Abstract
Face and body perception is mediated by configural mechanisms, which allow the perception of these stimuli as a whole, rather than the sum of individual parts. Indirect measures of configural processing in visual cognition are the face and body inversion effects (FIE and BIE), which refer to the drop in performance when these stimuli are perceived upside-down. Albeit FIE and BIE have been well characterized at the behavioral level, much still needs to be understood in terms of the neurophysiological correlates of these effects. Thus, in the current study, the brain’s electrical activity has been recorded by a 128 channel electroencephalogram (EEG) in 24 healthy participants while perceiving (upright and inverted) faces, bodies and houses. EEG data were analyzed in both the time domain (i.e., event-related potentials—ERPs) and the frequency domain [i.e., induced theta (5–7 Hz) and gamma (28–45 Hz) oscillations]. ERPs amplitude results showed increased N170 amplitude for inverted faces and bodies (compared to the same stimuli presented in canonical position) but not for houses. ERPs latency results showed delayed N170 components for inverted (vs. upright) faces, houses, but not bodies. Spectral analysis of induced oscillations indicated physiological FIE and BIE; that is decreased gamma-band synchronization over right occipito-temporal electrodes for inverted (vs. upright) faces, and increased bilateral frontoparietal theta-band synchronization for inverted (vs. upright) faces. Furthermore, increased left occipito-temporal and right frontal theta-band synchronization for upright (vs. inverted) bodies was found. Our findings, thus, demonstrate clear differences in the neurophysiological correlates of face and body perception. The neurophysiological FIE suggests disruption of feature binding processes (decrease in occipital gamma oscillations for inverted faces), together with enhanced feature-based attention (increase in frontoparietal theta oscillations for inverted faces). In contrast, the BIE may suggest that structural encoding for bodies is mediated by the first stages of configural processing (decrease in occipital theta oscillations for inverted bodies).
Highlights
Humans can identify hundreds of faces with ease, all share a common 3D structure
The results revealed significant interactions between face inversion and house inversion over a bilateral occipitotemporal cluster (29 electrodes; p = 0.008) and a frontal cluster (44 electrodes; p = 0.003), suggesting that face inversion generates an increase in N170 amplitude that is significantly higher than the increase generated by house inversion
A statistically significant interaction between body inversion and house inversion was found over a right occipito-temporal cluster (19 electrodes; p = 0.025) and a left frontal cluster (20 electrodes; p = 0.017), indicating that the increase in N170 amplitude generated by body inversion was significantly higher than the increase generated by house inversion
Summary
Humans can identify hundreds of faces with ease, all share a common 3D structure (i.e., two eyes above the nose, which is in turn above the mouth). The reduced accuracy (and increased latency) in recognizing faces when they are perceived upside-down rather than in their canonical orientation is known as the ‘‘face inversion effect’’ (FIE; Yin, 1969), and has traditionally been considered as (indirect) evidence for the existence of configural processing for upright faces only. Since this effect is much smaller for non-face objects (Valentine, 1988) and objects of expertise (Robbins and McKone, 2007), it has been suggested that FIE might be face-specific, leading to the conclusion that configural processing only mediates upright face perception (ibidem). Inversion was proven to affect all these configural processing stages (Maurer et al, 2002), and face processing is thought to require and rely on the last stage (i.e., the holistic stage; ibidem)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
