Feature and Configuration in Face Processing: Japanese Are More Configural Than Americans

Previous work suggests that Asians allocate more attention to configuration information than Caucasian Americans do. Yet this cultural variation has been found only with stimuli such as natural scenes and objects that require both featureand configuration-based processing. Here, we show that the cultural variation also exists in face perception—a domain that is typically viewed as configural in nature. When asked to identify a prototypic face for a set of disparate exemplars, Japanese were more likely than Caucasian Americans to use overall resemblance rather than feature matching. Moreover, in a speeded identity-matching task, Japanese were more accurate than Americans in identifying the spatial configuration of features (e.g., eyes). Together, these findings underscore the robustness of culture’s influences on cognition.