Two experiments were conducted to investigate how color and stereoscopic depth information are used to segregate objects for visual search in three-dimensional (3-D) visual space, Eight observers were asked to indicate the alphanumeric category (letter or digit) of the target which had its unique color and unique depth plane, In Experiment 1, distractors sharing a common depth plane or a common color appeared in spatial contiguity in the xy plane. The results suggest that visual search for the target involves examination of kernels formed by homogeneous items sharing the same color and depth. In Experiment 2, the xy contiguity of distractors sharing a common color or a common depth plane was varied. The results showed that when target-distractor distinction becomes more difficult on one dimension, the other dimension becomes more important in performing visual search, as indicated by a larger effect on search time. This suggests that observers can make optimal use of the information available. Finally, color had a larger effect on search time than did stereoscopic depth. Overall, the results support models of visual processing which maintain that perceptual segregation and selective attention are determined by similarity among objects in 3-D visual space on both spatial and nonspatial stimulus dimensions.An important function performed by the visual system in processing visual information is to select information from the visual scene which is relevant to the current behavior. To achieve this, the visual system has to segregate stimuli in visual space so that parts belonging to the same whole can be linked together and boundaries can be drawn between different wholes. It has been shown that the visual system uses information such as spatial location (see, e.g., Eriksen & Eriksen, 1974;LaBerge & Brown, 1986), perceptual organization (e.g., Banks & Prinzmetal, 1976;Baylis & Driver, 1992;Driver & Baylis, 1989), and feature similarity (e.g., Duncan & Humphreys, 1989;Kramer, Tham, & Yeh, 1991) in order to achieve segregation. The purpose of the present study is to extend previous work on segregation in two-dimensional (2-D) space and investigate how observers use stereoscopic depth and color information to segregate objects and search for a target in three-dimensional (3-D) visual space.Humphreys (1981) summarized two models of how multidimensional I stimuli are processed. The first is the independent dimensional processing model, which assumes that some dimensions can be analyzed indepen-The authors thank Arthur F. Kramer, George 1. Andersen, Marc Carter, and an anonymous reviewer for their invaluable comments on an earlier draft of this manuscript. Correspondence concerning this article should be sent to A. W. Chau, Department of Psychology, University of Hong Kong, Pokfulam Road, Hong Kong (e-mail: awlchau@ hkucc.hku.hk). dently and in parallel. A stimulus is processed on the basis of the first relevant dimension that is resolved. Garner (1974) labeled these dimensions "separable," and it is easy to limit process...