A user-interactive graphics-based computer system for analysis of the EEG

Computer technology provides a powerful high-resolution tool for the analysis and understanding of the electroencephalogram (EEG). A system of user-interactive graphics-based computer programs analyzing the power spectral density (PSD), coherence, and phase-angle relationships of the EEG is described, along with the mathematical algorithms used. The analytic variables inherent in coherence analysis are reviewed. These variables include choices of sampling parameters, auto- and cross-spectral algorithms, smoothing techniques, and use of the γ²(f) or γ(f) value for coherence. These variables in the analysis itself have a net interactive effect on the calculated results that often makes comparability of data across studies impossible. Thus, it is proposed that studies include a graph showing the authors’ derived coherence values for a set of standard test signals to enhance scientific communication.