Face recognition in emotional scenes: observers remember the eye shape but forget the nose

Face recognition is believed to be a highly specialized process that allows individuals to recognize faces faster and more accurately than ordinary objects. However, when faces are viewed in highly emotional contexts, the process becomes slower and less accurate. This suggests a change in recognition strategy compared to recognition in non-arousing contexts. Here we explore this finding by using a novel paradigm to determine which face dimensions are most important for recognizing faces that were initially encoded in highly emotional contexts. Participants were asked to recognize faces from a 3-alternative display after viewing a similar face that was embedded in either a neutral, positive, or negative emotional scene. Results showed that individuals rely on eye shape when recognizing faces that were encoded while embedded in either positive or negative emotional contexts, and ignore nose shape when recognizing faces that were encoded while embedded in negative emotional scenes. The findings suggest that, after encoding that face during heightened emotional arousal, individuals are more likely to commit errors when identifying a face on the basis of nose shape, and less likely to commit errors when identifying a face on the basis of eye shape.