Poster C22, Sunday, March 25, 1:00-3:00 pm, Exhibit Hall C
Probing the Time-course of Face Representations with Time-resolved Multivariate Pattern Analyses of EEG Signals
Laurie Bayet1, Rachel Wu2, Benjamin Balas3, Richard N. Aslin4; 1Boston Children's Hospital, Harvard Medical School, 2University of California, Riverside, 3North Dakota State University, 4Haskins Laboratories
Time-resolved MVPA (decoding) of MEG data has successfully uncovered the time-course and temporal dynamics of neural representations in adults. EEG, a low-cost alternative to MEG, is generally better suited for developmental research. Thus, a question is whether decoding from EEG signals would similarly uncover the timing of representations in adult and developmental populations. This talk presents two uses of EEG decoding to uncover the time-course of neural representations associated with face perception in adults and children. In a first study, adults completed a visual search task for non-human animal (N = 20) or human (N=21) faces. Target location (i.e., right or left) could be decoded from EEG signals from approximately 200ms onwards, with similar accuracy for both human and animal faces. Neural patterns associated with individual versus categorical searches were more robustly differentiated for human than animal faces from 176-204ms. These findings suggest that perceptual narrowing affects task-dependent neural representations during visual search, but surprisingly preserves the attentional selection of targets. In a second study, 5-7 year-olds (N=18) and 8-10 year-olds (N=18) passively watched pictures of real or doll faces presented in an upright or inverted orientation. Information about face orientation (inverted vs. upright), but not face animacy (real vs. doll), could be decoded from the EEG signal in 5-7 year-olds from 248-316ms. The same time-course was evident in 8-10 year-olds, although decoding accuracy for face orientation only marginally reached statistical significance (272-316ms). Taken together, these findings demonstrate the potential of time-resolved EEG decoding for developmental and cognitive neuroscience research.
Topic Area: EMOTION & SOCIAL: Person perception