TONOTOPIC REPRESENTATION OF AUDITORY RESPONSES USING SELF-ORGANIZING MAPS

A self-organizing neural network model called CORMAP has been develope d to rep resent the tonotopic organization of the auditory cortex of a n FM bat. The map self-organizes purely based on external input stimul us parameters. The weights in the competitive layer self-organizes int o a topological map of the input space. This process in turn allows th e representation of the similarity of inputs into the proximity of exc ited neurons. An additional output layer interconnected with the self- organized layer in the CORMAP model provides a mechanism to observe th e functional characteristics of individual neurons. The CORMAP model i s biologically motivated and the functional characteristics of the neu ron used in the self-organizing layer were obtained directly from neur ophysiological experiments. CORMAP demonstrates a plausible model for representing cortical activities in response to external stimulus inpu ts.