Item

Abstract

Experimental and computational studies suggest that complex motor behavior is
based on simpler spatio-temporal primitives. This has been demonstrated by
application of dimensionality reduction methods to signals from electrophysiological and EMG recordings during execution of limb movements. However, the existence of
such primitives on the level of the trajectories of complex human full-body movements remains less explored. Known blind source separation techniques, like PCA and ICA, tend to extract relatively large
numbers of sources from such trajectori
es, which are difficult to interpret. For
the analysis of emotional human gait patterns, we present a new non-linear
source separation technique that treats te
mporal delays of signals in a more
efficient manner. The method allows the accurate modeling of high-dimensional
movement trajectories with very few
source components, and is significantly
more accurate than other common techniques. Combining this method with
sparse multivariate regression, we identified primitives for the encoding of emotional gait patterns that match features, which have been shown to be important for the perception of emotional body expressions in psychological studies. This suggests the existence of emotion-specific motor primitives in human gait.