Synoptic Connections, Receptive Fields, and Patterns of Activity in the Tiger Salamander Retina A Simulation of Patterns of Activity Formed at Each Cellular Level From Photoreceptors to Ganglion Cells [The Friedenwald Lecture]

The function of the retina is to transform the pattern of the visual world impinging on the photoreceptors into patterns of neural activity at the ganglion cells. These patterns are then further processed by higher centers for perception. The retinal transformations appear to take place in a series of stages: the initial neural pattern at the photoreceptors is transformed by interactions at the outer retina into patterns of activity at the bipolar cells. Bipolar patterns, in turn, are transformed by interactions at the inner retina into more complex patterns at the ganglion cells. These patterns of activity are dynamic: retinal cells are most sensitive to changes in space and time. Dynamic patterns have never been measured or, as far as I can tell, even suggested. Viewing the patterns can provide some valuable insights into retinal signal processing. This report is an attempt to predict and generate the dynamic patterns of cellular activity within each retinal cell mosaic. Most physiologic studies record the electrical effects of convergent inputs to single retinal cells, a tradition established more than 5 decades ago by Hartline and further developed by Kuffler and Barlow in the cat and frog. They characterized the socalled "receptive fields" by measuring ganglion cell activity elicited by simple stimulus patterns. In an alternative single-cell approach, Maturana et al showed that retinal ganglion cells in the frog responded exclusively to a limited number of more complex stimulus patterns representing important