Background
s and objective : EEG reflect dynamic changes of continuous neuronal activities by internal and externalstimuli. The aim of this study is to quantify nonlinearly the local dynamic differences among EEG data correspondingto different states of brain.Methods : EEG was recorded from twelve healthy normal subjects(mean age, 29.7 years; 8 men and 4 women) usingdigital EEG machine. 18-channel EEG data were selected during eyes closed(EC), eyes open(EO), and mental arithmetic(MA) in each subject. Correlation dimension(D2) and largest Lyapunov exponent(LLE) were calculated fromthree states and average value was mapped 2 dimensionally and compared with each other.Results : The distribution of D2 was relatively symmetric and its value was higher in frontal than in parieto-occipitalregion during EC. These findings were reversed during EO. Bilateral centro-temporo-parietal region showed high D2value in MA compared with those in EC, which was more prominent in left side. LLE was larger than zero in all stateand showed significant differences among EC, EO and MA(p=0.000).Conclusion : These results suggest that nonlinear analysis of EEG can quantify dynamic state of brain.