Nonlinear Dynamic Analysis of EEG in Patients with Positive and Negative Schizophrenia

Chae, J H; Pak, E J; Kim, D J; Jeong, J S; Kim, S Y; Kim, K S

Abstract

OBJECTIVES
The hypothesis that the brain is a nonlinear dynamical system exhibiting deterministic chaos has offered new perspectives to the investigation of information processing in the brain of schizophrenic patients. It seemed worthwhile to estimate nonlinear measures of the electroencephalogram (EEG) in positive and negative schizophrenics, because nonlinear measures might serve as indicators of the specific brain function in schizophrenia according to specific psychopathologies. METHOD: Previous studies which estimated the chaoticity in the brain of schizophrenia with nonlinear methods recorded the EEGs at limited electrodes, so we tried to record EEGs from 16 channels for nonlinear analysis in 8 positive and 9 negative schizophrenics and 8 healthy control subjects. We employed a new method to calculate the nonlinear invariant measures. For limited noisy data, this algorithm was strikingly faster and more accurate than previous ones.
RESULTS
Our results showed that the patients with negative schizophrenia had lower the first positive Lyapunov exponents (L1) than the positive schizophrenics and control subjects at T3 lead. Positive symptoms were positively correlated with L1, in C3, O1 leads, and negatively correlated with C4 lead.
CONCLUSION
These results suggest that if clinical variables such as psychopathology or neuroleptic medications would be well controlled, the nonlinear analysis of the EEGs in patients with schizophrenia seems to be a useful tool in analyzing EEG data to explore the neurodynamics.