Abstract
It is well known that observer race can influence performance when individuating faces. Inspired by recent studies of “ensemble” face processing, we asked whether similar effects occur when making decisions about groups of faces. Here, in each trial, 16 faces were continuously shuffled within a moving 4 x 4 grid which remained visible for 4 seconds. The proportion of Asian and Caucasian faces were varied from trial to trial and observers (16 Asian
/ 16 Caucasian) were instructed to indicate which race was in the majority. Between subjects, we varied the “target” response that should be given when observers were uncertain. We fitted cumulative normal functions to the response distributions and extracted the PSE and JND as dependent measures. For PSE, observers consistently weighted other-race faces more heavily than own-race faces, requiring more faces of their own race to perceive equality. For JND, Asian observers were generally more sensitive, but both groups had smaller JNDs when the target
response was the other-race. Overall, our findings indicate that observers can rapidly estimate the racial
composition of a group of faces, but these estimates are influenced by their own race.
