Abstract
Distraction occurs when the perceptual system fails to avoid processing task-irrelevant stimuli (distractors) that are presented together with task-relevant ones (targets). The question whether distractors are processed before, concurrently with, or after targets are processed, bears crucial theoretical implications. Yet, direct behavioral investigations of the temporal loci during which distractors are processed remain largely absent in the literature. Theoretical models have instead adopted divergent assumptions regarding the time of distraction, which have led to disparate theoretical propositions about when and how attentional mechanisms operate. Early-selection and dual-process models assume that distraction occurs in the beginning of the presentation. Late-selection and single-process models assume that distractors are continuously processed for as long as presented. Perceptual load theory proposes that distractors are processed only after targets have been processed, when surplus resources remain available. We introduce the mutations paradigm, which allows for a direct assessment of the time window during which distractors are processed. Participants were presented with a central target flanked by two identical distractors. Distractors’ identities could be either disruptive or neutral (non-disruptive). Distractors mutated once during each trial. There were three mutation types; disruptive distractors that mutated to neutral ones, vice-versa, or neutral distractors that mutated to different neutral distractors (control). In each trial, the mutation randomly occurred at one out of 11 times (between 17 ms and 187 ms after stimulus onset) (Fig. 1). Results revealed that (a) distraction occurred exclusively during the first ~50 ms following stimulus onset; (b) after ~50 ms, neither the presentation duration of distractors nor their identities affected performance (Fig. 2). That is, after ~50 ms all distracting stimuli were effectively disregarded. In accord with early-selection and dual-process theories of attention, we conclude that distraction occurs at extremely early stages, prior to the operation of attentional mechanisms that modulate processing.
Meeting abstract presented at VSS 2013