This project defines features of primate retinal function in the magnocellular (MC) and parvocellular (PC)pathways at the retinal ganglion cell (RGC) level. Using both physiological and psychophysical techniques, it tests hypotheses about retinal physiology and about how retinal signals underlie central processing of visual information. Precise knowledge of the retinal output provides a reference for interpreting cortical and behavioral data, and for interpreting visual deficits in retinal disease. Specific goals to be investigatedare: (1) When a background changes in time, color or spatial pattern, as must occur in natural environments, stimuli must be detected and analysed under dynamic conditions. The goal is to determine and model the RGC signal with such stimuli, and to determine how visual stimuli are detected and encoded under complex, changing conditions, ranging from simple modulated backgrounds to natural scenes. (2) All information leaving the retina is encoded within RGC spike trains. The goal is to specifically assess the spatiotemporal information inherent in the RGC signal with specially designed achromatic or chromatic patterns; by comparison with human performance, one will then be able to constrain models of how central mechanisms handle the retinal output. (3) Among mammals, red-green opponency is unique to humans and their primate relatives; it is based on the presence of L and M cones, which with S cones form the basis of trichromacy. Knowing how their signals combine in different cell types is central to our understanding of retinal wiring and function. A new method is described for specifying cone inputs to visual neurons. Also, the L,M-cone opponent input which is characteristic of PC cells provides some signal to MC cells; the origin and behavioral significance of the chromatic response component to MC cells is to be described and modeled.
Showing the most recent 10 out of 39 publications
