and guide gaze.
B53
Priming and Visual Processing Under Binocular Suppression: Investigating the Effects of Delay on Implicit Memory for Fully and Partially Processed Information in Younger Adults
Brandon Mayer
bmayer495@my.nipissingu.ca
▶ Show abstract
In this study, participants (younger adults ages 18–29) identified words that gradually appeared in one eye while under binocular suppression (BS) from a visual mask presented in the other eye from the beginning of the trial. Before each BS trial, participants identified a target word while ignoring a distracting word in a priming display. The target BS task word was either unrelated to any priming display or was a previously attended or ignored word from earlier priming. Delay effects were examined across multiple delays including: no delay (immediate), or delays of 1, 5, or 10 trials. Significant priming from attended stimuli was present immediately and remained stable across delays, indicating consistent benefit from fully processed target information. Priming from ignored distractors was initially significant but smaller than attended priming. However, a single trial delay erased the effect of priming from unattended words, suggesting a rapid decay of the benefit found from partially processed distracting material. Findings highlight initial and sustained implicit benefit for fully processed information and immediate benefit but then rapidly declining benefit for partially processed irrelevant information. Including older adults in a future study could provide information on how age affects such implicit benefits.
B54
Putting movement back into fixation models: the role of saccade trajectories in predicting exploratory gaze
Coleman Olenick
colenick@uoguelph.ca
▶ Show abstract
Computational models of visual attention have long recognised that where we look next depends on both prior fixations and the suppression of recently visited locations, encoding these as explicit fixation history and inhibition of return. However, these accounts overlook regions that were attended or inhibited without being directly fixated. Saccade trajectories encode this history directly by deviating toward covertly attended locations and away from inhibited ones, a pattern observed reliably in laboratory tasks but unexamined in natural scenes.To examine this generalization to natural scenes, we present a multi-stream convolutional model of free-viewing eye movements that incorporates saccade trajectory deviations alongside conventional spatial inputs, using limited prior eye movements to predict the next fixation. Rather than treating saccades as transitions between fixations, we assess whether trajectory features improve the behavioural plausibility of model predictions. Incorporating saccade deviations does not improve spatial fixation accuracy but significantly improves correspondence between predicted and observed saccade amplitudes, demonstrating that location and movement plausibility are dissociable model properties. These results position saccade deviation as an underutilized behavioural metric in attention modelling and motivate broader evaluation frameworks that assess not only where observers will fixate, but the exploratory range their eyes are likely to cover.
B55
Red Contextual Cues Facilitate the Perception of Happiness Across Cultures
Emma Yuan
yuane8@mcmaster.ca
▶ Show abstract
Colour has been shown to influence emotion perception, particularly with the colour red, which is commonly associated with anger and threat. However, red also has many positive associations, including love and passion. Furthermore, in Chinese culture, red is the dominant colour used in celebrations to signify fortune and auspiciousness. Previous research has found that red facilitates the perception of anger; however, its effect on the perception of happiness remains unclear. The present study examined whether red background colour facilitated the identification of happy facial expressions. Participants were presented with faces with happy or sad expressions on red or black backgrounds and categorized them as happy or not. Reaction time and accuracy were measured. The study included participants from university students from both China and Canada to examine whether the effect of red on happiness perception is consistent across cultures. It was observed that, in both culture groups, identification of happy faces was facilitated by a red background compared with a black background, but the perception of sad expression was little affected by the background colour. These findings contribute to research on the effect of contextual cues on emotion perception and have practical societal implications on visual communication.
B56
Seeing More Than Averages: a similarity test for ensemble stimuli
Mincheol Lee
► Show abstract
Ensemble perception refers to the rapid extraction of statistical information from multiple objects (ensembles), without individuating individual items, supporting tasks like rapid scene classification. This capacity is often explained in terms of representations of summary statistics, such as the mean and variability. However, summary statistics are insensitive to distributional shapes (e.g., distributions that share the same mean and variance but differ in shape). Recent findings suggest that people may retain richer information about feature distributions. Here, we test that possibility more directly using a similarity-judgment task. In a pilot study, participants viewed sets of oriented lines sampled from different distributions and judged which set was more dissimilar to a reference ensemble. This design asks whether perceived ensemble similarity is better captured by summary statistics-based measures or by the broader structure of the underlying distributions. Preliminary analyses indicate that participants’ judgments are better predicted by distributional similarity measures (e.g., Earth Mover's Distance) than by differences in summary statistics such as mean or variability of orientations. These findings are consistent with the view that ensemble representations are sensitive to global distributional structure. An ongoing main experiment will use a refined design and pre-specified analyses to test this hypothesis more rigorously.
B57
Seeking Structure in Uncertainty: Boredom proneness and meaning seeking
Ryan Chen
r449chen@uwaterloo.ca
▶ Show abstract
Boredom is a self-regulatory signal that motivates individuals to seek more meaningful activities, yet little is known about how this state translates into specific behaviors. The present study examined whether trait boredom proneness is associated with different pathways of meaning construction under ambiguity, including perceptual updating and endorsement of conspiratorial beliefs.Participants completed an ambiguous figures task in which one recognizable image gradually morphed into another. They indicated when they recognized the second object, providing a measure of perceptual updating in ambiguous contexts. Participants also completed questionnaires assessing boredom proneness, conspiratorial beliefs, perceived meaning in life, and sense of agency.Results showed that higher boredom proneness significantly predicted stronger endorsement of conspiratorial beliefs. Mediation analysis indicated that this relationship was partially explained by a diminished sense of agency: individuals higher in boredom proneness reported lower agency, which in turn predicted greater conspiratorial belief. Although boredom proneness did not predict perceptual switch points, exploratory analyses revealed that higher conspiratorial ideation and lower agency were associated with later pattern detection in the morphing task, suggesting reduced flexibility in perceptual updating.
B58
Single item priming in singleton visual search: Examining the role of prime masking
Alina Saad
faithful.scholar10@gmail.com
▶ Show abstract
In studies of singleton visual search, observers are faster at locating an odd-colored target when the target color repeats across trials, compared to when it switches. This performance benefit from repeating the target color is called priming of pop-out (PoP). Prior studies have consistently reported PoP across consecutive singleton search trials. However, PoP from a single item trial to a search trial is much less robust. We examined whether this pattern of results occurs because the target defining feature (e.g., colour) of a singleton search is attended only weakly when presented in the context of a preceding single prime. To explore this issue, we used a pattern masking method aimed at amplifying attention to single primes. In two experiments, we found that pattern masking of a single prime did not amplify the priming effect for a following singleton search target. However, we did find that postponing a response to the single item until after the response to the following singleton search did amplify this priming effect. We discuss these results in the context of dual process accounts of the inter-trial priming effects in visual search.
B59
Social Encoding Reduces False Alarms in Face Recognition
Salma Ben Messaoud
Sbenm025@uottawa.ca
▶ Show abstract
Introduction. Many studies have shown that performing social evaluations (e.g., personality attributes) leads to better recognition of target faces than performing perceptual evaluations (e.g., physical characteristics of faces). Based on the remember-know framework, we examined if this social encoding advantage extended to remember responses. Using a novel blocked design also allowed us to examine if encoding instructions influence false recognition of new distractor faces. Finally, we examined if gaze behaviours changed, as a function of encoding instructions. Method. Participants (to date N = 15) encoded target faces while performing social, perceptual, or absence (control) evaluations. During testing, participants were shown pictures of target individuals under different lighting and viewpoints, and new, distractor, faces. When test faces were identified as old, participants were asked to indicate if their response reflected a remember, know, or guess judgment. Eye movements were tracked during the study. Results. Results show that the social-encoding condition produced the highest proportion of hits, and lowest proportion of false alarms. Eye-tracking and remember-know analyses underway. Conclusion. To our knowledge, this is the first study to demonstrate that the social-encoding advantage applies, not only to recognition of target faces but also to rejection of distractor faces.
B60
Spontaneous attention towards faces of different races in adults and infants
Carie Guan
guanc7@mcmaster.ca
▶ Show abstract
Infants form categorical representations of faces. However, how categorical representations drive infants’ attention (spontaneous looking) remains unknown as traditional cumulative looking time paradigms neglect these processes.We developed a novel and highly effective paradigm investigating infants’ spontaneous attention towards own- and other-race faces by presenting a face in the peripheral visual fields while infants tracked a moving cartoon. Thus, gaze deviations from the cartoon’s moving trajectory indexed spontaneous detection. 71 Canadian infants (8 Asian, 63 White, 118 – 413 days, 37 females) and 73 adults (40 Asian, 33 White, 58 females) participated in the current study. We used Asian and White faces to examine race information’s impact on attention across 40 trials. Participants’ attention was measured by an EyeLink 1000 Plus (500Hz) eye-tracker.Preliminary analyses revealed that infants exhibited stronger spontaneous attention for own-race over other-race faces. This own-race bias, however, decreased significantly with age (r = -.38, p = .039). Conversely, adults demonstrated stronger attentional biases for other-race faces. We demonstrate that categorical representations modulate spontaneous attention across development. The transition from a robust infant own-race bias to an adult out-group alertness underscores the evolving nature of in-group/out-group face processing and validates the efficacy of our novel peripheral-attention paradigm.
B61
Statistical learning during visual search: feature-specific spatial regularity of the target
Rebecca Dong
dongr21@mcmaster.ca
▶ Show abstract
Implicit learning of spatial regularities has been widely established to enhance visual search performance. The spatial distribution of targets provides a reliable source of information that can be acquired without conscious awareness, unconsciously guiding attention toward high probability locations. However, it was unclear how implicit learning operates beyond location probability when the probability of the target features is also manipulated across trials. In this study, the visual search task involved one of the two target features, either colour (green) or shape (circle), among distractor stimuli (grey squares). Targets appeared in one of the eight possible locations, including two high-probability locations (HL, each 35%) in a diagonal arrangement and six low-probability locations (LL, each 5%). Each target feature was more likely to appear in a high probability feature (HF-HL, 30%), less likely to appear in a low probability feature LF-HL, 5%), and least likely in LLs (each 2.5%), creating three distinct experimental conditions. Reaction times were found to be faster (1) in HL than in LL conditions and (2) in HF-TL than in the LF-HL condition. These results suggest that participants can learn the location probability and also learn the feature-specific spatial regularity.
B62
The 3-dimensional advantage: Visual and tactile contributions to executive functioning in a 3-dimensional sorting task
Adam Szybunka-Ostopowich
szybunkaostopowicha@mymacewan.ca
▶ Show abstract
In the standard version of the Dimensional Change Card Sort (DCCS), children’s executive functioning is assessed by the ability to sort picture cards accurately according to one sorting dimension (e.g., colour) and then switching to another dimension (e.g., shape). In previous work, we demonstrated that a three-dimensional version of the DCCS enabled 3-year children to switch rules more flexibly than the standard version. In the present study, we examined the three-dimensional advantage directly by varying the availability of object features. Adult participants sorted polyhedrons with bivalent shape and texture dimensions in two shape form (simple, complex) and modality conditions (visual-tactile, tactile-only). In addition to accuracy, sorting time was used as a measure of executive functioning. While adult participants achieved perfect accuracy in switching their sorting behaviour in both shape form and modality conditions, their response times indicate that greater cognitive effort was required to switch between dimensions when sorting complex shapes on the basis of tactile features alone. These findings shed further light into the perceptual basis of cognitive flexibility and the use of timing to better characterize behavioural performance and age-related change in executive functioning.
B63
The Effects of Prioritization on the Allocation of Attention
Mimi Juffe
mj21zg@brocku.ca
▶ Show abstract
How is spatial attention affected by the allocation of visual working memory (VWM) resources at different times? To address this question, we used the capture-probe paradigm intermixed with a two-alternative forced choice (2AFC) VWM task. We manipulated VWM resource allocation by changing the priority of an item in the 2AFC task. The capture-probe letters were presented at the locations of the memory stimuli after a delay of either 700ms, 400ms or 100ms. Overall, VWM performance in the 2AFC and correct letter recall in the letter-probe increased as the likelihood of an item being probed increased. In the letter-probe trials, correct letter recall in the un-prioritized locations was lower at a 100ms delay compared to a 400ms delay, however, at a 700ms delay correct letter recall was not different from a 100ms or 400ms delay. Additionally, correct letter recall in the prioritized location at a 100ms delay was lower than at a 700ms delay, however, at a 400ms delay correct letter recall was not different from a 100ms or 700ms delay. These results reveal that the distribution of spatial attention follows the allocation of VWM resources and where these resources are distributed changes over time.
B64
The Impact of Auditory and Visual Noise on the Colavita Effect
Kate Weldon
kmweldon@mta.ca
▶ Show abstract
In multisensory integration research, visual information sometimes dominates auditory information. When auditory and visual stimuli are presented simultaneously, vision sometimes overshadows audition and participants only report perceiving the visual stimulus – something called the Colavita effect. Studies examining this effect are typically conducted in quiet laboratory settings, raising the question of whether the findings are ecologically valid. We investigated the impact of auditory and visual noise, meant to emulate everyday environments. We hypothesized that background noise would modulate the strength of the Colavita effect. Participants reported if they were presented with a visual, auditory, or audiovisual stimulus under three experimental conditions: no noise, visual noise, and auditory noise. The presence of noise – especially visual noise – generally slowed down response time and increased error rates for both unimodal and bimodal trials. This effect was stronger for visual noise than auditory noise, and visual noise generated a larger Colavita effect. Visual noise seemed to increase attention towards the visual stimulus, resulting in more ‘visual-only’ errors. These findings suggest that the Colavita effect may be stronger in natural conditions, particularly when visual noise is present.
B65
The Influence of Haptics and Vision on the Endowment Effect: An Eye-Tracking Study
Velika Kristianto
velika.kristianto@ubc.ca
▶ Show abstract
Haptics research, which explores the mechanisms and functions of touch, plays a key role in fields such as cognitive science, human-computer interactions, and consumer behaviour research. Despite its clear involvement in human experience, the mechanisms underlying haptic perceptual influences within applied settings are less understood than other sensory modalities. For instance, though the role of vision in influencing the endowment effect (where added value is gained by possessing an object) is established, the contribution of haptic input within this process of value construction is unclear. Recent work on multisensory integration suggest that touch interfaces may modulate subjective valuation (Brasel & Gips, 2014). However, the mechanisms through which this occurs remains to be determined. The present study explored whether haptic feedback (i.e., directly touching an object) influences both the magnitude of the endowment effect and patterns of visual attention during object evaluation. Participants (N = 56) either had both haptic and visual access to an object or only visual access. Eye-tracking data assessed visual attention and valuation was measured using endowment effect ratios. The findings indicate that haptic access is associated with differences in valuation and visual attention patterns compared to vision alone.
B66
The Influence of Lateral Biases in Spatial Judgments
Daria Chernova
chernovd@uregina.ca
▶ Show abstract
Lateralized brain function leads to spatial biases such as overestimating the leftward space. Typically demonstrated using the line bisection task, instructions are given to mark the centre of lines, but left-of-centre bisections are consistently made. These near space misperceptions appear to shift rightward in distant space, with more rightward bumps and deviations occurring when walking. Evidence suggests that lateral bias differences are robust in near space between left-to-right (LTR) and right-to-left (RTL) direction readers, with the present research investigating the gap in the literature around RTL readers’ lateral biases in distant space. Using the line bisection task and a walking and bumping paradigm, we measured near and distant space lateral biases of RTL and LTR readers, predicting rightward bumping among LTR readers and more left side bumping among RTL readers. 34 RTL and 70 LTR participants walked 1.9 metres, passing through a 68 cm aperture. Participants completed 20 trials, alternating starting foot and balancing a tray of glasses. The distance between the left-side of the doorway and the participant’s midline was measured each trial. Lateral bumps and deviations will be compared between groups and across tasks, with these findings contributing to understanding how perceptual biases are influenced by reading direction.
B67
The Influence of Posture and Valence for Body-Related Words in Visual Word Recognition
Cheleine Doyle
cdoyle@unbc.ca
▶ Show abstract
Embodied cognition postulates that our cognitive processes are at least partly rooted in our past bodily experiences (MacRae et al., 2022). For example, research has shown that maintaining upright postures tends to be associated with more positive affective states while slouched postures are associated with more negative affective states (Awad et al., 2021). Although posture and word-level characteristics such as valence have been shown to influence visual word recognition (VWR), there is a lack of research examining how one’s posture and the valence of body-related words interact in this process (Kuperman et al., 2014). To examine this question, we analyzed response latency data from a semantic categorization task using the criterion of “is the word body-related?”. Results demonstrated a significant main effect of posture where those who held an upright posture responded faster to both positive and negative body-related words compared to their slouched counterparts. However, neither the main effect of valence nor the valence X posture interaction was significant. These findings further substantiate the framework of embodied cognition, supporting the contention that the way one holds their body influences their ability to process incoming stimuli, including body-related words.
B68
Threshold-Dependent Biases in the Biological Motion Perception of Nervousness
Inci Eke
ieke052@uottawa.ca
▶ Show abstract
Humans readily extract social information from biological motion, but observer and stimulus characteristics can interact to bias these perceptions. This study investigates whether gender-related biases in recognizing nervousness are intensity-dependent. Male and female observers were presented with animated point-light walkers generated via MATLAB Psychtoolbox, featuring male and female models at four nervousness levels (control, mild, moderate, extreme). Participants rated perceived nervousness via keypress. Preliminary analyses (N = 28) reveal a significant three-way interaction between stimulus gender, nervousness intensity, and observer gender, F(3, 78) = 4.37, p = .007, η²ₚ = .14. Relative to female observers, male observers showed a systematic intensity- and gender-dependent underrating bias. Particularly, males rated male point-light displays as significantly less nervous than female displays at moderate (p = .011) and extreme (p = .012) intensities. This perceptual gap was absent at baseline (p = .932) and trended marginally at the mild nervousness level (p = .089). Data collection is ongoing to reach our target sample of 46 (determined via a priori power analysis). We expect that increased statistical power will further elucidate the interactions of observer and model gender in social cognition.
B69
Understanding Sound Intolerance: Investigating Measures of Hyperacusis and Misophonia in Emerging Adults
Lily Wortley
lily.wortley@gmail.com
▶ Show abstract
Decreased Sound Tolerance (DST) is a sensory experience characterized by unusually strong negative reactions to everyday sounds that do not bother most people. DST encompasses multiple subtypes, including misophonia, which involves severe emotional reactions to specific sounds, and hyperacusis, in which sounds are perceived as unusually loud or painful. DST can occur in individuals with normal hearing and reflects differences in how individuals perceive and respond to auditory stimuli. However, we lack clarity on how different DST subtypes present and relate to one another across commonly used measurement tools. Here, we examined DST in fifty undergraduate students (MAge = 20.19, SDAge = 5.06) from ages 18-35 using questionnaire measures and a psychoacoustic task. Participants completed multiple measures assessing misophonia and hyperacusis symptoms and related traits, followed by the Core Discriminant Sounds (CDS) task in which they rated the subjective loudness and unpleasantness of potential trigger sounds. These findings will explore relations between different DST measures and perceptual ratings of loudness and unpleasantness. Findings from this study shine light on how DST presents in emerging adults and inform how DST subtypes are best measured and distinguished in research and clinical contexts.
B70
Use of language and speaker cues during real-time language processing
Suevin Un
suevin.un@uwaterloo.ca
▶ Show abstract
Language processing is incremental and predictive – listeners use a variety of cues to predict upcoming content as each word unfolds.We explore adults’ use of language (i.e., verb) and speaker (i.e., age) cues to make predictions using an online Visual World Paradigm. In each trial, participants (N=60) heard a sentence spoken by an adult or child speaker (e.g., “I will drink the nice water” spoken by a child). The visual display included a target (e.g., child’s water bottle), a verb-consistent/age-inconsistent competitor (cup of tea), an age-consistent/verb-inconsistent competitor (hotdog) and a distractor (steak).Prior to the verb, participants directed their gaze to the age-consistent objects (target and age competitor). Following verb onset, looking increased to the verb competitor and decreased to the age competitor. These results show that adults use both cues to guide their interpretations. However, verb information is stronger: looks to the verb-consistent competitor increased even though it was inconsistent with the speaker’s age. In future analyses, we will examine whether listeners’ predictions were modulated by how strongly individual objects are associated with children vs. adults. This is the first work to show that listeners generate predictions based on a voice cue to a speaker’s social category (age).
B71
Visual imagery and its relationship to executive and precision-based components of visual working memory
Alyssa M.L. Thibeault
wy20rd@brocku.ca
▶ Show abstract
Multiple aspects of visual working memory (VWM) exist, each with possibly different relationships to visual imagery. This study examines the relationship between imagery and precision-based versus executive VWM. Given some studies describe imagery as “pictorial” (i.e., visually detailed), imagery strength was expected to correlate more strongly with performance in a continuous response task (i.e., VWM precision) than with filtering ability (i.e., executive VWM). Imagery was measured subjectively (Vividness of Visual Imagery Questionnaire/VVIQ; Vividness of Object and Spatial Imagery Questionnaire/VOSI) and objectively (binocular rivalry). Preliminary analyses (N = 29) revealed no correlations across any of the VWM and imagery tasks. Principal component analysis (PCA) with the preliminary dataset revealed two components representing executive VWM and self-reported imagery. Interestingly, VWM precision for load 1 (but not load 6) and imagery performance in the binocular rivalry task for specifically imagery of target houses both also loaded moderately onto the second component. Given the dissociation between imagery measures and executive measures of VWM, and evidence of a possible link between some measures of imagery and VWM precision, preliminary findings suggest that the relationship between imagery and VWM processes may vary depending on the aspect of VWM in question.
B72
Visual Profile Analysis in a Contrast Discrimination Task
Farhan A. Vaheed
abdulvaf@mcmaster.ca
▶ Show abstract
How do observers discriminate visual patterns? One view proposes that discrimination depends on changes in the distribution of responses across spatially localized filters tuned to spatial frequency, orientation, and motion. However, studies of spatial-interval (Morgan & Ward, 1985) and relative-phase discrimination (Badcock, 1984) suggest that observers rely on representations that explicitly encode spatial structure. Here, we tested whether observers use changes in the spatial luminance profile when performing a contrast discrimination task. Stimuli were f+2f compound gratings with contrasts of 0.1 and 0.05, respectively. A 3IFC task measured thresholds for detecting a contrast increment added to 2f. Thresholds were measured in a fixed baseline condition, where frequencies, contrasts, and orientations of the two components were constant, and in conditions where overall stimulus contrast (±0.3 log units), spatial frequency (±0.3 log units), or orientation (±15°) were randomized across intervals. In compound random conditions, both components were randomized by the same amount, preserving the spatial luminance profile. In component random conditions, only one component was randomized, altering the profile. Thresholds were significantly higher in component random conditions but not in compound random conditions. These results suggest that observers relied on changes in the spatial luminance profile to discriminate the contrast of 2f.
B73
"How Does That Make You Feel?": How Emotion Concepts are Impacted by Experiences of Trauma
Emma Power
emargaretp@mun.ca
▶ Show abstract
“How does that make you feel?” A question routinely asked in talk therapy, but how do we understand these feelings and the concepts like “Sad” or “Happy” we use to represent them? Embodied Cognition posits that the understanding of concepts, such as emotions, is facilitated by partial reactivation of sensorimotor information, suggesting that concepts are ‘grounded’ in bodily states. For example, when shown the word “kick,” corresponding motoric systems are reactivated. Importantly, interoception–the sense which enables perception of the body's internal states– grounds emotion conceptualization. If interoception is the foundation of emotion and crucial in comprehension, then what does this mean for those with altered interoception? Research has shown that experiences of trauma can alter interoception. In the present study, we investigated the grounding of emotional concepts in people with varied levels of traumatic experience. Participants were shown words and decided whether they were emotional, then indicated on a body map image where they felt each word. Preliminary results (n=12) indicate that participants with higher levels of traumatic experience– altered interoception– produce faster reaction times in word categorization for emotional words, and had differently organized body maps. Altogether, suggesting that disrupted interoception alters emotional understanding and conceptualization.
B74
A Loving Engagement: Exploring Predictors of Cognitive Engagement in Loving-Kindness Meditation
Scott McQuain
18slm10@queensu.ca
▶ Show abstract
Loving-Kindness Meditation (LKM) aims to foster feelings of love and warmth towards oneself and other living beings. Although originally a Buddhist practice, contemporary research finds consistent associations between LKM and improvements in psychological well-being. This archival study examines how personality (Big Five Personality Traits), meditation experience, social variables (attachment style, inclusion of the other in the self, desire for emotional closeness), as well as anxious and depressive symptoms predict average self-reported cognitive engagement in LKM. We analyzed data from 83 undergraduate women who received daily ~5-minute audio recordings of LKM instructions for one week, rating their engagement with LKM after meditating. Exploratory linear regressions found that only neuroticism (b = -0.71, SE = 0.20, p < .001), attachment anxiety (b = -0.46, SE = 0.13, p < .001), and anhedonic depressivity (b = -0.55, SE = 0.18, p = .002) predicted lower engagement in LKM; only conscientiousness predicted higher engagement in LKM (b = 0.56, SE = 0.25, p = .03). These results may suggest that individuals with maladaptive psychological patterns have more difficulty engaging in LKM. Future research could explore how LKM engagement influences psychological outcomes for these individuals as well as potential changes in LKM engagement over time.
B75
Building Trust One Character at a Time: How AI Trust is Impacted by Content Presentation
Monica Tsang
jacoblevioliveira10@gmail.com
▶ Show abstract
Trust plays an important role in how people interact with conversational artificial intelligence systems. In the present research we examined whether a common interface feature, typing animations that reveal responses gradually, can influence user trust in AI systems. Across two experiments, participants interacted with AI agents under different response presentation conditions in which answers appeared either instantly or with visible typing animations. In Experiment 1, participants (N = 71) asked a series of factual questions to three AI agents that provided pre-scripted responses, allowing response presentation speed to be manipulated while holding response content constant, such that responses appeared either instantly, with fast typing, or with slow typing. After each response, participants rated both their trust in the AI agent and the response content. Typing animations increased trust in the AI agent relative to instant responses even when response accuracy varied. Experiment 2 (N = 68) replicated and extended this effect using a large language model and allowing participants to generate their own questions, with typing animations increasing trust in both the AI responses and the AI agent. Together, these results suggest that simple interface features such as typing animations can influence trust in AI systems independently of response accuracy.
B76
Can We Trust Experts’ Data in Judgments of Where We are Heading? Evaluating Indicators used in Expert Judgment of Human Progress and Social Change
Peter Diep
pndiep@uwaterloo.ca
▶ Show abstract
Policymakers and the public increasingly rely on indicators to track societal progress around the world and guide interventions on critical challenges from climate change to economic inequality. Yet the validity and reliability of indicators of societal progress remain largely unexamined. We systematically bench-marked 99 popular indicators across eight domains critical to human flourishing, as identified by 24 experts from diverse disciplines. Evaluating indicators against six foundational measurement criteria of definability, quantifiability, data availability, global representativeness, transparency, and temporal resolution, only 31% of expert-endorsed indicators met the standards necessary for valid inference. Strikingly, experts’ rankings of domain importance were inversely related to the quality of available indicators, with even the highest-priority domain (climate) having substantial measurement gaps, suggesting insufficient development of reliable indicators in critical domains. These findings reveal a fundamental disconnect between what we need to measure and what we can reliably measure, raising questions about the empirical foundation of global policy initiatives. Our results underscore the urgent need for investment in robust, transparent, and globally representative measurement systems, particularly as political pressures increasingly threaten data integrity and availability.
B77
Compassion Fatigue: Caring to Capacity or Resource Depletion?
Carolyn Stone
carolynstone@trentu.ca
▶ Show abstract
Compassion is a multi-faceted construct involving cognitive, affective, and behavioural components. Although compassion often feels automatic, prolonged, intense caregiving can result in a state of physical, emotional, and spiritual exhaustion referred to as compassion fatigue. Emerging theory suggests that elements of compassion may require effortful self-regulatory processes. To test whether compassion draws upon the same limited-capacity cognitive resources as self-control, the present study used a modified version of the sequential task paradigm. A Go/No-Go task was used as the induction task. Self-control during the induction task was measured using boredom (Multidimensional State Boredom Scale) and effort (Hsu et al., 2017). The outcome task was a state compassion questionnaire. Results indicate that multidimensional components of boredom and effort differentially predict distinct components of compassion. The implications for theories of compassion and compassion fatigue are discussed. Keywords: Compassion fatigue, cognitive resources, self-regulation.
B78
Emotion Regulation and Well-Being: Consequences of mild traumatic brain injury in older adults
Tala Tayem
ttayem@uwaterloo.ca
▶ Show abstract
Long after experiencing a mild traumatic brain injury (mTBI), individuals report lingering cognitive and emotional regulation challenges. Known brain changes associated with aging, particularly within the frontal lobes, may exacerbate any long-term consequences of mTBI. This large-scale study examined consequences of mTBI in a large sample (N = 8,779 control and N = 2,010 mTBI) of older adults from the Canadian Longitudinal Study of Aging. We compared performance on assessments of cognitive functioning and self-report of mental health symptoms as well as life satisfaction, in individuals with and without a self-reported mild head injury in their past. We also investigated whether sex changed patterns of findings. Individuals with mTBI reported significantly lower life satisfaction, and had higher scores on assessments of psychological distress relative to controls. There were no reliable differences between groups on tests of cognitive functioning, though the mTBI group scored higher on verbal fluency. Regardless of group, females relative to males showed greater verbal fluency but endorsed lower life satisfaction and more symptoms of depression. Findings suggest poorer emotion regulation and mental well-being may be long-term consequences of a mild head injury.
B79
Event-related potentials to outcome valence are modulated by the decision to explore or exploit
Eunchan Na
eunchan@ualberta.ca
▶ Show abstract
Maintaining a balance between exploitation and exploration, also known as the explore-exploit dilemma, is the key to optimizing decisions in uncertain environments. While many studies have investigated the effect of the number of trials on the explore-exploit dilemma, little is known about the effects of response availability and its neurophysiological correlates. We investigated how exploitation, informative exploration, and random exploration would be modulated during 4- and 6-response versions of the same game while measuring electroencephalogram (EEG) activity. The game provided two types of feedback: early feedback, providing information regarding positive (win) or negative (loss) outcome valence, and late feedback, providing information on outcome magnitude. At the behavioural level, exploitation increased post-win and as response availability decreased, random exploration increased post-loss and as response availability increased. Informative exploration was not modulated by either outcome valence or response availability. At the neurophysiological level, we observed larger amplitudes for feedback-related negativity (FRN) and P3 when participants explored and exploited, respectively, but only in response to early feedback. These results suggest that 1) increased response availability promotes random exploration while discouraging exploitation, and, 2) event-related potentials in response to outcome valence are modulated by our prior decision either to explore or exploit.
B80
Evidence-Based Persuasion in Polarized Contexts: Testing the Impact of a Chatbot Intervention on Political Beliefs
Sze Yuh Nina Wang
► Show abstract
Evidence-based arguments are thought to be effective at persuasion: we update our beliefs in response to new information. However, motivated partisan reasoning accounts predict that evidence-based arguments may have no effect or even backfire in highly polarized contexts, leading people to become further entrenched in their initial beliefs. We test this hypothesis by having Republicans converse with a LLM chatbot. The chatbot presented participants with information about unconstitutional (Study 1)/corrupt (Study 2) actions taken by President Trump and the harmful effects of DOGE (Study 3; N ≈ 1,000 for each study with participants recruited from Prolific). On average, about a quarter of participants showed a change in beliefs about the topic of conversation of at least 15 percentage points, with a smaller decrease in general support for Trump. Additionally, we find that these effects persisted even when the chatbot was labelled as being trained on Democrat-leaning sources and was adversarial in tone, suggesting that people are much more receptive to information that counters their political beliefs than motivated reasoning theories would predict (Study 4). These studies demonstrate that people can be persuaded by evidence-based arguments even on highly polarized political topics.
B81
Examining Age Differences in Moral Information Search
Mane Kara-Yakoubian
mkarayakoubian@torontomu.ca
▶ Show abstract
Moral decision-making varies by age, with older (vs. younger) adults favouring deontological (duty-based) over utilitarian (outcome-based) approaches. Notably, prior work has overlooked the role of information search processes preceding moral judgment, and has relied heavily on sacrificial dilemmas. The present research examined age differences in moral decision-making and the role of information search in shaping moral judgments, using a set of non-sacrificial dilemmas. In Study 1A (N = 130), younger (18-35) and older (60+) adults revealed moral information (principle, outcome, general context) before rating moral concern and behavioural preference (deontological to utilitarian) across 20 dilemmas. Compared to younger adults, older adults spent more time searching, reported greater moral concern, and showed a stronger preference for deontological action. Information search predicted moral judgment, such that greater selectivity for morally relevant information was associated with increased moral concern, and preference for principle-based information predicted greater deontological responding among younger adults, but not older adults. In a follow-up experiment (N = 128) without an information search component, these age differences disappeared. Taken together, these findings suggest that age differences in moral judgment depend on how moral information is encountered, emerging under conditions that allow for active information search, rather than passive information exposure.
