Perception-Action Coupling During Goal-Directed Locomotion Without Vision.

The ability to walk to distant targets accurately without the use of vision has been demonstrated in a number of studies (e.g., Elliott, 1987; 1990; Loomis, Da Silva, Philbeck, & Fukusima, 1996; Rieser, Ashmead, Talor, & Youngquist, 1990; Steenhuis & Goodale 1988; Thomson, 1983). Processes underlying the mechanisms involved in this ability however, are not completely understood. One possible mechanism examined involves non-visual sensory flows during locomotion. In Experiment 1, manipulations to normal walking led to significant increases in error relative to the control group. Error expressed as a percentage of target distance revealed that target distance did not affect accuracy during normal walks. However, altering the method of locomotion led to significant differences in AE and VE. In Experiment 2, the influence of vestibular information was further examined by alterations in head orientation relative to the direction of travel. CE scores indicated that participants veered off course in the direction of head turning when attempting to walk a straight path to the target. This suggests a non-conscious effort to regain normal inertial flows that occur when the vestibular apparatus is not in line with the direction of travel. Non-visual sensory flows were also re-examined by comparing sidestepping to normal walks. Results indicated that participants accurately estimated distance employing either gait but that veering errors during sidestepping led to decreases in overall accuracy. Experiment 3 investigated whether certain deleterious combinations of gait and head positions noted Experiment 2 could ever be beneficial when walking a curved path. Results indicated that when required to travel along a curved path, sidestepping was significantly better than normal walks when estimating distance to the target. The findings from Experiments 2 and 3 support Rieser et al.\u27s (1990) notion that implicitly learned correlations among vision and non-visual sources of information allow for accurate performance when vision is removed. However, the findings from Experiment 3 are in contrast with this theory and suggest that externally imposed task constraints must be considered when examining the ability to perform accurate goal-directed walks without vision