Variability and long range correlations in human walking and running
Open Access
- Author:
- Jordan, Kimberlee
- Graduate Program:
- Kinesiology
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- July 10, 2006
- Committee Members:
- Karl Maxim Newell, Committee Chair/Co-Chair
John Henry Challis, Committee Member
Joseph Paul Cusumano, Committee Member
Stephen Jacob Piazza, Committee Member - Keywords:
- locomotion
detrended fluctuation analysis
variability
adaptability
gait cycle - Abstract:
- Fluctuations in the stride interval of human walking contain long range correlations that decay in a fractal-like manner (Hausdorff et al., 1995). Using Detrended Fluctuation Analysis (DFA), this thesis examines the structure of variability of the gait cycle in human locomotion. Three experiments were carried out to address several issues: 1) are long range correlations present in the fluctuations of the running gait, 2) are long range correlations present in gait variables other than the stride interval, 3) what is the influence of speed of locomotion on the scaling behavior of the gait cycle fluctuations, and, 4) what is the relationship between stability and long range correlations? Experiment 1 examines the fluctuations in a range of kinematic and kinetic gait cycle variables in walking at 60 through to 140% of preferred walking speed, while Experiment 2 investigates gait cycle fluctuations in running from 80 to 120% of preferred running speed. The results reveal the presence of fractal-like scaling behavior in all variables investigated. For many of both the walking and running gait variables, long range correlations follow a U-shaped function with speed and were minimized at preferred speeds of locomotion. Thus, at preferred speeds, there are a larger number of timescales present in the motor out put which is suggestive of greater adaptability. Experiment 3 examines the relationship between local dynamic stability, stability of the movement pattern, and the long range correlations of the gait cycle while walking and running at speeds close to preferred transition speeds. The results suggest that the scaling behavior of gait cycle fluctuations relates to the stability of the gait cycle. Collectively the findings indicate that DFA is revealing about the number of degrees of freedom available under given conditions, or conversely the degree of constraint that results from a set of conditions. An alternative but not mutually exclusive possibility is that the correlations are related to the degree of active control associated with locomotion under different circumstances. Thus, as the speed of locomotion moves increasingly away from preferred speeds, structure is introduced to the variability as a result of: a) increasing constraints, b) decreasing degrees of freedom and c) increasing levels of control.