B82
Mapping Cognitive and Emotional Outcomes After Traumatic Brain Injury: Comparisons with Orthopedic Injury and Healthy Controls
Jeremy Brand
jbrand9@uwo.ca
▶ Show abstract
Cognitive and emotional impairments often co-occur and are highly heterogeneous after traumatic brain injury (TBI). This study mapped long-term cognitive and emotional outcomes in survivors of TBI and orthopedic injuries compared with healthy age- and sex-matched controls. Sixty-four survivors (20 moderate-to-severe TBI, 17 mild TBI, 27 orthopedic injury) completed seven neurocognitive tests and assessments of anxiety and depression at four time points over one year post-discharge. At the first time point, cognitive outcomes showed short-term memory impairments in the moderate-to-severe TBI and orthopedic groups, reasoning impairments in the mild TBI group, and verbal impairments across groups. Linear mixed-effects models indicated short-term memory improvements in the moderate-to-severe TBI and orthopedic groups, persistent reasoning deficits in the moderate-to-severe TBI group, and verbal improvements across groups. Emotional outcomes showed low-to-mild anxiety and depression in mild TBI and orthopedic injury groups, but persistent moderate levels in the moderate-to-severe TBI group. Depressive symptoms partially accounted for reasoning scores in the moderate-to-severe TBI group relative to the other injury groups. Older age predicted worse cognitive outcomes, and females reported worse anxiety and depression. These findings highlight the need for an integrative model incorporating cognitive and emotional domains, injury severity, sex, and age to improve prognostic precision.
B83
More Than a Game: How Gaming Motivations Shape Online and Offline Social Capital
Noor Alyafei
nhuss012@uottawa.ca
▶ Show abstract
Social capital (i.e., networks, relationships, and norms of trust and reciprocity) plays a crucial role in forming new friendships and maintaining existing familial and peer relationships in emerging adulthood. This study examined the relationship between different modalities of social capital (online and offline) and motivations for video game play, while controlling for personality traits, hours spent gaming, age, and gender. A total of 342 first-year undergraduate students (220 women, 109 men, 9 gender+; M age = 20.25) participated in the study. Participants completed the Internet Social Capital Scale, the Ten-Item Personality Inventory, and the 12-item Game Motivation Inventory. Hierarchical linear regression analyses were conducted to examine whether gaming motivations were associated with online and offline social capital. The results indicated that social and immersion motivations, alongside extraversion, were significantly related to higher online social capital; immersion motivation and extraversion were associated with offline social capital. These findings suggest that specific gaming motivations and personality traits play an important role in the development of online and offline social networks. The results provide valuable insights into how underlying reasons for video game play foster social capital, offering implications for understanding the relationship between leisure activities, relationships, and well-being.
B84
On the Attribution of Criminality from the Rapid Presentation of Faces
Jason Phonchareon
jason.phonchareon@unb.ca
▶ Show abstract
There is growing interest in whether socio-evaluative traits (e.g., trustworthiness, competence, and criminality) can be accurately derived from faces. The present study assessed whether criminality could be correctly ascertained from the faces of a small sample of incarcerated and non-incarcerated individuals. Specifically, it examined whether participants can accurately judge a person’s criminality based on viewing an image of their face. Across two studies, 49 participants completed a criminality judgment task following the rapid presentation of a neutral face (i.e., belonging to a criminal or innocent person). A signal detection model and an ERP analysis of the face-sensitive N170 tested both the behavioural and neural responses when making a decision. Across two experiments, incarceration judgements were poor and not significantly different from chance. However, a larger N170 (an event-related potential component well-known to be sensitive to face processing) was observed for incarcerated individuals. This effect did not replicate in a second experiment. Our findings suggest that the accurate attribution of criminality from pictures is very unlikely, and are instead based on inaccurate implicit biases.
B85
Paranoia predicts perceptions of untrustworthiness, but not memory for faces
Ilana Davids
► Show abstract
Individuals with paranoid ideations tend to be mistrustful of others' intentions, which can make it difficult to navigate their social environment. Such biases may extend to other cognitive processes. For example, faces perceived as untrustworthy tend to be remembered better than trustworthy ones. While research supports a relation between paranoia and perceptions of trustworthiness, findings have been mixed in nonclinical samples. We investigated the relations among paranoia, perceptions of trust, and subsequent memory for faces in a sample of 224 undergraduates. Participants rated their perceptions of predefined sets of trustworthy and untrustworthy faces and subsequently completed a surprise recognition test in which they rated how confident they were that they had seen them before or were new. Results demonstrated a robust untrust memory effect: participants remembered the untrustworthy faces more confidently. While greater levels of paranoia related to lower perceived trustworthiness, paranoia did not significantly relate to the untrust memory effect. Moreover, the untrust memory effect was not explained by differences in trustworthiness ratings between the face sets. Our findings support that while subclinical paranoia relates to lower trustworthiness judgments, it fails to predict subsequent memory confidence, suggesting a social-cognitive bias in face perception that may not extend to memory.
B86
Predicting Knowledge of Psychological Misconceptions: The Roles of Actively Open-Minded Thinking and Intellectual Humility
Natalie Joly
nataliecjoly@gmail.com
▶ Show abstract
Psychological misconceptions are widely spread through media, social interactions, and online platforms. Identifying factors that reduce susceptibility to these misconceptions is critical in our modern world. This study examined whether actively open-minded thinking (AOT) and intellectual humility (IH) are associated with greater knowledge of psychological misconceptions, and whether AOT and IH are related to each other. Undergraduate students were recruited through York University’s Undergraduate Research Participation Portal. Participants completed the Actively Open-Minded Thinking Scale (AOT-13), the Comprehensive Intellectual Humility Scale (CIHS), and the Test of Psychological Knowledge and Misconceptions (TOPKAM). Correlational analyses revealed a strong positive relationship between AOT and IH, and a weak but significant association between AOT and knowledge of psychological misconceptions. However, no significant relationship was found between IH and misconception knowledge. Regression analyses indicated that AOT accounted for 2.7% of the variance in misconception scores. These findings suggest that being open to multiple viewpoints may slightly reduce susceptibility to psychological misconceptions, while intellectual humility does not appear to have a direct effect. This highlights the potential of fostering actively open-minded thinking in efforts to reduce psychological misinformation.
B87
Psychological Factors Associated with Depressive Symptoms Among Women with Polycystic Ovary Syndrome
Freya Anderson
fande057@uottawa.ca
▶ Show abstract
Women with polycystic ovary syndrome (PCOS) are more than two and a half times more likely to experience depression compared to individuals without the condition. Despite this elevated prevalence, the mechanisms underlying depressive symptoms in PCOS remain poorly understood and are often underrecognized in clinical settings. This study examined how emotional experiences, psychological pressures, and social circumstances associated with PCOS relate to depressive symptoms. A total of 179 women with a medical diagnosis of PCOS were recruited through online platforms, PCOS support groups, and gynaecology clinic networks to complete an online survey. This survey assessed demographic and clinical characteristics, depressive symptoms, and well-being across six PCOS-related domains: emotional well-being, fertility concerns, sexual function, weight-related concerns, menstrual irregularities, hirsutism, and coping. Results indicated that weight-related concerns represented the most distressing domain among participants. Overall, women with PCOS reported moderate levels of depressive symptoms and a mean body mass index (BMI) within the obese range. Regression analyses revealed a significant positive relationship between BMI and depressive symptoms. These findings suggest that cognitive appraisal of weight-related symptoms may play a central role in depressive outcomes among women with PCOS and highlight the need for care models that integrate mental health support.
B88
Referential Cues Reveal General Effects on Memory and a Unique Role for Gaze
Paris Y. Wang
► Show abstract
Previous studies have shown that communicative eye contact from a live speaker can facilitate memory among female but not male participants. The present study examined whether such effects generalize beyond eye gaze to other referential cues. Across two experiments with female pairs, pointing and verbal naming were used to signal the intended recipient of spoken words. In Experiment 1, an investigator pointed at the participant, the partner, or neither person while reading words. In Experiment 2, the referential cue consisted of calling the recipient's name. Across both experiments, recognition memory was highest when participants themselves were signaled and lowest when their partner was signaled. Because the memory effects of these cues were consistent with prior eye-gaze findings, Experiment 3 examined whether memory in male participants could be improved by a cue other than gaze. We found that recognition memory was higher when the investigator pointed at the participant than when no gesture was made. Together, these findings suggest that memory benefits are not specific to eye gaze but reflect a broader influence of communicative signals. At the same time, findings in male participants suggest that eye gaze may not be fully interchangeable with more overt referential cues such as pointing.
B89
Seeking or avoiding mutual gaze: Quantifying gaze coordination across communication media
Avantika Utam
utamavantika2@gmail.com
▶ Show abstract
Mutual gaze contains information that supports successful communication. Previous research has typically characterized mutual gaze using aggregate measures, such as total duration or frequency. While informative, these measures overlook the dynamic coordination of gaze between partners, including when mutual gaze is sought or avoided. Here, we examined the coordination of mutual gaze during communication and compared this behaviour between face-to-face interaction and Zoom where mutual gaze does not imply genuine eye contact. We recorded eye gaze and head orientation from dyads while they played a game of Heads-Up in three conditions: (1) regular face-to-face, (2) face-to-face through a screen-sized window cut into a barrier (to control for the smaller subtension), and (3) over Zoom using two iPads inserted into the same barrier. Results showed that the probability of initiating mutual gaze after direct gaze onset was greater than what was expected by chance in the face-to-face conditions, but not in Zoom, suggesting that communication over this platform affects typical mutual gaze seeking behaviour. Further, the probability of initiating mutual gaze after direct gaze onset was dependent on the gaze angle of the responder from the initiator. Together, these findings provide insight into dynamic mutual gaze behaviour across communication media.
B90
Social Competence and Hot and Cool Cognitive-Executive Performance in Typical Development and After Pediatric Brain Tumour Treatment
Katie Wade Alonso
► Show abstract
Social competence is a core component of adaptive functioning, but its relationship to cognitive-executive processes (i.e. processing speed and cognitive control) remain unclear, particularly after pediatric brain tumour treatment. Using a novel task that assessed hot (incentivized) and cool (unincentivized) processing speed, response inhibition, and interference control, along with multi-informant assessments of social skills, performance, and adjustment, we assessed typically developing children (TDC; n=25) and children treated for brain tumours (n=25). In TDC, response speed and accuracy improved with age, and participants responded more quickly in hot than cool conditions, indicating greater engagement when performance was incentivized. Hot cognitive control showed the most consistent associations with social competence, whereas cool functions provided modest protective effects. Children treated for brain tumours showed comparable task performance to TDC but significantly lower social competence across informants. Path modeling revealed that younger age, female sex assigned at birth, greater treatment intensity, and more treatment complications predicted poorer social competence both directly and indirectly through task performance. Across task measures, cool processing speed performance showed the strongest association with social outcomes. These findings suggest that motivational context is more informative for social development in typical populations, whereas processing speed is more relevant after brain injury.
B91
The Ecology of Extremism: How Compromised Meaning-Making Accelerates Authoritarianism
Neil Wegenschimmel
nhwegens@uwaterloo.ca
▶ Show abstract
In an era of information saturation, the ways individuals ascribe credibility to what they encounter are central to understanding ideological extremism. Drawing from Durkheim’s concept of anomie—normative breakdowns that leave individuals feeling ungrounded and disconnected—and classic theories of the authoritarian personality, this research investigates how affective instability, catalyzed by modern media ecologies, shapes authoritarian attitudes across the political spectrum. We conceptualize authoritarianism not merely as an outgrowth of political beliefs, but as a dynamic response to compromised meaning-making. Across three U.S. samples, we demonstrate that affective states like uncertainty and loneliness cultivate maladaptive epistemological orientations: conspiracy mentality (CMS) and existential nihilism (ENS). These orientations, in turn, differentially predict right-wing and left-wing authoritarianism (RWA and LWA). Crucially, information consumption functions as a powerful environmental accelerant in certain instances in predicting authoritarian outcomes. Results reveal that distinct information sources amplify these parallel epistemic pathways. Ultimately, this research demonstrates that meaninglessness and suspicion are as impactful as misinformation itself, but that information environments also matter, and that democratic stability depends heavily upon how media ecologies either buffer or accelerate the transformation of anomie into authoritarianism.
B92
The Power of the Pause: Executive Function and Social Problem Solving in Young Adults
Sabrina Perry
sperry@trentu.ca
▶ Show abstract
Social relationships become increasingly complex during young adulthood. The ability to resolve social conflicts, or social problem solving, influences the quality of these relationships and is critical for well-being in young adults. Executive function, higher-order cognitive processes that regulate thoughts and behaviour, also undergoes significant development during this time. Although social problem solving and executive function have been linked, few studies focus on young adults. Hence, the current study examined social problem solving and three components of executive function (updating of working memory, shifting of mental sets, and inhibition of prepotent responses) in 167 young adults aged 18-29 years (M = 19.95 years, SD = 2.29). Analyses revealed that inhibition predicted the problem identification step in the social problem solving process as well as the overall social problem solving score. Inhibition also predicted higher odds of responding to a social conflict by accommodating to the needs of others rather than the needs of the self. Shifting and updating were not associated with any aspect of social problem solving. These results suggest that the ability to inhibit prepotent responses may be beneficial for effective social problem solving by allowing individuals to consider the thoughts and emotions of others before their own.
B93
The Social Brain Endures: rTPJ Activation Predicts Real-World Network Size in Older Adults Despite Mentalizing Declines
Sarah Saju
► Show abstract
Social disconnection in later life is a growing public health concern. Both structural aspects of social connections (reduced number of social contacts) and experiential elements (perceived loneliness) are linked to poorer health, yet their neural bases may differ. Prior work showed that neural patterns associated with mentalizing—the ability to understand others' mental states—predict variance in the number of social contacts in young adults. Whether this generalizes to older adulthood—when mentalizing may decline—and if it extends to experiential aspects of social connection, remains unclear. To address this, we applied multivoxel pattern analysis to fMRI data from 41 older adults (65–77 years) performing a task eliciting mentalizing and empathy (EmpaToM). Mentalizing-related activation in the right temporoparietal junction predicted the number of social contacts, despite age-related declines in behavioral mentalizing, but did not reflect experiential measures such as perceived loneliness or social support. Supplemental analyses of empathy-related brain regions showed no associations, indicating specificity to mentalizing. These findings suggest that preserved neural representations of mentalizing may help maintain social ties in later life, while being distinct from subjective experience. By dissociating structural from experiential dimensions, our results clarify the neural mechanisms supporting social connectedness and resilience in aging.
B94
Thinking about Autism: Mixed Method Insights into Representations and Impressions of Autism
Natalia Van Esch
vane4950@mylaurier.ca
▶ Show abstract
Autism is a common and highly variable neurodevelopmental condition; however, it remains widely misunderstood and stigmatized. Autistic individuals are often evaluated more negatively than non-autistic individuals yet, little is known about the representations of autism that underlie these judgments. This study examined how children and young adults conceptualize and perceive autism. Sixty-five school-age children created autistic and non-autistic characters and rated them on social traits and behavioural intentions. Characters were then rated by two-hundred-fifty-one undergraduates, with all participants additionally providing open-ended responses describing and defining autism. Autistic character representations were similar in variability to non-autistic characters, though common autistic characters tended to be more odd, non-human, and stereotypical. Furthermore, autistic characters were consistently rated more negatively by children and undergraduates. Undergraduates were also able to identify autistic characters above chance, suggesting that shared stereotypes may guide these judgments, with undergraduates often citing the usage of sensory accommodations as a cue for autism. Qualitative analyses revealed diverse and sometimes contradictory definitions and descriptions of autism, though participants frequently emphasized broad variability as core to autism. Together, these findings suggest that while autism functions as a broad and diffuse social label, the label itself continues to evoke consistent negative social evaluations.
B95
Social media filters’ influence on the perception of attractiveness and sociability of male faces
Florence Mayrand
florence.mayrand@mail.mcgill.ca
▶ Show abstract
Social media include a diverse set of tools for ‘beautifying’ facial appearance. The impact of such tools on viewers’ impressions of attractiveness has so far only been studied for woman’s images. These data indicated that women were rated as more beautiful when moderate beautifying filters were applied, but as having less social qualities. Labels indicating image editing exerted no effects on ratings. Here we examined how increasing facial attractiveness via these apps influenced the perception of attractiveness and sociability of male faces. We also examined if knowledge of photo editing modulated these ratings. To do so, we digitally ‘beautified’ male face images sourced from validated databases using 0%, 25%, 50%, 75%, 100% slider modification in a common app. Half of the images were randomly labeled as digitally altered while the other half were marked as not edited. Participants reported attractiveness and sociability for each image. Participants’ ratings of attractiveness and sociability increased as the degree of photo editing increased. Editing labels did not affect these effects. Hence, photo editing appears to increase perceived attractiveness and sociability of males with information about photograph editing exerting no reliable effect on those ratings.
B95
Valenced stimuli bias the content but not the frequency of mind-wandering episodes
Charmi Pastagiya
cpastagiya@mta.ca
▶ Show abstract
Mind-wandering (MW) refers to task unrelated thought and is associated with heightened neuronal activity in the default mode network. The locus coeruleus norepinephrine system, on the other hand, is implicated in keeping thought on task. It is understood that MW can be intentional or unintentional, however what dictates the content of MW during unintentional episodes in unknown. The present study explores whether cues in the environment can bias the content of MW episodes. Participants rated a series of images on their valence (i.e., positive, neutral, or negative), with each image type paired with a distinct colour. They then completed the Sustained Attention to Response Task (SART). The SART was presented in blocks, with each block bordered by a colour that had been presented in the previous phase. Thought probes were used to assess the frequency and content of MW. The results show that overall frequency of MW did not differ across conditions. However, the frequency of negative MW was more pronounced on SART trials associated with emotional valenced colours than trials associated with the neutral colour. These results suggest encountering valenced stimuli does not influence the frequency of MW, but increases the affective nature of those episodes.
B96
Facial expression recognition: visual information utilization is more informative than ocular fixations
Pénélope Pelland-Goulet
penelope.pelland-goulet@uqo.ca
▶ Show abstract
ADHD affects 3-4% of Canadian adults. Its main symptoms are inattention, hyperactivity and impulsivity. ADHD is often associated with cognitive deficits, mainly sustained attention and working memory. There exists a large body of research comparing resting EEG power in different frequency bands between neurotypical individuals and individuals with ADHD. However, little research has focused on young adults with ADHD and the existing literature rarely accounts for age and gender. In addition, there is very limited data regarding the links between resting EEG and cognition for this population. The goal of this study was to examine the link between ADHD, cognition and resting EEG activity while controlling for, and describing, the effects of gender and age. Hierarchical regression analysis was used to evaluate how much variance in EEG power was accounted for by the variables of interest. Results indicated elevated power across frequency bands and electrode sites in women relative to men, and a reduction of delta power with age. Strikingly, ADHD diagnosis was not associated with any differences in EEG activity compared to neurotypicals. Stronger beta power was associated with poorer performance in sustained attention and working memory tasks. The results are interpreted in light of recent literature.
B97
Disregard disgust, focus on fear: Categorical emotion effects on L2 lexical decision times
Aya Amer
ayaamer@cmail.carleton.ca
▶ Show abstract
Although reading emotional words generally prompts an affective response, bilinguals seem to be “detached” from the emotional content of words in their L2 (Aguilar et al., 2024; Pavlenko, 2012). Previous studies of this phenomenon have defined emotion by a combination of valence and arousal, with no consideration of categorical differences between affective states. We investigated bilinguals’ reduced sensitivity to L2 emotional words using a lexical decision task. L1 English speakers (n = 148) and L2 English speakers (n = 119) performed lexicality judgements on words that differed in their valence, arousal, and relatedness to one of the five basic categorical emotions (happiness, sadness, fear, anger, and disgust). We found that happiness- and fear-related words facilitated response times in L1 speakers, while words that were more strongly related to disgust prompted slower reaction times. Happiness and disgust effects were absent in L2 speakers; however, the facilitatory effect of fear was preserved. Crucially, categorical emotion ratings captured the variance in reaction times better than valence and arousal considerations. Not all L2 participants displayed equal indifference to the emotional dimensions of words: a combination of early age of acquisition, high proficiency, and frequent daily use of English resulted in more native-like reactions.
C01
A Glimpse Into the Past: Assessing the Validity of Recalled Attentional States Using Pupillometry
Aahana Uppal
20au6@queensu.ca
▶ Show abstract
Prior work has established that we can broadly recall how attentive we are, with recent studies demonstrating that we can remember specific moment-to-moment fluctuations in recalled attentional states. However, it is unclear how accurately these recalled attentional states reflect objective measures of attention. One means of capturing objective measurements is through pupil dilation, which fluctuates with in-the-moment attentional states. Therefore, in this study, we investigated the temporal overlap between subjective and objective recalled attentional states using pupillometry. To do so, we recruited participants to complete (1) a Video-Viewing Phase, wherein they watched a 30-minute video lecture and intermittently reported their in-the-moment attentional states while having their pupil dilation recorded using an eye tracker, followed by (2) a Stimulated Recall Phase, wherein they reported their recalled attentional ratings when viewing clips from the original video. When assessing the temporal overlap between in-the-moment and recalled attentional ratings, preliminary results replicated prior work showing low-to-medium temporal overlap as time lag increased. Further analyses will investigate whether objective measures of attention (assessed through pupil dilation) overlap with in-the-moment and recalled subjective states of attention (assessed through ratings), providing critical information about the precision of attention and the specificity of memory.
C02
Aha! Investigating the effects of spontaneous and induced insights on incidental memory
Linda He
linda.he@mail.utoronto.ca
▶ Show abstract
Previous research has established that solving a problem through insight bestows a memory advantage. Insight can be self-generated or induced by revealing the problem solution. Despite exhibiting potential differences in affective salience and prediction error, characteristics known to modulate memory, a recent study indicates that both insight types exhibit equivalent incidental memory benefits. To explore this tension, we reinvestigated the insight memory advantage in a different memory modality (visual stimuli) and utilized the Mnemonic Similarity Test, a test of detailed memory. With compound remote associates as insight problems, participants spontaneously generated solutions on some trials, and were shown solutions on others. Following each problem, participants incidentally encoded object images, memory for which was probed at a surprise test. Counter to prior work, we found a detrimental effect of insight, regardless of type, on memory for incidentally encoded images. Overall, this investigation constitutes the first known replication study of this effect, with its findings underscoring the nuanced dynamics between insight and incidental memory.
C03
Associations Between Objective Instagram Behaviours and Psychological Traits
FengWei Pi
f2pi@uwaterloo.ca
▶ Show abstract
Research on the psychological impact of social media has yielded mixed results, possibly due to a reliance on self-reported social media use. The present study used a longitudinal design and custom data-processing scripts to extract and analyze participants' objective Instagram account data across two sessions (N=202). Participants also completed self-report measures of everyday (in)attention, affect, and perceived Instagram use. Between-session correlations for Instagram actions were high, but cross-action correlations were low, indicating reliable, but distinct behaviours. Objective Instagram actions did not correlate strongly with any self-reported usage measure. Self-reports of inattention were negatively correlated with browsing activity but positively correlated with both the proportion of high-engagement actions and interactions with followed users. These results demonstrate the value of granular, objective behavioural analysis in understanding the nuanced psychological effects of social media use. Future research should examine the psychological motivations underlying specific social media behaviours.
C04
Building a Better Battery: Developing Measures of Gaze Control
Natalia McCullough
natalia.mccullough@sasktel.net
▶ Show abstract
With eye tracking, our eyes can control user-interfaces— this is known as gaze control. Gaze control applications allow people to draw, make music, and play video games with their eyes. Given individual differences in eye movements, the relation between eye movements and frontal lobe functioning, and having to control eye movements to use these applications, the current study had two goals: (1) to develop and administer a battery of gaze control measures, and (2) to determine if individual differences in gaze performance are attributable to cognitive factors including attention control (AC) and working memory capacity (WMC). Participants (N= 92) completed a Gaze Control Battery (GCB) containing four gaze control measures. Participants also completed AC and WMC tasks. The results indicated that three of the GCB tasks were reliable. However, AC and WMC did not predict performance on the GCB. Since most participants achieved high scores on the battery, the tasks may have been too easy to find individual differences, and thus making the tasks harder would reduce ceiling effects. Nonetheless, the study provided more information for creating better measures of gaze control and improving user experiences. Keywords: Voluntary Eye Movements, Gaze Control, Human-Computer Interaction, Midas Touch Problem
C05
Does Posture Affect Sustained Attention?
Jenna M. Bolzon
jennabolzon@trentu.ca
▶ Show abstract
People who use standing desks in the workplace report that standing improves their cognitive performance (Straub et al., 2022), yet laboratory evidence remains elusive. Initial research examined whether standing improves selective attention (Rosenbaum et al., 2017; 2018; Smith et al., 2019). This research has been unable to produce consistent effects of posture on selective attention (Bolzon et al., 2024; Caron et al., 2020; 2022; Straub et al., 2022). Here we hypothesize that sustained attention may better capture the real-world cognitive demands of working for extended periods. In order to test the hypothesis that standing will improve sustained attention, participants completed the Sustained Attention to Response Task (SART) in sitting and standing postures with order counterbalanced across participants. Multiple indices of sustained attention were examined. There was no evidence that posture affected sustained attention. Future research will continue to examine what may be the critical connection between posture and attention. If posture can affect attentional state, there are serious implications for real-world situations where attention lapses could have detrimental outcomes such as healthcare, air traffic control, and the military.
C06
Doomscrolling your attention away: Investigating the effect of continuous versus interrupted TikTok content on inhibition and task switching performance
Dana Hayward
dana.hayward@ualberta.ca
▶ Show abstract
With the current rise of short-form videos on social media, concerns have emerged about their impact on cognitive function. However, experimental research in this area remains limited, specifically for functions such as inhibition control (ability to suppress distractions), and attention shifting (ability to alternate focus between tasks). Moreover, previous studies have not investigated how specific features of short-form videos, such as the continuous flow of content, influence cognitive processes. In this study, we asked participants to scroll on our custom “TikTok-style” app for ten minutes each, prior to completing two attention tasks; the Stroop, measuring response inhibition, and the number-letter task, measuring task switching. Half the participants saw a continuous feed of videos (flow), while the other half experienced grey screens that randomly interrupted the feed for roughly 15 seconds each (no-flow). Gaze behaviour, swiping metrics, and self-report social media usage were also collected. While our manipulation affected performance, a median split on self-report media usage didn’t: the “flow” participants showed worse inhibition and less accurate task switching than the “no-flow” participants. We also found some evidence that swiping patterns differ for our flow vs. no-flow groups. Taken together, our work suggests that doomscrolling could affect your subsequent attention.
C07
Examining variability in pupil responses to mutual gaze sub-events in face-to-face and video-mediated communication
Kristen Lott
klott@yorku.ca
▶ Show abstract
Pupil responses, an index of autonomic regulation, synchronize within dyads during face-to-face communication. One way synchrony may arise is through shared responses to the same social events. If a social event reliably elicits similar pupil responses across individuals, these responses may align across partners and produce dyadic synchrony. Mutual gaze (MG) has been suggested as one such event, as perceiving another’s gaze directed toward oneself increases pupil diameter. However, these effects typically emerge in closed-loop interactions where gaze signals are accurately sent and received. In videoconferencing, MG behaviour is disrupted, potentially weakening autonomic responses associated with MG. Here, we examined evoked-pupillary responses to MG while dyads communicated face-to-face or over videoconferencing. Rather than treating MG as a single event, pupil responses were analyzed time-locked to six sub-events marking socially meaningful gaze transitions. Eye gaze and pupillometry data were recorded while dyads played Heads-Up in three conditions: face-to-face, face-to-face through a screen-sized window, and over Zoom. The transition from no-interaction to direct gaze evoked pupil dilation and produced the most consistent changes in pupil size across participants, but not in Zoom. Consistent responses to direct gaze may contribute to pupillary synchrony, and these responses are weakened during videoconferencing.
C08
Instant or Gradual: The Impact of Social Attention on Accurate Impression Formation Over Time
Mridula Srikanth
20ms115@queensu.ca
▶ Show abstract
Social interactions are the cornerstone of everyday society, fostering feelings of belongingness while shaping social connections. Navigating these interactions requires forming accurate impressions of others, and prior video-viewing work suggests that social attention facilitates this process by preferentially directing our focus on social cues like faces. However, it remains unclear whether the relationship between social attention and impression formation accuracy emerges instantly or stabilizes gradually during one-on-one interactions. Therefore, this study examined how social attention impacts impression formation accuracy over time by recruiting dyads to complete a 20-minute conversation task together. We measured each participant’s social attention by capturing their fixations on their partner’s face using eye tracking. We measured each participant’s impression formation by (i) asking them to assess their own personality at the start of the task, and (ii) intermittently probing them five times during the task to assess their partner’s personality. Accuracy in impression formation was then calculated by subtracting each participants’ self-assessment from their partner’s assessment of them. Linear mixed models revealed that over time, social attention toward faces decreased, whereas impression formation accuracy increased. These results suggests that while social cues are gathered quickly, accurate impression formation occurs gradually, enabling efficient navigation of real-world interactions.
C09
Modulation of spatial attention does not affect Object Substitution Masking.
Amanjot Grewal
► Show abstract
Object substitution masking (OSM) occurs when the four dots that surround the target remain on screen beyond target offset. It has been suggested that spatial attention may not play a role in OSM. The present work pursues that hypothesis by manipulating the spatial distribution of attention. The display contained five locations: one in the centre and one in each of the cardinal peripheral locations. There were two groups of observers that differed in respect to the probability of target eccentricity. In the Central group the target appeared at fixation 80% of the time and 20% in the periphery. These probabilities were reversed for the Peripheral group. The results revealed significant OSM at both central and peripheral locations, with larger OSM in the periphery. The magnitude of OSM, however, did not differ between groups. We conclude that, consistent with the initial hypothesis, the OSM effect is not modulated by the spatial distribution of attention.
C10
Task Demands Shape the Structure of Transsaccadic Working Memory Error Across ADHD and Neurotypical Adults
Simar Moussaoui
simar.moussaoui@mail.utoronto.ca
▶ Show abstract
Transsaccadic working memory (tWM) enables information to be updated and maintained across saccades, but performance reflects both systematic and unsystematic error. Prior work identified a multicomponent error structure in spatial remapping tasks and a rightward saccade vulnerability in ADHD. The present study examined whether this component structure generalizes to a toggling paradigm that alters task structure and response demands by requiring maintenance of multiple items, sequential probe evaluation, and response selection. ADHD and neurotypical adults completed transsaccadic and fixational working memory tasks in colour and verbal domains. Using a principal component analysis (PCA), we characterized the structure of memory error and evaluated its cognitive relevance. Unlike earlier spatial-only paradigms, error variance converged onto a single dominant component capturing systematic and unsystematic error, and selection error across modalities, whereas a secondary component showed no cognitive association. The primary component correlated with global cognitive status (MoCA), and component scores did not differ between ADHD and neurotypical participants. Although saccades reliably disrupted performance, no directional asymmetries emerged. These findings indicate that the component structure of tWM error is sensitive to task architecture and response format. Under increased retrieval and selection demands, performance is governed by a unified, domain general component.
C11
The Effects of Divided Attention on Episodic Memory in Adults with ADHD
Wendy Gachagua
wgachagu@uwo.ca
▶ Show abstract
Prior research has documented episodic memory difficulties in adults with ADHD, yet it is unclear whether these impairments occur primarily during encoding or retrieval. The current study investigates whether divided attention differentially disrupts encoding versus retrieval in adults with and without ADHD. Participants were assigned to the ADHD group or neurotypical control group based on performance on the Adult ADHD Self-Report Scale (Kessler, et al. 2005). All participants completed a memory task either under full attention, or while attention was divided with an auditory animacy decision task during either encoding or retrieval. As expected, we found that memory performance was better in the full attention condition than the divided attention conditions and was better when attention was divided at encoding than retrieval. While memory performance in the ADHD group was numerically worse than the control group in the full attention and divided attention at retrieval conditions, there was no significant Group × Condition interaction. This indicates that the divided attention costs were not significantly greater for adults with ADHD than for controls. This study offers a novel approach to examining the mechanisms underlying episodic memory dysfunction in adult ADHD, particularly the role of attentional resources at encoding and retrieval.
C13
Time-on-task during video lectures: Separating the influences of time and lecture content on mind wandering and performance
Melissa van Dijk-Allen
m.vandijk@uwaterloo.ca
▶ Show abstract
While viewing online lectures, people sometimes become more inattentive as the lecture progresses (i.e., time-on-task effect). This phenomenon has typically been investigated using standard recorded lectures from university courses. While this approach enhances ecological validity, it also introduces several potential confounds. Since lecture content is presented in a particular order, it is unclear whether temporal increases in inattention (i.e., mind wandering) truly reflect an effect of time-on-task, or are instead driven by differences in lecture content. To address these concerns, we developed a video lecture consisting of six discrete subtopics that could be viewed in any order. Across two experiments, participants watched the lecture in a unique order and intermittently reported on their mind wandering throughout. Following the lecture, participants completed a memory test on the lecture content. Preliminary results showed that mind wandering increased over time-on-task but was not influenced by lecture content, indicating that these temporal increases were not driven by changes in content. In contrast, memory test scores were not influenced by time-on-task but were influenced by lecture content, indicating different temporal trajectories of mind wandering and performance. This work provides a methodological approach for isolating time-on-task effects during online lectures.
