Abstract
The 3-state model of visual working memory (VWM) proposes three states of memory: Focus of attention (FA), direct access region (DAR), and activated LTM (aLTM), with the FA and DAR representing VWM and aLTM representing a relatively active portion of LTM. Previous research has found the item in the FA to be the most accessible, followed by the item in the DAR, and then task relevant items in aLTM are passively maintained. The goals of the present study were (1) to look for evidence for the 3-state model, and (2) to examine how information is forgotten from VWM and aLTM by using a directed forgetting paradigm. In previous research, the directed forgetting paradigm resulted in a directed forgetting benefit – memory for remember items is significantly better when item(s) are directed to be forgotten. Within the 3-state model, it could be that information is dropped completely from all three states resulting in a directed forgetting benefit across all states. Using eye tracking and gaze-contingent trials, participants were presented six faces on each trial in sequential order. On cue-to-forget trials, after viewing all faces in a trial, an arrow pointed at one face for 500 ms and following a visual mask, participants were shown one face and asked, “Did this face change?” All cue trials were valid and participants were never tested on a cued-to-forget face. The present experiment found support for the 3-state model of VWM, but did not find a directed forgetting benefit for any of the states. This experiment suggests that a directed forgetting benefit is not found with complex stimuli and that the 3-state model of VWM can be robust against directed forgetting.