Monocular Motion Adaptation Affects the Perceived Trajectory of Stereomotion Kevin R. Brooks University of Sussex Perceived stereomotion trajectory was measured before and after adaptation to lateral motion in the dominant or nondominant eye to assess the relative contributions of 2 cues: changing disparity and interocular velocity difference. Perceived speed for monocular lateral motion and perceived binocular visual direction (BVD) was also assessed. Unlike stereomotion trajectory perception, the BVD of static targets showed an ocular dominance bias, even without adaptation. Adaptation caused equivalent biases in perceived trajectory and monocular motion speed, without significantly affecting perceived BVD. Predictions from monocular motion data closely match trajectory perception data, unlike those from BVD sources. The results suggest that the interocular velocity differences make a significant contribution to stereomotion trajectory perception. Object motion along the z-axis (i.e., toward or away from the observer) can be sensed either by means of monocular cues (e.g., change of image contrast, rate of retinal image expansion, etc.) or by binocular cues (e.g., stereomotion). This study concerns human perception of motion in depth from binocular cues alone. As pointed out by Rashbass and Westheimer (1961), as an object’s binocular disparity changes, its image translates at different veloc- ities on the two retinas. Beverley and Regan (1973) showed that the ratio of image velocities in the left and right eyes can uniquely specify any 3-D trajectory in the horizontal meridian. For a stim- ulus located near the median plane, the trajectory,  (measured in degrees of clockwise angular deviation from the median plane, as seen from above) is given by   tan 1  I{[(d/dt) R /(d/dt) L ]  1} 2D{[(d/dt) R /(d/dt) L ]  1}  (1) or equivalently by   tan 1  I[(d/dt) R  (d/dt) L ] 2D[(d/dt) R  (d/dt) L ]  , (2) where D is the viewing distance, I is the interpupillary separation, and (d/dt) L and (d/dt) R represent left and right monocular image velocities, respectively (Portfors-Yeomans & Regan, 1996). I refer to the potential cue to an object’s trajectory that this presents as the interocular velocity difference cue. Alternatively, stereomotion trajectory could be sensed by a combination of the rate of change of disparity (d/dt) and the rate of change of binocular visual direction of the fused image (d/dt). Since (d/dt) is mathematically equal to (d/dt) R  (d/dt) L , and (d/dt) is equivalent to [(d/dt) R  (d/dt) L ]/2, trajectory is given by   tan 1  I(d/dt) D(d/dt)  (3) (Portfors-Yeomans & Regan, 1996). In the interests of brevity, I refer to this cue as the changing disparity cue. Though these two binocular correlates are mathematically equivalent and correspond perfectly for natural examples of mo- tion in depth, they lead to the possibility that the visual system could encode the trajectory of objects in 3-D by either one or both of two entirely independent means. In the interocular velocity difference system, monocular motion signals (i.e., [d/dt] L and [d/dt] R ) are derived before binocular combination. In the case of the changing disparity cue, binocular combination precedes the derivation of any motion signal. This study aimed to determine the relative contributions of each of these cues to the perception of stereomotion trajectory. Regan (1993) showed that subjects could discriminate stereo- motion trajectories on the basis of the changing disparity cue alone. Regan’s study used a cyclopean stimulus—the dynamic random dot stereogram—to investigate the discrimination of 3-D trajectories. This stimulus is similar to a conventional random dot stereogram, except that a novel random array of dots is presented in every frame, specifying the appropriate binocular disparity for the central target area. Hence, each monocular image remains camouflaged for its entire duration. Any motion percept can be explained only as the result of positional signals gained through binocular combination, from which a motion percept is derived. Regan concluded that subjects were able to accurately discriminate different stereomotion trajectories, even when the stimulus con- tained no monocular motion signals, and hence no interocular velocity difference cue. However, no data were reported. The experiment reported here investigated the relative impor- tance of interocular velocity difference and changing disparity cues to stereomotion trajectory perception for a standard random dot stereogram target moving through the horopter, purely within the horizontal meridian. By exploiting two separate visual illusions Kevin R. Brooks, Department of Experimental Psychology, University of Sussex, Brighton, United Kingdom. Correspondence concerning this article should be addressed to Kevin R. Brooks, who is now at the Human Information Processing Research Branch, Human Factors Research and Technology Division, National Aeronautics and Space Administration Ames Research Center, M-S 262-2, Moffett Field, California 94035-1000. E-mail: kbrooks@mail.arc.nasa.gov Journal of Experimental Psychology: Copyright 2002 by the American Psychological Association, Inc. Human Perception and Performance 2002, Vol. 28, No. 6, 1470 –1482 0096-1523/02/$5.00 DOI: 10.1037//0096-1523.28.6.1470 1470