Abstract
We have previously shown that adapting to a specific identity improves discrimination of structural changes pertaining to identity on that face but not on other faces (Oruc & Barton, 2011). In this study we ask whether adapting to faces can improve discrimination of facial expression. 175 subjects participated (91 female). Each participant was randomly assigned to one of four test faces (two female) that remained constant throughout the session. We measured accuracy for perceiving subtle deviations in facial expression after adapting to expressive faces in five conditions: (1) same-identity/same-expression, (2) same-identity/different expression, (3) different-identity/same-expression, (4) different-identity/different expression, and (5) blank. Gender remained the same between test and adapting faces. Adapting stimuli were video clips displaying a progression from neutral to the full extent of the adapting expression. Adapting duration was 8s during which the same video clip repeated four times. The task was 2-interval 2-alternative forced choice discrimination of expression in which subjects viewed two static images each displaying a subtle deviation equidistant from a base expression towards two different end expressions. We hypothesized that performance in expression discrimination would improve after adapting to the same expression, but not a different expression, compared to baseline (blank). We also hypothesized that adaptation-induced improvements would be maximal when adapting faces had both the same identity and expression with the test faces. Results show an unexpected own-gender effect: adaptation facilitated discrimination accuracy for other-gender test faces (M=2.17%, p<0.023) but impaired it for own-gender faces (M=-3.15%, p=0.002) compared to baseline, also shown by a significant main effect of gender-congruency (F(1, 684)=11.10, p<0.001). There was no main effect or interaction involving adapting condition. Unlike the context of the own-race effect, observers should have equal exposure to male and female faces through their daily experience. This suggests a genetically determined component to facial expression processing.
Meeting abstract presented at VSS 2013