Application of a neural complexity measure to multichannel EEG

We apply a recently suggested measure for neural complexity (G. Tononi, O. Sporns, G.M. Edelman, Proc. Natl. Acad. Sci. 91 (1994) 5033), that is hypothesised to capture the interplay between two fundamental aspects of brain organisation, functional segregation and integration, to human EEG recordings. This measure is based on a weighted sum of entropy differences evaluated at different length scales of the system. A strong prediction is that this measure correlates with the conscious state of the subject, having lower values if consciousness is reduced (G. Tononi, G.M. Edelman, Science 282 (1998) 1846). It is found, however, that this neural complexity measure increases in neurological disorders where consciousness is severely reduced or absent. We discuss several possible explanations for this observation and suggest directions for future work.