Identifying Differences in the P600 Component of ERP-Signals between OCD Patients and Controls Employing a PNN-based...

Area of Science:

• Neuroscience
• Cognitive Psychology
• Computational Psychiatry

Background:

• Working Memory (WM) deficits in Obsessive-Compulsive Disorder (OCD) have yielded inconsistent findings in neuropsychological research.
• Event-related potential (ERP) signals, specifically the P600 component, offer a potential avenue to investigate neural differences.

Purpose of the Study:

• To differentiate between OCD patients and healthy controls using P600 ERP component analysis.
• To identify brain regions associated with potential Working Memory deficits in OCD.

Main Methods:

• Utilized a computerized digit span Wechsler test for 18 OCD patients and 20 healthy controls.
• Recorded electroencephalogram (EEG) activity from 15 scalp electrodes.
• Developed custom software to analyze ERP P600 component features (500-800 ms) and employed a Probabilistic Neural Network (PNN) classifier.

Main Results:

• Achieved high classification precision (86.8%) using single-lead PNN at Fp2 and C6 electrodes.
• A Majority Vote Engine (MVE) integrating all leads achieved perfect classification of all subjects.
• Observed significant differences in ERP signals between OCD patients and controls in prefrontal and temporo-central brain regions.

Conclusions:

• The P600 component of ERP signals, analyzed via PNN and MVE, provides a robust method for distinguishing OCD patients from healthy controls.
• Neural differences related to working memory in OCD are localized to prefrontal and temporo-central brain areas.