Title: Physiological Clustering of Visual Channels in the Mouse Retina Authors and Affiliations 3

27 Anatomy predicts that mammalian retinas should have in excess of 12 physiological channels, 28 each encoding a specific aspect of the visual scene. Although several channels have been 29 correlated with morphological cell types, the number of morphological types generally exceeds 30 the known physiological types. Here, we attempted to sort the ganglion cells of the mouse retina 31 purely on a physiological basis. The null hypothesis was that the outputs of the ganglion cells 32 form a continuum, or should be divided into only a few types. We recorded the spiking output of 33 471 retinal ganglion cells on a multi-electrode array while presenting four classes of visual 34 stimuli. Five parameters were chosen to describe each cell’s response characteristics, including: 35 relative amplitude of the ON and OFF responses, response latency, response transience, 36 direction selectivity and the receptive field surround. We compared the results of four clustering 37 routines and judged the results using the relevant validation indices. The optimal partition was 38 the 12-cluster solution of the Fuzzy Gustafson – Kessel algorithm. This classification contained 39 three visual channels that carried predominately OFF responses, six that carried ON responses 40 and three carrying both ON and OFF information. They differed in other parameters as well. 41 Other evidence suggests that the true number of cell types in the mouse retina may be somewhat 42 larger than 12, and a definitive typology will probably require broader stimulus sets and 43 characterization of more response parameters. Nonetheless, the present results do allow us 44 to reject the null hypothesis: it appears that in addition to well known cell types (such as 45 the ON – OFF DS cells) numerous other cell classes can be identified in the mouse retina 46 based solely on their responses to a standard set of simple visual stimuli. 47 48