The dimensionality of the human visual evoked scalp potential

A small number of processes can account for most of the evoked potentials activity in the two subjects studied. Principal components analysis indicates that six independent processes can account for approximately 97% of the variability in the data. Moreover, the factor analysis and plots of the factor coefficients yield indications that the times during which these principal factors are active agree quite well with the times at which the equipotential maps show some organized activity. The question of dipoles being the underlying cause of the observed activity is not answered by the factor analysis. The principal factors are not unique, but models which have a small number of parameters are more justifiable in light of the results of this study.