Abstract:

Traditionally parameters of amplitude-frequency characteristics of recorded biopotentials are used in EEG. In this paper we show that they are still chaotic, which takes place for the autocorrelation function A(t). It is proposed the two new approaches for estimation of parameters of electroencephalograms. The first is based on the construction of the matrix of pairwise comparisons of recorded samples of brain potentials in subjects in different physiological (mental) states. The second method is based on calculating the parameters of quasi-attractors, which are built on the plane in the coordinates x1=U(t) – a function that changes due to registration of biopotential in current point and x2=dx 1/dt - rate of change of x1. It is possible to construct a three-compartmental model quasi-attractor in the phase space of states, which is also discussed.

Keywords: chaos self-organization theory, quasi-attractors, elektroentsefalo-grams.