Abstract
Visual information have long been thought to be processed in two separate visual pathways, with ventral stream including regions along the temporo-occipital cortex representing object information and dorsal stream including regions along the parietal cortex representing spatial information. This view has been challenged by studies indicating the involvement of the parietal cortex in object representation. In this study, using stimuli that were controlled for low-level features, we investigated object representation in human parietal cortex and explored the functional significance of this representation in a perceptual decision making task. Observers were asked to categorize three novel object categories. The images were controlled for luminance, contrast and spatial frequency and were rendered noisy to increase task difficulty. For each observer, two noise levels were determined individually to yield behavioral performance of 100% (Easy) and 75% (Hard) correct. Using multivariate pattern analysis (MVPA), we investigated fMRI response patterns for this object classification task from five retinotopic areas along the human intra parietal sulcus (IPS 0-5). We also examined response from lateral occipital cortex (LOC), a region previously shown to be involved in processing visual object information. These regions of interests were identified individually on each observer using independent localizers. We found that it was possible to classify the three novel object categories in both the "Easy" and the "Hard" conditions, using patterns of fMRI activity from LOC, IPS 3 and IPS 4, with performance being higher in the "Easy" than in the "Hard" condition. Classification performance from IPS 0-2 did not reach significance. Thus both the object selective area LOC and sub-regions of the human parietal cortex carry visual object information that track the quality of the sensory information available for object categorization. These results suggest possible role of the human parietal cortex in object categorization and perceptual decision related processing.
Meeting abstract presented at VSS 2012