Paul A. Watters, University of Newcastle, Callaghan NSW, Australia

Abstract: Recent studies in brain dynamics have utilized a dependent variable calculated from the electroencephalogram (EEG) known as dimensional complexity (DCX), where variables such as scalp locus, cognitive task difficulty, or cortical arousal, are manipulated to test quantitative hypotheses regarding brain-state changes. The technique has been criticised on technical and theoretical grounds, yet its application to experimental time series in many domains has succeeded in yielding information about cortical activity which either complements or surpasses spectral band analysis, and other linear stochastic techniques. The aim of this paper is to provide a pedagogical review of the contribution of dimensional complexity studies in understanding the psychophysiology of cortical arousal by outlining strategies for the successful estimation of DCX as an empirical measure.

Keywords: EEG, chaos, correlation dimension, dimensional complexity