C14
Trait Mindfulness and Collisions in a Driving Simulator: Associations Beyond Attention in Young Adults
Noor Alyafei
nhuss012@uottawa.ca
▶ Show abstract
Social capital (i.e., networks, relationships, and norms of trust and reciprocity) plays a crucial role in forming new friendships and maintaining existing familial and peer relationships in emerging adulthood. This study examined the relationship between different modalities of social capital (online and offline) and motivations for video game play, while controlling for personality traits, hours spent gaming, age, and gender. A total of 342 first-year undergraduate students (220 women, 109 men, 9 gender+; M age = 20.25) participated in the study. Participants completed the Internet Social Capital Scale, the Ten-Item Personality Inventory, and the 12-item Game Motivation Inventory. Hierarchical linear regression analyses were conducted to examine whether gaming motivations were associated with online and offline social capital. The results indicated that social and immersion motivations, alongside extraversion, were significantly related to higher online social capital; immersion motivation and extraversion were associated with offline social capital. These findings suggest that specific gaming motivations and personality traits play an important role in the development of online and offline social networks. The results provide valuable insights into how underlying reasons for video game play foster social capital, offering implications for understanding the relationship between leisure activities, relationships, and well-being.
C15
When Angry Faces Seem Most Frequent: The role of Encoding Focus
Makenna Tardif
mt18qu@brocku.ca
▶ Show abstract
The present research examined whether people overestimate the frequency of angry faces and whether this bias is influenced by their attentional focus during encoding. In Experiment 1, participants viewed a series of angry, happy, and neutral faces with a word under their chin. Task instructions required attention to the face only, the word only, both the word and the face, or allowed participants to view the displays freely. Participants consistently overestimated the frequency of angry faces relative to happy and neutral faces, and this bias was not modulated by task instructions at encoding. Recognition memory for individual faces did not differ across emotions, and individual differences in memory did not predict frequency estimates, suggesting that the angry-face frequency bias does not reflect enhanced memory for individual angry faces. In Experiment 2, participants explicitly classified each face’s emotional expression during encoding. Participants continued to overestimate angry faces, but the magnitude of the bias was reduced significantly compared to all conditions in Experiment 1. Furthermore, emotion classification tendencies during encoding did not predict subsequent frequency judgments. These findings suggest that the angry-face frequency bias is robust, not due to a memory advantage for angry faces, and flourishes when emotion is encoded incidentally.
C16
A predictive coding account of asymmetric cortical connectivity
Romesa Khan
romesa.khan@mail.utoronto.ca
▶ Show abstract
Seminal frameworks of predictive coding propose a hierarchy of generative modules, each attempting to infer the neural representation of the module one level below; predictions are carried by top-down feedback projections, while predictive error is propagated by reciprocal forward pathways. However, studies have also yielded evidence of asymmetric (non-reciprocal) cortical feedback connections, including long-range descending pathways, and findings suggesting a benefit of medium-range predictive feedback under noisy conditions. We investigated the contribution of neural feedback to visual processing during grasp planning, using convolutional neural network models augmented with predictive feedback and trained to compute grasp positions for real-world objects. After establishing an ameliorative effect of symmetric feedback on grasp detection performance for noisy stimuli, we characterized the performance effects of asymmetric feedback similar to that observed in cortex. Specifically, we tested model variants extended with short-, medium-, and long-range feedback connections either originating at the same source layer or terminating at the same target layer. We found that the performance-enhancing effect of predictive coding under adverse conditions was optimal for medium-range asymmetric feedback, especially when it originated from a level of representational abstraction proximal to the input layer. Overall, biologically realistic asymmetric predictive feedback improved robustness to noisy visual stimuli.
C17
Accessing the meaning of autobiographical memories promotes shared representational structure in anterior hippocampus
Sydney Lambert
sydney.lambert@mail.mcgill.ca
▶ Show abstract
The hippocampus plays a central role in reconstructing autobiographical memories. Prior work suggests that the anterior hippocampus supports the reconstruction of conceptual information, providing a scaffold for reinstating perceptual details represented in the posterior hippocampus. However, the extent to which these regions share representational overlap during similar retrieval processes remains unclear. Using data reported by Gurguryan et al. (2025), we examined hippocampal representational similarity during autobiographical memory recall (N = 26). During the final recall phase, participants were instructed to elaborate on memories by focusing on either conceptual or perceptual details, or to recall memories without a specific elaboration focus. Preprocessed fMRI data were analyzed using representational similarity analysis (RSA) within subject-specific anterior and posterior hippocampal masks. Paired t-tests comparing ROI-level overlap revealed significantly greater similarity in the left anterior hippocampus for conceptual memories compared to the left posterior (t(25) = 2.18, p = .039, d = 0.43) and right posterior hippocampus (t(25) = 2.51, p = .019, d = 0.49). No significant ROI differences were found for perceptual or no-shift conditions (ps > .05). These findings suggest that conceptual reconstruction engages shared representational patterns within the anterior hippocampus, reflecting a more generalized form of memory reconstruction. In contrast, perceptual elaboration appears to reflect more dissociation and content-specific posterior representations, consistent with functional specialization along the hippocampal
C18
Assessing the Weapon Focus Effect Using a Virtual Reality Crime Simulation
Mikayla Tat
mtat@mun.ca
▶ Show abstract
The Weapon Focus Effect (WFE) refers to the phenomenon that the presence of a weapon (e.g., gun or knife) reduces an eyewitness’ ability to remember other details of the crime (e.g., the perpetrator’s features) as compared to memory for a similar event not including a weapon. Competing accounts attribute the WFE to either (a) the weapon capturing attention by unexpected objects (i.e., the unusual item hypothesis) or to attentional narrowing under heightened arousal (i.e., the arousal-threat hypothesis). However, traditional laboratory paradigms using static or video-based stimuli may not elicit genuine emotional or attentional responses. To address this, the current study employed a Virtual Reality (VR) paradigm. Participants took on the role of a cashier in an interactive, simulated convenience store, and after serving several customers they were then robbed by a character holding either a gun (weapon), teddy bear (unusual object), or nothing (control); their eyewitness memory for the crime was evaluated. Heart rate and gaze were tracked throughout. Preliminary evidence was equivocal with respect to supporting the WFE. Findings are discussed in relation to the challenges inherent in studying realistic eyewitness experience and their implications for the generalizability of the phenomenon.
C19
Associations between socioeconomic variables and functional alpha connectivity in toddlers
Bianca Sirbu
flavia21@my.yorku.ca
▶ Show abstract
Socioeconomic status (SES) and early life stress are linked to variations in neural development in children; yet, few studies have examined whether SES and stress predict EEG functional connectivity at preschool age, nor whether caregiver responsiveness mediates these associations. This study investigated whether SES and early-life stress predict frontal-parietal functional alpha connectivity (FFAC) at 24 and 36 months, and whether caregiver responsiveness mediates these associations. 167 caregiver-infant dyads were recruited and observed longitudinally, at 24 months (N=120), and 36 months (N=93). Baseline EEG was recorded at both time points. Caregiver responsiveness was coded during free-play. Linear regression demonstrated that financial insufficiency, a subjective marker of SES, predicted higher FFAC at 24 months, when adjusting for SES, neighbourhood poverty, and caregiver perceived stress (B= -0.0294, p < 0.05). Caregiver responsivity did not mediate associations between SES variables and FFAC at 24 months (indirect effect = −0.0000366, 95% CI [−0.00082, 0.000412], p = .81). There were no significant associations between SES, stress, caregiver responsiveness, and FFAC at 36 months. These findings indicate that early socioeconomic adversity may be associated with faster brain maturation. Future research should examine long-term effects and implications of this developmental pattern.
C20
Attention to Sincerity in Social Communication: The Role of Autistic Traits
Antonella Feeny
19aif@queensu.ca
▶ Show abstract
Successful communication relies on interpreting cues such as facial expressions and tone of voice to discern sincerity and intent. Based on the Autism Spectrum Quotient (AQ), Individuals higher in autistic traits have difficulty integrating multimodal cues that benefit social understanding. We investigated whether individuals with higher versus lower autistic traits differ in their visual attention to facial cues when interpreting sincerity and intent. Participants (N = 79; age range = 17-40 years) were divided into High and Low AQ groups. Eye movements were recorded while participants viewed videos from the Relational Inference in Social Communication (RISC) dataset. Significant interactions emerged between sincerity type and AQ group for first fixation duration and total dwell time. The High AQ group fixated longer on sarcastic exchanges (positive words with negative intent), whereas the Low AQ group fixated longer on joking control exchanges (negative words with matching intent). These findings suggest that autistic traits modulate visual attention to cues conveying sincerity during communication. Individuals with lower autistic traits attend more to sincere, negative exchanges, whereas those with higher autistic traits show enhanced processing of insincere statements. These differences emerged rapidly and persist, which reflects variability in how social multimodal cues are integrated.
C21
Beyond Reach-to-Grasp: Distinct Neural Processes for Object Placement
Petros Georgiadis
gewpetros1@gmail.com
▶ Show abstract
Most studies of human reaching treat grasping as the endpoint of an action. In natural behavior, however, grasping is typically followed by purposeful object use, such as placement. Object placement introduces additional demands on visuomotor planning, spatial updating, and sensorimotor integration. While grasping requires detailed specification of object properties before contact, placement may rely more heavily on online control during movement execution. We hypothesized that placement actions recruit neural processes distinct from grasping during both planning and execution. To test this, EEG and motion tracking were recorded while twenty-two participants performed alternating grasp and place movements. Behaviorally, placing showed faster reaction times but longer movement durations than grasping, consistent with differential reliance on feedforward and feedback control mechanisms. EEG data revealed task-specific differences with stronger preparatory activity for grasping, up to one second before action initiation. Source imaging localized these preparatory task differences to pre-motor, parietal, and occipital regions, while during execution, additional task differences were observed in ventral premotor and somatosensory cortices. Time-frequency analyses indicated that these effects were primarily driven by low-frequency activity (<8 Hz), consistent with motor-related cortical potentials. Together, these results indicate that object placement engages neural processes distinct from grasping during both preparation and execution.
C22
Blink-related neural responses in older adults: task effects, sex differences, and test-retest reliability
Shao-Yang Tsai
stsai@research.baycrest.org
▶ Show abstract
Spontaneous blinking is modulated by cognitive demands and may be affected in aging. Modulations in the delta-band (0.5-4Hz) reflecting post-blink neural processing (blink-related oscillations, BROs) have been proposed as a marker of brain health. Here, we evaluated the test-retest reliability of BROs in older adults during active and passive tasks. Cognitively normal older adults (n = 81, 43 females, 60–84 years) completed two EEG sessions 1–2 weeks apart, comprising four conditions: contour integration task (CIT), simple reaction task (SRT); dot motion viewing (DOT), and resting state (RSO). Blink-related EOG activity was removed from the EEG, and global field power (GFP) of delta-band (0.5–4 Hz) activity time-locked to blinks was quantified in two time windows (P1: 50-150 ms; P2: 230-280 ms). Average global field power (GFP) and ERP amplitude in the POz channel were analyzed within each time window. Condition modulated BROs: SRT produced the largest amplitudes, while DOT showed the smallest. Females showed larger GFP than males across all tasks and windows. Test-retest reliability ranged from moderate-to-good, with higher reliability in SRT and RSO. Delta-band BROs show adequate reliability and demonstrate condition and sex effects on post-blink neural processing in healthy aging.
C23
Brain Waves in Noise: Do Background Speech Language and Semantic Context Matter?
Setareh Dorood
sdoro102@uottawa.ca
▶ Show abstract
This study used a sentence reading task in the presence of background speech, employing ERP measurements. The disruptive effect of background speech on non-auditory sentence processing is known as the Irrelevant Sound Effect (ISE) (Grieco-Calub et al., 2019). We examined whether semantic context facilitates easier semantic integration during reading and whether language dominance modulates this effect in irrelevant speech environments. Rapid Serial Visual Presentation (RSVP) was used (Potter, 2018) to test French-dominant bilinguals (N=8) under two conditions: English and French two-talker conversational background. Each block included both high-predictability (HP) and low-predictability (LP) sentences. The background noise consisted of audio recordings of two individuals engaged in problem-solving conversations with natural hesitations and pauses, simulating real-life listening environments (extracted from Gosselin, 2025). The dependent variable was the N400 amplitude at the target words. A significant main effect of semantic context (p = .016), indicating differences between high- and low-predictability conditions. LP sentences elicited more negative N400 amplitudes compared to HP sentences. There was no significant main effect of noise (p = .402). Overall, the results support that semantic context facilitates easier semantic integration during reading in noisy environments.
C24
Characterizing critical bands and absolute efficiency of neural networks using ideal observers
Anne Peiris
peirisa@yorku.ca
▶ Show abstract
Which spatial frequencies are critical for deep neural networks (DNNs) to perform object classification? Previous work found that DNNs have critical bands 2–4 times wider than those of humans in an object classification task (Subramanian et al., 2023). Wider channels may be a shortcoming of DNNs, or might mean that they exploit more of the useful spatial frequencies in images. We used ideal observer analysis to test these hypotheses. We ran five ideal observers and 18 DNNs in the ImageNet classification task. We presented images in filtered Gaussian noise and measured object classification thresholds at each noise frequency to identify critical bands. DNNs’ critical bands were, on average, around two octaves wide, consistent with previous work. Interestingly, the ideal observers’ critical bands were low-pass rather than band-pass, with a 1/f fall-off. Thus, DNNs do not exploit the entire range of informative spatial frequencies in natural images. Indeed, a separate analysis revealed that DNNs’ thresholds in white Gaussian noise were orders of magnitude higher than ideal observers, meaning that DNNs’ absolute efficiency is extremely low (η ≈ 7e-5). These results inform DNNs’ use as models of human vision and can help make them more efficient for computer vision
C25
Charting Spatial Memory Over Time in Developmental Amnesia
Tolu Faromika
tolufaro@yorku.ca
▶ Show abstract
Spatial memory enables our navigation and orientation in the world. Cognitive Map Theory posits that the hippocampus is critical for allocentric spatial memory; however, evidence from the literature presents a mixed picture. While adult-onset hippocampal damage is associated with impaired formation of new allocentric representations, individuals with hippocampal damage have been shown to retain coarse, gist-like representations of highly familiar environments and even newly experienced environments. This raises the question of whether extended learning can support the development of gist-like spatial representations. We investigated this question through a longitudinal examination of spatial memory in an individual with developmental amnesia, H.C. Building on prior work (Rosenbaum et al., 2015), H.C. was retested on a battery of mental navigation tasks based on familiar environments. Performance was compared to that of control participants with similar environmental exposure. Results indicate that extended learning supports the development of gist-like spatial representations in developmental amnesia. H.C. demonstrated preserved relational knowledge sufficient for coarse navigation and landmark anchoring, but continued to show reduced flexibility, limited detail, and swap errors. These findings suggest that while neocortical systems may support schematic spatial representations, the hippocampus remains critical for binding fine-grained relational detail within a coherent and flexible map.
C26
Decisions under Uncertainty: A Statistical Framework for Evaluating Practical Relevance in Interval-Based Hypothesis Testing
Paul Riesthuis
paul.riesthuis@kuleuven.be
▶ Show abstract
Psychological researchers are increasingly encouraged to move beyond a narrow focus on detecting statistically significant effects that differ from zero under the traditional null-hypothesis significance testing framework. To interpret findings more meaningfully, best practices emphasize evaluating the magnitude of effects to determine whether they are practically meaningful or negligible. This requires specifying a smallest effect size of interest and conducting interval-based hypothesis tests, such as minimum-effect or equivalence tests. Although these approaches improve current practices by explicitly taking into account effect sizes, interval-based hypothesis tests can still yield inconclusive results, leaving uncertainty about whether an effect is practically relevant or negligible. In this article, we first introduce interval-based hypothesis testing and highlight the challenge posed by inconclusive outcomes. We then propose a set of complementary tools—implemented in an accompanying Shiny application (https://paulriesthuis.shinyapps.io/SESOIdecisions/)—to support more informed decision making under such uncertainty, including threshold alpha, the robustness index, the practical relevance replication probability, Bayesian posterior probabilities, and a meta-analytic approach. By explicitly incorporating uncertainty and replication potential, these tools aim to help researchers draw more nuanced statistical and practical decisions and improve the interpretation of results obtained from interval-based hypothesis testing.
C27
Developmental Maturity and Sex Linked to Altered Functional Brain Network Topology in Neonatal HIE: An fNIRS Study
Anagha Vinod
vinoda@mcmaster.ca
▶ Show abstract
Neonatal hypoxic-ischemic encephalopathy (HIE) causes long-term neurodevelopmental impairments, yet structural MRI often fails to capture functional impairments. Assessing functional connectivity (FC) during the tertiary phase of injury may provide insights into recovery and response to therapeutic hypothermia. We used functional near-infrared spectroscopy (fNIRS) to characterize resting-state functional connectivity in 22 neonates with HIE between 5 and 15 days of life. We hypothesized that altered network topology is associated with developmental and clinical factors, reflecting compensatory or maladaptive mechanisms.The relationship between graph-theory metrics, including assortativity, hierarchy, and efficiency, and clinical factors was analyzed using partial least squares correlation (PLSC). PLSC revealed a significant latent variable (p = 0.02) linking developmental maturity to network topology. Higher gestational age, birth weight, and male sex covaried with greater network efficiency and small-world organization, while female sex and higher Apgar scores covaried with hierarchical, less integrated networks.These findings suggest that developmental maturity and sex influence functional brain network topology in neonates with HIE. fNIRS-derived graph-theory metrics may represent sensitive biomarkers of functional brain status. By detecting maladaptive, inefficient network organization at the bedside, clinicians may help identify infants at higher risk for adverse outcomes, guiding targeted interventions.
C28
Does the directionality of semantic associations affect working memory?
Sho Ishiguro
sho.ishiguro@umoncton.ca
▶ Show abstract
Semantic relatedness affects working memory: it improves item memory but impairs order memory. For example, if “apple”, “banana”, and “peach” are targets in an immediate serial recall task, participants are likely to recall them but more prone to recall them in the wrong order. This pattern of beneficial and detrimental effects resembles the phonological similarity effect. However, unlike phonological similarity, semantic relatedness can be directional. Word association norms have indeed documented directional associations: “stork”, for example, has a forward association to “baby” (stork -> baby), but not vice versa (stork <- baby). Building on the directionality of associations, we tested whether semantic relatedness could have a beneficial effect on order memory, contrary to the status quo. In the current study, participants completed an immediate serial recall task with six-word lists containing forward associations, such as “governance -> ruling -> judge -> robe -> shower -> hot”. Lists with forward associations produced better order memory than lists with no associations in Experiments 1–3 and than lists with backward associations in Experiment 4. These novel findings are well explained by the classic notion of associative chaining, which highlights interitem associations.
C29
Early emotional expression processing is not influenced by face gender nor participant sex – a mass univariate analysis of ERP data
Amie Durston
ajdurston@uwaterloo.ca
▶ Show abstract
Perceiving faces and the emotions they convey is the cornerstone of social life. Previous work suggests that the neural decoding of faces may be influenced by participants’ biological sex (Proverbio, 2017, 2021; Stasch et al., 2018; Tiedt et al., 2013) and stimuli gender (Proverbio et al., 2010; Zheng & Han, 2025). Further, females and males may perceive negative and threat-related emotions differently (Hampson et al., 2006; Huang et al., 2022; Olderbak et al., 2019). To date, no ERP study has considered how these three factors might interact. In the present study, participants (N=166; 84 Males) viewed female and male faces portraying happy, angry, fearful, and neutral expressions while ERPs were recorded. Participants rated faces on valence and arousal using 9-point Likert scales. Behaviorally, we found that participant sex and stimuli face gender did interact with facial expression to influence valence and arousal ratings. Full-scalp ERP analyses revealed traditional emotion effects, and more positive N170-P2-EPN amplitudes (~150-250ms on posterior sites) in females relative to males. However, no effect of face gender nor any interactions emerged, suggesting early (<300ms) emotional expression processing is immune to face gender and participant sex. Future research needs to test whether sex effects are general or face-specific.
C30
Evidence for Independent Processes Underlying Addition and Multiplication in Mathematical Decision Making
Heather Jordan
michaelasluceno@gmail.com
▶ Show abstract
The mathematical cognition literature is divided on whether addition and multiplication rely on shared cognitive mechanisms. Previous work has shown that competition between competing items causes saccades curvature towards the distracting object. Given sufficient time to resolve the competition and inhibit the distractor, saccades curve away from distractors (Giuricich et al, 2023). We tested competition between sum and products of single digit expressions, expecting earlier/greater inhibition of product than sum distractors due to rote-learned multiplication facts. Forty-five undergraduates were presented with single-digit addition/multiplication expressions and made single saccades to the correct choice between the sum and product. We observe two separate competitive influences of addition/multiplication. Initially (175-250 msec), we observe larger saccade deviation toward distractors when the target was the sum rather than the product. When competition between the numerals is resolved and the distractor inhibited (275–400 msec), we observed curvature away from the sum when the product target was correct. Inhibition of the product is observed on trials that require addition but varies with the magnitude between competing numerals. Our results are consistent with independent addition and multiplication processes. Multiplication uses fact-retrieval, but addition relies on an approximating process where the magnitude between two possible answers influences choice.
C31
fMRI PLS analysis supports a novel Thought Content x Autobiographical Memory System (TCAMS) multidimensional framework of major rumination domains
Scott Squires
9sds@queensu.ca
▶ Show abstract
Rumination fMRI research has largely been limited to paradigms that induce a single kind, typically depressive rumination. Our recent factor analytic work has identified four broad kinds of rumination in the general population: negative self-focused (NS), negative other-focused (NO), constructive (C), and positive (P). The Default Mode Network (DMN) underlies rumination, with subsystems processing self-relevant thoughts in semantic (e.g., dmPFC) and episodic (e.g., medial temporal lobe) domains. To bridge these findings, we tested a novel 4 × 2 framework (Thought Content × Autobiographical Memory System; TCAMS) using an updated induction paradigm that elicits eight kinds of rumination on idiographic topics alongside two distraction conditions (Abstract vs. Concrete Thinking). Thirty psychologically healthy young adults completed the TCAMS paradigm, and fMRI PLS analysis found that: 1) 16 clusters distinguished Rumination from Distraction, with four DMN clusters (PCC/precuneus, dmPFC, vmPFC, left aMTG/aSTG) associated with Rumination and 12 non-DMN clusters associated with Distraction; 2) 15 clusters differentiated Maladaptive (NS+NO) from Adaptive (C+P) Rumination, found in emotional/social regions (dmPFC, vlPFC, aIns/OFC), cerebellum, and visual cortex; and 3) 7 clusters distinguished Episodic from Semantic Rumination, including parahippocampal cortex, precuneus, frontal pole, dlPFC, and left amygdala. This evidence provides preliminary support of TCAMS as a rumination framework.
C32
From Steering to Screening: Predicting Mature Driver Health Status from On-Road Behaviour
Tal Friedman
talfriedman@cmail.carleton.ca
▶ Show abstract
Mature drivers are a high-risk group, and issues that affect their driving ability are often unrecognized or only understood after a serious incident occurs (Pachana & Petriwskyj, 2006). Changes in cognitive, physical, and visual health can be directly tied to altered decision-making and worsened driving skills (Anstey, 2005). Thus, it may be possible to analyze driving data to estimate a driver’s health status. Previous studies in this area utilized curated populations with selective inclusion criteria and clean divides between ‘healthy’ and ‘unhealthy’ groups, resulting in datasets that do not represent the general population. This project involved constructing composite health scores based on seven years of longitudinal health assessments and questionnaires from 793 mature drivers. Machine learning classifiers were trained on drivers’ vehicle data to predict physical, cognitive, and perceptual health scores. Preliminary KNN models have reached 75.8% accuracy predicting physical functioning from a reduced version of the driving dataset (n = 254). These analyses are representative of a real world sample, and thus generalize to the broader Canadian population, and can therefore scale for general use. This work can be extended to develop tools for drivers to increase self-monitoring capabilities, allowing for early warnings concerning health, safety, and driving ability.
C33
High-frequency hearing loss predicts medial-lateral sway in healthy young adults
Breann Krygsman
► Show abstract
Young adults are increasingly exposed to unsafe listening environments, leading to noise-induced hearing loss and its consequences. Despite hearing loss being known to affect balance in older adults, the relationship between subtle auditory changes and balance in the young adult population remains unclear. This study investigates whether elevated hearing thresholds, including those within the subclinical range, predict postural sway in a sample of young adults. Sixty-six university students completed pure-tone audiometry testing and a quiet-stance postural sway assessment under eyes-open and eyes-closed conditions. Hearing thresholds were averaged into low- (125, 250, and 500 Hz) and high-frequency (2000, 4000, and 8000 Hz) categories. Postural sway was measured using a Nintendo Wii Balance Board and quantified as center-of-pressure variability in medial-lateral and anterior-posterior directions. Correlational analyses revealed a significant positive relationship between high-frequency hearing thresholds and medial-lateral sway with eyes closed, suggesting that hearing loss at high-frequencies may compromise lateral stability when visual input is removed. No significant relationships were found between average low-frequency hearing thresholds and anterior-posterior sway. These findings suggest that high-frequency auditory deficits are associated with greater medial-lateral sway when visual cues are unavailable, underscoring the need for early detection and public awareness of unsafe listening behaviours.
C34
Individual Variability in Visual Evoked Potentials Predicted by Cortical Morphology
Isimeme Okonofua
isialero@yorku.ca
▶ Show abstract
EEG signals measured at the scalp and the underlying cortical areas that produce the signals, both have substantial variability between individuals. These signals arise from cortical pyramidal cell activity, generating current dipoles. Dipoles perpendicular to the scalp produce stronger signals than other orientations. Here, we assess the impact of cortical folding on EEG signals using EEG and MRI data from a study where participants viewed symmetry-containing stimuli in a steady-state paradigm. This paradigm isolates responses to image updates and symmetry in distinct frequency components. We investigated the contribution of cortical folding on both the image update response, and the symmetry response. For each participant, we approximate dipoles capturing average cortical folding within relevant regions of interest to generate forward models. Spatial correlations between the models and empirical topographies reveal individual differences in area contribution. V3 was the most representative morphological model for the image update response, and V4 for the symmetry response. The location and orientation of these regions may produce dipoles more predictive of the measured scalp EEG signal, despite likely computational contributions from other areas. This study informs methods for interpreting individual variability in EEG data.
C35
Investigating hippocampal predictive processing in speech-based statistical learning: an intracranial EEG study
Daniela Herrera-Chaves
dherrer3@uwo.ca
▶ Show abstract
Statistical learning (SL) is a fundamental learning mechanism that supports the extraction of regularities from the environment. The hippocampus has been proposed to play a central role in SL, but evidence regarding its involvement is mixed. While neuroimaging evidence suggests hippocampal engagement in SL, prior intracranial EEG data do not support this. To reconcile these conflicting findings, we hypothesized that the hippocampus does not directly support learning but rather retrieves already learned regularities to generate predictions of upcoming stimuli. In our study, 21 epilepsy patients implanted with intracranial electrodes listened to a continuous speech stream containing repeating “hidden” trisyllabic words. In a following test, participants made speeded responses to specific target syllables embedded within shorter snippets of the speech stream. Behaviorally, we found faster reaction times to syllables in later, more predictable positions within a word. At the neural level, preliminary event-related potential (ERP) analyses revealed a subset of electrode contacts showing an ERP effect prior to the onset of target syllables, suggesting task-relevant predictive processing. These contacts were mainly located within the hippocampus, insula, and middle temporal cortex, suggesting that the hippocampus is part of a network of brain regions that supports prediction as a consequence of statistical learning.
C36
Investigating public-health-related optimism bias using survey data and agent-based modelling
Abhishek Dedhe
ananya.jogalekar@duke.edu
▶ Show abstract
Optimism bias or (illusory) self-superiority is the tendency to perceive oneself as more desirable than others. We analyzed pre-registered survey data from a non-WEIRD population during the “psychological theater” of the COVID-19 pandemic (N = 631) and in a post-pandemic setting (N = 1621), and used an agent-based model (ABM) to simulate self-superiority about public-health-related behavior (mask wearing, mobility, etc.) across these two periods. We tested whether threat perception and connectedness to one's social network are predictors of self-superiority. We found self-superiority gaps in both time periods, though significantly higher during the pandemic. Regression analyses showed that during the pandemic, these gaps were driven by threat perception and connectedness. In the post-pandemic period, the influence of threat perception decreased, replaced instead by socioeconomic factors such as gender. Notably, connectedness remained a significant inverse predictor of self-superiority across both time periods. Phase transitions in ABM results showed a decrease in self-superiority after the pandemic ended and highlighted how lower connectedness drove higher self-superiority across time periods. In summary, our findings suggest that self-superiority may be a well-calibrated, fluid response to psychosocial environments rather than a static trait representing cognitive biases.
C37
Investigating the role of the hippocampus in statistical learning in patients with temporal lobe epilepsy
Emily Cordeiro
ecordei@uwo.ca
▶ Show abstract
Statistical learning (SL) is the process by which we extract regularities from the environment, producing knowledge that can be expressed either explicitly or implicitly (with or without conscious retrieval). Whether the hippocampus, a brain structure critical for associative learning and episodic memory, is necessary for SL remains unclear. A recent computational model suggests that both implicit and explicit SL depend on the hippocampus, whereas traditional memory frameworks propose that implicit knowledge may rely on structures outside the hippocampus. Supporting the latter view, a patient with a selective hippocampal lesion demonstrated impaired explicit SL but intact implicit SL (Wang et al., 2023), suggesting dissociable underlying neural mechanisms. We extend this work to patients with temporal lobe epilepsy (TLE), a group often characterized by broad hippocampal lesions and memory impairments. Consistent with our previous patient study, TLE patients (n=23) showed comparable facilitation on our implicit measure relative to age-matched controls (n=25), but performed significantly worse on our explicit measure. These preliminary results support that the hippocampus may contribute to explicit but not implicit SL. To further clarify the role of the hippocampus and its subfields, ongoing analyses of high-resolution Magnetic Resonance Imaging scans will relate hippocampal structure more specifically to task performance.
C38
Laminar Organization of Medial Frontal β-Bursts During a Stop Signal Task
Pranavan Thirunavukkarasu
thirunap@yorku.ca
▶ Show abstract
Recent studies in humans have shown that brief bursts of β-oscillations over the medial frontal cortex— known as β-bursts—may play a critical role in cognitive control. However, how these β-bursts influence stopping behavior and how they are generated remains unclear. To address these questions, it is necessary to examine the functional properties of β-bursts across the different cortical layers of the medial frontal cortex, a region strongly implicated in cognitive control. To address this, we collected intracortical local field potentials (LFP) within the SEF of four macaque monkeys and cingulate cortex of two macaque monkeys performing a saccadic stop-signal task. β-bursts were elevated during periods requiring cognitive control, including successful response inhibition, post-error processing. Additionally, they were elevated during intertrial intervals. We observed that these intracortical β-bursts were observed across all layers of cortex. These findings will provide important constraints for biophysical and cortical circuit models of medial frontal function in relation to EEG measures and invite further considerations of β-burst function in cognitive control.
C39
Language Exposure and Neurodevelopment: Linguistic and Cognitive Outcomes in Children Post-Stroke
Kai Ian Leung
kaiian.leung@mail.utoronto.ca
▶ Show abstract
Pediatric arterial ischemic stroke (AIS) occurs during a sensitive period of neurodevelopment, often leading to lasting neurocognitive impacts. Outcomes may be shaped by neurological and environmental variables, including language exposure. Given that half of the global population is bilingual (Grosjean, 2010), understanding how language exposure shapes post-injury neurodevelopment has significant theoretical and clinical implications. This cross-sectional study examined cognitive and linguistic functioning in 29 children following AIS, grouped by age at stroke onset (presumed perinatal, neonatal, childhood). Caregiver questionnaires assessed language exposure and socioeconomic status; standardized assessments measured expressive language, metalinguistic abilities, non-verbal cognition and executive function. Multivariate analyses examined the effects of percentage English exposure and stroke timing on neurocognitive performance, controlling for SES. Language exposure did not impede neurodevelopment in any domain. For expressive language and metalinguistic abilities, effects were domain-specific and varied by stroke onset: less English exposure was associated with stronger performance in the neonatal group, no relationship in the childhood group and an opposing pattern in the presumed perinatal group. No executive function effects were observed. These divergent patterns suggest that the relationship between language exposure and neurodevelopment is moderated by injury timing, consistent with the view that neuroplasticity shapes sensitivity to early language experience.
