Date of Award

Summer 1-1-2012

Document Type


Degree Name

Doctor of Philosophy (PhD)


Integrative Physiology

First Advisor

Rodger Kram

Second Advisor

Roger M. Enoka

Third Advisor

Alaa A. Ahmed


The first three chapters of my dissertation focuses on human running emphasizing step width and arm swing as primary balance control strategies. The fourth chapter tests the hypothesis that arm swing during human walking is primarily passive. In my first study, I found that when subjects ran (3.0 m/s) at step widths other than their preferred narrow step width or without arm swing, both net metabolic power demand and step width variability (indicator of lateral balance) increased. I interpret greater step width variability as a decrease in lateral balance. My findings suggest that humans prefer to run with a narrow step width and swing their arms so as to minimize energetic cost and improve lateral balance. In my second study, I found that when running (3.0 m/s) with or without arm swing, external lateral stabilization (LS) results in similar reductions in net metabolic power (~2.0%) and step width variability (~12.0%). I infer that the 2% reduction in the net energetic cost of running with external LS reflects the energetic cost of maintaining lateral balance. Furthermore, while eliminating arm swing increased the energetic cost of running overall (~8%), arm swing does not appear to assist with lateral balance. In my third study, I found that compared to non-amputees, sprinters with trans-tibial amputations ran with greater step width and medio-lateral (M-L) foot placement variability, indicating that they have greater challenges with maintaining lateral balance. At faster running speeds up to maximum sprint speed, variability of both step width and M-L foot placement increased in all sprinters, indicating progressive decreases in lateral balance. In my fourth study, I quantified arm swing amplitudes and shoulder muscle activity while subjects walked with 1) their biological arms and 2) with free-swinging, anthropomorphic, passive mechanical arms. I found that passive mechanical arm swing resembled the behavior of a horizontally driven pendulum, reaching its largest amplitude as step frequency approached the arm's natural frequency; however, the swinging amplitudes of the passive mechanical arms was much less than the swinging amplitudes of the biological arms. My findings demonstrate that arm swing during human walking is a hybrid system comprising active muscular actuation and passive pendulum dynamics.