Abstract
In the real world, objects are typically embedded in complex backgrounds. Certain backgrounds make object processing difficult while others facilitate it, especially in conditions where the intrinsic object information is degraded. The neural correlates of these influences are, at present, largely unknown. The goal of this study is to address this important issue by examining whether the activity of areas believed to be involved in object processing can be modulated by different kinds of background information. We focus on contextual influences on the perception of faces. A large body of convergent evidence has implicated a small region of the fusiform gyrus (the ‘FFA’) in the visual processing of faces. We hypothesized that if context information is incorporated in facial representations, we will observe fusiform activity to stimuli in which faces are implied by context, even in the absence of intrinsic facial information (for example, given a body with a highly blurred face, the body implies a face, even though the degraded face image by itself is unrecognizable as a face). Using fMRI techniques, we found that a relevant context can activate face-related areas as strongly as clear images of faces themselves. This finding provides evidence for the role of contextual cues in face processing. In order to further elucidate the mechanisms of such contextual influences, we are conducting analogous experiments using magneto-encephalography (MEG) that provides much finer temporal resolution. These experiments are designed to probe whether the temporal characteristics of face-specific responses (N170) differ depending on intrinsic or contextual definition of faces. The results have implications for the nature of processes that underlie face-perception in complex scenes.