The influence of locomotion on the axis-aligned motion bias in large situated display environments

Abstract

Work by Morikawa (1999) and Dolgov, et al. (2005) confirmed that stationary observers reliably misjudge projected destination of axis-misaligned figures in the direction the axis is tilted, a phenomenon we named “Axis-Aligned Motion” (AAM) bias. The current study evaluates whether or not the AAM bias is diminished if observers can respond with a navigational action versus stationary pointing. Participants observed moving symmetric and asymmetric figures in a large, overhead-projection, situated-display environment. The task was to judge projected destination of figures that move along the floor and then disappear midway across a colored 3 meter diameter circle. Eight participants completed 24 trials in each of the 3 experimental conditions: 1) stationary observers estimating projected destination of shapes using a laser pointer; 2) mobile observers that begin walking to the projected destination after the shape finishes moving; and 3) mobile observers that begin walking as soon as the shape begins to move. Replicating previous research, stationary observers reliably misjudge projected destination of axis-misaligned moving figures in the direction the axis tilts. Across all conditions, final destinations of axis-aligned symmetric shapes were judged significantly more accurately than asymmetric and axis-misaligned symmetric shapes (p[[lt]]0.05). The impact of locomotive action on the AAM bias was supported by the absolute magnitude of error being significantly greater (p[[lt]]0.05) in the stationary versus mobile conditions, which did not significantly differ from one another. The findings confirm the existence of the AAM bias in a large situated-display environment and demonstrate that its influence on motion perception can be mitigated by allowing observers to locomote to the estimated destination. They support that AAM is a natural regularity for which observers have developed a bias, and that this bias is diminished for navigational action tasks such interception, consistent with previous research demonstrating differences in perceptual phenomena for stationary versus active tasks.