Failures in adaptive locomotion: trial-and-error exploration to determine adequate foot elevation over obstacles

Lifting the limb sufficiently to clear an obstacle seems like a straightforward task, yet trips are a common cause of falls across all ages. Examination of obstacle contacts in the lab revealed a progressive decrease in foot elevation with repeated exposures, ultimately resulting in failure (Heijnen et al. Exp Brain Res 23:219-231, 2012). The purpose of this study was to determine if the progressive decrease in foot elevation continued when knowledge of obstacle contact was removed. Twenty-one young adults (mean 20.0 ± 1.0 years; 8 males) crossed a 20 cm obstacle in a 12 m walkway for 150 trials. The obstacle was covertly lowered between the lead and trail limb crossing of the obstacle, which eliminated obstacle contact with the trail limb if the limb was too low. The average failure rate was 8%, substantially higher than the 1-2% observed for stationary, visible obstacles. Therefore, tactile information from obstacle contact was instrumental for guiding the trail limb; visual information and joint angle information were insufficient for most participants. Foot elevation change over successive trials varied across participants, and was categorized as (1) asymptotic decrease (N = 11, 52%), with foot elevation converging to obstacle height, (2) linear decrease (N = 7, 33%), and (3) stable (N = 3, 14%). The asymptotic and stable groups appeared to have reasonable knowledge of obstacle height; the linear group did not. The asymptotic behavior is consistent with participants exploring the region above the obstacle through trial-and-error to determine appropriate foot elevation.