Abstract
The goal of this study was to isolate repetition suppression effects for each part of a whole face stimulus. To do so, the left and right halves of face stimuli were flickered at different frequency rates (5.88 Hz or 7.14 Hz) while changing face identity or not at every stimulation cycle (Figure 1). Recording high-density electroencephalogram (EEG) in 11 human participants fixating in the centre of the face, robust responses were observed to each face half at these specific frequency rates. These part-based EEG responses were larger in amplitude when different as compared to repeated face half identities were presented at every stimulation cycle. Contrary to whole-face repetition suppression effects, which are usually found over the right occipito-temporal cortex (Rossion & Bormemanse, J Vis. 2011 Feb 23;11(2)), these part-based repetition suppression effects effects were found over all posterior electrode sites. Most importantly, they did not decrease when the two face halves were manipulated by separation, lateral misalignment, or inversion (Figure 2). Critically, there were also robust intermodulation (IM) components in the EEG spectrum (e.g., 7.14 – 5.88 = 1.26 Hz), which are unequivocally produced by neuronal populations that interact or integrate the two face parts nonlinearly. These IM components were found mainly over the right occipito-temporal cortex and were significantly reduced following the aforementioned manipulations. Additionally, the IM components decreased substantially for face halves belonging to different identities (Figure 3), which form a less coherent face than when they belong to the same face identity. These observations provide objective evidence for dissociation between part-based and whole-based responses to faces in the human brain, suggesting that only whole-based responses may reflect high-level, possibly face-specific, visual representations.
Meeting abstract presented at VSS 2014