Abstract
Previous studies have demonstrated that in a visual search task, observers are faster to locate targets when these are presented in repeated rather than random contexts, an effect termed contextual-cuing. However, we have recently shown that this effect reflects interference. In the present study, we used event related potentials (ERP) to investigate whether this interference is reflected in target selection processing and in visual working memory (VWM). Three groups of participants completed a visual search task while EEG was recorded. For the Consistent-only group, targets and distractors appeared in predefined spatial locations, for the Random-only group the task contained no regularity, and for the Mixed group the task contained both consistent and random conditions intermixed. Behaviorally, we replicated our previous results, such that performance in a random visual search, without any regularity, was better than performance in a mixed-design search that contained a beneficial regularity. This interference was reflected by an earlier N2pc component with higher amplitude in the Random-first group than in the Random-mixed condition, suggesting that the efficiency of target selection in a random search deteriorated when regularity was introduced into the task. More generally, we conclude that the interference to performance under mixed conditions begins during the search process itself, long before the response. Additionally, the ERP marker for VWM (the contralateral delay activity; CDA) was higher when the task contained regularity (Consistent-only group, Consistent-mixed condition) than during a random search (Random-only group, Random-mixed condition), presumably because search templates (i.e., context) were continuously held in VWM.