C40
Learning Rule Activations via Backpropagation within a Neuro-Symbolic Cognitive Architecture
Mani Setayesh
mani.setayesh@mail.utoronto.ca
▶ Show abstract
In hybrid neuro-symbolic artificial intelligence, a prominent issue is the integration of neural and symbolic processing pathways. An instance of a solution to this issue is the Clarion cognitive architecture, which combines rule-based systems and neural networks together in a principled and synergistic fashion. Taking advantage of the tight integration between sub-symbolic and symbolic components in this architecture, we pose that, through reward-driven gradient signals, it is possible to optimize rule activations so as to better integrate neural and symbolic processing pathways. We investigate the properties of the resulting model and compare it to the theoretical standard in a simple bandit-style task as well as a more complex grammar-learning task.
C41
Mind Wandering Without the Media Multitasking: Validation of an Online Visual Rhythm Task
Josephine Nicholson
nichol40@myumanitoba.ca
▶ Show abstract
Mind wandering is often examined using simple, low‑demand tasks that encourage task‑unrelated thoughts and allow researchers to track subtle lapses in attention. The Metronome Response Task (MRT) is one such measure. However, its highly automatic auditory format may encourage multitasking and instructional noncompliance in online settings, making both reduced performance and reports of off-task thought difficult to interpret. The current study aimed to (1) validate a novel visual rhythm task as an effective measure of mind wandering and inattention, and (2) test whether the visual task reduces multitasking via gamified elements, visual stimuli, and task difficulty. Participants (N = 300) completed either an MRT or one of two visual versions with fixed or variable intervals. Both visual tasks produced higher performance, lower mind‑wandering rates, and less multitasking than the MRT, while preserving similar mind‑wandering–performance associations. Performance declined over time in the MRT but remained stable in the visual tasks, and interval variability had little effect. These findings support the online visual tracking task as a valid measure of mind wandering and as an effective way to maintain engagement and decrease multitasking during online mind wandering studies.
C42
More than Reach: an fMRI investigation of action intention and execution in object grasp and placement
Gaelle Luabeya
gaellenl@yorku.ca
▶ Show abstract
Studies of hand control focus more on prehension but often overlook the final placement to end the interaction with the object. Our study investigates the neural correlates of object placing and grasping. Despite their reliance on equivalent target information (e.g., its location and orientation), we expected to see a difference: grasp relying more on the actual visual input, and place relying more on the mental representation of the target outcome. Our event-related fMRI study asked twenty participants to alternate between grasping and placing a rectangular object on a template presented on an inclined platform. The target (object/template) was presented at varying locations (left/right) and orientations (±30°). After a delay, participants grasped or placed the object. We conducted a graph-theory-based functional connectivity analysis. The network was segregated into three modules: a Cerebello-Occipito-Parietal Module (Mod1), a Frontal Module (Mod2), and a Sensorimotor Module (Mod3). When quantifying their modularity strength, grasp was higher in Mod1, whereas place was higher in Mod2. Despite their similarity as goal-directed actions, grasp was more sensory-driven, whereas place was more cognitively driven. Indeed, movement intent affects brain state. In conclusion, grasp and placement share similar cortical mechanisms but emphasize different modules depending on action intention and sensory feedback.
C43
Neural and postural mechanisms associated with visually-induced motion sickness (VIMS)
Polina Andrievskaia
pandrievskaia@torontomu.ca
▶ Show abstract
The successful integration of our visual and vestibular systems is necessary when interacting with simulated environments, such as virtual reality (VR) technology. Yet, some studies indicate that up to 80% of individuals may experience visually-induced motion sickness (VIMS) while using VR, exhibiting symptoms akin to traditional motion sickness, including nausea, eyestrain, blurred vision, disorientation, and dizziness. The neurological mechanisms associated with these side effects have produced largely inconclusive results, whilst some work has found postural instability to be predictive of VIMS severity experienced during VR exposure. The proposed study will have participants engage in a popular VR exercise game for 40 minutes while cybersickness scores are recorded before, during, and after the exposure. Neurophysiological data will be recorded using a 32-channel EEG system during VR, and time-frequency analyses will be used to determine potential neural correlates associated with the development of VIMS. Postural control will be compared before and after engagement with the VR game, and differences of the root-mean-square and mean velocity of the centre of pressure will be calculated. Preliminary results reporting such variation in neurological and postural responses to the VR exposure will be presented.
C44
Neural object representations predict visual discrimination success in adolescents and adults
Merron Woodbury
merronwoodbury01@gmail.com
▶ Show abstract
Object perception involves processing both broad conceptual features as well as unique, item-specific ones. Developmentally, the ability to discriminate between unique but conceptually related objects improves through adolescence, yet, coarse-grained neural object representations appear mature by late childhood. Here, we asked whether behavioural improvements are linked to refinements in fine-grained object representations. We mapped representations in adolescents (N=36; 12-13 year olds) and adults (N=36) by having participants view exemplars of real-life object concepts while collecting functional magnetic resonance imaging data. We then assessed participants’ ability to visually discriminate between these exemplars using a match/mismatch task. Using a representational similarity analysis searchlight, we identified substantial overlap across age groups in the brain regions representing objects at both the exemplar-specific and concept level, primarily within the visual cortex. Further, while adolescents performed worse overall than adults on the visual discrimination task, we found developmental consistency in the neural representations supporting behaviour. Namely, discrimination performance for highly similar exemplars was linked to stronger neural coding at the exemplar, but not concept, level. Together, our results suggest that fine-grained object representations in the brain are relatively mature by early adolescence and inform object perception in a similar manner to adults.
C45
Oculomotor dance learning task: Implications for audio-visual cued spatial learning
Michael Petrovski
michael.petrovski00@gmail.com
▶ Show abstract
This study aims to address whether a new visual-motor-based learning paradigm with music can potentially promote neuroplasticity and create new interventional tools, building upon prior research that shows behavioural and putative neural changes fol-lowing dance-based neurorehabilitation in people with Parkinson’s disease (PD) [1,2]. Eye movements of 10 participants were tracked using the Eyelink 1000 Plus system during a 68-second eye-dance sequence. The experiment consisted of a learning phase, where participants observed the sequence five times with 30-second breaks, and a performance phase, where they performed the sequence five times from memory on a grey screen without visual cues. Results showed a significant improvement in perfor-mance accuracy between the first session (g1; M = 40%, SD = 7.2%) and the last session (g5; M = 69.7%, SD = 22.8%). Similarly, there was significant improvement in timing accuracy between the first session (g1; M = 0.29, SD = 0.06) and the fifth session (g5; M = 0.46, SD = 0.12). These findings align with prior studies, which suggest that visu-al-motor-based interventions have the potential to enhance motor and non-motor symptoms like depression and anxiety for neurodegenerative diseases such as PD.
C47
Prefrontal and Parietal Local Field Potentials Employ Different Visuospatial Codes for Reach: A Complex-Valued Network Classification Approach
Ashkan Karimi
ashkan70@yorku.ca
▶ Show abstract
Understanding how cortical oscillations coordinate spatial memory and motor planning is a central challenge in systems neuroscience. We tested whether phase–amplitude dynamics in cortical local field potentials (LFPs) encode distributed versus region-specific signals for spatial memory and planning under varying visuospatial conditions. We developed a deep Complex-Valued Neural Network (CVNN) to decode landmark-dependent spatial states from LFPs recorded in the posterior ventrolateral prefrontal cortex (pVLPFC, 128 channels) and intraparietal sulcus (IPS, 32 channels) of a female rhesus monkey performing memory-guided reaching tasks in which visual landmarks were stable, shifted 8° in one of eight directions, or absent. Preprocessed LFPs were transformed into complex-valued time series using the Hilbert transform to preserve phase and amplitude information. We trained separate CVNN models on IPS or pVLPFC signals which classified the three landmark conditions with >90% training accuracy and more than 51% overall validation accuracy, significantly above chance (33%). However, validation performance revealed inter-regional specialization: while IPS model performed best for no-landmark trials (88.35%±6.99), the pVLPFC model showed superior performance for shifted-landmark trials (71.73%±8.59). These findings suggest that IPS preferentially encodes egocentric spatial representations, whereas pVLPFC is more engaged during dynamic landmark conditions, with potentially competing spatial codes emerging during integration.
C48
Prefrontal LFP activity is associated with visuospatial context and task performance in head-unrestrained, memory-guided reaching
Jennifer Lin
linj68@yorku.ca
▶ Show abstract
The purpose of this study was to understand how prefrontal local field potential (LFP) signals encode visual landmark information during a memory-guided reach task, and how this is modulated by performance and spatial variables. We implanted a 128-channel Plexon Array over the ventrolateral prefrontal cortex (vLPFC). Single unit activity, LFPs and behavioral signals were recorded simultaneously in a Rhesus monkey. The task began with central gaze fixation, after which a landmark was presented. A visual target then appeared transiently, followed by a visual mask. After the mask, the landmark either reappeared at the same location or shifted. The fixation light then extinguished, signaling a reach to the target. Controls were the same, but without the landmark. Preliminary time-frequency analyses revealed frequency-specific decreases in LFP power during the memory delay, followed by increases peaking near movement onset across all conditions. Correct stable landmark trials and landmark shift trials showed a second power increase between acquisition and reward. These landmark-related modulations were reduced during error trials. Low-frequency LFP power also exhibited spatial tuning, with stronger modulation for targets located ipsilateral to the recording site. These preliminary findings suggest that vLPFC LFP signals reflect task-relevant spatial information and performance-dependent modulation during memory-guided reaching.
C49
Reading in Multiple Sclerosis: An Exploratory Study of Naturalistic Eye Movements
Courtney Stacey
courtney.stacey@unb.ca
▶ Show abstract
Multiple sclerosis (MS) is a neurological autoimmune disorder involving inflammation, demyelination, and neurodegeneration within the central nervous system. Although physical symptoms, sensory disturbances, and visual impairments are well-documented, cognitive dysfunction has received comparatively less attention. Deficits in executive functioning (e.g., inhibition, working memory) and language (e.g., verbal fluency) have been reported; however, reading (an ability that relies on both cognitive and linguistic systems) has yet to be investigated. Importantly, reading is central to nearly all aspects of daily life, including economic, occupational, and social activities, as well as routine tasks. As such, we employed eye-tracking to investigate naturalistic reading performance in MS. A total of 25 native English-speaking adults (11 with relapse-remitting MS, 14 controls) silently read texts (4 x ~250 words) for comprehension while their eye movements were recorded with an SR Research EyeLink 1000 Plus eye-tracker. Linear mixed-effects models examining global aspects of reading fluency (e.g., average fixation durations, average saccade amplitudes, average reading rates) revealed that the MS group exhibited more effortful eye movement reading behaviour, paired with lower reading comprehension scores. Although preliminary, our findings suggest that oculomotor control during reading is disrupted in MS and may serve as a sensitive marker of cognitive change.
C50
Repeated Exposure Modulates Auditory-Frontal Coupling as a Function of Musical Preference: An fNIRS Study
Harley Glassman
hglassman@torontomu.ca
▶ Show abstract
Music is consistently ranked as a highly rewarding activity. Familiar music is often considered more rewarding, such that increased exposure to preferred pieces produces greater enjoyment. However, the neural mechanisms of musical preferences on reward as a function of familiarity remain unknown. This study used functional near-infrared spectroscopy (fNIRS) in a repeated exposure paradigm to investigate the extent to which familiarity influences the hemodynamic activation of the auditory cortex and prefrontal cortex for preferred and non-preferred music. Additionally, it examined the hemodynamic activation and self-reported liking for preferred and non-preferred pieces in distinct reward sensitivity groups, as measured by the extended Barcelona Music Reward Questionnaire (eBMRQ). A significant interaction was found between musical preference and repetition, where overall activation was greater with repeated exposure to preferred music. Additionally, individuals with higher reward sensitivity showed greater overall activation for preferred music. Together, these results suggest that musical preference and repetition interact to modulate activity across auditory–frontal networks, which was most prominent in individuals with higher music reward sensitivity. This novel study demonstrates the dynamic influence of musical preferences on the neural mechanisms of musical reward, indicating a greater need for future studies to consider its integral role in personalized listening.
C51
Response-Locked Time–Frequency Dynamics of Error Processing and Their Association with Anxiety Traits
Zelin Chen
z439chen@uwaterloo.ca
▶ Show abstract
Performance monitoring is critical for adaptive behaviour, and errors are known to elicit stronger monitoring than correct responses. Error neural responses measured by event-related potentials have been linked with greater anxiety traits. However, whether a similar relationship is seen in the time–frequency domain remains unclear. The present study examined response-locked oscillatory activity associated with error versus correct responses, and whether these neural signatures are modulated by individual differences in self-reported anxiety measured by the STAI and STICSA questionnaires in an arrow flanker task (N=75). Time–frequency decomposition was performed on response-locked data and oscillatory power was compared between error and correct trials across frequencies. Errors elicited stronger oscillatory power in the theta band (4–8 Hz) from 50-150 ms, and in the theta (4–8 Hz), alpha (8–13 Hz), and beta (12-27 Hz) bands from 300-450 ms. However, except for the early theta band which correlated weakly with STAI scores at the frontal central site, no correlation with anxiety was seen. These findings suggest temporally distinct oscillatory signatures of error monitoring, characterized by early and sustained theta activity and later alpha and beta activity, that operate relatively independently of individual differences in self-reported anxiety in the non-clinical population.
C52
Sarcastic Cue Perception Within ADHD and Neurotypical Populations
Kari Kinnear
kkinn675@mtroyal.ca
▶ Show abstract
Sarcasm is a complex colloquial tool that relies on the integration of prosodic, facial, contextual, and multimodal cues. While neurotypical adults generally detect sarcasm with ease, emerging evidence suggests that neurodivergent populations, particularly individuals with Attention-Deficit Hyperactivity Disorder (ADHD), may experience difficulty interpreting these cues. The present study examined whether adults with ADHD differ from neurotypical individuals in their ability to detect sarcasm across communicative cues. A sample of undergraduate students (N = 148) completed a computerized task assessing sarcasm detection across written, facial, intonational, and multimodal video stimuli, with accuracy and reaction time measured. Regression analyses revealed that, among neurotypical participants (n = 74), reaction times to facial expressions (p = .001) and intonation (p = .05) significantly predicted video sarcasm reaction time. In the medicated ADHD (n = 41) group, only intonation reaction time emerged as a significant predictor (p = .003), whereas no significant predictors were observed in the unmedicated ADHD (n = 33) group. Differences in cue-processing patterns suggest that ADHD medication may facilitate the integration of communicative cues involved in social interpretation, rather than solely enhancing attentional control. These findings highlight the potential role of ADHD medication in facilitating social communication processes.
C53
Testing an Affective Arousal Account for Temporal Binding
Defne Tuncer
defne.tuncer@mail.utoronto.ca
▶ Show abstract
When participants are asked to estimate the time interval between two events, they tend to give smaller estimates when the first event consists of a voluntary action. This temporal binding effect is one of the most widely used implicit measures of the sense of agency. However, despite its widespread use, the cognitive mechanisms that generate this effect remain unclear. The rational analysis of memory instructs us that the strength of a memory trace should track the recency and frequency of its usage. Furthermore, emotional arousal is known to produce stronger memory traces. In this study, we hypothesize that participants perform interval estimation by comparing the availability of events in memory. In this light, temporal binding effects may be attributed to differences in emotional arousal at the time of encoding. We present an online interval-estimation paradigm aiming to test this hypothesis as well as a computational model of this process.
C54
The development of the fronto-parietal network for number processing in early elementary school.
Aymee Alvarez-Rivero
► Show abstract
The fronto-parietal network supporting numerical processing emerges early in development and continues to scaffold the acquisition of mathematical skills later in life. Cross-sectional studies suggest age-related increases in parietal recruitment alongside decreases in frontal engagement; a pattern interpreted as reflecting growing specialization of numerical representations that require less reliance on cognitive control mechanisms. However, studies exploring within-participant changes in this network are severely lacking. The present fMRI study addresses this gap by examining longitudinal changes from kindergarten to first and second grade during a number-matching task. We characterized developmental trajectories within frontal and parietal regions and tested hemispheric differences in the coupling between neural and behavioral change in the parietal cortex. Consistent with prior work, we observed age-related increases in parietal recruitment. But contrary to prevailing hypotheses, frontal regions also showed increasing, rather than decreasing, engagement over time. Our findings suggest that prefrontal regions may play a sustained, supportive role in the development of parietal numerical representations during the acquisition of mathematics in elementary school, and the transition toward reduced frontal reliance may occur later, across longer developmental windows.
C55
The face and house inversion effects are robust at the individual level, but not related to behavioural measures
Calla Mueller
c7muelle@uwaterloo.ca
▶ Show abstract
The face inversion effect (FIE) is a highly replicated finding in which inverted faces elicit greater amplitudes on early visual ERP components (P1, N170) than upright faces. The inversion effect is not reliably reported for objects, but a recent study reported an inversion effect for houses (HIE) with a different time course than the FIE. We analyzed these FIE and HIE at the individual level (N=80). Mass univariate statistical analyses performed separately for each participant revealed a robust and more reliable FIE than HIE expressed in 75/80 participants vs. 66/80. Individual analyses also reflected the same patterns found at the group level in which the FIE is stronger than the HIE between P1-N170 while the HIE is stronger than the FIE between N170-P2. Furthermore, in this orientation discrimination task, these neural FIE and HIE did not correlate with the F/HIE at the accuracy or reaction time levels. These findings suggest that early visual processing is not directly related to behaviour and that the early neural inversion effects are independent from the behavioural ones. Therefore, caution should be used when attempting to draw parallels between these two separate literatures and their implications for our understanding of face and object perception.
C56
The neural indicators of grammatical co-activation in adverb placement processing of English-French bilinguals
Yubin Xing
yxing035@uottawa.ca
▶ Show abstract
This study explores the neural indicators of bilingual grammatical co-activation. Based on Friederici’s neurophysiological model of sentence processing (Friederici, 1995, 2002), we hypothesized that bilinguals’ neural correlates of L1 grammatical processing are modulated by their unique language experiences which monolinguals do not share. Specifically, we investigate how ERP signatures of L1 sentence processing (such as ELAN and P600) are modulated by L2 age of immersion (AoI). Three groups of participants were tested with English stimuli manipulated with the two opposing rules of adverb placement in English and French: early immersion bilinguals (n=20), late immersion bilinguals (n=16), and functional English monolinguals with no French immersion (n=23). EEG recording was time-locked to the onset of the adverb in a trial. Preliminary analysis (mean amplitude) on 39 participants revealed a marginally significant effect (p=0.079) of the group × condition × RoI interaction in the time window of 500~1000ms. Early bilinguals may be facilitated by the co-activated French grammar when processing the ungrammatical condition. They are also expected to show more inhibition in the early time window (150~400ms) of the grammatical condition due to a stronger French grammar compared to late bilinguals. Time-frequency analysis will be used to complement these findings.
C57
Whole-Brain Anatomical and Temporal Patterns for the Stop Signal Task
Bailey Thompson
baileyj.thompson12@gmail.com
▶ Show abstract
Response inhibition (RI), a key cognitive process underlying self-regulation, enables the suppression of prepotent or ongoing actions that no longer align with current goals. RI is commonly studied with the Stop Signal Task (SST), where participants must inhibit a response following a delayed stop signal. Evidence suggests RI involves multiple interacting cognitive processes supported by distributed neural networks. Similarly, SST-related fMRI-BOLD responses reveal coordinated activation across brain regions. Cognitive modes, which reflect task-general cognitive, sensory, or motor processes producing reliable BOLD patterns across tasks, provide a framework for examining these networks. We examined cognitive modes associated with successful and unsuccessful RI in 126 adults (ages 21–50). In line with our preregistered hypothesis paper (Thompson et al., 2025), results yielded the auditory attention for response (showing greater activation for unsuccessful than successful stops, consistent with response execution when inhibition fails) and re-evaluation mode (with activation patterns consistent with error monitoring and reinterpretation, peaking late in the trial and stronger for unsuccessful than successful stops). The focus on visual features mode also emerged, showing strongest deactivation during go trials, possibly reflecting disengagement in preparation for the upcoming visual cue. This study informs understanding of the complex neural processes underlying RI.
C59
Curiosity and Information Seeking: Effects of Age, Culture and Relevance
Shadini Dematagoda
sdematagoda@torontomu.ca
▶ Show abstract
Epistemic curiosity – the intrinsic drive for knowledge – is a key motivator of cognition and information seeking. Although relevance is thought to guide epistemic curiosity, current literature overlooks individual differences, assuming that what is “relevant” is universal. Self-construal offers a critical cultural lens: individuals with independent self-construal (common in Western contexts) often view themselves as autonomous, whereas those with interdependent self-construal (common in Eastern contexts) define themselves in relation to others. These differences are suggested to shape motivated cognition and may amplify with age as older adults prioritize meaningful content to manage cognitive resources. To address this gap, we examined how age and culture influence information seeking across personally-relevant, relationally-relevant, collectively-relevant and socially-irrelevant information. Younger and older adults (N=128) of Eastern and Western descent completed an information search task involving eight topics. For each topic, participants viewed an initial fact before choosing to reveal additional facts, providing a measure of information search. Multilevel modelling revealed significant effects of relevance, topic-level curiosity, and a three-way interaction between age, interdependent self-construal and relevance. Older adults with higher interdependence showed increased curiosity for relational information, but this did not extend to search behavior, suggesting a dissociation between culturally-shaped motivation and behaviour.
C60
Health Literacy as a Predictor of Quality of Life in Women with Polycystic Ovary Syndrome
Eleni Dubé-Zinatelli
edube024@uottawa.ca
▶ Show abstract
Health literacy encompasses four key processes involved in health-related decision making: acquiring, understanding, evaluating, and applying health information. In chronic conditions such as polycystic ovary syndrome (PCOS), these competencies shape how individuals interpret medical information, navigate healthcare systems, and engage in health behaviours, making health literacy an important determinant of health outcomes. The present study examined the role of PCOS literacy and its underlying competencies in predicting quality of life (QoL) among women with PCOS. A sample of 263 women with PCOS was recruited through online platforms, PCOS support groups, and personal networks. Participants completed an online survey assessing demographic and clinical characteristics, PCOS literacy, and PCOS-related QoL. Greater PCOS literacy was associated with better QoL (β = .385, t = 5.63, p < .001). Among the four literacy dimensions, acquisition (β = .279, t = 3.39, p < .001) and application (β = .178, t = 2.32, p = .021) were the strongest predictors of QoL, suggesting that the ability to access relevant information and apply it to self-management may be particularly important for well-being. Overall, these findings identify PCOS literacy as a potentially modifiable target for future interventions aimed at improving QoL in this population.
C61
Investigating the Relationship Between ADHD Symptoms, Video Game Experience, and Simulated Driving Ability
Cameron Dobko
camerondobko@bell.net
▶ Show abstract
ADHD symptoms are associated with poorer simulated driving performance, while video game play has been linked to improvements in areas like speed management and lane control. This study examined the relationships between ADHD symptoms, video game experience, weekly gaming, and simulated driving ability. One hundred undergraduate participants participated in the study (44 men, 53 women, 3 nonbinary, M age = 19.51). Video game experience was measured using a questionnaire, ADHD symptoms were measured using the ASRS V1.1, and driving performance was measured using STISIM Drive 3 driving simulator and was scored using the CRSD road test. Presence of ADHD symptoms was significantly correlated with speeding infractions (r = .212, p = .017), speeding duration (r = .203, p = .021), turning infractions (r = .281, p = .002), and time spent out of lane (r = .236, p = .009). Weekly gaming had a significant relationship with collision frequency, such that those reporting increased weekly play had fewer collisions (F = 2.600, p = .03). ADHD symptoms were not correlated with video game experience or weekly gaming in relation to simulated driving ability. This suggests that ADHD symptoms do not moderate the relationship between video game play and simulated driving.
C62
Mechanistic Creativity: An Active Inference Model of the Alternative Uses Task
Sarah Keating
smkeating@mun.ca
▶ Show abstract
Human innovation is supported by the ability to generate and evaluate novel and usefulideas. For instance, we know that the handle of a hammer can be used to roll breaddough. What are the mechanisms of this ability? Theories of grounded and embodiedcognition suggest that simulations of motor actions underly the ability to generate novelobject uses. Further, to overcome highly learned actions with actions, interoceptivesignals can act as sets points to guide action selection away from common actions. Weseek to provide a proof-of-concept mechanistic model of these embodied processesusing the mathematical framework of active inference, a novel computational techniquefor modelling perception and action. We show that agents who have learned actionaffordances of objects can readily use them in appropriate ways. Critically, bymanipulating the strength of known associations, and the interoceptive signals used toguide actions, our agents will select novel actions to known objects. To the best of ourknowledge, our simulations are the first to show mechanistically processes of embodiedcreativity.
C63
Similarity Ratings Reveal Expert-Novice Differences in Knowledge Organization
Paul Minda
pkalra7@uwo.ca
▶ Show abstract
Classic studies on the nature of expertise and expert-novice differences established that expert knowledge is structured differently from novice knowledge of the domain (i.e. hierarchically and taxonomically) and that experts selectively attend to relevant (diagnostic) features and ignore irrelevant features. However, these findings rely on semi-qualitative methods that require explicit self-reports by experts. These methods may overlook tacit or implicit forms of expert knowledge. In this study, we examine expert-novice differences in knowledge organization using similarity ratings. Geology novices and experts provided pairwise similarity ratings for images of typical and atypical geological specimens. The pairwise similarity ratings for each group were used to construct multidimensional space representations as well as network models. Both network and MDS analyses revealed clear taxonomic differentiation by the experts. The novice MDS dimensions were easily matched to perceptual feature ratings provided in the Rocks-30 dataset. However, some of the expert dimensions did not match either the perceptual feature ratings or expert self-reported features. These results reveal expert-novice differences in knowledge organization using a quantitative methodology including implicit aspects of expert knowledge. They also provide a baseline for future studies comparing the effects of instructional approaches on students’ knowledge organization.
C64
Cognitive Offloading to GenAI: Effects of Scaffolded Versus Unrestricted ChatGPT Use on Student Retention
Lindsay Richardson
LindsayRichardson@cunet.carleton.ca
▶ Show abstract
Generative artificial intelligence (GenAI) tools such as ChatGPT are rapidly becoming integrated into how students study and complete academic work. While some evidence suggests AI can support learning, other findings indicate that reliance on AI-generated answers may promote superficial engagement with course material. The present study examined whether the manner in which students engage with GenAI tools influences learning outcomes. Undergraduate participants were randomly assigned to one of three conditions: No AI, AI Instructed, or Unrestricted AI. Participants studied a neuropsychology passage and completed an initial comprehension test under their assigned condition. They then completed a surprise retention test without AI access. The dependent measure was percentage of correct responses on each test. This mixed-factorial design allowed direct comparison between AI-assisted performance and unaided retention. Unrestricted AI use is expected to produce the highest comprehension scores during AI-assisted testing but the lowest retention once AI access is removed. In contrast, scaffolded AI use is expected to produce stronger retention outcomes. These findings will help clarify when AI supports learning and when it may instead encourage unwanted cognitive offloading.
C65
Do students over-estimate their career-readiness skills? The Dunning-Kruger Effect in work-integrated learning courses
Cheryl Techentin
ctechentin@mtroyal.ca
▶ Show abstract
In 2021, the government of Alberta mandated that all university undergraduate students be given a work-integrated learning (WIL) experience prior to graduation. WIL pair students with employers to allow them to apply their course learning while developing career competencies. To accommodate this mandate, Mount Royal University began embedding WIL opportunities into individual courses.The present study examined the ability of undergraduate students to assess their career-readiness skills prior to and after taking embedded work-integrated learning courses WIL across courses offered at three different levels. One hundred students were asked to complete an anonymous survey delivered in Qualtrics assessing their skill development in a first-, third-, and fourth-year WIL course. For each course, students were asked to rate their perceived competence on each of the seven NACE core competencies prior to beginning and upon completion. Frequency analyses revealed that students reported continued growth and development of each of the skills across the first-, third-, and fourth-year courses, however, results suggest that, whereas first year students over-estimate their skill levels, students preparing to graduate tend to underestimate their readiness for employment. Discussion focuses on a possible Dunning-Kruger effect (Kruger & Dunning, 1999) and the importance of helping undergraduate students develop stronger metacognitive skills.
C66
Explaining categorical thinking: Specialized ability to form category prototypes?
Julia Schirmeister
► Show abstract
A common concern we often hear about human thinking is that it is categorical. We put things in boxes; we create artificially discrete classes of things. Categorical thinking can be explained as a particular form of optimal information compression: reducing distributions to their most-representative members (i.e. category prototypes). To demonstrate a human natural aptitude for identifying and using category prototypes, we designed a new classification learning task. Stimuli were abstract, coloured shapes. Participants learned to classify visual stimuli as either being category members or being random in appearance. Categories were contiguous subsets of the feature-space from which category members were drawn with uniform probability. Participant classification performance was best when the category was based around a category prototype (i.e. formed a hypersphere in feature space around a central point). Performance was worse when the category was stretched along one rotated dimension, creating linear dependence between features. Stretch preserved Shannon entropy, a measure of how predictable a distribution is. The stretched distribution was best summarized, not by a prototype, but by recognition of the linear relationship between features. Participants' improved performance in the prototype condition is used as evidence to support human pattern-recognition ability being specialized for identifying recurring similar instances.
C67
Culture and Recognition Memory Response Bias
Steve Lindsay
slindsay@uvic.ca
▶ Show abstract
Response bias on tests of old/new recognition memory can be liberal (more false alarms than misses), conservative (fewer false alarms than misses), or neutral (equivalent hit and false alarm rates). Prior research in our lab yielded evidence of a stable individual difference in recognition memory response bias for common English nouns: With equal numbers of old and new items and a flat payoff matrix, most undergraduates we tested were approximately neutral, but some tended conservative and others tended liberal across repeated tests and changes in context (Baldassari et al., 2019; Kantner & Lindsay, 2012, 2014; Takarangi et al., 2014). We have unpublished findings suggesting that Japanese undergraduates tend to exhibit conservative bias on tests of old/new recognition memory that elicit neutral bias on average among Canadian undergraduates. This poster summarizes that prior work and reports a new high-powered comparison of response bias among Canadian and Japanese undergraduates tested on recognition memory for abstract visual images
C68
Does encouraging distinctiveness heuristic enhance the between-subject production effect?
Brendan Redmond
bmr785@mun.ca
▶ Show abstract
The production effect refers to the finding that produced items (e.g., those read aloud) are better remembered than items that were not produced (e.g., those read silently), although this difference is smaller for between- compared to within-subject designs. One perspective is that participants use memory of having produced a given item to guide recognition judgements at test (the distinctiveness heuristic), giving rise to the performance benefit. It is theorized that when participants only read words e aloud (as in between-subjects studies), they are less likely to use this strategy. The current study examined whether providing participants with instructions to use a distinctiveness heuristic before completing a recognition test increased the production effect in a between-subjects design. Participants were assigned to one of three conditions: (a) Reading silently, (b) reading aloud , or (c) reading aloud with the addition of specific instructions to think about whether each item at test had been read aloud as a useful strategy to differentiate between “old” and “new”. Participants then completed a recognition test. Although we observed higher memory accuracy for produced items compared to non-produced item, our current findings suggest minimal to no additional effect when participants receive instructions to use a distinctiveness heuristic.
C69
Does reminding improve source memory accuracy?
Katherine R. Churey
kchurey@uoguelph.ca
▶ Show abstract
Repeated exposure to information is beneficial for memory, yet it can also lead learners astray. Indeed, it has been found that repeating items from a non-target source can be both beneficial and harmful for subsequent source discrimination (Jacoby, 1999). This may reflect the repetition-induced familiarity of non-target items increasing source confusions, unless that familiarity is opposed by veridical recollection of the item’s source. However, it remains unclear which factors promote the use of recollection over familiarity in source memory decisions. One potential factor might be the degree to which repetitions elicit reminding, as we recently found that reminding at study facilitates later recollection of reminded items thereby reducing false recognition of similar lures (Churey & Fiacconi, 2026). Extending this idea, repetition-induced reminding could harbour similar consequences for later source memory accuracy. Here, using the same method as Jacoby (1999), we found that noticing repetitions of items from a non-target source reduced source memory errors whereas failing to notice some repetitions increased those errors. These findings suggest that reminding during study is critical for downstream source recollection which enables more accurate source discrimination. Importantly, these findings highlight the potential utility of remindings in reducing false memories across several domains of memory.
C70
Does the reactivation positivity (PR) survive response execution?
