Abstract
Face identity and facial expression are believed to be represented by separate neural mechanisms in the human brain. Both face identity and facial expression are adaptable and perceptual aftereffects can be observed in each. Further, expression aftereffects can be generated with faces of different individuals, consistent with an expression representation independent of identity representation. However, given the well known “other-race” effect in face perception, it remains possible that faces from different races are encoded by distinct neural mechanisms or different neuronal subpopulations. Consequently, the question arises on whether there is a single expression analysis system across faces from different race categories.
We investigated this issue by testing whether or not a race-contingent expressional aftereffect could be observed, despite evidence that expression aftereffects are insensitive to identity within the same race. Participants adapted to alternating faces of two different races (Caucasian (C) and African American (AA)). During adaptation, faces from the two races were consistently rendered in two opposing expressions (e.g., C-happy and AA-sad or the other way around). Following this combined adaptation, briefly presented faces with neutral expression were perceived as slightly sad or happy, depending on the racial category of the test face and the race-expression pairings during adaptation. The results show that facial expression adaptation could be made contingent on the racial identity of the faces. These results suggest that facial expression analysis is not completely independent of the face identities. At the neuronal level, the observation of a race-contingent expression aftereffect implies the existence of neurons with joint tuning sensitivity to expression and racial category of the face.
This research was supported by the James S. McDonnell foundation, the US National Institutes of Health Grant R01 EY015261-01.