Long-range temporal correlations in violence and psychosis: An EEG-based study

Area of Science:

• Neuroscience
• Psychiatry
• Biomedical Engineering

Background:

• A subset of individuals with psychosis exhibit elevated violence risk.
• The relationship between neural dynamics, specifically long-range temporal correlations (LRTC) in electroencephalography (EEG), and violence risk in psychosis is not well understood.

Purpose of the Study:

• To investigate whether LRTC in resting-state EEG differentiates individuals with psychosis based on violence history.
• To explore the association between LRTC and symptom severity in psychosis.

Main Methods:

• Resting-state EEG data were collected from healthy controls (HC), individuals with psychosis with a history of violence (violent-PSY), individuals with psychosis without a history of violence (nonviolent-PSY), and violent individuals without psychosis (violent non-PSY).
• LRTC was quantified using detrended fluctuation analysis (DFA) in alpha and beta frequency bands during eyes-closed and eyes-open conditions.
• Eight LRTC metrics were derived from mean and peak amplitudes for both bands and conditions.

Main Results:

• Eyes-open condition LRTC did not differ significantly across groups.
• Eyes-closed condition showed significantly reduced LRTC in the nonviolent-PSY group compared to HC for alpha and beta mean/peak amplitudes.
• No significant differences in LRTC were found between violent-PSY and nonviolent-PSY groups.
• Lower alpha-band LRTC correlated with increased symptom severity, including total PANSS and depressive scores.

Conclusions:

• Reduced LRTC in the eyes-closed condition was observed in nonviolent psychosis patients compared to controls.
• LRTC did not differentiate psychosis patients with or without a violence history.
• Current findings offer limited evidence for LRTC as a predictor of violence risk in psychosis.