Nadja Jankovic
nadja_jankovic@sfu.ca
▶ Show abstract
Presentation of a temporally leading stimulus can either facilitate (alerting) or impair (deviance distraction) the speed at which a visual target is detected. In studies of alerting, the leading stimulus has typically been presented on 50% of the trials. Studies of deviance distraction have typically employed two auditory tones that differed in pitch and probability of occurrence. One (e.g., the low-pitch tone) was presented frequently (e.g., 80% of the trials), the other (e.g., the high-pitch tone) was presented less frequently (e.g., 20% of the trials). In two previous experiments with visual cues, we found that cue probability alone failed to predict whether alerting or distraction would occur. In the present experiment, we asked whether match/mismatch in sensory modality would provide better predictive base.The leading stimulus was either visual or auditory. As such, it either matched or mismatched the trailing stimulus which was always visual. Contrary to expectations, neither condition showed any evidence of deviance distraction. These results are discussed in the context of current accounts of deviance distraction and differences between the alerting and the deviance distraction paradigms.
C71
Effects of Orthographic Neighbourhood Density on Naturalistic Reading in Schizophrenia and Dyslexia: An Eye-Tracking Investigation
Gabrielle Levasseur
glevasse@unb.ca
▶ Show abstract
Reading relies on rapid and efficient word recognition, shaped by several lexical properties, including orthographic neighbourhood density (OND; number of words that can be formed by changing, adding, or deleting a single letter in a target word). Recent work from our group has found that individuals with schizophrenia and those with dyslexia exhibit comparable impairments in lexical processing, including larger word frequency effects relative to controls during naturalistic reading (Whitford et al., 2026). Here, we examined whether OND similarly affects reading in these conditions, given its role in lexical support. Twenty adults with schizophrenia and 16 matched controls, as well as 18 adults with dyslexia and 16 matched controls, silently read four ~250-word texts for comprehension while being eye-tracked. Linear mixed-effects models revealed larger OND effects in both clinical groups, driven by slower processing (longer fixation durations) for lower-OND words. In schizophrenia, these effects occurred for both early- and late-stage reading (reflecting lexical access and semantic integration), whereas in dyslexia they were restricted to late-stage reading. Direct comparison of the clinical groups revealed larger early-stage OND effects in schizophrenia. Taken together, our findings suggest that reduced orthographic support differentially affects reading in schizophrenia and, to a lesser extent, in dyslexia.
C72
Enhanced Item but not Source Memory for Inverted Text
Aditi Roy
asroy1@ualberta.ca
▶ Show abstract
Previous research found that sentences displayed upside-down were recognized better (Kolers, 1973). Kolers suggested that memory benefits from re-enacting mental actions associated with material engagement. That, and other experiments of Kolers’, not only revealed an inverted-advantage for old/new recognition but sometimes for source judgment, judging whether the probe sentence had been studied inverted or upright. We address two limitations: First he used sentences from a single science textbook, leaving the research vulnerable to idiosyncratic effects of the materials. Second, participants judged a sentence as old or new and its orientation all in a single response. Our participants typed random lists of 40 words, half upside-down and half upright, and were tested with old/new recognition. Item-memory was better for initially upside-down words, replicating and extending Kolers’ findings (three experiments). Experiments 2 and 3 added a source judgement. An apparent advantage was explained by participants having a bias to think words were initially inverted in Experiment 2, which was absent in Experiment 3 with a change of instruction wording. These conceptual replications suggest that enacted features produced while reading inverted text act to strengthen item memory while not offering information diagnostic of the inversion, itself.
C73
From one thought to another: Chained memories and future thoughts originating from event and location cues
Mackenzie Bain
mbain29@uwo.ca
▶ Show abstract
People’s thoughts frequently originate from event (going shopping) or location (mall) cues. Previous work demonstrates that these cues influence the characteristics of people’s deliberately generated memories, and the chaining of spontaneous past and future events. In the current study, we investigated how cue type influences the characteristics of deliberate episodic simulations of past and future events, and how cues influence the chaining of multiple events. Forty participants completed an event fluency task followed by a chain connections task. Participants were shown 12 event or 12 location cues for 90 seconds each, and they generated as many cue-related past and future events as possible. Participants then were shown their generated events, and they identified how adjacent pairs of events were connected. Event cues prompted a greater number of past and future events than did location cues. The connections participants identified differed across cue type, but the pattern of connections was similar across temporality. These results demonstrate the influence that cues have on episodic simulations, and crucially, that this influence is consistent across past and future events. The present findings support theories in which deliberate simulations of past and future events are generated through the same underlying mechanism.
C74
How Culture Alters Autobiographical Memory Recollection
Christiane Marie Canillo
christianecan@gmail.com
▶ Show abstract
There are large differences in how people recount autobiographical memories that depend both on the characteristics of the rememberer and the listener of these memories. Prior work has shown that the cultural background of a person will affect autobiographical memory recall, with members of individualistic cultural groups recollecting with more specific details than members of a collectivistic group. This work, however, has not examined the impact of the cultural identity of the person listening to a memory on autobiographical memory. Prior studies have defined culture as a categorical variable (individualistic vs. collectivistic) rather than appreciating culture on a continuum. To address these limitations, we ran an online experiment where young adults from Canada (N = 140) provided a continuous measure of cultural identity, then performed an autobiographical memory task. Participants described personal memories under three conditions: unprompted, and to an audience from an individualistic and collectivistic culture. Linear mixed models examined the relation between cultural identity and recollection condition (unprompted, individualistic audience, collectivistic audience) on the details within the memories. Results show that culture significantly altered the details within these memories, both with respect to rememberer and audience. These findings contribute to a greater understanding of how culture impacts cognition.
C75
How do Subjective Cognitive Difficulties Affect Quality of Life During Perimenopause?
Denelle O'Neil
antmand@yorku.ca
▶ Show abstract
The menopausal transition is often associated with significant, and distressing, cognitive difficulties that impact daily life. These cognitive struggles are described as ‘brain fog’, encompassing everyday failures in memory and concentration. Although these difficulties are one of the most common, and unexpected, symptoms of the menopausal transition, the impact of subjective cognitive difficulties (SCDs) on quality of life is yet to be fully elucidated within the perimenopausal context. Thus, the present study examined whether the relationship between SCDs and quality of life in perimenopause can be delineated across specific domains of functioning. A sample of 269 participants, who were in perimenopause, were recruited from popular social media platforms and invited to complete a brief online survey. Controlling for age, SCDs were significantly negatively associated with all subdomains of quality of life, across relationships and material well-being (b=-.55, p<.001, adjusted R2=.11), health and functioning (b=-.66, p<.001, adjusted R2=.12), and personal, social, and community commitment (b=-.57, p<.001, adjusted R2=.10). These results are consistent with previous findings indicating SCDs as being associated with significantly poorer quality of life during perimenopause. Importantly, these findings underscore the deleterious impact of SCDs on a broad range of functioning during perimenopause.
C76
Implicit memory for unattended background information appears different in two different tasks.
Dana Murphy
kbassi464@my.nipissingu.ca
▶ Show abstract
In this study, we used measured implicit memory for words with a binocular suppression (BS) task and a fragment completion (FC) task. The priming phase of the BS task, participants identified a target word and ignored a distractor word. In the BS phase, participants identified a gradually appearing target word in one eye while under binocular suppression caused by a visual mask presented in the other eye. One-third of the BS target words were new words, 1/3 were presented as attended Prime targets, while the remaining 1/3 were ignored priming distractors. In the FC task, participants completed word fragments with 1/3 of the presented fragments for words previously presented as attended primes in the BS task, another 1/3 were ignored primes in the BS task, while the final 1/3 were new words. Participants were significantly faster to identify BS targets that were previously attended than BS targets that were previously ignored and both were identified significantly faster than new words. In the FC task, only previously attended words resulted in more accurate completion than unpresented words. Either the effects of previous presentation are different in the two tasks, or implicit memory fades with time for partially processed ignored words.
C77
Individual differences in metamemory ability predicts fluid intelligence
Ria Sahota
riasahota03@gmail.com
▶ Show abstract
To act intelligently in a complex environment, we need to represent task-relevant information in our working memory. To this end, individual differences in working memory capacity are known to predict individual differences in fluid intelligence. At the same time, not every working memory representation is accurate. Thus, it is imperative to evaluate the accuracy of working memory representations to prevent acting on inaccurate ones. To test whether this metamemory ability also predicts individual differences in fluid intelligence, we measured young adults’ (Experiment 1; n = 197) fluid intelligence, working memory capacity, and metamemory ability. The results showed that individual differences in fluid intelligence were reliably predicted by both working memory capacity (r = .66, p < 001) and metamemory ability (r = .41, p < 001). Furthermore, the preliminary results from Experiment 2 (n = 36/64) suggest that this metamemory ability may be trainable by providing participants with performance feedback. Taken together, our results reveal the importance of metamemory ability in supporting intelligent behaviors and suggest a possible means to improve it.
C78
Inter-item spacing of short lists is judged in two different ways
Jeremy Caplan
► Show abstract
Repeated items within lists are better serial-recalled for immediate repetitions whereas spaced repetitions are impaired (Ranschburg Effect). But serial recall confounds item and order memory. With procedures close to most Ranschburg studies, we asked participants to estimate the spacing between a pair of probe items. Accuracy for rapidly presented lists of 7 consonants fell along a Weberian scaling function above spacing=1, while spacings 0 and 1 diverged from that relationship. Repetitions were facilitated at spacing 0 in one experiment and extending to spacing 1 in another experiment. At larger spacings, there was no effect of repetition. The presence of massed repetitions did not influence accuracy when the non-repeat items were probes, ruling out the possibility that an immediate repeat is functionally like a single list-item. Finally, response time increased monotonically with spacing, suggesting participants were sequentially recalling the list. In sum, these findings suggest that participants are not readily able to make direct comparisons of positions within short lists, compatible with inter-item associative models and some relative-order coding models but potentially problematic for models that presume high-precision positional information. Rather, at large lags, spacing has to be derived and as Henson (1998) suggested, immediate repetitions are tagged as such.
C79
Investigating Memory Integration as a Mechanism for the Testing Effect: Inconsistent Evidence and Its Implications
Lauren Homann
lauren.homann@mail.utoronto.ca
▶ Show abstract
The "testing effect" (enhanced memory after retrieval practice versus restudy) is considered among the most robust findings in memory research, and one account proposes that retrieval strengthens memory by integrating related memories. We examined this account across three experiments (N=122). On Day 1, participants encoded person-object-location events; related event pairs shared an object and scene type but featured different people. Pairs were assigned to retrieval practice (reimagining the event from a face cue, with feedback) or restudy (full re-exposure), and participants answered questions probing each event's location and object category. On Day 2, we assessed memory for directly-reviewed associations (location and object category) and integration, operationalized as associative inference (inferring relationships across related event pairs) and misattribution errors (misattributing background details from a related event). Across studies, condition-level (retrieval vs. restudy) testing and integration effects were inconsistent. Trial-level analyses showed more integrated memories had better-retained directly-reviewed associations. These results suggest that testing effect variability is not noise, but a signal that retrieval and restudy are not unitary processes. We discuss contextual factors that may shape which processes retrieval practice and restudy engage and the representational reactivation they elicit, which may determine when and why their mnemonic outcomes differ.
C80
Is Metacognitive Monitoring Vulnerable to Working Memory Load?
Leili Rouhi
leili.rouhi@usask.ca
▶ Show abstract
This research examines how working memory (WM) load affects an individual’s accuracy in metacognitive monitoring. Although metacognitive monitoring is central to adaptive decision-making, existing findings on the role of WM are inconsistent, partly due to methodological confounds in which WM demands are embedded within the same tasks used to assess monitoring.To address this limitation, we isolate WM load from metacognitive processes using a dual-task approach. In a large-scale study (N=499), participants performed a reasoning task under conditions of either time pressure, WM load or both. Metacognitive accuracy was indexed by the ability to discriminate between conflict (mismatch between logical validity and conclusion believability) and non-conflict (aligned logical validity and believability) trials. Contrary to our hypothesis, results did not show a significant interaction between conditions and conflict detection, suggesting that metacognitive monitoring of conflict items remains stable regardless of WM demands.Building on these findings, a follow-up study (N=100) combines an analogy reasoning task with a concurrent WM task while manipulating cue diagnosticity. Our prediction is that individuals are less accurate in their monitoring judgement and less sensitive to valid cues while they are under WM load. Together, this work clarifies how cognitive constraints shape self-evaluation in complex problem-solving contexts.
C81
Language Based Statistical Learning is a Stable Individual Trait
Amiya Aggarwal
aaggar7@uwo.ca
▶ Show abstract
Statistical learning (SL) is the process of becoming sensitive to patterns in the environment and occurs simply through exposure to input. Some prior studies have shown reliable domain-specific individual differences, though evidence is primarily limited to explicit measures of learning. Given that explicit memory abilities are generally stable at the individual level, results from these prior studies may reflect differences in explicit memory performance, rather than sensitivity to patterns in input. To further understand whether SL is a stable trait that reliably differs among individuals, we tested 97 participants’ ability to learn syllable patterns in two distinct artificial language streams separated by a two-week delay, using multiple measures of learning. Both testing sessions included exposure to a unique artificial language composed of repeating trisyllabic words. We found performance on both implicit and explicit SL measures significantly correlated across sessions. This evidence of stable individual differences encourages further investigations targeted at understanding whether implicit statistical learning performance predicts individual differences in real-world language learning.
C82
Learning From Experience: Studying Semantic Memory Formation using a Naturalistic Paradigm
Casey Aurin
▶ Show abstract
Semantic memory refers to the accumulation of general knowledge about the world acquired across the lifespan (Tulving, 1972; Nelson & Fivush, 2004). Semantic knowledge structures support learning by facilitating the integration of newly learned information with existing knowledge. Relatively few studies have examined semantic memory formation in real-world, but experimentally-controlled, settings. In the present study, we examined how well semantic information is acquired from naturalistic experiences, the types of semantic information retained, and relations between semantic and other types of memory. Young adults (n = 85; 18–22 years old; M = 19, SD = 0.99) took part in a 1-hour guided walking tour at a living history museum, during which they learned facts about 10 distinct historical locations. After a five-day delay, participants were asked to recall everything they learned during the tour. We developed a novel semantic coding scheme and coded transcripts for (1) accuracy and (2) the types of semantic information retained (e.g., biographical, artifact details). We found meaningful variability in the amount of information retained and that certain types of semantic details were better recalled (e.g., historical figures). Relations between semantic fact retention and episodic memory accuracy (e.g., temporal, spatial) will also be reported and discussed.
C83
Listener age affects age-related differences in the episodic specificity and flexibility of autobiographical recollection
Bryan Hong
► Show abstract
When recalling events from the personal past, older adults typically produce fewer episodic details than younger adults. However, this pattern is often assessed in interview-style studies in which participants recall events to younger adult experimenters, raising the possibility that older adults shape autobiographical recall to accommodate the knowledge of a younger listener. Here, we investigated this question by having younger (n = 76) and older adults (n = 80) initially recall personal events with no audience in mind, before recalling these again to target listeners who were either younger or older adults. Notably, when the listener was an older adult, there was no age-related episodic memory decline, with both younger and older adults producing a comparable number of episodic details. Moreover, although both groups downshifted episodic details when recalling to a different-aged listener, a natural language processing analysis showed that older adults also varied the similarity across their retellings and incorporated more listener-specific information at recall, indicating greater flexibility. Altogether, these findings demonstrate how the social context can mitigate changes in autobiographical memory across the lifespan.
C84
Organization of the Bilingual Mental Lexicon: Memory Effects of Translations Versus Semantic Relations in Simultaneous and Late Bilinguals
Laura Sabourin
asuur049@uottawa.ca
▶ Show abstract
We are investigating whether the bilingual mental lexicon shows evidence of integration, to what degree this integration exists, and whether age of acquisition and short-term memory play roles in variability. The semantic interference effect was compared to translation interference in word production (Tomoschuk et al., 2021), but not word processing. We compare reaction times for semantic relations and translations in a word recognition task, highlighting cognitive connections not yet clearly defined. This way, we infer whether simultaneous and late bilinguals have stronger lexical connections within-language or between-languages. Participants memorize 30-40 words and then complete a recognition phase where they identify 60-80 words as old (seen before) or new (unrelated, semantically related, or translation of a memorized word). The word lists are divided into 4 sets: two unilingual (one English; one French) and two bilingual. We predict that simultaneous bilinguals will take longest to reject translation equivalents, followed by semantically related words, then unrelated words, showing evidence of mental lexicon integration. Late bilinguals will reject translation equivalents and unrelated words equally quickly and semantically related words will be slower, showing evidence of mental lexicon separation. Online testing will be completed by mid-April.
C85
Recasting the Past as the Future in Younger and Older Adults
Jasmine Collins
jcollins@research.baycrest.org
▶ Show abstract
Research on episodic future thinking (EFT) primarily uses experiments designed to elicit the imagining of novel events. However, these future events are often unrealistic and unrelated to personal plans. We hypothesize that in everyday life people frequently re-imagine past events as occurring again in the future, a process we have called “recasting.” In particular, when imagining the near future, recent experiences should provide an informative template. Additionally, given age-related declines in EFT, older adults may be more likely to recast. This online study used an unconstrained EFT task to investigate the degree to which younger (N=87) and older (N=86) adults use recasting when imagining near vs. distant future events, and whether recasting influenced the cognitive attributes of the future events imagined. Overall, 67% of future events were reported to be based on a past event. Participants were more likely to recast when imagining near compared to distant future events. Recast events were rated as easier to imagine, more detailed, and more plausible, especially when the template memory was a regularly experienced event. All of these effects were evident in both age groups, with no significant interactions. These findings support recasting as a more naturalistic form of future thinking.
C86
Robustness of Prediction-Based False Memory in Naturalistic Contexts
Mahveen Salman Mubarak
mahveen.mubarak@mail.utoronto.ca
▶ Show abstract
The human brain constantly predicts upcoming events to interpret the world in real time. While predictive processes are known to influence memory for simplified stimuli, their impact on memory for naturalistic events remains unclear. To test this, participants viewed first-person videos of everyday events that paused midway, prompting them to predict what would happen next. The unfolding event then either confirmed the prediction, violated it, or ended without revealing the true outcome. Critically, when memories of actual outcome were later assessed, predicted events were misremembered as actual outcome even when the prediction was contradicted or no feedback was presented. In subsequent experiments, we showed these prediction-based false memories amplified in incidental encoding (Experiment 2) and persisted across a 24-hour delay (Experiment 3 & 4). We further demonstrated that subjective surprise (Experiment 5) or objective unexpectedness (Experiment 6) of the actual outcome could not explain the robustness of prediction-based false memories. Finally, we demonstrate that memory for predicted and actual events could coexist, suggesting that prediction can become integrated into memory representations alongside real outcomes (Experiment 7). Together, these findings highlight prediction as a reliable driver of memory distortions in everyday tasks, underlining the role of expectations in shaping episodic memories.
C87
Semantic Similarity Disrupts Order Recall: The Key Role of Task Difficulty
Ian Dauphinee
► Show abstract
Similarity is known to affect short-term order memory across many stimulus domains, typically impairing recall. Semantic similarity, however, has long been treated as an exception: it reliably benefits item memory while often appearing to leave order memory unaffected. Three studies nevertheless reported a detrimental effect of semantic similarity on order recall. Here, across five large-scale experiments, we revisited these findings under stricter methodological conditions and again observed a detrimental effect of semantic similarity on order recall. We then conducted a systematic review and random-effects meta-analysis, which revealed a small but reliable detrimental effect across the literature. To explain the apparent inconsistencies in previous findings, we identified task difficulty as a likely moderating variable. Across four additional experiments, we manipulated task difficulty through presentation speed and list length. As predicted, the detrimental effect of semantic similarity on order recall emerged under more difficult task conditions but disappeared when the task was easier. These findings show that semantic similarity is not a special case for short-term order memory and that its effect depends in part on task difficulty, helping reconcile previously mixed results and clarifying constraints on models of serial order memory and their treatment of semantic representations in immediate recall.
C88
The Effect of Retrieval Suppression on the Probability and Fidelity of Recall
Nathaniel Wells
nkw020@mun.ca
▶ Show abstract
Retrieval suppression is often treated as an all-or-none process, evaluated with categorical measures such as cued recall or recognition. This leaves open a mechanistic question: What changes when people repeatedly stop retrieval? Does suppression reduce the probability that a memory can be accessed, degrade its fidelity when accessed, or both? We adapted the Think/No-Think (TNT) paradigm to a continuous-report framework to dissociate these components. Participants learned word-object associations, with each object encoded in a unique colour and spatial location later reproduced on continuous response wheels. Learning was reinforced with test-feedback and a criterion phase. Participants then completed a TNT phase with direct-suppression instructions: Think trials required retrieving and holding the object and its features in mind, and No-Think trials required preventing the object and its features from entering awareness without thought substitution. Baseline objects were not cued during this phase. At final test, participants recalled object names and reproduced features. Object-name recall was analyzed with Bayesian mixed-effects logistic regression and showed evidence of suppression-induced forgetting, whereas Bayesian mixture modelling of colour and location errors showed no clear evidence that No-Think altered retrieval probability or precision. Results suggest direct suppression can impact categorical access while leaving feature fidelity largely unchanged.
C89
The False Fame Effect and Memory Processes through Online Delivery
Grace Bucci
grace.bucci@unb.ca
▶ Show abstract
Researchers (e.g., Yonelinas, 2002) propose that conscious and unconscious memory processes are independent yet cohesive. These processes can be examined using a task, in which participants initially read a list of nonfamous names. Next, 24 hours later, participants complete a fame judgment task on the previously presented old nonfamous names, new nonfamous names, and famous names (Jacoby et al., 1989). The false fame effect occurs when old nonfamous names are judged as being famous, demonstrating a weakening of conscious memory and a strengthening of unconscious memory. The present study aims to replicate this task via online delivery. Participants (N=50) will complete the false fame task through the Qualtrics platform during either a morning or evening time slot. This pilot study is part of a larger project that investigates whether time of day affects the false fame effect. It is expected that participants will judge famous names most frequently as famous, followed by the old nonfamous, and lastly the new nonfamous names. Completion of the study through an online delivery should increase ecological validity; related work on the spread of misinformation (thought to be elicited by these memory processes) has typically been conducted online (Ruggieri et al., 2023).
C91
The Influence of Physical and Virtual Puzzles on Parents’ and Children’s Spatial Language Production
Marie Pier Grégoire
mgreg097@uottawa.ca
▶ Show abstract
Spatial language refers to words used to describe objects’ location, features, and relation to other objects. Parents vary in how much spatial language they produce during puzzle play. When parents produce more spatial language, it increases their children’s production of spatial language. In turn, children’s spatial language production predicts their spatial skills–recalling, encoding, and mentally manipulating objects or space. With the emergence of technology, children are shifting from physical (3D) to virtual (2D) play, yet little is known about how this impacts spatial language production between parents and children. Thus, we investigated whether puzzle format (physical, virtual with rotation, and virtual without rotation) moderates the relation between parent and child spatial language production. Parents and their 3- to 5-year-old child (N = 162) completed one of three puzzles, and their proportion of spatial language production was coded and analyzed. Results showed that parents’ spatial language production was significantly positively related to children’s spatial language production during both virtual puzzles, but not during the physical puzzle. These findings suggest that children may depend more on their parents’ spatial language cues during virtual puzzle play than during physical puzzle play.
C92
The Spatial Production Effect: Is Embodied Cognition the Key?
Monika Daigle
emd3330@umoncton.ca
▶ Show abstract
The production effect refers to better memory for information that is actively produced rather than passively encoded. In verbal memory, this advantage is typically observed for words read aloud, and the Revised Feature Model (RFM) explains it as the joint outcome of two opposing consequences of production: it adds distinctive modality-dependent features to memory traces, but disrupts rehearsal. Recently, Saint-Aubin et al. (2024) extended this logic to a spatial dot task in which locations are presented one at a time and later reconstructed in order. In that study, clicking each dot with a mouse impaired performance relative to passive viewing, but a positive production effect emerged when the comparison condition also disrupted spatial rehearsal. This study asked whether a more natural, body-anchored response—touching each dot on a touchscreen—would enhance spatial memory, as suggested by embodied cognition accounts. Across two experiments, 48 participants encoded sequences of eight spatial locations and later reconstructed their order. Touch-based production again impaired memory relative to remaining still, but a clear benefit emerged when an irrelevant finger-tapping task reduced rehearsal availability. These findings support the RFM account: production can enhance distinctiveness, but its net effect depends on how much it also disrupts spatial rehearsal.
C93
Trajectories of Cognitive and Functional Performance in Subjective Cognitive Decline: Evidence from the Canadian Longitudinal Study on Aging
Stefanie Gard
► Show abstract
Subjective cognitive decline (SCD), especially with worry, may signal early risk for cognitive and functional decline. Not all individuals with SCD progress to dementia, but subtle functional changes in daily activities may precede measurable cognitive deficits, emphasizing evaluation of everyday functioning alongside cognition. Participants (n = 18,971) were drawn from the Canadian Longitudinal Study on Aging and completed testing at two follow-ups three years apart. SCD was classified as no SCD, SCD without worry, or SCD with worry. Standardized measures of cognitive performance across multiple domains (memory, executive function, processing speed, fluency) and functional ability were analyzed using a two-step cluster analysis at follow-up 1; clusters were compared on SCD status, cognitive and functional performance, and demographics. Two clusters emerged: one (82%) with preserved cognition and functional independence, and another (18%) with lower cognition and functional limitations. Cluster 2 showed slightly declining cognition, reduced independence over time, and a higher prevalence of SCD with worry. This group was older, more often female, and had lower educational attainment. Functional changes accompanying SCD with worry may precede measurable cognitive decline. This highlights the importance of combining subjective and functional assessments with objective cognitive measures, which may improve early identification of at-risk individuals.
C94
Visual memory of coloured shapes
Brooke Vachon
bvachon1@laurentian.ca
▶ Show abstract
Participants (n = 30) completed 900 trials of a continuous recognition task. On each trial an image was presented and participants indicated whether or not they had seen the image previously. The images were a series of textured and coloured shapes each composed of three lines. In the instructions and practice trials images were defined as new if either the shape had not been previously seen or if it had been seen in a different colour. Some shapes were repeated up to four times in a given colour. Additional repetitions took place in other colours. Accuracy and reaction time improved over repetitions. Most importantly, when a shape was presented in a new color performance was reduced relative to a repetition in the same colour; as if a new stimulus was presented. This finding is mostly consistent with the integration of colour and shape.Abstract in French too.
C95
Issues of Generalization from Unreliable or Unrepresentative Psycholinguistic Stimuli: A Case Study on Lexical Ambiguity
Jiangtian Li
jiangtianli91@gmail.com
▶ Show abstract
When a speaker describes a unique banana to a listener, saying “the banana” is sufficient; adding any other information would be redundant. Previous work has focused on the production of redundant color adjectives, as in “the yellow banana.” We extend this research to state adjectives, such as “dirty” and “open.” In a free production experiment, speakers saw four pictures and had to tell their partner which one to click on. Objects in marked states elicited far more redundant state adjectives (dirty stovetop or open gate: 32%) than objects in unmarked states (clean stovetop or closed gate: 1%). Parallel to research on color, we demonstrate that modification rates greatly depend on how typical the state is for the particular object. Surprisingly, however, typicality does not replace markedness: a further significant portion of the variance is explained by the more abstract concept of markedness. Using a Bayesian rational speech act model, we show that the production of redundant state adjectives cannot be fully explained by the informativity of the words alone, but relies on other information encoded in markedness, potentially the negative valence encoded in marked states such as “dirty,” “broken,” and “bent.”
D01
Pharmacological Modulation of Sleep-Like State Alternations in Urethane Anesthetized Rats
Abigail Riesen
► Show abstract
Sleep's function and the related patterns of brain activity remain incompletely understood. Studying sleep is methodologically challenging due to its fragility, necessitating a more tractable model. Our lab has shown urethane anesthesia to be an unparalleled model of sleep in rats, producing spontaneous, rhythmic alternations between rapid eye movement (REM)-like and non-REM (NREM)-like states. Extensive central and peripheral measures indicate that these alternations closely resemble those observed in natural sleep.
Both sleep, and the sleep cycle, are affected by centrally acting drugs. The most common over-the-counter sleep aid is the antihistamine diphenhydramine (DIP), which promotes drowsiness and biases brain activity toward NREM-like patterns. Here we evaluated several antihistamines, including DIP, on the states produced during urethane anesthesia.
Bipolar electrodes were used to collect LFP recordings from the cortex and hippocampus of urethane-anesthetized rats before and after I.V. administration of DIP, loratadine (peripheral control), or PBS (vehicle control). Initial results have indicated that DIP produces a shift into a transitional state between activated and deactivated patterns, with broad increases in slower power in cortex and slow and theta frequencies in hippocampus. These results suggest that DIP promotes NREM-like patterns but not deep slow-wave activity states in the urethane model.
D02
A Spreading-Activation Model of Causal Reasoning
Dominic Le
dominic.le@mail.utoronto.ca
▶ Show abstract
People reason causally in two directions: predicting effects from causes and diagnosing causes from effects. The direction of causal reasoning affects the certainty of people's inferences and whether people integrate alternative explanations. Few studies have investigated how these directions differ in memory retrieval during inference, possibly due to the scarcity of causal reasoning process models. We propose a mechanistic computational cognitive model of causal reasoning that explains observed asymmetries in diagnostic versus predictive inference. We hypothesize that diagnostic reasoning follows depth-first search, elaborating single causal chains while avoiding alternatives, whereas predictive reasoning follows breadth-first search, readily exploring alternative effects. We present BLISA (Bidirectional causal reasoning using Lateral Inhibition in a Spreading Activation process), formulated using the cognitive architecture CLARION and capable of simulating individual-level choices and response times. Simulations (N=24,000) reveal that diagnostic inferences take longer when multiple causes compete, while predictive inferences slow when multiple effects branch. Retrieval failures occurred only during diagnoses with branching causes. These results demonstrate that directional asymmetries depend on causal structure and emerge from elementary retrieval mechanisms operating through distinct search patterns in spreading activation.
D03
Breaking Bad Agreement: Intervening on Correlated Errors in Human Judgment
Hamza Tariq
h33tariq@uwaterloo.ca
▶ Show abstract
When multiple analysts provide similar numerical estimates, people often treat agreement as evidence of accuracy and consensus, overlooking the possibility that the estimates could be sharing common methods and sources of error. We examined how the potential for correlated errors influences judgments when multiple estimates inform a single decision. Participants evaluated sets of Initial Public Offering (IPO) estimates that systematically varied in agreement spread — from high agreement to high disagreement around a similar mean. We tested interventions aimed at reducing overreliance on agreement by informing participants about the possibility of correlated errors and by showing them the sources of the estimates. Results revealed a robust bias toward high agreement, but also showed that this bias can be attenuated through interventions. Ongoing work builds on these findings by developing and testing additional debiasing strategies to improve sensitivity to correlated errors and helping people better distinguish between misleading consensus and genuine agreement.
D04
Influence of Experience and Training on Go/Reject Decision-Making During Airplane Takeoff
Sam Clément-Coulson
coulson.sam@outlook.com
▶ Show abstract
Pilot experience may vary demands on attention, situational awareness, decision-making, and workload management, making it a powerful context for studying human cognition under dynamic, high-stakes conditions. Thus, we examine the respective contributions of experience (flight hours) and formal training to rapid decision-making in performance-limited simulated aircraft take-offs. Pilots holding private, commercial, and airline licences completed 30 simulated take-offs in X-Plane 12. The aircraft simulated was a Beechcraft Bonanza, where density altitude was manipulated to produce scenarios classified as safe, marginal, or unsafe based on aircraft performance data. Safe conditions met both runway length and climb performance requirements; marginal conditions met runway length but not climb performance requirements; unsafe conditions met neither.Data collection included flight performance, flight data recordings, go/no-go decisions, and pilot electrocardiography (ECG), eye-tracking, and post-trial subjective evaluations. Analyses focused on differences between pilots as a function of experience and formal training, despite their natural correlation. Differences in decision-making, objective safety margins and subjective evaluations are examined between groups of pilots. ECG and ocular measures are treated as exploratory indicators of workload and attentional processes. The study aims to contribute to a research program aimed at disentangling the roles of experience and structured training in pilot expertise development.
