Synchrony Unbound

In the early stages of visual processing, objects and scenes are represented by neurons with small visual receptive fields. Each neuron provides information about local features of a scene, but to describe a scene in terms of objects requires that these features be combined. Objects can cover wide areas of visual space and be partially occluded by other objects, so the problem of binding the separate representations of parts into coherent wholes is not a simple one. This "binding problem" has received considerable attention. Gestalt psychologists articulated a number of principles for grouping and organizing scene elements (Kö hler 1930; Koffka 1935; Kanisza 1979), and more recently the related problem of image segmentation has received a good deal of attention in computer vision (see, for example, Pal and Pal 1993, for a review). The binding problem is really best considered as a series of related problems, all of which require the combination of information from multiple sources. Information must be integrated across the visual field and combined according to specific attributes. Some of those attributes are pictorial features like line orientation, texture, color, simultaneity of appearance, and common motion, but others require more complex information about such things as 3D shape, lighting, and object surface properties (Gregory 1970; Marr 1982; Shimojo et al. 1989; Adelson 1993; Ullman 1996; Adelson 1999; Kersten 1999).