Visual cortex in the macaque monkey contains at least 32 distinct areas in a mosaic that occupies more than half of the cerebral cortex. These areas are interconnected by hundreds of cortico-cortical pathways to form a distributed hierarchical network that includes at least 10 stages of cortical processing. This proposal involves anatomical and physiological experiments aimed at (i) clarifying the connectivity and hierarchical relationships among specific extrastriate visual areas, and (ii) characterizing receptive field properties in extrastriate visual cortex that are relevant to form vision and pattern recognition. One set of experiments will involve injections of retrograde and anterograde tracers into several areas (V3A, PIP, VIP, and the putative dorsal and ventral subdivisions of LIP) whose connections have not been fully characterized in previous studies. The major inputs and outputs of each area will be examined, and the laminar pattern of its connections with other areas will be used to infer hierarchical relationships. Particular emphasis will be placed on analyzing irregularities in laminar patterns that do not conform to current criteria for distinguishing ascending, descending, and lateral pathways. Tracer injections will also be made into MT and V4 in order to examine hierarchical irregularities that might be related to the modular organization of V4. Another set of anatomical experiments will involve paired tracer injections in which one tracer is restricted to superficial cortical layers and the other to deep layers. Injections will be made into area V4 in initial experiments and into areas V2, DP, and MT in later experiments. The results should reveal whether or not projections that originate from both superficial and deep layers have similar patterns of termination for the two components. This will improve our understanding of the circuitry associated with ascending, descending, and lateral directions of information flow. The proposed physiological experiments are aimed at studying higher-order receptive field properties in extrastriate cortex. In one project, single unit activity in V2 and V4 will be measured in response to concentric, radial, and hyperbolic grating patterns as well as conventional straight gratings. Preliminary results suggest that these stimuli can be highly effective in activating extrastriate neurons and will be a valuable way of analyzing neural response selectivities. Another project will involve the analysis of neural responses in V4 to simulated 3-dimensional surfaces, including frequency modulated gratings and non-planar surfaces such as bumps and saddles. These experiments should improve our understanding of how 2-dimensional and 3-dimensional form information is processed in extrastriate visual cortex. Collectively, the proposed research should elucidate several information processing strategies used in visual cortex of non-human primates. These principles are relevant to our understanding of human cerebral cortex and will help in diagnosing and ultimately treating the deficits in cortical function that result from strokes and other neurological disorders.
Showing the most recent 10 out of 40 publications
