Comparison of shape coding in somatosensory and visual cortex

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): The goal of this research proposal is to investigate neural mechanisms underlying 2D form processing in the brain of the awake rhesus monkey (Macaca mulatta). Specific Aim 1 is to characterize the responses of neurons in primary (SI) and secondary (Sll) somatosensory cortex to a large set of indented 2D shapes comprising arcs and angles. In Aim 1 a, neural responses will be collected using standard neurophysiological techniques. Haptic shapes will be presented to the animal's finger pad using a custom-built tactile stimulator. Responses will be quantified using standard statistical analyses. Specific hypotheses regarding the nature of shape processing (i.e. what shape information is represented?) will be tested using competing mathematical models. The models will test the degree to which neurons represent contour position, orientation, and curvature. The goal of Aim 1b is to understand how the shape signal is derived in individual neurons. To this end, the time course of the shape representation will be characterized using models comprising temporally weighted parameters. A final goal of Specific Aim 1 is to understand how shape information is transformed from SI to Sll. Aim 1 c will combine the results from Aims 1 a and 1 b to model the interactions between SI and Sll. Our working hypothesis is that higher-level neurons (area 2/SII) integrate information about simple orientation features (processed in areas 3b/1) to form explicit representations of angles and curvatures. Specific Aim 2 is to test whether visual neurons in area V4 use similar coding strategies to represent curvature. In these experiments, the same types of models as those developed in Specific Aim 1 will be applied to previously collected data from V4 neurons. These neurons were presented with a set of 2D visual stimuli analogous to the haptic shapes to be used in Specific Aim 1. The shape models will be iteratively refined on the responses of the visual neurons. We will test the hypothesis that neurons in V4, which reportedly represent curvature, process shape in a manner similar to neurons in area 2 and Sll; namely, by combining simple orientation features to form representations of complex contour geometry. The ultimate goal of this proposal is to understand how shape information is generated and transformed in the brain. Knowledge of this transformation is key to understanding how neural circuits work on a systems level in sensorimotor functions. It has major implications on the design of future generations of neural prosthetics for amputees and blind subjects. This knowledge also impacts clinical approaches to sensory deficits caused by stroke, diabetes, and developmental disorders like autism.