Brightness Decoding in the Fly Visual System: Spike timing and Linear Approximation

The fly visual system is a popular system of study in neuroscience, because of its reduced scale and the robustness of the experimental preparation. These favorable conditions have made it possible to readily identify individual neurons in the fly's visual information processing system (1). One prominent neuron of study is the H1 neuron, a neuron roughly analogous to a neuron in primary visual cortex in humans. This neuron responds preferentially to horizontal motion across the fly's visual field, obeying a tuning curve dependent on the angular velocity, contrast and average intensity of a visual stimulus. Recording from the H1 neuron while manipulating the full-field visual stimuli presented to the fly, for protracted periods of time, provides a means of examining how the neuron, and presumably the fly, encode visual information in spike trains. Because the tuning curve is dependent on a multitude of parameters, it is necessary to judiciously select only a subset of parameters to examine experimentally while holding the rest constant. This set of experiments chooses to focus on how the H1 neuron encodes full-field luminance changes. Although probably only a peripheral function of H1, luminance changes do provide visual clues about the full field horizontal motion in the environment.