Nonlinear Analysis of EEG in Various Mental States of Normal Person

Chae, Seung Hee; Oh, Jee Young; Joo, Eun Youn; Park, Eun Mi; Choi, Kyoung Gyu; Kim, Eung Soo

Abstract

BACKGROUND: In order to compare the differences between various mental activities and to determine the significance of chaotic patterns, we performed a fractal dimension and Lyapunov exponent analysis of digital EEG data which are known to be non linear biological signals.
METHODS
During the EEG recordings, different kinds of tasks were performed, including eye closing, eye opening, calculation, listening to music, and remembering a picture. Performance signals were recorded for each test as a digital EEG for more than one minute. We used our own software, CHASIM, to calculate the fractal dimension and Lyapunov exponent. The statistical analysis performed was an ANOVA using SAS 6.12 for Windows (9.0).
RESULTS
The fractal dimensions during the "listening to music and noise", and "moving both toes" tasks were increased. During the "eye opening", "calculation", and "moving both fingers" tasks, the fractal dimensions were decreased.
CONCLUSIONS
Auditory stimulation generated a higher correlation dimension than visual stimulation. The interpretation of these results is still not totally clear, but we hope to find future applications for non-linear chaotic analysis for the functional evaluation of the central nervous system.