Abstract
Theoretically, one can recover instantaneous direction of 3D translation (heading) from a single 2D retinal velocity field created by the instantaneous motion of points in a rigid 3D environment. However, recovering the trajectory through the environment (path) requires more (e.g. motion over time of environmental points, displacement with respect to recognizable landmarks, or eye-rotation information). Confusion between heading and path has made interpretation of many earlier studies difficult (see Royden,1994). Although Stone & Perrone (1997), careful to avoid this problem, showed that humans can estimate heading in the absence of landmarks, static depth cues, or eye-movement information, they did not resolve whether heading was estimated directly from the velocity field or indirectly by first estimating path and then inferring its tangent. To isolate heading from path, we used dynamic random-dot motion stimuli in which environmental points are periodically redrawn. The display simulated an observer travelling on a circular path (yaw rate: 5–20°/s) through a random-dot 3D cloud (depth range: 6–50 m) at three translation speeds (7.5, 10, & 15 m/s) under two conditions: “static scene” in which dots were displayed until they left the field of view and “dynamic scene” in which dot lifetime was limited to 100 ms to match the integration time of human motion processing. Observers received feedback during initial practice only with static scenes. On each trial, observers used a joystick to rotate their line of sight until deemed aligned with true heading (method of adjustment). For 5 observers (4 naïve), performance was similar in both conditions (mean heading bias ± mean heading uncertainty across observers: 1.9° ± 4.1° and 0.1° ± 4.3°, for static and dynamic scenes, respectively). Humans can accurately adjust their heading with no visual path information (from the velocity field and an a priori assumption of a curved path), showing that heading is available for active steering.
Supported by: NASA's Aviation Operation Systems (711-80-03) & Biomedical Research & Countermeasures (111-10-10) programs