A hybrid system for computer analysis of EEG data



Journal Title

Journal ISSN

Volume Title



In the past few years the field of neuropsychology has begun to make use of some mathematical methods of wave analysis developed for use in information theory and signal processing systems. The present study relates these information theory concepts to regression analysis, expands upon the concepts, discusses a hardware computer system for the realization of the mathematical methods and presents a few examples of some data analysis. The first chapter discusses some basic time function terms in the framework of statistics as well as engineering, making relations between the two. This done, the Fourier wave analysis method is derived from the least squares regression model. Various innovations and extensions of Fourier analysis are discussed. Chapter two discusses the design of a general purpose hybrid computer. The design philosophy is presented, the hardware capability discussed and programming methods are introduced. Much basic hardware theory is presented. Chapter three is an extension of chapter two in the sense that it covers the basic programs to date. Some mathematical model innovations and interpretations are also discussed here. Chapter four is a discussion of the results of the analysis of some EEG data from the olfactory bulb and amygdala of three cats. The emphasis here is upon the analysis methods and the conclusions which may be drawn from them. Comparisons were made between EEG traces under different odor stimulation conditions. No significant differences across odors were found. The relationships between the two central nervous system structures were also discussed in an information theory framework. Chapter five discusses the implications of this work for future research.



Electroencephalography, Data processing