Abstract des Autors

The effects of training on human balance performance with different supra-postural tasks were investigated. We present a method for stochastic data analysis, the socalled Langevin Approach, and discuss the application to measurement data of human balance pedormance. The presented analysis framework is based on a Langevin model of the dynamics of balance control. We show that our approach can give deeper insight into the balance performance with different supra-postural tasks than just considering the overall variability, i.e. the standard deviation, of the data. Namely, we explicitly separate deterministic and stochastic dynamics. By this means we find quantitative differences in the deterministic and in the stochastic dynamics of movement control. Moreover, the impact of training and supra-postural tasks on both aspects is quantitatively derived.