Spatial distribution of face and objects representation in the human brain.

Numerous imaging studies have reported a consistent activation in the fusiform gyrus (in a portion localized between the occipital and temporal lobes) when contrasting the response to faces to other categories of objects. This region has been called “Fusiform Face Area”, FFA (Kanwisher et al. 1997). Since this pivotal finding, further work has investigated the role of the fusiform gyrus in the perception of object categories other than faces. With this purpose, many experiments have contrasted the response to different types of objects and other areas have been identified by their maximal response to specific categories. For example, an area in the parahippocampal gyrus responds maximally to interior spaces, buildings or landscapes, the so-called parahippocampal place area (PPA) (Epstein and Kanwisher 1998), and an area localized in the inferior temporal gyrus responds maximally to body parts other than faces, the so-called extrastriate body part (EBA) (Downing et al. 2001). This preferential activation in specific regions in response to specific categories has contributed to a controversy about how the brain represents and differentiates among categories of objects. One line of thought proposes a modular theory according to which the brain is organized in ‘modules’ each dedicated to the representation of a specific type of information (Kanwisher J Neurosci 1997). Another line of thought proposes, in contrast to previous hypothesis, that the representation of different categories of objects in the brain is shaped by experience. According to this view, the “fusiform face area” is specialized for the representation of ‘visual expertise’ – which refers to the capacity for finer-grained discrimination among individuals within category – not just for face perception (Gauthier et al Nat Neurosci 2000). A third line of thought argues that the organization of object-responsive cortex can be explained by a coarse retinotopy that biases the representation of object categories according to how consistently those categories are perceived in central or peripheral vision (Malach et al. 2002). We have proposed that the representations of faces and objects in ventral temporal cortex are distributed and overlapping (Haxby et al. 2001). We showed that the distinctiveness of the response to a given category is not due simply to the regions that respond maximally to that category by demonstrating that the category been viewed can still be identified on the basis of the pattern of response when those regions were excluded from the analysis. Thus, representations of faces and objects are distributed locally within ventral temporal cortex. Faces and other categories of objects also evoke neural activity in cortical areas outside the ventral object vision pathway. These responses indicate the spontaneous activation of other types of information that are associated with objects, such as the emotion associated with a facial expression, the direction of attention indicated by eye gaze, or the motion associated with tool use. Thus, the representation of a visually-presented object appears to be distributed not only locally but also across multiple cortical areas. Finally, the temporal course for the development of face and object representations can be studied with electro- and magnetoencephalography (EEG and MEG). Although early neural responses (< 200 ms) can differentiate among some categories, only later responses show the effect of familiarity or reflect finer discriminations within a category (Bentin et al. 1998; Sugase et al. 1999; Eimer 2000). In this chapter, we review the functional neuroanatomy of face and object recognition with emphasis on explicating how representation are distributed in space and time.