Abstract
Perceptual advantages for own-race compared to other-race faces have been demonstrated for the recognition of facial identity and expression. However, these effects have not been investigated in the same study, and to date, substantial procedural differences between the tasks used to investigate identity and expression have precluded such a comparison. To address this issue, we used a photo sorting task in which Chinese and Caucasian participants were asked to sort photographs of Chinese or Caucasian faces by identity or by expression. This paradigm matched the task demands of identity and expression recognition and avoided constrained forced-choice or verbal labelling requirements. In addition, this method can also determine the extent of cross-cultural agreement as well as differences. Other-race effects of comparable magnitude were found across the identity and expression tasks. Caucasian participants made more confusion errors for the identities and expressions of Chinese than Caucasian faces, while Chinese participants made more confusion errors for the identities and expressions of Caucasian than Chinese faces. However, analyses of the patterns of responses across groups of participants revealed a considerable amount of underlying cross-cultural agreement. These findings suggest that widely repeated claims that members of other cultures "all look the same" overstate the cultural differences.
Meeting abstract presented at VSS 2016