Long-range lateral connections in the primary visual cortex (V1) are known to link neurons with similar orientation preferences, but it is not yet known how color-selective cells are connected. Using a self-organizing model of V1 with natural color image input, we show that realistic color-selective receptive fields, color maps, and orientation maps develop. Connections between orientation-selective cells match previous experimental results, and the model predicts that color-selective cells will primarily connect to other cells with similar chromatic preferences. These findings suggest that a single self-organizing system may underlie the development of orientation selectivity, color selectivity, and lateral connectivity.

PDF

Neurocomputing, Vol. 65--66 (2005), pp. 69-76.

@Article{bednar:neurocomputing05, title={Self-organization of color opponent receptive fields and laterally connected orientation maps}, author={James A. Bednar and Judah B. De Paula and Risto Miikkulainen}, volume={65--66}, journal={Neurocomputing}, pages={69-76}, url="http://www.cs.utexas.edu/users/ai-lab?neurocomputing05", year={2005} }