CLASSIFYING FACES BY RACE : THE STRUCTURE OF FACE CATEGORIES

This article explored the finding that cross-race (CR) faces are more quickly classified by race than same race (SR) faces. T. Valentine and M. Endo (1992) modeled this effect by assuming that face categories can be explained on the basis of node activations in a multidimensional exemplar space. Therefore, variations in exemplar density between and within face categories explain both facilitated classification of CR faces and the relationship between typicality and classification RT within face categories. The present findings from classification and visual search tasks suggest that speeded classification of CR faces is instead caused by a quickly coded race feature that marks CR but not SR faces. Also, systematic manipulations of facial typicality cause no variation in classifiability aside from slowed classification of very distinctive faces. These results suggest that the exemplar model cannot explain important aspects of face classification. Although face perception is usually studied from the standpoint of our amazing ability to differentiate a large number of faces, representations of face categories are also important. The process of categorizing individual faces has a number of implications both for general models of classification and for understanding face identification. The focus here is on the apparently paradoxical finding that participants are faster to classify faces they have difficulty recognizing. In the present case, this means that White participants classify Black or Asian faces faster than White faces (Levin, 1989; Valentine & Endo, 1992). In attempting to understand facilitated classification of cross-race faces (hereinafter referred to as the CR [cross-race] classification advantage), the present research considers explanations for the CR classification advantage as they relate to the basic structure of face categories, both in terms of discrimination between categories and in terms of their internal structure. Three explanations for the CR classification advantage are tested here. The first, stemming from Valentine's (1991) multidimensional space framework, places the advantage in the context of an exemplar model of face classification and recognition. This model uses simple assumptions based on