Effects of travel speed and attentional load on visual control of steering toward a goal

Abstract

Human use both optic flow and target egocentric direction cues to guide selfmotion at walking speed. The first study examined whether people change their reliance on the two cues when they are controlling high-speed steering, e.g. driving a car? In Experiment 1, I simulated two viewing conditions. In the opticflow only condition, the participant's virtual gaze direction was fixed on the target which was placed 10° away from their straight ahead. The target egocentric direction was fixed during steering and was thus unavailable for steering control. Participants steered to align their heading but not their path of forward travel with the target at all travel speeds tested. In the optic flow-plus-egocentric direction condition, the participant's virtual gaze direction was aligned with heading which was displaced 10° away from their straight ahead. The target egocentric direction changed during steering and was thus available for steering control. Participants’ steering was affected by the heading displacement at all travel speeds tested. The faster and larger reduction of heading error observed at higher travel speed for both viewing conditions was mirrored by an increase in the precision of heading judgment in the corresponding heading perception experiment (Experiment 2). The findings support that while people are increasingly more accurate and efficient in using heading specified by optic flow for steering when travel speed increases, high-speed travel does not affect the type of visual strategy used for the control of steering toward a goal. The second study examined how different attentional loads affected people’s steering toward a goal at both low and high travel speeds. In Experiment 3, I used the same display setting as the optic flow-plus-egocentric direction condition in Experiment 1 for the steering task. Participants were asked to steer toward a red post target with (1) no attention tracking task, (2) concurrently tracking one dot (low attentional load), or (3) three dots (high attentional load) among eight dots that randomly moved inside the red circle on top of the target post. I found that attentional load affected the early stage of steering control but did not affect the overall visual strategy. Experiment 4 replicated Experiment 3 except that participants were specifically instructed to center the target straight ahead or align the target with heading for the steering task. I found that attentional load only affected the early stage of steering for the centering-the-target instruction group, but affected the steering performance throughout the trial for the aligning the-target-with-heading instruction group. In both Experiments 3 and 4, while the tracking accuracy was high (> 90%) and not affected by the travel speed when the attentional load was low, it decreased more at higher travel speed when the attentional load was high. The findings suggest that people have more difficulty in dealing with high attention demanding task at high than low travel speed. Increasing attentional load affects the accuracy in using heading specified by optic flow but does not change the natural visual strategy for goal-directed selfmotion control.