Abstract
Other-race-effect (ORE) refers to the observation that we can recognize own-race faces better than other-race faces (Meissner & Brigham, 2001). Yet, whether featural or configural face processing might contribute to other-race effect is still unclear. In the present study, we tested Taiwanese adults with faces of four ethnic groups (Taiwanese, Philippine, Caucasian, African) and each with four levels of discriminability: Easy (change configuration and component: change identity), Medium (change component: change eyes), Hard-I (change configuration: widen eye spacing), and Hard-II (change configuration: mouth moved up). We adopted the visual paired-comparison task with two-alternative-forced-choice (2AFC) procedure. The overall results showed that accuracy decreased and response time increased as the stimulus difficulty increased for each race. The accuracy was highest and the response time was lowest for the Taiwanese easy condition, which suggests an own-race advantage. In addition, the pattern of response time for Philippine faces was similar to that of Taiwanese faces and was shorter than Caucasian faces in the medium and Hard-I conditions. In conclusion, our study had two main findings. First, Philippine faces were seen as more like own-race faces rather than other-race faces. Second, both featural and configural face processing contribute to the other-race-effect.