Control of adaptive gait: Effect of experience and light level on action and perception

Christopher K Rhea, Purdue University

Abstract

Visual information is used for perceiving potential environmental threats to stability during locomotion (i.e. uneven terrain) and controlling the action to avoid such threats, but it is unclear if the perception and action processes share a common representation of the visual information. In two studies, the association between the perceived height of an obstacle and the action taken to step over the obstacle during gait was examined. Two obstacle conditions were used in each study; a full obstacle (entire surface of obstacle was visible) and a perimeter obstacle (only perimeter of obstacle was visible). The obstacles were of identical height and width, but the full obstacle provided the illusion of being taller. In both studies, subjects estimated the obstacles’ height five times before (pre-action estimation) and after (post-action estimation) the gait trials, which consisted of crossing an obstacle placed in the middle of an 8 m walkway. The gait trials were repeated 50 times for each obstacle in a blocked format and the obstacle order was randomized between subjects. Two studies were conducted with the same tasks, the first study was conducted in the low light (n = 15) and the second study was conducted in the full light (n = 15). Different subjects participated in each study. The association between perceived obstacle height and toe clearance (vertical distance of the toe from the obstacle when the foot was directly over the obstacle) was examined, as well as other gait variables, such as horizontal distance of foot placement prior to stepping over the obstacle and stride length in the approach phase. In the low light experiment, the full obstacle was estimated as 2.0 cm taller than the perimeter obstacle in pre-action estimation and toe clearance was 2.5 cm higher in gait trials 1-5 (p < 0.01). After 50 obstacle crossings the full obstacle was estimated as 2.6 cm taller (p < 0.01), but toe clearance between obstacles was not different in gait trials 46-50 (p > 0.05). Therefore, an association between perceived height and toe clearance was observed when the gait and perception tasks were novel, but a dissociation was observed with increased experience with the tasks. In the full light experiment, a dissociation between perceived height and toe clearance was observed at the beginning and end of the experiment. The perceptual manipulation in the full light may not have been adequate to influence toe clearance. The possibility of a context-dependent relationship between perception and action is discussed. The context-dependent relationship may be due to education of attention (Gibson, 1966; 1979), where visual attention is directed to visual cues specific to the task due to increased task experience – a term Gibson called attunement. Additionally, perception-action relationships observed in the low light experiment may have resulted from an open-loop visual process early in the task (leading to a perception-action association) and a closed-loop visual process latter in the task (leading to a perception-action dissociation).

Degree

Ph.D.

Advisors

Rietdyk, Purdue University.

Subject Area

Physical education|Experimental psychology

COinS