Abstract
Visual search is expedited in a repeatedly encountered spatial context. A much-debated question is whether attentional guidance in visual search is facilitated by contextual cueing, and if so, how early this facilitation arises. The current study investigated the facilitation of attentional guidance by examining the time course of contextual cueing benefits. The participants in an experiment engaged in an SAT (speed-accuracy tradeoff) task after learning the spatial contexts in a standard visual search task in which they searched for a rotated T target among Ls. In the SAT task, participants were required to search for a target and respond immediately when a sound probe was presented, even if they had not found or identified the target. The inter-stimulus interval between a search display and the probe varied from 40 ms to 2,000 ms. Participants completed two blocks of the SAT task, in which they searched either under “learned” or “new” contexts. The result of the SAT procedure indicated that responses were more accurate in repeated contexts than in new contexts, even when only a brief time had elapsed after the search display presentation (>90 ms). We also conducted an analysis of the time course of contextual cueing effect using Bayesian hierarchical modeling. Our model was based on the conventional SAT function that describes the increase of accuracy as a function of elapsed time after stimulus presentation. This analysis demonstrated that the rate of accuracy increase was higher than in new contexts than in repeated contexts. These findings suggest that attentional guidance is enhanced by learning the context, and this enhancement begins at a very early stage of the visual search.