Abstract
Faces and body-parts are amongst the most salient visual stimuli in our environment. Research from human and non-human primates has reported multiple clusters of selectivity for faces and body-parts in the ventral visual pathway. Body-selective regions are often found close to face-selective regions, suggesting some sort of organizational principle. However, the nature of this organizational principle is not well established. Here, we investigated the topographical organization and specificity of body- and face-selective regions in visual temporal cortex at high resolution (1.2 mm isotropic voxels) using a 7T MRI scanner. First, we identified regions selective for faces and body-parts. Given prior reports of a body-part topography in lateral occipital cortex, we separately tested both hands and feet, which might be expected to have the most distinct representations. Second, we mapped population receptive fields in each participant to determine the extent to which the location of face- and body-selectivity reflect underlying retinotopic biases. Third, we tested the representational structure in face- and body-selective regions in a condition-rich event-related experiment. As expected, we found that faces elicited strong responses along the mid fusiform sulcus, in a region that has often been referred to as the Fusiform Face Area. This face selectivity coincided with a foveal representation of the visual field, while hands and feet produced robust responses adjacent and lateral to the face selectivity. We found little evidence for alternating patches of face and limb selectivity. Instead, we observed parallel streams of limb and face selectivity, extending from lateral to medial areas of the ventral cortex. Further, analysis of the representational structure of limb and face selective regions revealed striking differences. Our findings highlight the fine-grained organizational structure in ventral temporal and the importance of underlying retinotopic biases.
Meeting abstract presented at VSS 2015