Model-guided discovery of a retinal chromatic feature detector that signals visual context changes

The retina transforms patterns of light into visual feature representations supporting behaviour. In the mouse, these representations are distributed across more than 30 types of retinal ganglion cells (RGCs), whose spatial and temporal tuning properties have been studied extensively. However, a systematic approach for linking the often nonlinear retinal transformations of natural visual inputs to specific ethological purposes is lacking.
Here, we trained a convolutional neural network model (CNN) on large-scale functional two-photon recordings of RGC responses to natural mouse movies, and then used this model to search in-silico for stimuli that maximally excite distinct types of RGCs.
This procedure predicted centre colour-opponency in transient Suppressed-by-Contrast RGCs (tSbC), a cell type whose function is being debated. We confirmed experimentally that these cells indeed responded very selectively to Green-OFF, UV-ON contrasts, which we found to be characteristic of transitions in the visual scene from ground to sky, as might be elicited by head- or eye-movements across the horizon. Indeed, we found that tSbC reliably detected these transitions. Based on these findings, we suggest a role for tSbC RGCs in providing contextual information (i.e., sky or ground) necessary for the selection of appropriate behavioural responses to other stimuli, such as looming objects.
Beyond these specific results, our approach sketches a fast-forward path towards a better understanding of the feature channels encoded by the retina by creating a continuous link from natural stimulus to feature selectivity to ethological relevance.