In this work we investigate and compare a number of time series of stabilograms of healthy subjects and Parkinsonians. This is carried out by means of the chaos paradigm through the preliminary computation of the first minimum of the mutual information function and the embedding dimension (using false nearest neighbours) in order to obtain the correlation dimension as well as the largest Lyapunov exponent. We show that the postural act is indeed chaotic and especially that the latter two parameters do not allow to discriminate healthy subjects from parkinsonians. Moreover we report a discrepancy of our values with those found in previous works. (c) 2004 Elsevier Ltd. All rights reserved.
