The present work examines the impact of irrelevant auditory tones on visual performance (reaction time, accuracy, recognition memory) in a lexical decision task. Research has shown that deviant irrelevant items capture attention and impair memory for serially ordered word lists. Memory models have suggested that interference occurs with order memory, rehearsal processes, or phonological encoding. Using a non-serial surprise recognition memory task where no rehearsal process is required, reaction time and memory performance costs were observed for irrelevant oddballs. The memory deficit for deviant trials will be examined in light of working memory capacity (OSPAN).