TALK: Attentional capacity limitations in face perception - processing face sets

Distracter faces are difficult to ignore. Faces that are entirely irrelevant to an ongoing task are being processed even in ‘high load’ conditions that promote distracter inhibition for non-face stimulus categories. Distracter face processing can, however, be reduced by adding a second face to a display. In my talk, I will report ERP- and behavioral evidence for such capacity limitations in face processing. Additionally, I will present new data from a series of experiments on ‘set averaging’ for famous faces. It has recently been suggested that information from multiple faces, presented simultaneously for a brief time (‘set’), may efficiently proceed in form of a statistical value such as the ‘mean/average’ from the set (e.g., mean emotion, Haberman & Whitney, 2009). Our data suggests that such abstractive coding can occur for facial identity information, even when famous faces were shown in the set. In contrast to earlier set averaging studies, where memory for the set instances was extremely poor, participants in our experiments were highly accurate for famous set faces. This seems to suggest that participants can extract abstractive and instance-based representations simultaneously.

Person Perception Seminars - ARC Centre of Excellence in Cognition and its Disorders