The estimation of distance, direction and self-position through nonvisual locomotion

Abstract

Five experiments examined the accuracy with which individuals could utilize visual and non-visual sensory information to either return to, point towards or verbally estimate the distance and direction to previously learned targets. Gender differences in spatial performance were also explicitly evaluated. All five experiments indicated that prior visual information about a target's location facilitated accurate performance and showed that gender differences exist for perceptual spatial tasks but disappear when performance is evaluated. Non-visual performance was more variable but the errors generated in experiments I, 2 and 3 revealed that both the direction and the distance of a target might be determined with some accuracy using non-visual information. Whether participants pointed or walked to a named target, experiment 4 showed that prior vision enabled them to maintain the integrity of their representations of space after they were led without vision to a novel location. Walking heading was more accurate than pointing heading for participants in the non-vision group of this experiment. We postulated that accurate pointing was related to the intimate link between the eye-head arm system and might be affected by manipulating retinal and extra-retinal cues. Altering the congruence of retinal and extra-retinal signals in experiment 5 did not significantly influence pointing or walking performance. Overall, performance accuracy appeared to depend on whether the response output used was pointing, walking, map drawing or a verbal response. The results suggest that there is not one central representation of space that each response system may tap into but that each response output likely employs separate representations of space that are governed by different rules.