D05
Metacognitive Beliefs and Confidence
Sofia Sierra
sgsierra@uwaterloo.ca
▶ Show abstract
Our beliefs about how our minds work play an important role in how our metacognitive judgements are formed. Although much work has examined the content of our metacognitive beliefs, less work has focused on other characteristics of those beliefs. Here we focus on one of these neglected characteristics – the confidence in one’s metacognitive beliefs. According to Koriat’s self-consistency model of confidence, when answering questions about what one believes, individuals construct their judgments by sampling relevant information from memory, with confidence being a product of the consistency of the information retrieved. Here, we tested this account by examining the relation between the consensuality of participant beliefs about a diverse set of cognitive phenomena and their confidence in those beliefs. Consistent with Koriat’s self-consistency model, a strong positive correlation between item consensus and average item confidence was observed.Koriat, A. (2012). The self-consistency model of subjective confidence. Psychological Review, 119(1), 80–113. https://doi-org.proxy.lib.uwaterloo.ca/10.1037/a0025648
D06
Previous responses bias present decisions
Ver-Se Denga
► Show abstract
Recognition memory tasks assess whether participants can discriminate "old" (studied) from "new" (unstudied) items. A fundamental assumption in many models of recognition is that each old/new decision is independent of all others. However, there is mounting evidence that the likelihood of responding "old" on a given trial increases if an "old" response was also made on the previous trial, indicating that memory decisions are dictated by information apart from that provided by study phase experiences. This response carryover constitutes a bias effect, with potentially detrimental impacts on memory performance. Across two experiments, we explored the mnemonic properties of this bias effect in two ways. First, we examined whether the response carryover effect was sensitive to the subjective difference between remembering and knowing. Second, we examined whether the response carryover effect was sensitive to judgments of recognition confidence. The results of two experiments revealed that the response carryover effect was sensitive to both of these mnemonic properties, with particularly strong response carryover effects for "remember" responses, and for consecutive trials responded to with similar confidence levels. These results support a view in which the response carryover effect reflects the carryover of memory-related evidence across trials, rather than a mere response perseveration effect.
D07
A Wedge in Perception: Bilingual Advantage in the Detection of Coarticulation Violations
Anna Cole
annamcole@cmail.carleton.ca
▶ Show abstract
Over development, monolingual infants undergo perceptual narrowing, whereby auditory speech processing becomes specialized for their native language. By contrast, bilingual infants retain perceptual sensitivity, laying the foundation for the Perceptual Wedge Hypothesis. This theory proposes that exposure to multiple languages modulates perceptual narrowing, thereby increasing phonological sensitivity.While existing literature on the Perceptual Wedge Hypothesis has focused on infants and toddlers, it remains unclear whether this perceptual advantage persists into adulthood. The present ERP study investigates differences in phonological processing between English monolinguals and native English speakers proficient in a second language.Using an auditory adaptation paradigm to measure both behavioural responses and two ERP components, the mismatch negativity (MMN) and the P3a were measured. Participants were presented with English words manipulated through cross-splicing that differed only in their final consonants. Cross-spliced stimuli were created by swapping onset–nucleus segments, resulting in words with incorrect phonological representations. Sequences ended with either a correctly produced or cross-spliced word, and participants judged whether the final sound deviated from preceding stimuli.Both behavioural performance and MMN amplitudes were comparable across groups. However, Bilinguals exhibited larger P3a responses, suggesting equivalent early detection of phonological mismatches but enhanced attentional evaluation following deviation processing.
D08
Baby Talk: A Scoping Review of Language and Reading Measures in Infantile Spasms Research
Mia Collins
miacollins@trentu.ca
▶ Show abstract
Infantile spasms is a type of epilepsy that presents before age two; a critical developmental period for language acquisition and prereading skills. There is limited research examining the language and reading outcomes of those with a history of infantile spasm. Although some evidence indicates age of onset and early speech and language support are associated with communication skills in this group, a comprehensive understanding of developmental outcomes is scarce. The present study is a qualitative analysis of articles published between 1990 and 2024 that included measures of language and word reading in participants with a current or past diagnosis of infantile spasms (e.g., West Syndrome). Fifteen articles were identified with only one study conducted in Canada and the majority being conducted in the USA (k = 6). Most studies included the Bayley Scales of Infant and Toddler Development as a measure for language (k = 9), while only one study included a measure of reading skill. No studies investigated language or reading beyond 42 months of age and no studies included a control group. These findings highlight a critical need for Canadian research examining language and reading outcomes of those with a history of infantile spasms, particularly later in life.
D09
Blunted Engagement or Increased Effort? Ocular Markers of Naturalistic Reading in Depression
Akshaya Kirithy Baskar
akshayakirithybaskar@cmail.carleton.ca
▶ Show abstract
Major depressive disorder has been linked to alterations in language usage yet its influence on naturalistic reading behavior remains unexplored. The present study investigates depressive symptoms associated with difference in reading speed, effort allocation, and sustained engagement during text processing. 34 participants with depressive symptoms and 34 matched controls completed a story-reading task, while eye tracking was performed. The task embedded within a broader cognitive assessment of vocabulary, grammar and nonverbal IQ. This design enabled us to disentangle language ability from engagement-related processes indexed by eye movements and pupil dilation. We examined whether depression predicts overall reading speed, within-subject variability in fixation duration, and rereading frequency, while statistically controlling individual differences in language and cognitive skills. Parallelly, task-evoked pupil dilation is being analyzed as a physiological marker of effort allocation. Two competing possibilities under consideration are, MDD associated with (1) reduced dilation, reflecting blunted engagement, (2) increased dilation reflecting greater perceived effort required to maintain comparable performance. By integrating behavioral, ocular and cognitive measures, this study aims to answer observed differences in reading behavior reflects altered cognitive capacity, changes in effort mobilization, or variability in sustained engagement. This approach offers new insight into depression shaping real-time language comprehension.
D10
Crosslinguistic Effects of Literacy Interventions in Elementary School Children: A Scoping Review
Laura MacGrath
laura.macgrath@mail.utoronto.ca
▶ Show abstract
Background and Rationale:About one-fifth of Canadian homes are multilingual. While crosslinguistic effects of oral language interventions have been reviewed, evidence for written language is less clear. Literacy is a central focus of education policy, highlighted by the Ontario Human Rights Commission’s Right to Read Report (2022), which identified bilingual children as underserved. Understanding crosslinguistic effects can clarify how bilingual children draw on their full linguistic repertoires.Purpose:What evidence exists for crosslinguistic transfer resulting from reading and writing interventions for elementary school–age children?Methods:Following PRISMA ScR guidelines, we searched six databases. Title and abstract screening is underway. Ancestral and Google Scholar searches will supplement database searches. Eligible studies examine bilingual children aged 5–12 receiving literacy instruction/intervention and report crosslinguistic effects. Extracted data will include intervention characteristics, targeted skills, language of instruction, participant traits, and outcome measures.Anticipated Results:Existing evidence suggests crosslinguistic transfer for certain language pairs and literacy subskills. Instruction in an additional language often supports, and does not hinder, literacy in another language.Significance:Literacy is a social determinant of health, yet one in five Canadian adults has low literacy. Synthesizing this literature will inform future research into equitable literacy supports for linguistically diverse children.
D12
Do language learners show a bias for transparent vocabulary? Insights into the development of compositional language structure.
Aya Amer
ayaamer@cmail.carleton.ca
▶ Show abstract
Although reading emotional words generally prompts an affective response, bilinguals seem to be “detached” from the emotional content of words in their L2 (Aguilar et al., 2024; Pavlenko, 2012). Previous studies of this phenomenon have defined emotion by a combination of valence and arousal, with no consideration of categorical differences between affective states. We investigated bilinguals’ reduced sensitivity to L2 emotional words using a lexical decision task. L1 English speakers (n = 148) and L2 English speakers (n = 119) performed lexicality judgements on words that differed in their valence, arousal, and relatedness to one of the five basic categorical emotions (happiness, sadness, fear, anger, and disgust). We found that happiness- and fear-related words facilitated response times in L1 speakers, while words that were more strongly related to disgust prompted slower reaction times. Happiness and disgust effects were absent in L2 speakers; however, the facilitatory effect of fear was preserved. Crucially, categorical emotion ratings captured the variance in reaction times better than valence and arousal considerations. Not all L2 participants displayed equal indifference to the emotional dimensions of words: a combination of early age of acquisition, high proficiency, and frequent daily use of English resulted in more native-like reactions.
D13
Does speaker age impact preschoolers’ predictive use of disfluencies?
Brooke Johnston
bajohnst@uwaterloo.ca
▶ Show abstract
Disfluencies (such as “uh” and “um”) provide listeners with useful information, including helping them anticipate the type of linguistic material that will follow. When a speaker is disfluent, adult listeners predict they will change topics or talk about something difficult to describe (e.g., an unfamiliar object). Preschoolers similarly use disfluencies to predict reference to discourse-new objects, but existing work suggests that they do not anticipate difficult-to-describe objects. However, these studies have only used adult speakers. In the present study, we ask whether children are more likely to use disfluency to predict reference to unfamiliar objects for child speakers, as they may have a better sense of which objects are more difficult for their own age group. Children aged 3.5- to 4.5-years were presented with pairs of familiar and unfamiliar objects while listening to recordings of fluent and disfluent adult and child speakers. Looking behavior to the objects was coded offline. Preliminary data suggest that preschoolers use disfluencies to predict reference to unfamiliar objects for both child and adult speakers, with no interaction between speaker age and fluency. This pattern challenges previous findings with preschoolers and suggests that children make the same types of disfluency-based predictions as adults.
D14
English Exposure Impacts Polish Lexical Categories
Chelsea McKenzie
cmcken7@uwo.ca
▶ Show abstract
Lexical category structures differ across languages despite the common belief that translations refer to equivalent concepts. Therefore, acquiring another language requires conceptual restructuring. We aimed (1) to understand how learning a second language impacts lexical conceptual representations in the first, and (2) to investigate individual differences that may influence how L1 representations shift, such as length of immersion and age of acquisition. The bilinguals of interest were forty native speakers of Polish living in Poland, and sixty who had immigrated to an English-speaking country. Monolingual speakers of each language served as comparison groups. To examine the impact of bilinguals’ L2 on L1 lexical categories, we used the picture naming task. The object category of the task stimuli was seating. We found evidence of conceptual restructuring as a result of English immersion, such that some L1 lexical categories expanded, whereas others contracted with more time spent in an English-speaking country. Interestingly, well-established L1 naming patterns were not exempt from L2 influence, even with brief immersion. The results provide clear evidence for the plasticity of L1 concepts.
D15
How Does Gender Stereotype and Pronoun Congruence Impact Sentence Reading in Bilingual Adults?
Michelle Yang
► Show abstract
Gendered word associations pervade language (Charlesworth et al., 2023) and impede comprehension when sentences violate stereotypes (Carreiras et al., 1996). Yet most work focuses on binary pronouns and rarely considers non-binary forms (they/ze; Rose et al., 2023) or bilingual experience (Jankowiak et al., 2024 To address this, we recruited 101 French-English bilingual adults (65F, 36M; testing ongoing) to read 128 occupation-pronoun sentences (e.g., “The housekeeper washed the sheets because he/she/they/ze thought it was dirty.”). Subjects subsequently made a grammaticality judgement and confidence rating. Self-paced reading data suggest that readers accessed gender cues during reading: stereotypically incongruent sentences yielded slowdown effects, and non-binary pronouns slowed reading overall, consistent with lower frequency and newer lexical integration. Percentage English usage modulated these effects, with greater English exposure facilitating processing by decreasing stereotype effects and nonbinary pronoun integration. Judgement data also showed complementary pattern. Non-binary pronouns received lower acceptability ratings, with no effect of occupation gender, suggesting that explicit evaluations remain anchored in binary norms. Here, English usage also sped up judgment responses, indicating that language experience modulates real-time processing. Collectively these results highlight how the linguistic environment shapes gender-stereotype processing in a manner that varies with bilingual experience.
D16
How similar are mowing the lawn and taking a college exam? Predicting human ratings of event similarity based on content and temporal structure
Malak Osman
mosman52@uwo.ca
▶ Show abstract
Event knowledge is fundamental to people’s ability to think about their past, predict their future, and understand language. Theories of event cognition emphasize similarity when people think about events. Humans are sensitive to the extent to which events overlap in content, defined as overlapping activities, people, objects, and locations. Temporal structure, or how an event unfolds over time, is also a key component of event cognition. However, because no measure of temporal structure similarity has existed, it is unknown if it influences people’s notions of event similarity. We constructed temporally structured network science models of events to compute temporal similarity for 3,160 event pairs. We also computed several measures of content similarity using the Large Language Model, ModernBERT. We used these measures to predict human participants’ similarity ratings. Preliminary data indicates that people’s event similarity ratings are predicted by the computational measures. These results suggest, for the first time, that people are sensitive to similarity based on both content and temporal structure. Therefore, the degree to which two events are deemed similar depends on the activities, people, objects, and locations involved in the events, in conjunction with how the events unfold over time.
D17
Influence of autocomplete interference on typing words embedded in pseudo-compounds.
Alexander Taikh
rplayfor@student.concordia.ab.ca
▶ Show abstract
Typing a word involves recognizing it, planning, and executing the sequence of keystrokes. Keystroke speed reflects processing linguistic information from the word and its context, where the initial keystroke reflects recognizing the word and planning its keystroke sequence, while typing the non-initial keystrokes reflect executing the sequence. Interactive theories of typing posit that linguistic information influences all these stages, and should influence the initial and non-initial keystrokes. Morphology refers to the internal structure of words, and allows for the segmenting of words into meaningful subunits (i.e., morphemes). Playford et al. (2025) found faster initial and non-initial keystroke latencies when typing words embedded in compounds than pseudo-compounds (e.g., arm embedded in armpit vs. in armour) suggesting that embedded morphemes are more available than pseudo-morphemes.We examine whether the typing of words embedded in a pseudo-compound is influenced by a concurrently presented interfering compound with a shared embedded word. Neither the initial or non-initial keystroke speeds of the shared embedded word differed, suggesting that interfering linguistic information from the compound distractor did not influence the typing process of the initial constituents. Pseudo-compounds may be more likely to be typed as whole words rather than as subunits.
D19
Keystroke dynamics in bilingual typing: Evidence from a phrase-copying task in the KLiCKe Corpus
Leah Gosselin
► Show abstract
Despite the pervasiveness of digital communication (e.g., texting, emails, social media), few psycholinguistic studies have explored the intersection between typing behaviours and bilingualism. As such, we leveraged available keystroke data from the KLiCKe corpus (Tian et al., 2025) by extracting a phrase copying test for a subset of 24 L2-English typists (with Latin-alphabetized L1s) and 24 matched L1-English typists (M=36 years, SD=9 years). During the test, participants typed iterations of three-word phrases with high frequency bigrams (“seven wonderful surprises”) and low frequency bigrams (“some awkward zigzags”). We computed inter-key intervals (the time between releasing one key and pressing the following key) as a proxy for planning processes (Van Waes et al., 2021). Our analyses revealed that phrases with infrequent bigrams were typed more slowly than those with frequent bigrams (F=110.63, p<.001). This effect interacted with the typists' objective proficiency (F=6.38, p=.002): increased LexTALE scores were associated with faster keystrokes, and this slope was significantly steeper for infrequent bigrams (z=3.57, p=.001). Collectively, these results demonstrate that typing is not merely motor; language-based factors (bigram frequency) and person-based factors (language proficiency) shape keystroke dynamics. Our future work aims to expand these analyses to a group of L2-typists with more diverse writing systems.
D20
Language Dominance in Multilingual Young Adults
Vincent Hou
vincent.hou@mail.mcgill.ca
▶ Show abstract
Language dominance has been assessed through subjective and objective means, but how these measurements relate to each other has not yet been examined. We assessed language dominance in 60 French-English multilingual young adults (Mage = 24 years) to triangulate the relationships between three markers of language dominance: (1) subjective self-report of dominant language; (2) self-reported proficiency in English and French; and (3) objective English and French assessments. The language with higher reported proficiency and objective performance was coded as the dominant language in (2) and (3). Results showed that subjective language dominance and objective dominance had a fair agreement (Cohen’s κ = 0.39, p=0.10). Logistic regression showed that self-reported proficiency across speaking, understanding, reading, and writing in their dominant language (either English or French) did not predict objective language dominance (zs < 1.1, ns). Surprisingly, self-reported proficiency did not predict self-reported dominant language (zs < 1.8, ns). Objective assessment of the dominant language in receptive (z = 2.2, p=0.02) and expressive (z = 2.2, p=0.03) vocabulary predicted self-reported dominant language in the logistic regression, but discourse comprehension did not (z < 0.1, ns.). These findings indicate that language dominance is multidimensional, and that the different measurements provide divergent information.
D21
Lexical Conceptual Representations For Seating and Tools in Vietnamese-English Bilinguals
Ashley Tran
atran329@uwo.ca
▶ Show abstract
Second language (L2) learners often assume that translations (e.g., cup-taza), refer to the same set of objects. However, research shows that languages often categorize objects differently. These mismatches can create challenges for bilinguals, who must learn language-specific naming patterns to achieve native-like competence in their L2. Here, we examined the conceptual processing involved when Vietnamese speakers learn English. Vietnamese–English bilinguals with different lengths of English immersion and monolinguals of each language named pictures from two lexical categories–seating and tools–in an online naming task. We found that monolingual speakers of Vietnamese and English categorized seating differently. Vietnamese speakers mainly used “ghế,” while English speakers used a broader range of labels, meaning that Vietnamese learners of English need to restructure their lexicon, moving from a single seating category in Vietnamese to multiple categories in English. Results for bilinguals showed greater alignment with English categories with increased English exposure. Conversely, naming patterns for tools were similar across languages, implying restructuring is not required for this category. Bilinguals, therefore, must navigate which categories transfer directly and which require conceptual restructuring, highlighting challenges in second language learning.
D22
Moral Evaluation Across Languages: Developing Moral and Psycholinguistic Norms for L1 and L2 Readers
Esteban Hernandez-River
esteban.hernandezrivera@mail.mcgill.ca
▶ Show abstract
Moral evaluation is central to everyday language comprehension, yet few normative resources have attempted to capture how readers judge the morality of simple actions, particularly across languages. In this two-part study, we aimed to develop behavioral norms of morality for 244 actions (122 negative, 122 positive) for first-language (L1) and second-language (L2) readers. In part one, we asked whether morality norms could be extracted for L1 and L2 using measures of moral decision accuracy and morality ratings. We then evaluated the validity of these norms by examining their relations with established lexical and semantic variables, including frequency, concreteness, emotional arousal, emotional valence, and number of senses. We also explored whether these relationships differ between L1 and L2 readers. Crucially, These relationships were broadly similar across L1 and L2 readers, though some differences in their strength emerged across language contexts. Then, in part two, we conducted an exploratory analysis to systematically benchmark our human-derived moral norms against morality ratings generated by GPT-5.4. Together, this work provides a new resource for studying moral language and offers preliminary evidence about the extent to which model-generated moral judgments align with human evaluations.
D23
New metric, same effect: Final-word surprisal and whole-phrase frequency comparably impact binomial processing
Naima Mansuri
► Show abstract
Native-like fluency extends beyond vocabulary and pronunciation to include mastery of formulaic language, such as idiomatic multiword expressions (MWEs). Binomials (e.g., "salt and pepper") are a relatively understudied subtype characterized by a preferred, canonical word order. Like other MWEs, canonical binomials are typically processed more efficiently than reversed forms, likely due to their strong lexicalization. Two frequency-based mechanisms may underlie this advantage: whole-phrase frequency, reflecting how often a binomial occurs as a unit, and conditional frequency (surprisal), reflecting how expected a word is given its context. Comparing these measures is theoretically important because they capture distinct sources of information: stored lexical knowledge versus online predictability. We tested 74 English–French bilinguals to examine how these factors affect online and offline binomial processing. Whole-phrase frequency and surprisal were moderately correlated (r = .59), with highly frequent and canonical binomials showing lower surprisal than infrequent and reversed binomials. Across measures, higher-frequency binomials were processed faster and rated as more comprehensible, with greater canonical advantages for highly frequent binomials. Surprisal patterned with these results, showing diminishing canonical advantages for binomials with high surprisal. These findings suggest that differences among binomials which reflect whole-phrase frequency and conditional frequency of the binomial-final word are linguistically aligned. Crucially, they may have a psychologically indistinguishable impact on bilingual readers.
D24
Not so simple: Language, executive function, and reading comprehension in school-aged children
Mia Collins
miacollins@trentu.ca
▶ Show abstract
Infantile spasms is a type of epilepsy that presents before age two; a critical developmental period for language acquisition and prereading skills. There is limited research examining the language and reading outcomes of those with a history of infantile spasm. Although some evidence indicates age of onset and early speech and language support are associated with communication skills in this group, a comprehensive understanding of developmental outcomes is scarce. The present study is a qualitative analysis of articles published between 1990 and 2024 that included measures of language and word reading in participants with a current or past diagnosis of infantile spasms (e.g., West Syndrome). Fifteen articles were identified with only one study conducted in Canada and the majority being conducted in the USA (k = 6). Most studies included the Bayley Scales of Infant and Toddler Development as a measure for language (k = 9), while only one study included a measure of reading skill. No studies investigated language or reading beyond 42 months of age and no studies included a control group. These findings highlight a critical need for Canadian research examining language and reading outcomes of those with a history of infantile spasms, particularly later in life.
D25
Orthographic Structure Shapes Eye Movements and Perceptual Span During Reading: Evidence from Korean-English Bilinguals
Erika Guedea
e.guedea@unb.ca
▶ Show abstract
The perceptual span (i.e., breadth of parafoveal processing during reading) is a key contributor to reading efficiency and fluency (Schotter et al., 2012). However, limited research has examined the perceptual span in bilingual readers, especially those whose first-language (L1) and second-language (L2) differ substantially in orthographic structure. To address this gap, we investigated both L1 and L2 perceptual span in Korean-English bilinguals relative to L1 perceptual span in native English speakers. Participants (native English speakers: n = 21; Korean-English bilinguals: n = 26; aged 18-37) silently read 75 sentences while being eye-tracked. Parafoveal visibility was manipulated using a gaze-contingent moving window paradigm, allowing us to examine both global measures of reading fluency (e.g., average reading rates) and perceptual span across languages. Linear mixed-effects models revealed reduced L1 reading fluency (e.g., more fixations, slower reading rates) and reduced L1 perceptual span in bilinguals versus native English speakers. Additionally, bilinguals exhibited a larger L2 (English) versus L1 (Korean) perceptual span. Taken together, our findings suggest that orthographic structure modulates reading behaviour and perceptual span. Korean, a syllabic-block script that clusters phonemes into compact visual units, appears to constrain parafoveal processing relative to English, an alphabetic script that maps graphemes onto individual phonemes.
D26
Reading Direction Shapes Global Scanpath Structure in Bidirectional Traditional Chinese Reading
Yaqian Bao
baoy47@mcmaster.ca
▶ Show abstract
Traditional Chinese can be read both horizontally and vertically, offering a rare opportunity to test whether reading direction shapes global reading behavior while holding language constant (Pan et al., 2026). We examined how reading direction influences scanpath similarity and standard reading measures in bidirectional reading of Traditional Chinese. Participants read matched texts in horizontal and vertical layouts. We compared scanpaths of horizontally (HH) and vertically (VV) oriented pairs of texts, as well as across reading directions (HV), quantifying global scanpath similarity using pairwise scasim measures (von der Malsburg et al., 2015; von der Malsburg & Vasishth, 2011). Vertical reading (VV) showed reliably greater within-direction regularity of behavior than both HH and HV pairs. Word-level measures corroborated this pattern: compared with horizontal reading, vertical reading involved longer reading times, more and longer fixations, and slower reading rates. It follows that reading direction does not merely alter local oculomotor behavior, but reorganizes global reading strategy. Familiarity with vertical reading was not predictive of reading patterns, but we found evidence of individual reading strategies permeating both horizontal and vertical reading. Rather than being a simple rotated version of horizontal reading, vertical reading demonstrates a distinct and more homogeneous scanpath pattern across readers.
D27
The effect of filled pauses on subsequent word fluency in adults who stutter.
Alexander Taikh
tgranson@student.concordia.ab.ca
▶ Show abstract
Stuttering is a developmental speech disorder that is characterized by disruptions in the normal flow of fluent speech, including repetitions, prolongations, blocks, and filled pauses (e.g., uhhs and umms). According to the EXPLAN theory of speech fluency (Howell, 2002), planning the subsequent word and executing the word being said occur concurrently, and insufficient planning time may result in articulating it disfluently, for example if there is too much linguistic information associated with the subsequent word. Filled pauses contain little linguistic information and could provide temporal buffers for planning the subsequent words (Clark & Fox Tree, 2002).We examined the fluency of words spoken after filled pauses in the FluencyBank Timestamped database (Romana et al., 2024), which contains disfluency annotations and word timings of interview answers given by adults who stutter in the FluencyBank database (Bernstein-Ratner & McWhinney, 2018). Surprisingly, disfluencies were more likely following longer filled pauses. Our findings suggest that in people who stutter, filled pauses may signal a persistent difficulty planning the production of the subsequent word. Specifically, the planning difficulty is not less likely to be resolved during a longer filled pause, in contrast to EXPLAN.
D28
The effects of contextual diversity on lexical processing: A scoping review
Rebecca Norman
rebeccaemma.norman@utoronto.ca
▶ Show abstract
Research into the effects of contextual diversity on lexical processing has flourished in the past 20 years, encompassing diverse tasks, populations and languages, and underpinning influential theories of word learning (Nation, 2017). However, there is no current synthesis of the findings. We conducted a scoping review to provide a comprehensive synthesis of this field. 86 articles (145 experiments) met pre-registered inclusion criteria, including behavioural studies (N=111), computational modelling studies (N=20), and corpus validations (N=14). We show that the terminology used to refer to different diversity metrics has been applied inconsistently across operationalisations. For behavioural studies, high diversity facilitates written form processing regardless of operationalisation or task. However, effects on meaning-based tasks are more mixed and may depend on whether precise semantic selection is required. Modelling studies show that the social structure of language influences lexical organization. Corpus validations show that diversity explains more variance in word recognition data than word frequency across languages. This review confirms that diversity in linguistic experience is a key organizational principle of the lexicon, although diversity effects may vary across types of linguistic knowledge and task demands. We recommend that future studies categorise diversity metrics based on how they are calculated, avoiding arbitrary naming conventions.
D29
The Influence of Phonological Processing on L1 and L2 Reading in Monolingual and Bilingual Children and Adults: An Eye-Tracking Investigation
Isabelle Boucher
isabelle.boucher@unb.ca
▶ Show abstract
Reading is central to academic achievement and long-term socioeconomic success, with phonological processing (i.e., abilities involved in representing, storing, and accessing a language’s sound structure) playing a crucial role. However, the extent to which phonological processing influences in-the-moment reading behaviour remains poorly understood, particularly among developing readers and bilinguals. To address this gap in the literature, we examined how differences in first-language (L1) and second-language (L2) phonological processing (specifically, phonological awareness, phonological memory, and rapid naming via English and French versions of the Comprehensive Test of Phonological Processing/CTOPP) modulated L1 (English) and L2 (French) eye movement reading behaviour for naturalistic texts (4 x ~120-word stories). Participants included children (aged 7-12): 34 English monolingual and 33 English-French, and young adults (aged 18-21): 30 English monolingual and 30 English-French bilingual. Linear mixed-effects models revealed a stronger facilitatory effect of L1 phonological processing on L1 reading in bilinguals versus monolinguals, regardless of age. Additionally, among bilinguals, L1 and, more notably, L2 phonological processing exerted a stronger facilitatory effect on L2 versus L1 reading, especially in adults (who had reduced current L2 exposure levels). Taken together, our findings suggest that phonological processing differentially supports reading in bilinguals and in the weaker/lesser used language.
D30
Voice-based Social Judgments in Bilinguals
Mariam Mazen
mmazen@uwaterloo.ca
▶ Show abstract
Although most work on how we form rapid impressions of others involves judgments of faces, we also use information in the voice. Previous research has suggested that listeners of different language backgrounds may make different voice-based judgments. However, little is known about how bilingual listeners make voice-based judgments. In the present study, four groups of listeners (monolingual English-speakers and bilingual speakers of English plus Spanish, Chinese, or Arabic; n=40 per group) heard sentences produced by native English speakers and were asked to make judgments about the speakers’ dominance, trust, kindness, and intelligence. Following their initial judgments, participants viewed culture-neutral primes (advertisements) in either English or their L1 (Spanish, Chinese or Arabic) and then completed a second rating phase with new sentences produced by the same speakers. Preliminary analyses reveal that Arabic-English bilinguals had significantly higher judgments of dominance overall than participants from other groups. Ongoing analyses will reveal whether there are effects of priming language (L1 or L2) on listeners’ judgments, as well as whether individual factors (including L2 proficiency and cultural affiliation) modulate the effects. The results will contribute to understanding how bilinguals' cultural contexts impact their interpersonal judgments
D31
Vowel Classification in Nonwords: A Comparison of Automated, Crowdsourced, and Expert Responses
Irys-Amélie Champagne
iryschampagne@outlook.com
▶ Show abstract
Reliably identifying phonemes in speech has been an ongoing challenge for the field of psycholinguistics. Vowels are particularly difficult to classify consistently, with inter-transcriber agreement falling substantially below that of consonants, even among trained phoneticians working on native speech. The challenge is compounded when stimuli are nonwords, which lack the lexical context, and when speakers are demographically diverse, as non-native phonology introduces additional acoustic variability at phoneme boundaries. We investigated three approaches to vowel classification based on data from reading nonwords aloud: 1) we trained an automatic vowel classifier on Wav2Vec 2.0 embeddings; 2) we crowdsourced vowel classifications from the same diverse population that produced the nonwords (mean = 245 trials each); and 3) we trained a set of research assistants to manually classify vowels (mean = 23892 trials each). We found that each method carries significant trade-offs: naïve crowdsource classifications have relatively low IRR; IRR is higher but not at ceiling for proficient classifiers, but their classifications may drift as a function of experience; and automated methods are no panacea, as they sometimes struggle even in cases where multiple human classifications agree.
D32
When bilinguals are more than that: distinguishing between non-monolinguals in executive function research
Ariane Senécal
asene076@uottawa.ca
▶ Show abstract
The authors seek feedback on a proposed project aiming to contribute to best practices for conducting psycholinguistics research in today’s largely non-monolingual world, and provide valuable data to the growing body of executive function-related linguistics studies. The basic premise is that many participants in bilingual studies actually know more than two languages, which can pose a challenge to conclusions drawn about bilingualism. A preliminary behavioural study revealed a potential domain-general monitoring advantage (cf. Costa et al., 2009) for multilinguals compared to bilinguals, as well as asymmetrical language-specific inhibitory control skills. A second study found significant distinctions between bilinguals and trilinguals in reaction times and interaction with language entropy and English proficiency. Trilingualism also seemed to offer a particular protective effect against age-related decline (Schroeder & Marian, 2017; Pot et al., 2018). These initial forays point to multilinguals being speakers with distinct characteristics rather than just bilinguals with bonus languages. Therefore, we propose to conduct novel research combining comprehensive language background questionnaires with electrophysiological measurements (EEG) of bilingual and trilingual participants completing executive function-related tasks. The aspiration is that analyses will lead to empirical recommendations for research involving non-monolingual populations.
D33
Worth the Weight: Modern LLMs Demonstrate Accurate Metacognitive Knowledge of Decision Weights in Multi-Attribute Choice
Trent N. Cash
tcash@uwaterloo.ca
▶ Show abstract
Metacognitive judgments are informed by both privileged cues and observable cues. Across three studies implementing a variant of the learner-observer-judge paradigm, we explore the extent to which decision makers and observers can leverage these cues to infer the processes underlying a decision maker’s multi-attribute choice behavior. In Study 1, we demonstrate that decision makers (n = 221) more accurately estimate the weights they applied to each attribute in a multi-attribute choice task than observers (n = 220) who studied their choices. However, observers’ estimates were still weakly correlated with decision makers’ true weights, suggesting that reasonably accurate inferences can be made without access to privileged metacognitive cues. In Study 2, we demonstrate that the relative (but not absolute) accuracy of observers’ (n = 218) inferences decreases when the context of the decision is obscured, suggesting that domain-specific beliefs slightly improve metacognitive inferences. In Study 3 (n = 662), we replicate these findings and investigate differences in subjective experiences (e.g., motivation) across conditions. Together, our findings suggest that privileged metacognitive cues improve the accuracy of decision makers’ metacognitive knowledge, but that observers can make reasonably accurate inferences about decision makers’ choice behavior, particularly when they combine statistical monitoring with domain-specific beliefs.
D34
A Balloon Model of Bilingual Plasticity: From Development to Cognitive Reserve
John Anderson
► Show abstract
Like a balloon stretching as it fills with air, the cortex expands and thins as white matter grows. We propose that this Balloon Model provides a unifying biomechanical account of how experience, particularly bilingual language use, reshapes brain structure and function. By conceptualizing cortical morphology as dynamically coupled to subcortical expansion and mechanical tension, this model explains why bilingual experience produces distributed, nonuniform effects across grey and white matter. We argue that tension-driven expansion fosters functional segregation and network specialization, providing a mechanistic pathway to cognitive reserve. The framework links sustained attention, mediated by the cholinergic system, to microstructural changes in grey and white matter, to morphometric cortical remodelling, and, finally, to the reinforcement of large-scale functional connections. The Balloon Model thus yields testable predictions connecting white matter microstructure, cortical morphometry, and network integration across the lifespan.
