Abstract
Despite some neurophysiological findings concluding that visual features such as color and motion are processed in separate regions of the brain, all visual features appear to be bound together in an object. At what level of process does binding occur and what could serve as the linkage for the binding? One possibility is that visual features are bound together based on each feature's depth information. We examined this possibility using the paradigm employed by Wu, Kanai and Shimojo (2004). Basically, two sheets of random dots were manipulated so as to move in opposite directions. The central dots were red and the peripheral dots were green on the upward moving sheet, but were reversed on the downward moving sheet. The observers reported on whether the majority of the peripheral red dots were moving up or down while gazing at the central part of the stimulus. The proportion of peripheral red dots moving upwards and the direction of the central red dots were manipulated. When the two sheets were not separated by depth, the observers' responses were modulated by the direction of the central red dots and misbinding of color and motion occurred. When they were separated, however, the responses were modulated only by the physical characteristics of the motion of the peripheral red dots and misbinding did not occur. This implies that color and motion are bound together based on each feature's depth information and that binding occurs after binocular disparity is processed.
Supported by the MIC, Korea, under the ITRC support program supervised by the IITA (IITA-2005-C1090-0502-0038)