Abstract
Faces are visually represented as mental templates, and people have discrete templates for different social categories such as sex and race. These templates develop as an average all of the faces one has encountered (Leopold et al., 2001; Valentine et al., 2004) and are malleable: they can be updated when an individual views new faces (Hurlbert, 2001). Face templates can be experimentally manipulated to test whether opposing aftereffects can be created, suggesting that the faces of social groups are represented by discrete templates (Little et al., 2005, 2008). The current study used an opposing aftereffects paradigm to determine which cues lead individuals to view Christians and Muslims faces as different social categories. We tested whether 1) explicit religious labels, 2) food preferences, and 3) country of origin would support opposing aftereffects. Ninety-three participants were assigned to 1 of 3 conditions: while viewing face images, they heard audio that either 1) stated the character’s religious identity explicitly, 2) named a preferred food, or 3) named their country of origin. The countries and foods that were used in the audio descriptions were validated, and only those which were strongly associated with Christianity or Islam were chosen. Participants in all 3 conditions viewed 60% contracted Christian faces and 60% expanded Muslim faces during the training phase. Opposing aftereffects were found in the religious explicit (t(54) = 2.27, p = 0.03, Cohen’s d = 0.58) and food audio conditions (t(57) = 3.23, p < 0.01, Cohen’s d = 0.82), but not in the country audio condition (t(57) = 1.21, p = 0.23, Cohen’s d = 0.31). This suggests that explicit religious labels and food preferences create a socially meaningful distinction between religious groups, but country of origin does not. Among other inferences, this is evidence of an impact of social categorization on visual processing.