The process of obtaining visual information about the world begins in the retina. The layers of computation in the retina extract multiple distinct features of the visual scene, which are sent to the brain by different types of retinal ganglion cells. Anatomically, 20 types of primate retinal ganglion cells have been identified, each communicating distinct information about the image to its set of central targets. The five highest-density cell types are well studied, but the function of the other 15 low-density types is largely unknown. The goal of this research is to characterize the physiology of the low-density primate retinal ganglion cell types as well as determine each cell type?s anatomical identity. A thorough characterization of primate retinal ganglion cell types will enable the advancement of retinal prosthetics and treatment of visual disorders. The features extracted by low-density retinal ganglion cell types will be investigated using large-scale multi- electrode recording because with this technique it is possible to obtain information simultaneously about hundreds of neurons. The cell types? responses to motion stimuli, natural scenes, and different light levels will be tested. Then, a novel alignment and matching method will be developed for combining multi-electrode recording with anatomical measurements. This is critical for determining the projections of the low-density cell types in the brain, but currently obtaining both physiological and anatomical measurements of the same cells is only possible with experiments on one cell at a time. The functional information will be connected with the anatomical literature by imaging the morphology of the recorded low-density cells. A microscope will be used to localize the recorded cells, and then fluorescent dye will be injected into the cell bodies in order to image the morphology. These data will draw a connection between a retinal ganglion cell type?s functional properties and where it projects in the visual system and build a fundamental understanding of the output signal of the primate retina.
The goal of this project is to understand the computations that occur in the retina by the many types of retinal ganglion cells, which send information to many targets in the brain. Using large-scale multi-electrode recording in combination with cell filling, many cell types can be simultaneously recorded and anatomically characterized. Determining the visual features extracted by different primate ganglion cell types will advance the development of retinal prosthetics designed to mimic natural patterns of retinal activity, with the goal of restoring high- resolution vision to blind patients.
