Abstract
Expectations based on previous experience shape our perception by modulating attention. The effect of expectations on selective attention has been well-studied with regards to task-relevant stimuli, but less is known about the role that expectations play in suppressing task-irrelevant, distracting stimuli. We investigated the effects of distractor expectations using three experiments, all of which were based on a modified version of the classic Eriksen flanker task. We manipulated the frequency of specific flanker configurations, producing Standard and Oddball flankers. Experiment 1 tested how block-level flanker configuration probabilities (Standard/Oddball: 0.95/0.05, 0.90/0.10, 0.75/0.25, 0.50/0.50) affected accuracy and response time. Experiments 2 and 3 tested the interaction between “runs” of successive trials (length: 2, 4, 7, 11, 16) and block-level flanker configuration probability (Standard/Oddball: 0.90/0.10, 0.50/0.50) to better understand the dynamics of expectation, attention, and task performance. Across all studies, we found robust evidence for sensitivity to the probability of flankers, with an approximately logarithmic relationship between the likelihood of a particular flanker configuration and the accuracy of subjects’ responses. The magnitude of this effect was larger for visually similar chevron stimuli, than for visually dissimilar letter stimuli. Subjects were sensitive to length of runs of repeated targets, but minimally sensitive to length of runs of repeated flankers. Our results demonstrate that the presence of statistical regularities across multiple time-scales and stimulus characteristics can affect performance. This is most likely accounted for by expectation-mediated distractor inhibition and from attentional capture resulting from its disruption. As a whole, these results show that expectations strongly influence attention in complex cognitive settings via multiple, nested factors.