Hanane Ramzaoui¹ (), Melissa R. Beck²; ¹Louisiana State University, ²Louisiana State University

Working memory (WM) serves current goals while enabling future ones to be planned and maintained. This latter role of WM is crucial, as everyday activities require finding task-relevant objects in a timely manner (visual search) and therefore prioritizing WM representations with respect to moment-by-moment task-relevance of items held in WM. How prioritization in WM affects the different processes during visual search remains unanswered. Here, we compare current and prospective search templates in a sequential search task. The current search template is the first target to be searched for, and the prospective search template is the target to be searched for in a subsequent display. On each trial, the participants perform two consecutive searches. Before the search displays, both targets are presented, and the one to be searched for in subsequent displays is indicated by the nature of its outline. Each array includes six colored objects, each unique in color and semantic category, arranged in a circle. Using an eye-tracker, response times were segmented into distinct processes during visual search: initiation time (first saccade latency), scanning time (elapsed time between the first saccade and the first fixation on the target) and verification time (elapsed time between fixation on the target and manual response). Search priority affected all three processes, resulting in shorter times for the first search than for the second search. This finding suggests that search priority affects both the process of setting up the template and the process of matching a fixated object to an internal representation of the target. Specifically, the information currently relevant provides a quickly accessible and precise template, enabling search to begin as soon as the array appears, as well as faster comparison of the fixated object to the template.