The investigations described in this application are designed to determine the extent to which well known principles of information processing can be applied to establish the methods used by the central nervous system to efficiently extract information from its environment. The approach in the application employs computational techniques with the expectation that existing physiologic results will guide and test the computational ones. Preliminary results suggest that many of the well known features of the primary visual cortex can be obtained from a simple visual system model which simultaneously optimizes information transfer in the cortex along with the neural costs of doing so. Outputs of the model will be compared with existing multidimensional physiologic descriptions of cells in the visual cortex which have been obtained with nonlinear systems identification techniques. The ensemble properties of the cells in the model cortex will be used to obtain functions which can be compared with established psychophysical results. The results from these studies will be used to guide the development of more complex models of neural information processing, and may be applicable to other regions of the central nervous system. Additionally, this material provides an important quantitative link between biologic and synthetic methods for image representation.
| Gottschalk, Allan; Sexton, Matthew G; Roschke, Guilherme (2004) Multiplicative neural noise can favor an independent components representation of sensory input. Network 15:291-311 |
| Gottschalk, Allan (2002) Derivation of the visual contrast response function by maximizing information rate. Neural Comput 14:527-42 |
