IS WALKING A RANDOM-WALK - EVIDENCE FOR LONG-RANGE CORRELATIONS IN STRIDE INTERVAL OF HUMAN GAIT

Complex fluctuations of unknown origin appear in the normal gait patte rn. These fluctuations might be described as being 1) uncorrelated whi te noise, 2) short-range correlations, or 3) long-range correlations w ith power-law scaling. To test these possibilities, the stride interva l of 10 healthy young men was measured as they walked for 9 min at the ir usual rate. From these time series, we calculated scaling indexes b y using a modified random walk analysis and power spectral analysis. B oth indexes indicated the presence of long-range self-similar correlat ions extending over hundreds of steps; the stride interval at any time depended on the stride interval at remote previous times, and this de pendence decayed in a scale-free (fractallike) power-law fashion. Thes e scaling indexes were significantly different from those obtained aft er random shuffling of the original time series, indicating the import ance of the sequential ordering of the stride interval. We demonstrate that conventional models of gait generation fail to reproduce the obs erved scaling behavior and introduce a new type of central pattern gen erator model that successfully accounts for the experimentally observe d long-range correlations.