Gait speed influences aftereffect size following locomotor adaptation, but only in certain environments

Movements learned in one set of conditions may not generalize to other conditions. For example, practicing walking on a split-belt treadmill subsequently changes coordination between the legs during normal ("tied-belt") treadmill walking; however, there is limited generalization of these aftereffects to natural walking over the ground. We hypothesized that generalization of split-belt treadmill adaptation to over-ground walking would be improved by maintaining consistency in other task variables, specifically gait speed. This hypothesis was based on our previous finding that treadmill aftereffect size was sensitive to gait speed: Aftereffects were largest when tested on tied-belts running at the same speed as the slower belt during split-belt adaptation. In the present study, healthy adults were assigned to a "slow" or "fast" over-ground walking group. Both groups adapted to split-belts (0.7:1.4 m/s), and treadmill aftereffects were tested on tied-belts at the slow (0.7 m/s) and fast (1.4 m/s) speeds. All participants were subsequently transferred to the over-ground environment. The slow and fast groups walked over-ground at 0.7 and 1.4 m/s, respectively. As in previous work, we found that the size of aftereffects during treadmill walking was speed-dependent, with larger aftereffects occurring at 0.7 m/s compared with 1.4 m/s. However, over-ground walking aftereffects were less sensitive to changes in gait speed. We also found that aftereffects in spatial coordination generalized more to over-ground walking than aftereffects in temporal coordination across all speeds of walking. This suggests that different factors influence aftereffect size in different walking environments and for different measures of coordination.