Abstract
There are obvious advantages to a mechanism for the efficient detection and monitoring of social threat. Indeed, angry faces elicit greater recall accuracy in working memory (WM) tasks than happy and neutral faces. However, the processes that facilitate this ‘angry benefit’ are not fully understood. Here we present evidence for a facilitatory process occurring early during memory maintenance. We presented study arrays of either two angry or two happy faces for 2000 ms followed 1000 ms later by a probe face. Probe faces had neutral expressions and were either the same individual as one of the faces in the study array, or (on 50% of trials) different individuals. Participants indicated by key press whether probe individuals had been present in the study array. During the memory maintenance interval (between study array offset and probe onset), scrambled face masks were presented at the same locations as the faces in the study array. Using a blocked design, the interval between study array offset and mask onset was manipulated. Intervals were: 17 ms, 117 ms, 300 ms, 500 ms and 700 ms. Masking early in the maintenance interval was done to interrupt the process of consolidation, i.e., the formation of durable, robust memory traces from the initial ‘fragile’ contents of memory. Later masks were predicted to disrupt consolidated WM representations. We observed the typical ‘angry benefit’ when masks were presented at timings of 117 ms or later. However, when the pre-mask interval was brief (17 ms), angry and happy face identities were recalled with equivalent accuracy. Interestingly, memory performance for happy faces was equivalent at all masking intervals, suggesting that early masks did not interrupt face consolidation, but instead disrupted a process that would otherwise result in angry faces developing their usual ‘extra robust’ memory representations.
Meeting abstract presented at VSS 2012