Abstract

Visual working memory (VWM) is a limited-capacity resource for the temporary storage of visual information (Cowan, 2001). Selective visual attention can protect VWM capacity by filtering relevant from irrelevant information during encoding or maintenance (Griffin & Nobre, 2003).The aim of the current study was to investigate changes in the accuracy of face recognition memory when attention was reflexively oriented to the location of a face before (pre-cue) or after (retro-cue) it was disappeared. The results indicate that VWM capacity for unfamiliar faces is poor and declines as a function of the number of faces in the display (load). When attention was oriented by a pre-cue, recall accuracy was independent of load, suggesting attention captures VWM resource in a reflexive manner. This advantage was not observed for retro-cued faces, suggesting reflexive attention operates by biasing VWM resources during encoding rather than the maintenance of remembered faces. These findings have a number of implications in real-world eyewitness scenarios. For example, the accuracy of eyewitness identification is likely to depend upon the number of individuals at the crime scene and differences in their salient visual characteristics.