CAREER: How populations of cortical neurons represent visual motion and binocular disparity

University Of Wisconsin-Madison, Madison WI

Investigators

Abstract

The goal of the research is to understand how a population of neurons acting together mediates visual perception using a multidisciplinary approach that includes neurophysiology, psychophysics and computer simulations. Population coding - the ability of neuron populations to represent information - is a fundamental problem that is especially critical when multiple stimuli or features are encoded by the same population of neurons. Binocular disparity and visual motion are detected through separate retinal mechanisms, but they form a joint representation in several brain regions. The proposed research will study how motion and disparity are jointly represented at a distributed level. The first aim of this research is to determine the number of stimuli one population represents. Specifically, the hypothesis will be tested that a single population represents one direction of motion, and that two directions of motion are thus represented by two separate populations of neurons. The second aim will determine whether correlations of neuronal firing rates between neurons are used as a mechanism to represent information about the direction and disparity of a stimulus over and beyond what is available in the information of the firing rates of two neurons alone. Together this research will elucidate the basics about how groups of neurons encode multiple stimuli and features and thus give rise to visual perception. The proposed work will further the development of a novel laboratory course by developing a computer-simulated visual organism (VISOR). VISOR will produce simulated neuronal and behavioral responses based on simulated visual stimulation. VISOR will be used as an integral part of the course to encourage students to formulate and test hypotheses about visual perception.