Adaptive Resonance in V1–V2 Interaction
A model for visual cortical boundary detection and contour grouping is proposed that takes into account the structure and functionality of the primate visual system. The architecture relates to visual cortical areas V1 and V2 which are bidirectionally interconnected via feedforward as well as feedback projections. It is suggested that their functionality is primarily determined by the measurement and integration of signal features that are continuously matched against neural codes of expectancies generated on the basis of long-range integration of compatible arrangements of initial measurements. Feedforward signal detection and the generation of feedback expectances is dedicated to different visual layers or areas. Thus, the bidirectional interaction between cortical areas can be understood as an active and continuing mechanism for the prediction and selection of elements in the visual input data stream. The net effect produces contour grouping and illusory contour completion as well as context-sensitive shaping in the tuning of orientation selective cells.
KeywordsBipole Cell Input Activation Illusory Contour Orientation Selectivity Adaptive Resonance Theory
Unable to display preview. Download preview PDF.
- R. von der Heydt. Form analysis in visual cortex. In M.S. Gazzaniga, editor, The Cognitive Neurosciences, chapter 23, pages 365–382. MIT Press (Bradford Book), Cambridge (MA/USA), 1995Google Scholar
- R. von der Heydt and E. Peterhans. Mechanisms of contour perception in monkey visual cortex. I. Lines of pattern discontinuity. The Jorunal of Neuroscience, 9 (5): 1731–1748, 1989Google Scholar
- P.J. Kellman and T.F. Shipley. A theory of visual interpolation in object perception. Cognitive P.ssrcholo,sty. 23 (2): 141–221. 1991Google Scholar
- H. Neumann and P. Mössner. Neural model of cortical dynamics in resonant boundary detection and grouping. In C. von der Malsburg, W. von Seelen, J.C. Vorbrüggen, B. Sendhoff, editors, Lecture Notes in Computer Science 1 112 “Artificial Neural Networks - ICANN 96”, (Proc. Int. Conf. on Artificial Neural Networks, Bochum, Germany, July 16–19. 1996 ) Springer, Berlin, 1996Google Scholar