همایش بین المللی هوش مصنوعی

صفحه اصلی / دومین همایش بین المللی هوش مصنوعی

Deep Learning-Based Prediction of Personality Profiles from Resting-State EEG Signals

نویسندگان :

Marjan Nosouhi¹ Moein Esghaei²

1- University of Tehran 2- Faculty of Computer Science and Engineering, Shahid Beheshti University

کلمات کلیدی :

electroencephalography (EEG)،Millon Clinical Multiaxial Inventory (MCMI)،deep learning،Convolutional Neural Networks (CNN)،Personality Neuroscience،Computational Psychiatry

چکیده :

The integration of neuroscience and artificial intelligence has opened new avenues for understanding the relationship between brain activity and psychological characteristics. In this study, a deep learning framework based on Convolutional Neural Networks (CNNs) is presented to predict personality and clinical profile parameters derived from the Millon Clinical Multiaxial Inventory (MCMI) using resting-state electroencephalography (EEG) data. EEG recordings were obtained from 30 participants using a standard 19-channel montage under eyes-closed resting conditions. The data were preprocessed and segmented into fixed-length epochs, and a multi-output CNN was trained to map multichannel time series to 28 MCMI parameters simultaneously. Training was performed with subject-wise cross-validation to avoid leakage. Experimental results demonstrate that the proposed CNN model achieves meaningful predictive performance across several MCMI parameters. Notably, certain scales—such as Raw 7 (related to compulsive personality) of the MCMI analysis—exhibited a high and statistically significant correlation with the corresponding ground-truth scores (p < 0.05), indicating that specific neural activity patterns are strongly associated with distinct psychological traits. Statistical significance was assessed with false-discovery-rate control across scales. This study highlights the potential of deep learning approaches in advancing computational psychiatry and personality neuroscience, offering a non-invasive framework for objective psychological assessment based on neurophysiological data.