D35
A Psychological Experiment on Lay Intuitions About Knowledge From Falsehood
Paul Emanuel Stan
past13@yahoo.com
▶ Show abstract
Can one attain knowledge about a conclusion by reasoning from a falsehood? Such “knowledge-from-falsehood” (KFF) cases abound in epistemology as extensions of the Gettier problem (see papers in Borges, de Almeida, & Klein, 2017). Experimentally, however, the only available KFF study reports that unlike philosophers’ disagreements, the lay population is highly forgiving of false premises by still endorsing any ensuing true conclusions (Turri, 2019).We expanded Turri’s study in a within-subjects experiment to evaluate people’s patterns of knowledge attribution under more stringent epistemic conditions. Participants (N = 96) read multiple vignettes about an agent inferring a conclusion from a stated premise. In addition to manipulating the truth value configurations of premise-conclusion pairs, we also contrasted the agent’s possession of knowledge to possession of belief. Our results indicate that in the “knowledge” condition, people are consistent in devaluing the agent’s knowledge of false premises while preserving their truthful conclusions; however, in the “belief” condition, we observed a recovery of knowledge attribution that increased participants’ credibility of false premises. Our study highlights that a discernable boundary emerges in folk epistemology when it comes to people’s adherence to formal deductive constraints versus practical knowledge attribution.
D36
Autonomy and Mental Effort
Theresa Waclawek
waclawek@yorku.ca
▶ Show abstract
The ability to exert mental effort is important for flourishing. Yet, exerting effort can feel aversive, and thus lead to limited task engagement and persistence. Therefore, understanding what shapes the experience of exerting mental effort is important. One factor that may influence the experience of mental effort is autonomy. Theories of motivation suggest that actions undertaken under conditions of autonomy, such as those aligned with one’s interests or freely chosen, may be associated with less aversive and more engaging experiences of mental effort, potentially through increases in autonomous motivation.To investigate this relationship, we conducted a preregistered experiment with 500 adults in which interest and choice were experimentally manipulated, and autonomous motivation and multiple mental effort outcomes were measured. Interest increased positive mental effort experiences and decreased negative mental effort experiences, whereas we did not find evidence for a significant effect of choice. The relationship between interest and mental effort was mediated by autonomous motivation. These findings suggest that interest shapes how mental effort is experienced, and that this relationship may operate through autonomous motivation.
D37
Did Video Kill the Radio Star? Exploring the Influence of Beauty and Motion in Subjective Music Appraisal
Michael Hopkins
Mphhopkins@mun.ca
▶ Show abstract
What makes us enjoy the music? Research within neuroscience and musicology has attempted to uncover what musical elements predict enjoyment. However, music requires a creator. Thus, the features of an artist may play a role in the reception and appraisal of music. In the present study we attempt to explore the influence the artist has on the reception of their art. Participants (n=22) were shown a set of 10 musicians unfamiliar to them and were asked to judge their music on 3 metrics, liking, aesthetic value, and involvement. Each participant was shown ten different songs with one of four photos being presented while listening. Each photo depicted the musician as either active or still, and was edited or unedited to reduce attractiveness. We hypothesized that participants shown an edited photo will appraise the music more negatively than those presented with unedited photos of the artists, and that depictions of active musicians would increase ratings. Preliminary results show that unedited still photos are rated more positively than those in the other condition groups; further edited active photos outperform edited still photos. This suggests that attractiveness and the activity of musicians shape the appraisal of a given piece of music.
D38
Do we attend less to detail in foreign-accented speech?
Katherine White
S.Medimorec@tees.ac.uk
▶ Show abstract
Previous research suggests that listeners process foreign-accented speech in less detail than speech from native speakers. We examine whether this is triggered by the speech signal itself, or whether knowledge of a speaker’s expertise can influence the amount of detail encoded. Participants (N=369; half in Canada, half in the UK) were assigned to one of 4 between-subject conditions in which they were introduced to a native (UK accent) or non-native (Ukrainian) English speaker, who was either a professor or undergraduate student. They heard a short interview with the speaker about a scientific topic, then saw a transcript of the interview and were asked to identify which words had changed from the original. This was followed by a working memory task. Change detection rates were consistent with increased attention to detail for both experts and native speakers (Foreign-accent expert: 15.3%; Foreign-accent non-expert: 12.7%; Native-accent expert: 18.1%; Native-accent non-expert 16.1%). However, a generalized linear model showed no significant effects of expertise (b=.19, p=.200) or accent (b=.25, p=.104), with the pattern similar for listeners in Canada and the UK. Future analyses will explore whether working memory modulates these effects. Our findings have implications for understanding how bottom-up and top-down cues influence language processing.
D39
Evidential Structures: An Axiomatic Core
Can Mekik
can.mekik@utoronto.ca
▶ Show abstract
Spreading activation and sequential sampling are fundamental and ubiquitous elementary mechanisms in computational models of cognition. This paper develops the core of an axiomatic theory of evidential structures with the goal of identifying a minimal and unifying set of foundational assumptions for cognitive models making use of these mechanisms. It is argued that bodies of evidence, as ordinarily understood, satisfy the axioms of closed extensive structures and that they may therefore be represented on an additive ratio-scale. Furthermore, a notable decomposition of the relevant ratio-scale representation is presented, under which spreading activation and sequential sampling emerge as elementary computations on bodies of evidence.
D40
Familiarity as a potential cue for strategic media multitasking behaviour
Colin Kwiatkowski
c2kwiatk@uwaterloo.ca
▶ Show abstract
In the context of an undergraduate lecture, media multitasking refers to the use of lecture-unrelated digital media during the class. We examined the possibility that people can choose to multitask with media strategically, in a way that prevents significant declines in lecture comprehension. Specifically we explored (1) whether people would choose to multitask more during familiar than unfamiliar segments of a lecture, and (2) whether they would remain engaged in multitasking after a segment of familiar content ended and novel content began, or whether they would disengage from digital media once novel content returned. Participants first watched a brief excerpt from a lecture (taken from the second or third quarter of the lecture) and then viewed the full lecture while being allowed to volitionally toggle on or off a secondary video showing an entertaining recording of a videogame being played. Results showed that (1) participants toggled on the secondary videogame (i.e., media multitasked) more frequently during the familiar segment (the one they previewed) than the other unfamiliar segments, and that (2) multitasking decreased immediately after the familiar segment. These findings suggest that content familiarity may serve as a strategic cue for both initiating and curtailing multitasking behaviour.
D41
From color to state: Production of redundant adjectives and limits of informativity-based models
Jiangtian Li
jiangtianli91@gmail.com
▶ Show abstract
When a speaker describes a unique banana to a listener, saying “the banana” is sufficient; adding any other information would be redundant. Previous work has focused on the production of redundant color adjectives, as in “the yellow banana.” We extend this research to state adjectives, such as “dirty” and “open.” In a free production experiment, speakers saw four pictures and had to tell their partner which one to click on. Objects in marked states elicited far more redundant state adjectives (dirty stovetop or open gate: 32%) than objects in unmarked states (clean stovetop or closed gate: 1%). Parallel to research on color, we demonstrate that modification rates greatly depend on how typical the state is for the particular object. Surprisingly, however, typicality does not replace markedness: a further significant portion of the variance is explained by the more abstract concept of markedness. Using a Bayesian rational speech act model, we show that the production of redundant state adjectives cannot be fully explained by the informativity of the words alone, but relies on other information encoded in markedness, potentially the negative valence encoded in marked states such as “dirty,” “broken,” and “bent.”
D42
Generation Speed of the Intuitive Response: Its Influence on the Likelihood of Rationalization Versus Decoupling
Emily Schwartz
► Show abstract
During decision-making, the choice to rationalize versus decouple from an intuitive response can be influenced by several factors. Previous research has shown that during decision making, faster generation speed of intuitive responses is associated with a preference for intuitively processed information. However, the impact of intuitive generation speed on the choice to rationalize versus decouple remains largely unexplored. The current experiment explores this question. 120 undergraduate students were asked to respond intuitively to base-rate questions under time pressure. They were then given the option to rethink their answer, with unlimited time to do so. Finally, participants completed the Cognitive Reflection Test. It is hypothesized that quickly generated intuitive responses will be rationalized at a higher frequency than more slowly generated intuitive responses. Additionally, it is hypothesized that a significant positive correlation between intuitive response generation time and CRT performance will be found. Preliminary data currently supports both hypotheses.
D43
How the relationship between accuracy and difficulty shapes perceived task competence
Matthew Yeung
m32yeung@uwaterloo.ca
▶ Show abstract
When observing one’s task behaviour, what information do we use to form an impression of their task competence? In many domains, we judge others’ competence based on how accurate they are. Another aspect of task behaviour, however, is how one’s accuracy changes in response to the difficulty of a task. For example, we assume that one should perform worse when faced with more difficult parts of task. Here we asked how the relationship between accuracy and difficulty might influence impressions of competence on a memory task. Across 11 experiments, participants read vignettes which described how a memory task works and the task behaviour of one or multiple agents who, hypothetically, completed the tasks. All agents had the same overall memory performance but, critically, some remembered more easier-to-remember items than harder-to-remember items while others remembered more harder-to-remember items than easier-to-remember items. Despite controlling for performance, we found that participants tended to judge the agents who had better memory for harder-to-remember items than for easier-to-remember items as being more competent. These results suggest that people may attribute greater task competence to others when their accuracy changes in an unexpected pattern relative to difficulty compared to when they exhibit a more expected pattern.
D44
Human Hippocampal Replay as Search
Mishaal Kandapath
mishaal.kandapath@mail.utoronto.ca
▶ Show abstract
Human-like systematic generalization remains a challenge for artificial systems. Recent work on hippocampal replay has suggested replay as a mechanism for online planning that affords generalization to novel, changing, and complex environments. Recent models, based on spatial tasks and data from rodents, suggest a serial notion of planning. In contrast, recent work on humans found replay sequences traversing multiple trajectories at the same time on a compositional generalization task. Such a possibility for parallel execution does not interface well with serial planning. We present a model of replay as planning with parallel search capabilities - a natural extension to a recent model of replay. The new models display human-level generalization faster than baselines. We find that only models actively using parallel search produced replay-like internal dynamics resembling neural sequences in humans. Our findings present search as a powerful general computation leveraged during replay for quickly learning good policies from little data.
D45
Is There a Relationship Between Freezing of Gait and Freezing of Speech In People Living with Parkinson’s Disease?
Liana Brown
bellasutherland@trentu.ca
▶ Show abstract
Parkinson’s disease (PD) is a debilitating hypokinetic neurodegenerative disorder that hinders quality of life (QOL). A common sign that arises in advanced PD is freezing, where someone is unable to continue voluntary movement despite their efforts. Freezing of gait (FOG) and freezing of speech (FOS) are common and often co-occur, yet their relationship is under-researched. We hypothesized that these two signs may share a common neuropathology. We investigated the relationship between FOG presence and severity and FOS by characterizing speech in people with PD (PwPD) using measures of voice quality and articulation. FOG questionnaire data and samples of natural speech were collected from 14 PwPD and 16 age-aligned control participants. We found that freezers produced a greater number of words and syllables per second but their speech was less recognizable than PwPD who do not experience FOG (non-Freezers), indicating that Freezers have distinctive patterns of speech articulation. FOG severity was not significantly correlated with speech measures. These findings support that there may be overlapping but not identical neuropathologies for FOG and FOS in PwPD. Through identifying the relationship between FOG and FOS, we may be able to better understand freezing and inform the development of novel treatments for PD.
D46
The association between self-reported sleep quality, physical activity and cognition in middle-aged females
Harleen Rai
harleen.k.rai@torontomu.ca
▶ Show abstract
Cognitive control declines in neurodegenerative diseases that disproportionately affect females, such as Alzheimer’s disease (AD). Midlife, which is marked by menopause-related neuroendocrine changes, represents a critical window for the early detection of cognitive decline in females. Furthermore, lifestyle factors may offer protective benefits. However, limited research has examined how modifiable factors like sleep quality and physical activity influence cognitive trajectories during this period. This study aims to investigate the interaction between menopause status and lifestyle factors on cognitive control performance in middle-aged females (40–65 years). This project utilizes data from the ongoing Brain Health at Midlife and Menopause (BHAMM) study conducted in Montreal and Toronto. BHAMM is a large cross-sectional study examining cognitive health at midlife. Through secure online platforms, participants complete questionnaires assessing sleep quality and physical activity, followed by an in-person neuropsychological assessment assessing cognitive control. In the 242 middle-aged females, physical activity moderated performance on the number of categories completed in premenopausal females only. Sleep quality did not significantly moderate age-related changes in cognitive performance. Given females’ heightened risk for AD, identifying lifestyle factors that promote cognitive resilience during midlife is essential for informing early prevention strategies.
D47
The Self After Cancer: Illness Identity and Continuity of Self Across Time
Sydney Smith
sydney.smith2@mail.mcgill.ca
▶ Show abstract
Living through major turning points in life, such as a diagnosis of cancer, can reshape an individual’s self-concept. Previous research suggests that experiencing cancer can lead to the adoption of new self-concepts. However, it remains unclear whether and how these self-concepts extend across representations of the self at different points in time. To address this question, we recruited participants with a history of cancer who were currently in remission. Participants completed the Illness Identity Questionnaire to assess the extent to which cancer was incorporated into their identity, and an adapted version of the I AM task. In this task, participants provided brief text descriptions describing their current self (present), their past self from 5–10 years ago (past), and their anticipated self in 5-10 years (future). These descriptions were compared using natural language processing techniques that quantified semantic similarity. Results revealed reduced similarity between past and present self-descriptions compared to future and present descriptions. Moreover, the similarity across life period descriptions varied as a function of illness identity. Greater similarity was associated with stronger rejection of cancer as central to one’s identity. These findings highlight how major life experiences like illness can influence perceptions of the self across time.
D48
Age differences in the effect of congruency in extrapolating beyond visual boundaries of scenes
Maria Orlando
morlando@my.yorku.ca
▶ Show abstract
We sample the world through our vision and can quickly and automatically extrapolate beyond what is visible to form a coherent, continuous understanding of a scene. This extrapolation process draws on conceptual knowledge structures, or schemas, to guide perceptual processing, leading observers to remember more than what was originally seen. Boundary extension (BE) is an error in scene memory, such that individuals retrieve details beyond the boundaries of a scene image, effectively “extending” the view in their internal representation. Schemas inform expectations about what likely exists beyond the arbitrary boundaries of a scene image, contributing to predictive processing underlying BE. Since typical aging is associated with more reliance on schemas to support memory, we examined age differences in BE. Young and older adults viewed scene images in which a focal object was either matched (scene-congruent) or mismatched to the surrounding scene context (scene-incongruent). BE was measured using a recognition task in which participants judged whether the view of the test scene image was the same or different than a target scene image. Scene-object incongruence may disrupt schema-based extrapolation processes that contribute to BE.
D49
Age-Related Difference in Peripheral Information Benefits for Face Perception
M. Eric Cui
mcui@research. baycrest.org
▶ Show abstract
While younger adults typically fixate on the eyes for face identification, older adults tend to fixate on the nose, which contains less identifying information. Whether these age-related fixation differences reflect covert sampling mechanisms, such as differential reliance on peripheral information, remains unclear. We recruited 76 younger (M = 26.2 years) and 34 older adults (M = 63.2 years) to complete a face identification task under two gaze-contingent conditions: a tunnel vision simulation (Window condition) and a blindspot simulation (Mask condition). Window and mask diameters ranged from 1.5° to 8.5° to control for available information in central and peripheral regions.Performance was substantially higher in the Mask than the Window condition. As more information became available (smaller Mask or larger Window), accuracy improved for both age groups but at different rates: the age gap slightly reduced in the Mask condition but moderately widened in the Window condition. These findings suggest that older adults are less efficient and flexible in integrating peripheral information for face identification. Ongoing analyses focus on using eye-movement data to explain the mechanisms underlying these behavioral patterns.
D50
Approaches to Visualizing Eye Movement Data From a Point of Care Ultrasound (POCUS) Study
Christian Riegel
katherine.robinson@uregina.ca
▶ Show abstract
POCUS is an essential diagnostic tool across emergency medicine and critical care enabling clinicians to rapidly evaluate conditions. We conducted a pilot study to examine differences in eye movements from 3 groups of POCUS users (medical students with little-to-no experience, health care workers with some experience, and experts) using three types of lung image video clips. The goal of our study is to develop an efficient training and competency measurement process. Conventional eye tracker visualizations such as heat maps, beeswarms, and dynamic heat maps generate valuable insights to the data alongside typical data analyses. Understanding of eye movement data is enhanced by artistically oriented visualization techniques (Riegel et al., 2017; Sharma et al., 2025). Using Godot, an open-source gaming engine, we sought to create visualizations that account for the embodied nature of POCUS imaging in a clinical setting affecting patients and healthcare workers. We employed a choreographic and interactive design approach (Dario and Rodighiero, 2025) to conceive meaningful visualizations that account for the dynamic nature of POCUS image generation, the interactions between healthcare worker and patient, and the data set from our study. Implications of using eye gaze data visualization alternatives to those provided via conventional eyetracking software are discussed.
D51
Assessing the generalization of perceptual learning using mental imagery
Helena Pimentel
pime9080@mylaurier.ca
▶ Show abstract
Perceptual learning improves performance in a perception task. Perceptual learning may be assessed as improvements in accuracy on an alternative forced choice texture identification task. Critically, accuracy improvements are diminished when the textures are modified (e.g., rotated by 180deg); a phenomenon known as stimulus-specificity. We attempt to reduce stimulus-specificity by using mental imagery. Mental imagery shares similar brain activity with visual perception. This may provide an opportunity to effectively rotate learned perceptual representations. Participants (N=55) completed 300 trials of a 6-AFC texture identification task. One day later, participants identified the textures in the same orientation (NoRotation), or with a 180-degree rotation. Prior to experimental trials, participants were either not told about the rotation (RotationOnly), saw a demonstration of each texture rotating (RotationDemo), or prompted to mentally imagine rotating each texture (RotationImagery). Preliminary results reveal that the accuracy decreased for the RotationOnly and RotationImagery (reflecting stimulus-specificity), but not for the RotationDemo. Independent measures of mental rotation abilities were associated with the magnitude of stimulus-specificity. Our results suggest that perceiving and mentally imagining a stimulus transformation may have different effects on the generalization of learned perceptual representations, providing insight in using mental imagery to affect seemingly basic perceptual mechanisms.
D52
Attentional Lapses in Time Estimation Across Prospective and Retrospective Paradigms
Veronica Bodea
veronicabodea@cmail.carleton.ca
▶ Show abstract
Problem solving is one of the most fundamental cognitive abilities. We propose that people allocate effort to solving a problem based on subjectively estimated solving time. As a result, individual differences in subjective time estimation should affect performance in a complex problem solving task. To test for this hypothesis, participants were recruited to perform a target time detection (TTD) task and solve the travelling salesperson problem (TSP). In the TTD, participants were shown a black circle for a target time (12 s) and asked to determine if subsequent circles were presented for the target time (or not). In the TSP, participants were asked to find the shortest path visiting 50 different cities exactly once and coming back to their starting point. The results in the TTD show that participants were generally able to distinguish between target and non-target times, and the deviations in the TSP replicated earlier findings from the literature. Preliminary results show a trending negative correlation between sensitivity (d’) in the TTD and deviations in the TSP. However, no relationship was observed between the decision criterion (beta) and problem solving performance. These results suggest that more accurate time estimation may facilitate optimal problem solving.
D53
Attentional prioritization in visual working memory can be guided by emotional retro-cues
Madeline Sinnamon
sinnamom@mcmaster.ca
▶ Show abstract
Visual working memory (VWM) is a limited-capacity system that supports the temporary storage and manipulation of information. Directing attention to a subset of memorized items during retention via retro-cues improves memory performance, a phenomenon known as the retro-cue benefit (RCB). However, most previous studies have used spatial cues to direct attention to target locations, and it remains unclear whether semantic cues based on emotional information can similarly enhance the maintenance of emotional face representations in VWM. In the present study, participants completed two tasks in which they memorized arrays of three faces displaying angry, happy, and sad expressions. During retention, retro-cues directed attention either to the target location (spatial retro-cue task) or to the relevant emotional category (emotional retro-cue task). Results showed a significant RCB, with higher accuracy in both spatial and emotional retro-cue trials than in neutral trials requiring the maintenance of all items. Critically, emotional retro-cues were as effective as spatial retro-cues, suggesting that attention in VWM can be guided by affective information. In addition, an anger superiority effect emerged, with higher accuracy for angry faces than for happy or sad faces. Together, these findings indicate that emotional information can prioritize representations in VWM after encoding.
D54
Audiovisual Speech Accessibility Study (ASAS)
Kareena Malhotra
kmalhotra@torontomu.ca
▶ Show abstract
Older adults with age-related hearing loss (ARHL) show improved performance in speech recognition when given audio-visual formats compared to audio-only formats. The mechanism of this benefit is unknown but is theorized to involve experience dependent cross-modal plasticity. This plasticity upregulates healthy senses (e.g., vision) when others (e.g., hearing) are lost. Those with ARHL may use visual cues to aid auditory communication and navigation. However it is unclear how those with ARHL use visual strategies to accommodate themselves. This study is an audiovisual speech perception task on 80 individuals with either ARHL or typical hearing. Participants view a video of a talker speaking a monosyllabic word, and verbally report the word they perceived. Stimuli are presented in background noise set to participants’ 50% word recognition during audio-only listening. Part-way through the study, participants are instructed to exclusively fixate on the speaker’s mouth. We will measure speech perception accuracy before and after instruction. Data are currently being collected, but we hypothesize that the mouth-fixation strategy will improve word perception most for ARHL participants due to cross-modal plasticity. Results can be used to improve the built environment (providing visual information in public spaces) and hearing healthcare (audiovisual strategies for aural rehabilitation).
D55
Collective Cognitive Estimation for Public-Health Related Forecasting
Aryanna Lee
ary4lee@gmail.com
▶ Show abstract
Public perceptions of social crises such as wars and pandemics are complex. Historians of infectious disease outbreaks highlight both medical and social “ends” of pandemics, with the former being driven by “objective” epidemiological trends, whereas the latter is based on “subjective” perceptions (life going “back to normal”). We conducted a computational cognitive science study involving aggregating predictions about total pandemic length. We compiled 445 high-precision human-annotated temporal predictions (~30 per 4 month period) derived from 2000+ automatically scraped online news articles between March 2020 and December 2023. These predictions came from diverse contexts (including epidemiologists, public health agencies, economists, journalists, politicians, and the lay populace). We plotted the trajectory of how these predictions about the end of the pandemic evolved over time, thereby assessing the optimism bias displayed by different social groups and by the general population overall. Preliminary results suggest that perceptions about the “end” of the pandemic were both subjective yet systematic, varying by profession and temporal anchoring (the time at which the prediction was made). Our findings about how humans navigate global crises with differing mental models of "normalcy” have implications for both fundamental cognitive science and applications such as epidemiological forecasting.
D56
Comparing visual and auditory deviance distraction in visual search
Rachel Yapp
nadja_jankovic@sfu.ca
▶ Show abstract
Presentation of a temporally leading stimulus can either facilitate (alerting) or impair (deviance distraction) the speed at which a visual target is detected. In studies of alerting, the leading stimulus has typically been presented on 50% of the trials. Studies of deviance distraction have typically employed two auditory tones that differed in pitch and probability of occurrence. One (e.g., the low-pitch tone) was presented frequently (e.g., 80% of the trials), the other (e.g., the high-pitch tone) was presented less frequently (e.g., 20% of the trials). In two previous experiments with visual cues, we found that cue probability alone failed to predict whether alerting or distraction would occur. In the present experiment, we asked whether match/mismatch in sensory modality would provide better predictive base.The leading stimulus was either visual or auditory. As such, it either matched or mismatched the trailing stimulus which was always visual. Contrary to expectations, neither condition showed any evidence of deviance distraction. These results are discussed in the context of current accounts of deviance distraction and differences between the alerting and the deviance distraction paradigms.
D57
Control of LTM Reinstatement Across Individual Differences in Working Memory Capacity
Jessica Kespe
jkespe@uoguelph.ca
▶ Show abstract
When an attentional template is formed in visual working memory (VWM), previously associated but task-irrelevant features can be reinstated from long-term memory (LTM), causing distraction. Although this process appears automatic, recent evidence suggests that individuals can exert control over when reinstatement occurs. One question is whether this control is constrained by individual differences in working memory capacity (WMC). Prior work suggests individuals with lower WMC are less able to mitigate distraction, raising the possibility that they may also have reduced control over reinstatement. To test this, participants completed a memory interference paradigm and the Operation Span (OSPAN) task. Participants were divided into low, medium, and high WMC groups. In Experiment 1, reinstatement effects were observed across all OSPAN groups, suggesting that reinstatement occurs across WMC levels. In Experiment 2, we used a paradigm designed to encourage strategic control over reinstatement by inserting a diamond search task before the object search task on every trial. Replicating previous findings, reinstatement appeared delayed, but no differences emerged across OSPAN groups. This suggests that both the reinstatement of task-irrelevant information and strategic control over its timing occur across all individuals with varying WMC, indicating that reinstatement may reflect a general mechanism of memory-guided attention.
D58
Crowding is stronger for edges than bars
Emma Neto
netoe@mcmaster.ca
▶ Show abstract
Relative phase discriminations are made with even-symmetric and odd-symmetric visual mechanisms (Field & Nachmias, Vision Res, 1984). In central vision, even- and odd-symmetric mechanisms are approximately equally sensitive. In peripheral vision, increasing the number of stimulus features, and decreasing inter-feature spacing, reduced odd- but not even-symmetric sensitivity (Neto et al., JOV, 2025). These results suggest that there are phase-specific effects of visual crowding (difficulty identifying objects in clutter). Here we examine the crowding phenomenon by measuring the effects of adjacent flankers on phase discrimination thresholds. Phase discrimination thresholds were measured in four observers. Target stimuli consisted of a single phase-reversing feature (i.e., bright/dark bars or left/right edges). We found evidence of crowding in both tasks: bar and edge discrimination thresholds increased with increasing flanker contrast and with decreasing target-flanker spacing. However, the effects of flanker contrast and spacing were greater for edge discrimination than for bar discrimination. Importantly, bar and edge detection thresholds varied little with flanker contrast and spacing, so the inability to discriminate relative phase was not due to a failure to detect the target stimuli. These results are consistent with the hypothesis that odd-symmetric mechanisms are especially sensitive to crowding in the visual periphery.
D59
Degree of Reentrant Processing Indexed by the N2pc
Graham Voiles
grahamjvoi1@gmail.com
▶ Show abstract
The N2pc is thought to represent the allocation of attention to lateralized stimuli in a visual search display. Evidence suggests that N2pc amplitude is affected by task complexity. We examined this relationship within and across two experiments. The first experiment employed single-step search tasks requiring either a ‘yes/no’ response to an oddball target (complex task) or a ‘go/no-go’ response to a specific target colour (simple task). The second experiment employed a similar complexity manipulation, but the number of steps required to perform the task was increased to two. The complexity manipulation did not modulate the N2pc within either experiment, but it did so between the experiments. We suggest that complexity does indeed modulate the amplitude of the N2pc. However, this depends on the way in which complexity is manipulated. The manipulation of complexity within each experiment may have been too subtle to affect the N2pc. In contrast, the difference in the number steps required to perform the task between the two experiments was sufficiently large to reveal the effect of complexity on the magnitude of the N2pc. This agrees with a reentrant view of visual information processing where the N2pc reflects the additional iterations required for processing the two-step tasks.
D60
Detection and discrimination of depth from local and global stereopsis across display methods
Tara Nichols
tara.nichols@mail.mcgill.ca
▶ Show abstract
Some individuals with clinically normal vision show selective deficits in stereopsis tasks, particularly when required to discriminate the direction of depth rather than simply detect its presence. This phenomenon, often referred to as stereoanomaly, has been reported in previous studies, with prevalence estimates ranging from approximately 2% to 30%, depending on the task and stimuli used. Prior work has employed stimuli that preferentially engage either local stereopsis (monocularly identifiable edges or shapes) or global stereopsis (random-dot patterns). There has been little work comparing performance (to identify stereoanomaly) between tasks using these two stimulus types. The present study directly compares local and global stereopsis across detection and discrimination tasks. Preliminary results indicate that performance on the local discrimination task diverges from the other three tasks, suggesting that it may rely on partially distinct mechanisms. A strong relationship was observed between global detection and global discrimination performance, exceeding the level of agreement reported in some previous studies, though a larger sample will be required for direct comparison. Together, the relationships between these four tasks can form the basis of a minimal set of measurements to characterise the stereopsis phenotype of a participant.
D61
Does Optimal Processing Explain Differences in Vision for Perception and Action?
Ethan Churchill
► Show abstract
Garner interference has been used to distinguish holistic object processing in perception from more analytic processing in action. Although grasping often appears resistant to interference from task-irrelevant object dimensions, it remains unclear whether this is equally true for unimanual and bimanual grasping. In the present study, participants grasped objects under baseline and filtering conditions while kinematic measures were recorded. Preliminary analyses from the first 10 participants examined reaction time (RT), movement duration, time to maximum grip aperture (MGA), MGA, peak velocity, and time to peak velocity. Repeated-measures ANOVAs revealed significant main effects of condition and grasp type on RT, such that filtering trials were slower than baseline trials and bimanual responses were slower than unimanual responses. Significant main effects of grasp type also emerged for time to MGA and MGA, indicating reliable kinematic differences between unimanual and bimanual actions. No grasp type × condition interactions were observed, and filtering did not significantly affect later movement kinematics. These preliminary findings suggest that task-irrelevant information may influence early movement initiation more than later grasp execution, while unimanual and bimanual grasping differ robustly in grip-related timing and aperture. Data collection is ongoing toward a final sample of 25 participants.
D62
Effect of emotional expression on the visual working memory of face identity
Yifan Han
han94@mcmaster.ca
▶ Show abstract
Past research by Ye et al. (2019) examined the two-phase resource allocation model of visual working memory (VWM) to determine how people balance precision with the number of items stored. Their findings suggest that shorter exposure times push memory into an early, involuntary, stimulus-driven phase. If negative emotion captures the attention in the early stage of the processing, the faces with negative emotion should lead to better memory compared to faces with neutral emotion, even when the encoding duration is short. Meanwhile, Long et al. (2020) showed that negative emotional states can improve memory precision, but at the cost of storing fewer items; thus, the effect of emotion appeared only during the later consolidation phase. The present study revisited this issue by examining how emotion and encoding duration interact to influence VWM accuracy. Participants completed a face identity change detection task, in which encoding duration (400ms vs. 1000ms), emotional expression (angry vs. neutral) were manipulated across trials. The results showed an interaction between emotion and encoding duration. Angry faces had a better performance, but only in the longer encoding condition. This suggests that emotional effects on VWM are not automatic but depend on sufficient encoding time.
D63
Effects of Dual-Task Listening While Driving on Speech Perception and Auditory Memory in Older Adults
Lauren Hong
lauren.hong@uhn.ca
▶ Show abstract
Successful communication requires both accurate perception of spoken language and the ability to retain that information for later use (memory). Word repetition accuracy is commonly used to index speech perception, yet it may not reflect the deeper processing required for successful communication in complex, everyday situations. The present study leverages a realistic driving paradigm to examine how listening difficulty, driving difficulty, and dual-task demands influence word repetition accuracy and subsequent auditory recognition memory in older adults. Forty-nine older adults repeated spoken sentences under single-task (listening only) and dual-task (listening while driving) conditions. Task difficulty varied, with easier and harder listening conditions (4 vs. 0 dB SNR) and driving conditions (rural vs. city). Word repetition accuracy and subsequent recognition memory were evaluated. Dual-task driving, particularly in harder listening and driving conditions, predicted poorer word repetition accuracy and generally predicted poorer memory, compared to easier listening and driving conditions, and listening only conditions. Interestingly, word repetition accuracy was less predictive of memory performance under greater task demands (i.e., harder listening, driving, and dual-task conditions). Overall, driving paradigms may provide a more ecologically valid model for understanding speech perception and associated memory in older adults under complex, realistic conditions.
D64
Effects of Feedback on Individual Differences in Attentional Asymmetries
Zainab Haseeb
zainab.haseeb@mail.utoronto.ca
▶ Show abstract
People differ substantially in how effectively they can attend to different locations across the visual field. Even with fixed gaze, covert attention is distributed unevenly across space, with participants showing “strong” and “weak” areas of performance that vary with polar angle around fixation, following stable and idiosyncratic patterns. We tested whether these attentional asymmetries can be modified through feedback and whether metacognitive awareness is associated with feedback-related improvements. Participants completed a visual search task in which a rotated target (T shape) appeared among distractor Ls at eight isoeccentric locations. A baseline session measured error rates and metacognitive judgments across locations; a subsequent session introduced trial- and block-level feedback. Both conditions revealed reliable and idiosyncratic asymmetries in error rates across locations. Feedback improved performance overall, but did not reliably improve uniformity, with observers showing comparable entropy in error rates between conditions. Moreover, individuals with higher baseline metacognitive accuracy did not show greater increases in entropy, suggesting accurate awareness at baseline does not compensate for attentional limitations. However, individuals who showed greater improvements in metacognitive accuracy in response to feedback showed greater gains in spatial uniformity, suggesting some degree of malleability of these asymmetries in covert attention.
