Meeting Abstract

P3.200  Sunday, Jan. 6  Evaluating models of locomotion dynamics: What complexity is adequate? HASANEINI, SJ; BERTRAM, JEA*; Univ. of Calgary, Calgary; Univ. of Calgary, Calgary jbertram@ucalgary.ca

One approach to understanding biological legged locomotion is to use theoretical mechanical models as a test bed to evaluate theories of why humans and animals move as they do, or as a means to predict their response to new environments. Simple models are more amenable to interpretation and are computationally fast, while comprehensive models have complexities that can obscure the underlying principles. Minimal analytical and numerical models (e.g. Kuo (J Biomech. Eng., 2002) and Srinivasan (Nature, 2006)) have shown good success in explaining some aspects of human and animal locomotion. However, the capability of the minimal models is limited, and thus greater complexity is necessary for such models to provide insight into the more subtle aspects of locomotion dynamics. The question remains as to what level of complexity is required in order to produce a functionally reliable model. Obviously the answer depends on the specific question at hand and the characteristic that is being investigated. Here we explore the required level of complexity in the human response to walking and running in simulated reduced gravity. Two different models that self-optimize for mechanical energy cost, each with a different level of complexity, are explored. The predicted optimum behavior for these models as gravity changes is compared with observations of human gait in reduced gravity. It is found that the model’s ability to predict human response to an unusual gravitational environment is often counter-intuitive. Through comparison of model and human in an experimentally manipulated physical environment, it is possible to determine the consequences of model simplicity and complexity.