Gait asymmetries are a common problem for clinical populations—such as stroke survivors and people with Parkinson’s disease—and are associated with increased gait instability and fall risk. Current methods to alter gait asymmetries rely heavily on split-belt treadmill training. Gait training using visual cues projected on a screen or in immersive virtual reality have been shown to produce greater improvements in gait asymmetries relative to traditional treadmill training alone. However, it is unclear the extent to which gait asymmetries can be systematically altered using an asymmetric visual cue, which represents a more cost-effective strategy relative to split-belt treadmill training or immersive virtual reality. Investigating whether using a visual asymmetric cue can alter gait symmetry in healthy adults is the first step in determining if this methodology is plausible for future research with clinical populations. The purpose of this dissertation was threefold:​ (1) to examine the extent to which healthy adults can synchronize to an asymmetric visual cue during treadmill walking;​ (2) to explore if the asymmetric walking pattern is retained once the visual cue is removed;​ and (3) to examine transfer of the asymmetric walking pattern to overground walking after the treadmill training session. Seventy-two healthy participants (age 23.89 ± 6.08 years) were enrolled in this study and quasi-randomized into four experimental groups (N = 64) or the control group (N = 8). All participants completed questionnaires related to health history/demographics, limb dominance, and physical activity. All groups completed three 10-minute walking sessions with wearable sensors (APDM Inc., Portland, OR) to record spatiotemporal gait measures. The first session was the same for all groups and consisted of walking at their self-selected speed on the treadmill. For session two, experimental groups 1 and 2 attempted to synchronize their gait to a visual cue (i.e., walking stick figure) exhibiting a small gait asymmetry presented on the projection screen in front of that treadmill, while experimental groups 3 and 4 attempted to synchronize their gait to a visual cue with a large gait asymmetry. For the third session, groups 1 and 3 walked on the treadmill for 10 minutes after the visual cue was removed, while groups 2 and 4 walked for 10 minutes overground after the treadmill training.

The dependent variables were calculated using the Symmetry Index (SI) equation:​ stride length SI %, step duration SI %, and single limb support SI %. Visually inspecting the data showed some participants responded to the visual cue stimulus, while others did not. Therefore, experimental groups were further divided into responders (N = 42) and non-responders (N = 22). Wilcoxon Signed Rank and Mann Whitney U tests were run to determine if the gait asymmetry metrics differed from baseline to adaptation, adaptation to post adaptation and between groups, and Wilcoxon effect sizes were calculated to determine the magnitude of the effect. The results reported in Manuscript I show gait asymmetries increased in the small and large asymmetry responder groups;​ the effect sizes were moderate to large. However, no changes were shown in the small and large asymmetry non-responder groups;​ the effect sizes were small to moderate. The results from Manuscript II show that the small asymmetry responder group has reduced gait asymmetries during retention, yet some gait asymmetry metrics remain elevated in the large gait asymmetry responder group during retention. The gait asymmetry responder groups revealed decreases in almost all gait asymmetry metrics indicating the adopted gait pattern did not transfer to overground walking. No gait asymmetry changes were observed for the non-responder groups for retention or transfer. Collectively, the results suggest that an asymmetric visual cue can be used to alter gait symmetry and retention may be observed as such training. However, further research should investigate why some participants did not respond to the visual cue.