D65
Examining Conscious and Unconscious Processing of Facial Attractiveness Using Standard and Reverse Breaking Continuous Flash Suppression
Micah Amd
micah.amd@proton.me
▶ Show abstract
The attractive face effect describes a common finding where attractive faces receive prioritized perceptual and attentional processing. Several breaking continuous flash suppression studies report that attractive faces break interocular suppression faster than less attractive faces, but whether this reflects pre-conscious perceptual processing or conscious attention capture remains heavily debated. This study outlines two within subjects experiments, with 44 females and 44 males, that compares standard unconscious to conscious breaking continuous flash suppression with reverse conscious to unconscious suppression. Target stimuli comprised 60 synthetic female faces systematically varied along attractiveness. It is predicted that attractive faces will demonstrate shorter breaking times and longer suppression times. Crucially, the magnitude of the attractive face effect is predicted to be larger in standard than reverse suppression paradigms. Support for this hypothesis will suggest that pre-conscious processing contributes toward facial attractiveness beyond conscious mechanisms. These hypotheses will be tested using two one-sided equivalence tests to distinguish meaningful from negligible effects. This research aims to clarify whether attractiveness processing precedes conscious inferential processes.
D66
Examining the Disfluency Effect with a Visual Masking Paradigm
Kamal Abou Chaaban
kamalkamal.kac@gmail.com
▶ Show abstract
Research has shown that reading from learning materials that are presented disfluently triggers deeper processing, which in turn fosters better learning outcomes. Disfluency has also been shown to affect judgments of learning (JOLs), a crucial part of the learning process. However, mixed results have been found for this disfluency effect because many of its moderators remain undiscovered. The proposed study aims to elucidate the boundary conditions of the disfluency effect and whether making learning materials perceptually more difficult to process results in better retention. In the proposed study, the fluency of learning materials will be manipulated by using a masking paradigm to explore the different contexts in which disfluency’s mnemonic benefits arise. Two potential moderators of the disfluency effect, test expectancy and JOLs, will be analyzed. Reading comprehension is the learning outcome that will be observed to test the disfluency effect and its moderators. Keywords: metacognition, desirable difficulties, disfluency effect, test expectancy, judgments of learning
D67
Here or there: Motor simulation and enactment boost memory for locations of objects, even without action during recall.
Suesan MacRae
smacra2@uwo.ca
▶ Show abstract
Remembering involves simulating the motor and sensory information elicited when we engaged with objects at encoding. Previously, we demonstrated that motor simulation improves memory for manipulable compared to non-manipulable objects, and that active placement of objects (enactment) at encoding boosts subsequent memory for their locations. Here, we investigated whether contributions of motor simulation and enactment can also be observed when the retrieval task does not carry motor demands related to active placement. Participants learned associations between presented items and locations on a computer screen either actively by clicking and dragging the items to their target location, or passively by watching recorded performance of other participants. Items varied on manipulability (hand-held objects vs. animals). In a forced-choice recognition test, participants were faster and more accurate in retrieving these associations when they had learned actively, and when the items were manipulable. Our findings suggest that motor information matters for remembering the location of objects even when the retrieval task does not require active manipulation. Moreover, they reveal independent contributions from lifetime manipulation experience as well as episode-specific information tied to enactment. Our study provides new evidence supporting memory theories that propose a critical role for motor simulation in remembering.
D68
Impaired contour integration in older adults with mild cognitive impairment
Eugenie Roudaia
eroudaia@research.baycrest.org
▶ Show abstract
Visuospatial deficits are a common early symptom of Alzheimer’s disease (AD), and visual perceptual tests may help identify individuals at risk in preclinical stages. The contour integration task (CIT) requires identification of a global contour composed of oriented elements embedded in clutter and engages a distributed network of feedforward and top-down recurrent processing. Although contour integration declines with age, preliminary evidence suggests it may be further impairment in mild cognitive impairment (MCI), a syndrome that often precedes AD dementia. Older adults with MCI (n=29, aged 60-85 years) and controls (n=83, aged = 60-84 years) completed the CIT task, judging the global orientation of a briefly presented spiral contour embedded in visual clutter. Clutter density varied across trials to estimate density thresholds. Participants completed two sessions 1-2 weeks apart to assess test-retest reliability. CIT density thresholds were significantly worse in the MCI group than controls in both sessions, with a small practice effect observed in both groups. Thresholds did not differ by sex, but worsened with increasing age. Test-retest reliability was moderate. These findings demonstrate impaired contour integration in MCI and support its potential as a screening measure. An ongoing longitudinal study will evaluate its predictive utility.
D69
Improving face discrimination in older adults through horizontal bias training
Jamie G.E. Cochrane
cochrj1@mcmaster.ca
▶ Show abstract
The ability to discriminate faces declines with ageing. Identifying effective ways to improve face perception may improve social interactions for older adults. Horizontal bias is a propensity for using horizontal information in the face during face discrimination. Stronger horizontal bias correlates with better face identification in younger adults, and this bias declines with ageing. The current study examined whether training to focus on horizontal information can increase horizontal bias and improve face identification in older adults, and if training benefits transfer to novel faces. Training consisted of a 10-alternative forced-choice face discrimination task where target faces were filtered to contain progressively narrower orientation bandwidth. Two training methods were compared across groups: one in which stimuli only contained horizontally-filtered target information, and another in which the horizontally-filtered target information was embedded in a non-informative vertical context. Participants completed testing and training over four consecutive days. Both training paradigms were found to improve identification, with some transfer occurring to new faces. Purely horizontal training showed more robust transfer to the new faces than training with vertical non-informative context. A longitudinal follow-up is planned to evaluate whether training benefits are retained in the long-term.
D70
Individual Differences in Future Thinking and Their Links to Mental Health
Reegan McCheyne
mcchertd@myumanitoba.ca
▶ Show abstract
This study examined relationships between key aspects of future thinking (FT) and (a) symptoms of depression and anxiety, and (b) facets of sensory processing sensitivity (SPS) and alexithymia, which are known risk factors for these common mental health disorders. A sample of 167 university students completed self-report measures and a future-oriented, cued imagery task. Depression and anxiety were both associated with reporting more vivid negative imagery, but only depression was linked to reporting less vivid positive imagery—a feature linked to strong expression of SPS and reporting few signs of difficulty identifying feelings (DIF; a feature of alexithymia). Both depression and anxiety were linked to experiencing intrusive thoughts, and anxiety was linked to experiencing intrusive imagery. These aspects of FT were linked to DIF and to “positive” features of SPS. Believing that negative events were likely to occur was liked to depression and to DIF. Finally, believing that future events would have a large impact was a feature of both depression and anxiety, and was linked to strong expression of multiple traits associated with alexithymia and SPS. These findings suggest ways in which specific facets of alexithymia and SPS may shape future-oriented cognition that is linked to depression and anxiety.
D71
Investigating the Frequency and Intensity of Illusory Self-Motion in the General Population
Sara Missaghi
saramissaghi15@gmail.com
▶ Show abstract
Vection, or the illusion of self-motion, is a sensory and perceptual phenomenon involving the perception of bodily motion in the absence of corresponding physical movement. While vection has been reliably induced and measured in controlled experimental settings, research on the prevalence and nature of this experience in everyday contexts and its relationship with individual factors remains limited. The present study aimed to address these gaps by (1) investigating the prevalence of vection in real-world contexts across the general population, and (2) exploring the relationship between selected individual factors and the frequency and intensity of vection experiences. 330 adults from the U.S. and Canada completed an online self-report questionnaire survey that collected information on individual and demographic factors as well as experiences of vection in various everyday, real-world situations. Results showed that approximately 90% of adults have experienced vection before, but that the majority of individuals rarely experience it in their day-to-day lives. On average, participants rated the intensity of their vection experiences as moderate. Further, greater vection frequency was significantly related to greater susceptibility to experiencing visually-induced motion sickness, depersonalization, migraines, and dizziness. These findings offer insights into the prevalence and nature of vection experiences in the general adult population.
D72
Perceived vs. lived history in times of crisis: Retrodictions and reality of societal change
Peter Diep
pndiep@uwaterloo.ca
▶ Show abstract
Policymakers and the public increasingly rely on indicators to track societal progress around the world and guide interventions on critical challenges from climate change to economic inequality. Yet the validity and reliability of indicators of societal progress remain largely unexamined. We systematically bench-marked 99 popular indicators across eight domains critical to human flourishing, as identified by 24 experts from diverse disciplines. Evaluating indicators against six foundational measurement criteria of definability, quantifiability, data availability, global representativeness, transparency, and temporal resolution, only 31% of expert-endorsed indicators met the standards necessary for valid inference. Strikingly, experts’ rankings of domain importance were inversely related to the quality of available indicators, with even the highest-priority domain (climate) having substantial measurement gaps, suggesting insufficient development of reliable indicators in critical domains. These findings reveal a fundamental disconnect between what we need to measure and what we can reliably measure, raising questions about the empirical foundation of global policy initiatives. Our results underscore the urgent need for investment in robust, transparent, and globally representative measurement systems, particularly as political pressures increasingly threaten data integrity and availability.
D73
Perceptions of public-health-related “normalcy” in a non-WEIRD setting
Jeonghu Lee
jeonghulee2000@gmail.com
▶ Show abstract
Understanding cognitive perceptions of "normalcy" is crucial for navigating prolonged social crises such as infectious disease outbreaks. While most psychological research relies on WEIRD populations, we conducted three rounds (total N = 2769) of pre-registered public surveys in a non-WEIRD setting during COVID-19 pandemic (December 2020, April 2021, and January 2022). We gathered temporal predictions from survey participants regarding the pandemic’s total length and the return to "normal." We employed multiple metrics to operationalize “normalcy" including behavioral mitigation and psychosocial factors such as predictions about social distancing, masking, vaccine uptake, economic recovery, herd immunity, being able to leave one’s home without thinking about COVID-19, and a return to “pre-pandemic times”. Using confirmatory factor analysis (CFA), we verified that these diverse variables converge onto a single underlying latent construct: “perceived normalcy”. The relative contribution of most variables was consistent over time, though the importance of some variables (e.g., vaccine uptake) declined over the course of the pandemic. Our results suggest that despite the complex nature of the COVID-19 pandemic, subjective human cognitive perceptions of crisis recovery are systematically structured. This study provides a rare non-WEIRD perspective on how mental models of "normalcy" are updated in response to shifting social realities.
D74
Reading Between the Lines: Personality Correlates of Sarcasm Detection
Danielle Harvey
dharv811@mtroyal.ca
▶ Show abstract
Previous research suggests that individuals reporting higher use of negative sarcasm score lower on agreeableness and conscientiousness (Dhensa-Kahlon & Woods, 2022; Ruch et al., 2018). Studies examining gender differences in sarcasm use yield mixed results (Johnson & Kreuz, 2022; Rockwell & Theriot, 2001). While individual and gender differences in sarcasm production have been studied, research on how these factors influence sarcasm perception remains scarce. This study addresses this gap by examining personality traits and gender in relation to sarcasm perception accuracy. A sample of 155 participants was exposed to isolated sarcasm cues (facial expression, intonation, and written context) and multimodal video stimuli, and were asked to classify each as sarcastic or sincere. Both accuracy and reaction time were recorded. Participants completed the NEO Five-Factor Inventory (Costa & McCrae, 1992) to assess personality. Results revealed significant correlations between neuroticism and both video accuracy (r = .19) and intonation accuracy (r = .20), and a negative correlation between openness and facial expression accuracy (r = -.18). Gender-specific effects emerged: extraversion positively correlated with contextual accuracy in men (r = .42) only. Findings suggest sarcasm detection is multimodal and draws on different personality resources depending on cue type and gender.
D75
Relations among posttraumatic stress symptoms, perceptions of trustworthiness, and face memory
Kaleigh McKye
kaleighmckye@gmail.com
▶ Show abstract
Our perceptions of others affect how we interact with and remember them. Prior research indicates that faces perceived as untrustworthy tend to be remembered better than trustworthy faces. Here, we further investigate the relations between perceptions of trustworthiness and the subsequent untrust memory effect for faces, and how these relate to posttraumatic symptoms following stressful life events. Undergraduate students (N = 307) first rated the perceived trustworthiness of 24 predefined trustworthy and 24 untrustworthy faces. After a distractor-filled delay, participants were shown the prior headshots along with 48 age- and gender-matched new face images, and indicated how confident they were that they recognized each item. Results confirmed the untrust memory effect: the predefined untrustworthy face set was remembered with significantly higher confidence. However, this effect was not explained by differences in participants’ ratings of trust. Moreover, analyses within face sets revealed an opposite relation: participants with higher average trust ratings also had higher subsequent average memory confidence ratings. The only observed relation with posttraumatic symptoms was lower confidence when detecting new faces. Despite replicating an untrust effect between face sets, the correlational findings raise questions about the underlying mechanisms that warrant further investigation.
D76
Semantic Change During Movie-Watching Is Associated with Default Mode Network Activation and Perceptual Decoupling
Samuel Ketcheson
20sbkk@queensu.ca
▶ Show abstract
Naturalistic tasks, such as controlled movie watching paradigms, approximate real-world experience and allow the study of ongoing cognition through the lens of brain activity and thought patterns. Studies using movie-watching have demonstrated that one group’s thought content reliably predicts changes in brain activity in another group, but little is known about how these thought-brain relationships are constrained by the objective features of the films and how people naturally segment them into distinct events. To address this, we combined multiple sources of data collected while participants viewed the same films: (1) existing brain data, (2) multidimensional experience sampling to quantify thought patterns, and (3) participant-identified event boundaries. To quantify the visual, auditory, and semantic features of the films, our research additionally uses embeddings from pretrained neural network models. We found that changes in semantic features were correlated with activity in the default mode network (DMN). Strikingly, the patterns of brain activity that were associated with semantic change systematically overlapped with the brain activity related to intrusive distracting thoughts, while being negatively associated with sensory-focused and verbally detailed thoughts. These results suggest that major semantic change may elicit a temporary disengagement with the external environment, consulting internal thoughts to facilitate narrative monitoring.
D77
Size Constancy in VR: How is Distance Measured?
Serena Trotti
sangelatrotti@gmail.com
▶ Show abstract
The visual system resolves size-distance ambiguity – the fact that a given retinal image size is consistent with infinite combinations of object size and viewing distance – by integrating available distance cues. Building on early work demonstrating a primacy of binocular disparity (BD) over motion parallax (MP) in distance perception (e.g. Ono, Rivest, & Ono, 1986), we used an Alberti frame paradigm in Virtual Reality (VR) (Wang & Troje, 2023) to examine whether this advantage extends to size perception. Two complementary experiments were conducted using similar Alberti frame methodology, one using a head-mounted display and one using a screen-based, auto-stereoscopic VR setup. A total of 48 participants judged whether virtual Rubik’s cubes appeared larger or smaller than a real-world Rubik’s cube across four viewing conditions: Picture (neither BD nor MP present), BD-only, MP-only, and Window (both present). For both display technologies, results indicated that BD, but not MP, supports size constancy across viewing conditions, paralleling earlier work on distance perception. These findings clarify the respective roles of BD and MP in resolving the size-distance ambiguity, and advance our understanding of the mechanisms underlying size constancy.
D79
Spatial Biases in Face-Based Visual Search: A Top-Left Advantage in Target Detection
Jack Simiao He
hes89@mcmaster.ca
▶ Show abstract
Prior research suggests that visual attention systems prioritize the upper and left visual fields. Building on these findings, the current study examined whether absolute target location influences face-based visual search. Participants completed a task in which six faces were displayed at equal distances around an imaginary circle, with one male face among five different female faces. The target (male) face appeared with equal probability along the vertical (top, middle, bottom) and horizontal (left, right) dimensions, and each face was tilted randomly to the left or right. Participants located the male face and indicated its tilt direction using a keyboard response. Analysis with a repeated-measures ANOVA for reaction time revealed (1) a significant main effect of laterality, with faster responses to left than to right targets, (2) a significant main effect of target row, with top-row targets faster than middle- and bottom-row targets, and (3) a significant interaction indicating a particularly strong top-left advantage. These results support an upper- and left-field advantage in face-based visual search, with performance strongest in the upper-left visual field.
D80
Squeezing to hear: Listeners readily take physical actions to avoid exerting cognitive effort
Jaime Lamont
jlamon03@uoguelph.ca
▶ Show abstract
Theoretical frameworks of auditory processing suggest that individuals take physical actions to reduce the need for cognitive effort when doing so can improve listening success. We tested this idea using a hand dynamometer that listeners (N = 83) could squeeze to reduce background noise on upcoming Revised Speech Perception in Noise (R-SPIN) sentences. We manipulated the initial noise level and sentence predictability to examine how these factors influenced listeners’ ability to report the final word of each sentence, their feelings of subjective effort, and their corresponding levels of physical exertion.Results showed that participants exerted significantly greater grip force under more challenging listening conditions, particularly at higher noise levels and for unpredictable sentences. This pattern indicates that listeners increased physical effort when speech was harder to understand. Additionally, participants exerted significantly more force following incorrect speech-identification responses and after trials rated as highly effortful, suggesting that effort allocation is dynamically adjusted based on recent listening experience.These findings support the view that hearing success involves adaptive responses to adverse auditory conditions involving different forms of effortful exertion. These results further suggest that physical-action measures, such as grip force, can provide a quantitative index of listening effort during speech perception.
D81
Statistical learning in visual search: A new paradigm for associative learning of relative locations between two targets
Arathi Shanmuganathan
shanma9@mcmaster.ca
▶ Show abstract
Visual statistical learning refers to the cognitive process by which humans unconsciously acquire the statistical regularities embedded in visually presented stimuli. Visual search research focused on statistical learning of the absolute feature/location of a single target or a singleton distractor. Associative learning between distractor layout and target location has also been established (e.g., contextual cueing, Chun & Jiang, 1998). The present study aimed to create a new associative learning paradigm in which participants learn the relationship between the locations of two targets. In the visual display, 8 search items were arranged at equal distances around an imaginary circle, with 2 diamonds serving as targets and 6 circles as distractors. Each search item contained a black dot on the left or right side. Participants searched for the two targets and judged whether the dots within them were on the same side. In half of the trials, the two targets were in direct opposite locations; in the other half, they were separated by 0, 1, or 2 distractors (with equal probability). The results showed that response times were significantly faster in the high-probability condition than in the low-probability condition, despite the greater distance between the two targets in the high-probability condition.
D82
Stimulus-level variability in eye gaze influences mental state perception but not gaze following
Siqi Yi
siqi.yi@mail.mcgill.ca
▶ Show abstract
While gaze following is one of the most fundamental forms of social communication, the mechanisms by which gaze conveys social information remains unknown. Recent work suggests that gaze following is resilient to gazers’ internal mental states. To extend this initial finding, here we examined the links between gazers’ eye direction cues, gaze following, and mental state attribution. Participants viewed videos of four different gazers making intentional and non-intentional eye movements and responded to a peripheral target that appeared on either the gazed-at or not gazed-at location. Gaze following showed a robust pattern of activity and yet again was not modulated by gazers’ mental states. The perception of gazer’s mental states however varied across individual gazer. Additionally, the magnitude of the gaze following was linked to the overall motion within the gaze signals only when the eye movements were non-intentional. Together, these findings suggest that gazer identity as well as kinematic differences in naturalistic gaze cues can influence the perception of mental states, highlighting the importance of understanding stimulus-level variability in social communication via gaze.
D83
Switching Between a Computer Monitor and Virtual Reality to Reduce the Sustained Attention Decrement
Maxime LeDrew
mledrew@uwaterloo.ca
▶ Show abstract
The present study examined whether changing presentation context can restore sustained attention during monotonous tasks. Participants (N=164) completed a sustained attention task (i.e., the visual metronome response task) with stimuli presented either on a computer monitor or through a virtual reality (VR) headset. Across two task blocks some participants switched devices (monitor to VR, or VR to monitor), while others used the same device throughout (monitor to monitor, or VR to VR). We observed a sustained attention decrement characterized by increasing response time variability and mind-wandering over time. Participants who switched presentation contexts experienced reductions in mind-wandering compared to those who maintained the same device, demonstrating contextual changes can effectively restore subjective attentional engagement. The type of device switched to (and from) had minimal impact, suggesting the benefit stems from the contextual change rather than the specific presentation mode. However, the effects of switching on task performance were not significant. Also, contrary to expectations, overall performance was poorer in VR than on the monitor. The results suggest changes in context surrounding a persistent task can temporarily reduce mind-wandering.
D84
The effect of race in visual search among faces
Fion Lee
leef22@mcmaster.ca
▶ Show abstract
Perceived membership in social categories, such as race, underlies group-based prejudice and influences social perception. Individuals are often motivated to conceal biases or are unaware of their own biases, making implicit measures that capture automatic responses a more objective approach to studying these processes. The human visual system prioritizes behaviorally relevant information, including social category membership. The present study adopted a task-irrelevant singleton paradigm (Theeuwes, 1992) to examine whether Black race faces capture attention in a visual search task even when race is task-irrelevant. In Experiment 1, participants either searched for a Caucasian female face among an array of Caucasian male faces or for a male face among an array of female faces. In half of the trials, the distractors contained one Black race singleton. In Experiment 2, we removed sex differences in the array, requiring participants to search through an all-male array for the rotated letter “C” among distractor “O’s” pasted on the forehead of the faces. We observed inconsistent differences in performance between the singleton-present and singleton-absent conditions, suggesting that multiple factors may influence visual search. Further research is necessary to dissociate the contributions of racial bias towards certain races and perceptual differences among facial features between races.
D85
The Impact of Emotional Valence on the Colavita Effect
Shannon St George
sstgeorge@mta.ca
▶ Show abstract
The Colavita effect is a perceptual phenomenon in which participants fail to report the auditory component of an audiovisual stimulus more frequently than the visual component. Stimuli semantics were recently shown to influence Colavita effect size: presenting incongruent animal and musical stimuli (concrete neutral objects) weakened the effect. This study therefore investigated the effect of emotional congruence, as valence is a stimulus characteristic inferred by the observer. Participants were presented with visual, auditory, and audiovisual stimuli and asked to report the stimulus modality. Crucially, bimodal stimuli were 50% congruent (e.g., happy face, happy sound) and 50% incongruent (e.g., happy face, sad sound). Given evidence that negative affect can influence perception through biased attention, the DASS-21 was administered to evaluate the predictive ability of mental health. As expected, participant responses to bimodal trials were slower than responses to unimodal auditory and visual trials. A significant Colavita effect was observed for incongruent trials but not congruent trials. Interestingly, participants’ anxiety levels also impacted responding. This unusual error pattern may reflect the impact of emotional content or stimulus duration on audiovisual integration. Results will be discussed in the context of arousal and sensory perception.
D86
The Role of Depth Cues in Shape Constancy
Iroshini Gunasekera
iroshini@yorku.ca
▶ Show abstract
Shape constancy supports stable 3D perception despite changes in retinal projections, but errors in distance or slant can disrupt it. While such distortions are well documented, how depth cues interact to support constancy remains unclear. In Experiment 1, we examined the contributions of monocular (pictorial) and binocular cues to shape constancy for planar surfaces rendered in perspective. Two rectangular tabletops were presented sequentially using a mirror stereoscope at different orientations: one with its long edge facing the observer (target) and one receding in depth (reference). Using a method of adjustment, experienced observers (N = 18) matched the target length to the reference. Four cue conditions were tested in 2D and 3D: baseline (untextured), added cast shadows, texture, and contextual cues (table legs, place settings). Shape estimation improved in 3D, with the highest accuracy seen in full cue conditions. In Experiment 2, novice observers (N = 14) completed the same task, with an additional condition that removed table legs. Although stereoscopic viewing improved performance, removing legs had minimal impact. Novices also consistently underestimated surface extent, even with full cues. These findings indicate that while multiple depth cues enhance shape constancy, their effective use depends on observer experience, not cue availability alone.
D87
The Role of Musical Absorption and Arousal in Mitigating Visually Induced Motion Sickness (VIMS)
Harley Glassman
hglassman@torontomu.ca
▶ Show abstract
Music is consistently ranked as a highly rewarding activity. Familiar music is often considered more rewarding, such that increased exposure to preferred pieces produces greater enjoyment. However, the neural mechanisms of musical preferences on reward as a function of familiarity remain unknown. This study used functional near-infrared spectroscopy (fNIRS) in a repeated exposure paradigm to investigate the extent to which familiarity influences the hemodynamic activation of the auditory cortex and prefrontal cortex for preferred and non-preferred music. Additionally, it examined the hemodynamic activation and self-reported liking for preferred and non-preferred pieces in distinct reward sensitivity groups, as measured by the extended Barcelona Music Reward Questionnaire (eBMRQ). A significant interaction was found between musical preference and repetition, where overall activation was greater with repeated exposure to preferred music. Additionally, individuals with higher reward sensitivity showed greater overall activation for preferred music. Together, these results suggest that musical preference and repetition interact to modulate activity across auditory–frontal networks, which was most prominent in individuals with higher music reward sensitivity. This novel study demonstrates the dynamic influence of musical preferences on the neural mechanisms of musical reward, indicating a greater need for future studies to consider its integral role in personalized listening.
D88
The Role of Physiological Arousal in Pilots’ Takeoff Decision-Making Under Simulated Conditions
Kelsey Szczerbickyj
K_szczer@live.concordia.ca
▶ Show abstract
Takeoff decision-making represents one of the most safety-critical moments in aviation, yet the influence of physiological arousal (e.g., unexpected go-around or system warning during approach) on pilot judgement remains underexplored. Arousal can impact the cognitive processes underlying decision-making, including risk perception and response inhibition. Thus, arousal could be a theoretically relevant factor in high-stakes operational contexts when flying a plane. The current study investigated whether pre- and post-trial heart rate (HR), as an index of physiological arousal, is associated with pilots’ go/no-go decisions across simulated takeoff scenarios. Licensed pilots completed 30 randomized experimental trials, and 3 practice trials in a Beechcraft Bonanza simulation using X-Plane 12, modeled after the St-Jean-sur-Richelieu (Québec) runway. Trial conditions varied systematically to alter aircraft performance and safety margins. Continuous HR was recorded via a HR variability monitor (Polar). Following each trial, participants completed a questionnaire assessing perceived safety margins, decision timing, and confidence of response. The analyses focused on trials in which objective conditions rendered rejection the more appropriate response. Findings are discussed in relation to arousal-performance frameworks, and their implications for evidence-based pilot training and aviation safety standards.Keywords: decision-making, physiological arousal, heart rate variability, risk appraisal, aviation
D89
Truth perception and memory for crossmodally presented information
Greg Shaikevich
► Show abstract
The ease with which we process information (i.e., processing fluency) is positively related to its perceived truthfulness, a phenomenon known as the Illusion of Truth Effect (IOT). Fluency may be manipulated through repetition, where repeated statements are typically judged to be more truthful than non-repeated statements. To date, most research on the IOT has focused on the visual domain. However, much of our exposure to information comes through the auditory modality. Furthermore, this means that, across several exposures, information may be presented in different modalities (i.e., crossmodally); you may hear something first and read it later. Therefore, it is critical to understand how processing fluency influences perceived truthfulness when the modality of the information changes across exposures and how this compares to when the modality remains the same. In addition, we must better understand the link between truth perception and memory, given that memory is often relied on when making truth judgements. Results from two experiments show that, while a change in modality between exposure and test leads to worse memory performance for the information presented, no difference in truth perception is observed. Future work aims to address this discrepancy and further explore the relationship between memory and truth perception.
D90
Vibrotactile stimulus detection during Fitts aiming: Implications for the use of biofeedback devices during skilled, manual tasks
Alice Atkin
aliceatkin@yorku.ca
▶ Show abstract
To mitigate musculoskeletal injuries in healthcare workers, we have developed a biofeedback system using vibrotactile motors and sensors. However, sensitivity to vibrotactile stimuli can vary depending on stimulus location, intensity, and whether a body part is moving. Vibration detection could also interfere with motor performance. 18 participants performed vibrotactile detection, Fitts aiming, and a dual-task condition. Vibration was delivered to the hand and the upper back. 50% detection thresholds, movement time and path length were measured. 50% detection thresholds were significantly higher for the dual-task compared to vibrotactile detection alone (+35.3%). This suppression effect was significantly greater for vibration on the hand (Back = +26.4%, Hand = +46.2%). Movement took significantly longer on the dual-task compared to the aiming task alone (+12.6%), and when vibration was delivered to the hand compared to the back (+9.0%). There were no path length differences. Detection thresholds for vibrotactile stimulation are increased during dual-task conditions. However, the suppression effect and movement cost was lower when vibration was delivered to the back. In practice, a vibrotactile biofeedback system worn on the upper back should have good sensitivity without impeding manual performance.
D91
Visual Search Termination When Salient and Memory-Matching Distractors Are Present
Leonie Ferber
leonie.ferber@mail.utoronto.ca
▶ Show abstract
When searching for a target stimulus in a visual scene, perceptual salience and working memory (WM) content influence how attention is allocated across the scene. Stimuli with high perceptual salience or similarities to WM content both receive preferential processing – regardless of whether they are the target. However, when terminating search of a target-absent display, salience and WM content have opposing effects. Salient distractors induce earlier search termination, a phenomenon termed the quitting threshold effect (QTE), while distractors that match WM content delay quitting. Here, we addressed this dichotomy by investigating the contributions of both perceptual salience and WM content on quitting during a single task. Participants completed a search task while holding a colour in WM on each trial. One distractor was manipulated such that it could be high or low in perceptual salience and match or mismatch the colour held in working memory. We replicated the finding that high perceptual salience leads to faster responses, however, this effect was not isolated to target-absent trials, thus does not represent a QTE. The similarity between distractor colour and the observer’s WM content also did not produce a QTE as indexed by response times, despite lower accuracy for a dissimilar distractor.
D92
Visual working memory: effect of task-irrelevant location information
Jingshu Hua
huaj13@mcmaster.ca
▶ Show abstract
We examined the role of task-irrelevant location information in visual working memory when participants were required to remember the colour of objects. During encoding, 3 or 4 colored dots were presented one after another (each for 300ms) in the visual periphery. A central fixation point was presented for 300ms during the time interval between the presentation of consecutive dots. The sequential appearance/disappearance of peripheral objects encouraged saccades from screen center to peripheral locations. During retrieval, a number prompt was displayed to request recall of the colour corresponding to the serial order of the encoding objects. The recall response was made on a colour wheel, overlapping in the position of the visual prompt. The spatial relations between the visual prompts during retrieval and the dots during encoding were varied between trials: (1) the original location during encoding (2) a location occupied by a different dot, (3) a location previously unoccupied and (4) central fixation. We failed to find a statistically significant difference among the four conditions, suggesting that location information (most likely combined with gaze shifts) contributes little to the memory of the objects' colours.
D93
When Layout Meets Syntax: The Role of Line Break Placement in Sentence Reading
Laura Franzoi
laura.franzoi@unitn.it
▶ Show abstract
Typographic features such as font type, character size, and spacing are well known to affect visual and cognitive processing during reading. The effects of other features, like layout and line breaks, remain relatively underexplored. The aim of this study is to investigate how line-break placement interacts with syntactic structure during reading.Short sentences were presented on screen, manipulating the alignment between line breaks and syntactic units. Breaks either preserved natural syntactic groupings or interrupted closely related constituents, creating a mismatch between visual segmentation and syntactic structure. University students read the sentences silently while sentence-level reading times were recorded, and comprehension questions ensured attentive reading.Results show that less predictable line breaks lead to longer reading times – particularly when they disrupt strong syntactic constituents – suggesting increased processing costs. Notably, this difference emerges even in skilled adult readers and in relatively naturalistic reading tasks. The effect remains significant when controlling for sentence length and is observed both across participants and across items. These findings highlight the role of typographic layout in reading and suggest that subtle formatting choices – such as line-break placement – influence reading fluency, with implications for the design of accessible texts, especially for readers with dyslexia.
D94
Illusory parallax in stereoscopic displays: Effect of Scene Context
Danial Kordmodanlou
► Show abstract
Under natural viewing conditions, stereopsis and motion parallax provide consistent information about three-dimensional structure. In virtual environments, however, these cues can be decoupled. When observers move in front of a stereoscopic display that provides binocular disparity without corresponding viewpoint-contingent updates, they often perceive illusory parallactic motion. This thesis investigates this phenomenon and evaluates the proposal that it arises from predictive mechanisms that use stereoscopic depth to generate expectations about motion parallax. Using a head-mounted virtual reality setup and two-interval forced-choice tasks, we quantified the illusion and tested its underlying mechanisms. The illusion depended on binocular disparity and reflected a signed mismatch consistent with negative parallax. Critically, scene context modulated illusion strength: it was strongest in minimal scenes and reduced in structured environments. Additional manipulations showed that viewing geometry and observer distance also influence the effect. These findings support an account of illusory parallax within a predictive framework, in which missing expected motion signals generate a percept of scene motion. More broadly, the results demonstrate that scene context shapes predictive depth processing and cue integration in virtual environments.