The ability to recognize individual faces, despite their structural similarity, has been a focus of investigation for both neuroscientists and psychologists. Yet, there is no current unified account of face representation that combines the knowledge from these two disciplines. Functional magnetic resonance imaging (fMRI) has identified a region in the fusiform gyrus (the FFA) that responds preferentially to faces, but the neural mechanism by which individual faces are encoded is still unknown. Several behavioral phenomena have inspired """"""""norm-based"""""""" models efface representation, where faces are coded as deviations from a norm, or prototype face. Until recently, the lack of a method for properly defining this norm and similarity relationships among faces has made it difficult to determine the neural basis of face representation. We propose a multidisciplinary approach combining the use of a new behaviorally validated face parameterization that affords these capabilities, high-resolution fMRI (HR-fMRI), and fMRI-adaptation, to parametrically probe the neural basis efface representation, addressing the following aims: - Aim 1: What is the response sensitivity of face-selective regions to the range of human faces? Using our parameterized model and HR-fMRI, we will characterize face-selectivity to a wide range of human face stimuli, expecting that typical faces will elicit the most activity, and distinctive faces the least. - Aim 2: How does the response sensitivity of face-selective regions change as a result of experience? Using a perceptual adaptation paradigm, we will induce face aftereffects that change the apparent distinctiveness of subsequently viewed faces. We predict that face-selective regions will respond more strongly to the same face when it is perceived as more typical than when it is perceived as more distinctive. - Aim 3: How sensitive are face-selective responses to changes in face similarity and gender? By parametrically manipulating similarity and measuring fMRI-adaptation, we will test the prediction that FFA is sensitive to face similarity and category, but other object-selective areas are only sensitive to similarity.
|Davidenko, Nicolas; Remus, David A; Grill-Spector, Kalanit (2012) Face-likeness and image variability drive responses in human face-selective ventral regions. Hum Brain Mapp 33:2334-49|
|Davidenko, Nicolas; Flusberg, Stephen J (2012) Environmental inversion effects in face perception. Cognition 123:442-7|