| Paper: | GS-5.3 |
| Session: | Contributed Talks V |
| Location: | Ormandy |
| Session Time: | Friday, September 7, 11:00 - 12:00 |
| Presentation Time: | Friday, September 7, 11:40 - 12:00 |
| Presentation: |
Oral
|
| Publication: |
2018 Conference on Cognitive Computational Neuroscience, 5-8 September 2018, Philadelphia, Pennsylvania |
| Paper Title: |
Familiarity Affects Early Perceptual Stages of Face Processing |
| Manuscript: |
Click here to view manuscript |
| DOI: |
https://doi.org/10.32470/CCN.2018.1104-0 |
| Authors: |
Katharina Dobs, Leyla Isik, Dimitrios Pantazis, Nancy Kanwisher, Massachusetts Institute of Technology, United States |
| Abstract: |
From a brief glimpse of a face, we extract not just the presence of a person, but their gender, age, familiarity and specific identity. How quickly are these specific dimensions of face information represented, and which dimensions are affected by familiarity? To find out, we used magnetoencephalography (MEG) and representational similarity analysis (RSA) to measure the time course of extraction of each of these dimensions of face information and their modulation by familiarity. Subjects viewed 80 face images, 5 of each of 16 celebrities, varying in lightning, pose, expression, and eye gaze. Celebrities varied orthogonally in familiarity, gender and age. Subjects (n = 16) performed a 1-back task on upright and inverted images in separate sessions. RSA analyses showed that we could decode identity, gender and age of face images at similar latencies within 130 ms after stimulus onset. We further found that familiarity enhanced face representations even at this early stage. Importantly, when identity decoding was analyzed within age and gender, early identity decoding remained significant only for familiar faces, suggesting qualitatively different early processing for familiar versus unfamiliar faces. |
