Towards the Application of Networks with Synchronized Oscillatory Dynamics in Vision

The functional relevance of synchronized oscillatory responses observed in the visual cortex of various species (Gray and Singer, 1989; Eckhorn et al., 1988; Kreiter and Singer, 1992) is a hotly debated topic in vision. A widespread assumption holds that the synchronized responses code the results of reintegration processes in the visual system, i.e. of image segmentation, figure-ground segregation, feature binding etc. (Malsburg and Buhmann, 1992; Engel et al., 1992). Many numerical simulations showed that oscillations in general can occur and that spatial synchronization can be mediated in physiologically plausible models (see, e.g., Bush and Douglas, 1991). In addition, the models allowed to qualitatively recover the experimentally observed responses when analogousstimuli, i.e. long bars, interrupted bars etc. are applied (Konig and Schillen, 1991; Niebur et al., 1991).