Nonlinear and adaptive properties of visual information processing in the fly nervous system

摘要

A simulation study of a neural network model for visual processing in the fly nervous system is presented. We refined the model of on-off units proposed by Ogmen and Gagne (1990) and studied adaptive properties of these units. The response of a neural network model for motion detection built on these on-off units is also obtained for drifting or flickering sinusoidal gratings. Predictions of the models are compared with electrophysiological data.