METHODS cont.

Subject Information

  • 23 subjects ages 18 to 62 (10 M, 13 F, height 148 to 184 cm, mass 46.5 to 90.7 kg) participated in this study
  • Participants had no history of neuromuscular or musculoskeletal impairment, all had a BMI less than 30
  • Institutional Review Board approval was received, and subjects provided informed consent

Constant Speed Baseline Trials

  • Subjects walked at five speeds adjusted for leg length at one minute each with rests in between bouts
  • The order of the speeds were randomized

Speed Transition Trials

  • In one speed transition trial, subjects walked on the treadmill as it continuously transitioned between five speeds until every possible transition combination (20 total transitions) had occurred
  • Subjects were notified what the post-transition speed would be, marking the beginning of a transition (step 0)
  • Three steps later (step 3), the treadmill accelerated to the post-transition speed during single support, at a constant acceleration of 2 m/s
  • Subjects walked at that speed for another 14 steps, concluding one transition (step 17) and resetting the step counter to 0

Transition Metrics

  • Steady-state behavior pre- and post-transition was quantified  by taking the mean and standard deviation of each parameter over the five steps prior to the notification (steps 13-17)
  • These steady-state means and standard deviations were then used to calculate the following metrics:
Converged Difference is calculated by subtracting the constant speed baseline mean from the steady-state mean for each parameter and each transition. This metric is used to determine if subjects converged to their baseline values after changing speeds.
Divergence means that a change in a given parameter actually occurred with a change in prescribed treadmill speed. An independent samples t-test (a = 0.05) compared the pre- and post-transition five step steady-state regions to determine if the parameter means were statistically different from each other.
Convergence was quantified using a moving window: When a given parameter value for three consecutive steps fell within two post-transition steady-state standard deviations of the steady-state mean, it was converged. Steps to converge (N) subtracted the first step that exceeded two standard deviations of the pre-transition steady-state mean from the step on which convergence occurred.

Transition Magnitude and Direction

  • Transition magnitude describes the size of the change in speed for a given transition (i.e. absolute value of the difference between the final and initial transition speeds)
  • Transition direction corresponds to whether the speed increases or decreases for a given transition
  • For categorical analysis, transitions are grouped by small, medium, and large magnitude

Categorical Analysis

  • In addition to the mixed effect models, Wilcoxon rank sum tests were used to compare the number of steps to converge across parameters within a given magnitude category
  • Two-proportion z tests were also used to compare the modes of convergence, or qualitative behavior between divergence and convergence
  • Possible modes of convergence include overshoot, direct convergence (parameter converges in one step), and indirect convergence (parameter requires more than one step to converge)