Abstract
Navigation of the visual world requires humans to rely extensively on the ability to attentionally filter distractors in the service of goal-directed action. Previously, we have shown that distractor filtering significantly improves over the span of several days as a result of task repetition (Greiner et al., 2021). Additionally, there have been many reports of visual statistical learning during which individuals unconsciously begin to suppress spatial locations in which distractors are present (Theeuwes et al., 2022). Here, we aimed to identify whether the improvement we observed in distractor filtering over several days is mediated by cue presentation, linking it to history-driven attentional selection. Participants were loaned an iPad containing an implementation of the Attention Network Test (ANT; Fan et al., 2002) to track attentional fluctuations across 42 sessions over a 12-day period. On the initial and final days of the study, participants completed a baseline ANT session. On the intervening days, participants were notified four times daily (morning, mid-day, evening, and night) to complete a brief ANT session. Our results showed significant distractor filtering improvement for both cue and no-cue trials (both p’s < 0.001, t(40) = 21.50 and t(40) = 17.56, respectively). However, trials containing a cue led to a significantly larger distractor filtering improvement than trials with no cue (p = 0.02, t(40) = 2.428). Additionally, there was no difference between distractor filtering improvement in trials with a (nonpredictive) center cue versus trials with a (100% valid) spatial cue (p = 0.45, t(40) = 0.7638). Taken together, these results indicate that visual statistical learning, particularly in regard to attentional selection, can be maintained over the course of several days whereby the presence of a cue leads to the enhancement of history-driven selection.