On-line control and model-based control are two general approaches to visual control of human actions. When intercepting a moving target, humans appear to use the constant bearing strategy (CBS): nulling the change in target bearing direction, consistent with on-line control. In this dissertation, I investigate on-line and model-based control by asking participants to walk to intercept a moving target in virtual environments. In Experiment 1, a target initially moved at 0.6 or 0.8m/s and after 3s randomly changed speed once. I simulated participants’ interception paths using the CBS with different input: current information about target motion or an internal model that averages target speed within a temporal window. The results indicate that current visual information is sufficient to account for locomotor interception, and the internal model impairs simulation performance, consistent with on-line control. In Experiment 2, a target moved at 0.6, 0.8 or 1.0m/s and after 2.5s moved behind a virtual occluder that degraded target visibility to different levels. The results indicate that degrading target visibility progressively impairs interception accuracy and precision, and total occlusion severely impairs performance. When the target was visible or even blurred, participants adaptively adjusted their heading to target speeds. In contrast, they turned a constant angle after the target was totally occluded. In Experiment 3, participants learned a target speed change (as in Exp 1) for 40 trials with the target always visible, then were tested with target totally occluded 2.5s after its appearance. Interception results indicate that target occlusion impairs interception accuracy and precision. Participants learned to anticipate target speed change by adaptively adjusting their heading for the upcoming speed change. After target speed change, participants adaptively adjusted their heading if the target was visible. In contrast, they turned a constant angle if the target was occluded. Overall, the results of the experiments suggest a primary role of current visual information in control of locomotor interception, no evidence for an internal model of target motion, which is consistent with on-line control.
Zhao, Huaiyong,
"Locomotor Interception of a Moving Target: On-line or Model-based Control?"
(2015).
Cognitive Sciences Theses and Dissertations.
Brown Digital Repository. Brown University Library.
https://doi.org/10.7301/Z0G44NP4
