Abstract
Humans recognize own-race faces more accurately than other-race faces (Malpass & Kravitz, 1969), suggesting differences in the nature of neural representations for these faces. We examined the spatial and temporal patterns of neural responses to own- and other-race faces. Functional magnetic resonance imaging data were obtained while Asian and Caucasian participants viewed blocks of Caucasian and Asian faces. Voxels from high-level visual areas, including fusiform gyrus and lateral-occipital areas were localized in a separate scan using faces (of the same race as the participant), objects and scrambled-images. We first applied a pattern-based classifier to discriminate neural activation maps elicited in response to Asian and Caucasian faces. A low dimensional representation of the brain scans based on their principal components was used as input to the classifier. We measured the ability of the classifier to predict the race of the face being viewed. We found above-chance discrimination of the neural responses to own- versus other-race faces for both Asian and Caucasian participants. Reliable discrimination scores were obtained only when the voxel selection process used a localizer that presented “own-race” faces. Next, we examined differences in time-course of neural responses to own- and other-race faces and found evidence for a temporal “other-race effect”. The neural response to own-race faces was larger than to other-race faces, but only across the first few time points in the block. The magnitude of the neural response to other-race faces was lower at first, but increased across the block to ultimately overtake the magnitude of the own-race response. This temporal activation pattern held across the broader range of ventral temporal areas, and for the FFA alone. Spatial discrimination, however, was reliable only across the broader range ventral temporal areas. The results highlight the importance of examining the spatio-temporal components of face representations.
UT-Southwestern Advanced Imaging Research Center Seed Grant.