Fig. 3

Gait metrics from analysis of Stage B video data. Participants in stage B each undertook a 1 h study session consisting of 4 activity circuits, with one of the following interventions applied during each circuit: no device (ND); device worn, no cueing (NC); device worn, responsive (FOG-initiated) rhythmic vibration cueing (RC); device worn, continuous rhythmic vibration cueing (CC). Video data were analyzed by 3 observers blinded to intervention (except ND) using GaitAnalyst software and a mean value generated for each metric during each circuit. a Mean step frequency (steps/min). b Total time to complete the circuit (seconds). c Step symmetry index (0 = perfect symmetry). d Number of freeze events detected per circuit. e cumulative freeze time per circuit (seconds). f Cumulative freeze time as a percentage of circuit completion time. Individual data points shown represent the mean response for an individual participant (n = 8). A Kolmogorov–Smirnov test was run to test parametric distribution. Data in (a) were normally distributed and analyzed using paired t-tests. Data in (b, c-f) were not normally distributed and were analyzed using Wilcoxon matched-pairs signed rank tests. Significant differences indicated by asterisks: P < 0.01**