Abstract
Same-race faces are easier to remember than other-race faces, a phenomenon known in the literature as the “other-race effect”. We investigated whether Caucasian participants use the same visual information when processing Caucasian and African-American faces using Bubbles (Gosselin & Schyns, 2001). We asked five participants to learn face-name associations for ten Caucasian and ten African-American males. Participants were then instructed to decide if a partly sampled (using Bubbles) face was part of the learned set (Caucasian and African-American faces tested in separate blocks). Multiple linear regressions between information samples and accuracy were performed for each race. For African-American faces, observers used mainly the nose in the spatial frequencies (SF) between 10 – 82 cycles per face width (cpf), the left eye in the highest SF band (between 41–82 cpf), and the region comprising the eyes and the nose in the second coarsest SF band (between 5–10 cpf). For Caucasian faces, observers employed the eyes in the SF between 10–82 cpf and the mouth in 5–41 cpf SF band. We argue that this difference in the information used to recognize Caucasian and African-American faces is the basis for the other-race effect observed in Caucasian observers.