High accuracy decoding of dynamical motion from a large retinal population

Abstract

Motion tracking is a challenge the visual system has to solve by reading out the retinal population. Here we recorded a large population of ganglion cells in a dense patch of salamander and guinea pig retinas while displaying a bar moving diffusively. We show that the bar position can be reconstructed from retinal activity with a precision in the hyperacuity regime using a linear decoder acting on 100+ cells. The classical view would have suggested that the firing rates of the cells form a moving hill of activity tracking the bar's position. Instead, we found that ganglion cells fired sparsely over an area much larger than predicted by their receptive fields, so that the neural image did not track the bar. This highly redundant organization allows for diverse collections of ganglion cells to represent high-accuracy motion information in a form easily read out by downstream neural circuits.

23 pages, 7 figures

Cited by in corpus (10)

• Optimal decoding of information from a genetic network
• Derivatives and Inverse of Cascaded Linear+Nonlinear Neural Models
• Learning to make external sensory stimulus predictions using internal correlations in populations of neurons
• Estimating information in time-varying signals
• Blindfold learning of an accurate neural metric
• Closed-loop estimation of retinal network sensitivity reveals signature of efficient coding
• Towards the Next Generation of Retinal Neuroprosthesis: Visual Computation with Spikes
• Linear response for spiking neuronal networks with unbounded memory
• Reconstruction of Natural Visual Scenes from Neural Spikes with Deep Neural Networks
• Evolutionary aspects of Reservoir Computing