Decoding the Position of a Visual Stimulus from the Cortical Waves of Turtles

The main purpose of this paper is to show how a population of excitatory and inhibitory interconnection of cells, dynamically interacting to produce a wave of activity can encode the input signal that generates the wave. The population considered are from the visual cortex of freshwater turtles and the input signals are visual signals through the retina of the animal. The wave of activity in the cell population can be converted to a spatiotemporal signal through low pass filtering, which in turn can be encoded both spatially and temporally using principal component analysis. The visual inputs are thus represented as a ‘low dimensional temporal strands' in an appropriate space described. In this paper, the visual input considered are ‘light flashes' localized at three different locations of the visual space. Detection of the input location using ‘statistical hypothesis testing' has been described as the main result of this paper