Abstract
The ability to differentiate a particular face from others or to generalize faces to a common feature such as gender or race is critical for social interactions. Then, how is face information represented in the visual cortex when face individuation is emphasized compared to when face generalization is emphasized? To address this question, we performed a functional magnetic resonance imaging (fMRI) experiment, comprising two generalization tasks – gender and race tasks -, and one discrimination task – matching task. On each trial of every task, a participant saw a sample face image, followed by a test face image. In the gender task, participants were asked to decide whether the gender of the sample face image is the same with that of the test face image, while in the race task, they had to determine whether the sample face and the test face are from the same race category. In the matching task, the test image was a face fragment image, and they were instructed to decide whether the fragment image belonged to the sample face. We independently localized face-selective cortical areas, and compared decoding of individual faces from the multi-voxel pattern of response for the sample face perception in each task. In the gender or race task, the face-selective cortex did not show significantly distinct representations for each face. However, in the matching task, the response of face-selective cortex could be used to decode the individual face information. These results suggest that top-down signals modulate the cortical representation of individual faces during perception. This work was supported by a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI) funded by the Ministry of Health & Welfare, Republic of Korea (HI15C3175), and the National Research Foundation of Korea(NRF) grant funded by the Korea government(MSIP) (2016R1C1B2010726).
Meeting abstract presented at VSS 2017