Biofeedback training effects on minimum toe clearance variability during treadmill walking

A number of variability analysis techniques, including Poincaré plots and detrended fluctuation analysis (DFA) were used to investigate minimum toe clearance (MTC) control during walking. Ten young adults walked on a treadmill for 10 min at preferred speed in three conditions: (i) no-intervention baseline, (ii) with biofeedback of MTC within a target range, and (iii) no-biofeedback retention. Mean, median, standard deviation (SD), and inter quartile range of MTC during biofeedback (45.57 ± 11.65, 44.98 ± 11.57, 7.08 ± 2.61, 8.58 ± 2.77 mm, respectively) and retention (56.95 ± 20.31, 56.69 ± 20.94, 10.68 ± 5.41, 15.38 ± 10.19 mm) were significantly greater than baseline (30.77 ± 9.49, 30.51 ± 9.49, 3.04 ± 0.77, 3.66 ± 0.91 mm). Relative to baseline, skewness was reduced in biofeedback and retention but only significantly for retention (0.88 ± 0.51, 0.63 ± 0.55, and 0.40 ± 0.40, respectively). Baseline Poincaré measures (SD1 = 0.25, SD2 = 0.34) and DFA (α1 = 0.72 and α2 = 0.64) were lower than biofeedback (SD1 = 0.58, SD2 = 0.83, DFA α1 = 0.76 and α2 = 0.92) with significantly greater variability in retention compared to biofeedback only in the long-term SD2 and α2 analyses. Increased DFA longer-term correlations α2 in retention confirm that a novel gait pattern was acquired with a longer-term variability structure. Short- and long-term variability analyses were both useful in quantifying gait adaptations with biofeedback. The findings provide evidence that MTC can be modified with feedback, suggesting future applications in gait training procedures for impaired populations designed to reduce tripping risk.