Reconstructing emotions in motion in prosopagnosia reveals separate representations for identity and emotion

Richoz, A., Jack, R.E. , Garrod, O.G.B., Schyns, P.G. and Caldara, R. (2014) Reconstructing emotions in motion in prosopagnosia reveals separate representations for identity and emotion. Perception, 43, p. 123.

Full text not currently available from Enlighten.

Publisher's URL:


The human face transmits a wealth of visual signals that readily provide crucial information for social interactions, such as identity and emotional expressions. Yet, a fundamental question remains unresolved: does the face information of identity and emotional expression categorization tap into single or separate representational systems? To address this question we tested PS, a pure case of acquired prosopagnosia, whom uses the suboptimal mouth to process identity and is impaired in categorizing many expressions of the (static) Ekman faces. We used a generative grammar of the Facial Action Coding System coupled with a reverse correlation technique to model 3D dynamic mental representations of the six basic facial expressions of emotion, in PS and healthy observers. Surprisingly, PS’s dynamic mental models of facial expressions were comparable to the controls. Subsequent verification tasks revealed that PS accurately categorized her own and average dynamic facial expression models, but not the very same static exemplars. Evidence that PS reconstructed dynamic facial expressions by using all facial features demonstrates that the face-system relies on distinct representations for identity and emotion, flexibly adapting to categorization constraints. Our data also questions evidence of deficits obtained from patients using static images, and offer novel routes for patient rehabilitation.

Item Type:Articles
Glasgow Author(s) Enlighten ID:Garrod, Dr Oliver and Caldara, Professor Roberto and Schyns, Professor Philippe and Jack, Professor Rachael
Authors: Richoz, A., Jack, R.E., Garrod, O.G.B., Schyns, P.G., and Caldara, R.
College/School:College of Medical Veterinary and Life Sciences > School of Psychology & Neuroscience
College of Science and Engineering > School of Psychology
Journal Name:Perception
ISSN (Online):1468-4233

University Staff: Request a correction | Enlighten Editors: Update this record