Abstract: The present study addressed the controversy around the effect of a working memory load and cognitive distraction, and more specifically the finding in past work that this load sometimes increases distraction, sometimes reduces it. SanMiguel et al. (2010) found that the effect of unexpected novel sounds on performance in a visual working memory task reduced when the task placed significant demands on executive processing by imposing a memory load approaching capacity limits. This load reduced the involuntary orienting of attention toward the sounds and, in turn, their impact (facilitation or distraction) on behavioral performance. In contrast, Lavie (2005) argues that a working memory load increases distraction because load competes for resources with executive control mechanism that attempt to inhibit prepotent responses towards distractors and help limit distraction. We argue that there are two important factors that might influence the deviance distraction: whether distractors are inhibited because they provide no relevant information for the task (e.g., when cueing targets with 50% validity), and whether a load depletes resources otherwise needed for this inhibition. We predicted deviance distraction when attentional resources are directed toward the distractors because they are informative and when a load monopolizes resources otherwise used to inhibit uninformative distractors. In our experiments, participants performed a two-alternative categorization task in which they indicated whether a target stimulus appeared above or below the vertical centre of a computer screen. These targets were preceded by a location cue and, shortly after, a bilateral auditory distractor (standard sound on most trials, deviant sound on the others). Across experiments we manipulated the degree of validity of the location cue (50% or 80%) and whether participants performed the task with a load (1-back task) or without. The location cue was uninformative (50% valid) in Experiments 1 and 2, and informative (80% valid) in Experiments 3 and 4. A load was introduced in Experiments 2 and 4 by use of a 1-back task in which participants responded to the previous trial instead of the current one. The results from Experiment 1 revealed no deviance distraction but longer response times for valid trials, suggesting the inhibition of the distractors. Remarkably, the introduction of a load in Experiment 2 led to observation of deviance distraction, suggesting that the load used resources no longer available for the executive control mechanisms inhibiting responses towards the distractors. In Experiment 3, where location cues were 80% valid, deviance distraction was observed in the absence of a load. Interestingly, it remained so in the presence of a load (Experiment 4). Our results demonstrate that with peripheral visual cues can mediate the extent to which auditory distractors are processed. Distraction seems to appear when attentional resources are available to be directed toward deviant sounds.

Key words: periphery visual cue, deviant distraction, working memory load, audio-visual oddball task