The Rough-Terrain Problem: Accurate Foot Targeting as a Function of Visual Information Regarding Target Location

ABSTRACT

The authors examined step-cycle regulation to accurately land on a single target. They also examined the effect of decreasing and increasing visual information regarding target location. Visual information was decreased with goggles that obstructed the lower visual field, removing information of the target and foot relative to target in the two steps before the target. Visual information was increased by adding 4 vertical poles (2.1 m tall) around the target location. A total of 14 participants landed with 1 foot on a flat target placed halfway down a walkway and continued walking. During target approach, step length variability increased and foot placement variability decreased. The final stride onto the target was longer and smoother than the penultimate stride, which may reflect that modifications were made earlier to reduce modifications needed in the final stride. Foot-target accuracy was reduced by lower visual-field obstruction. In the steps preceding the target, the presence of poles describing target location modified foot-placement variability, stride length, swing-trajectory smoothness, and head angle. However, foot-target accuracy was not modified by increased visual information regarding target location. That is, the presence of poles modified how the task was performed but did not alter the outcome. Therefore, view of the foot relative to the target is more important than view of target location in the control of a foot-targeting task.

Keywords:

• adaptive gait
• locomotion
• target
• vision

Notes

1. Note that the statistics were completed with and without this participant; the only difference in significance was observed for foot-placement bias: effect of visual obstruction. The large overshoot of the participant changed the visual-obstruction effect from insignificant (N = 14), F(1, 784) = 0.40, p = .530, to significant (N = 15), F(1, 840) = 8.07, p = .005.

2. Although the main focus of the present research was not on the effect of gender, we compared men's and women's accuracy results with a single-factor analysis of variance (ANOVA). We found no significant effects on absolute accuracy, F(1, 894) = 2.45, p = .12, or variable error, F(1, 110) = 0.01, p = .91. Therefore, we did not further investigate the effect of gender.