Neural signatures of deception: an explainable machine learning approach using EEG signals

Analyzing brainwave patterns related to cognitive functions have gained attentions in deception detection. Electroencephalogram (EEG) signals provide a direct, non-invasive means of capturing neural response related to truth and lie behavior that overcoming limitations of traditional lie detection methods. This paper proposes a effective and interpretable lie detection framework using Light Gradient Boosting Machine (LGBM). EEG signals corresponding to truth and lie were recorded from subjects using 16 electrodes. Relevant features are extracted using ANOVA F-score and fed to classifier for the truth and lie detection using proposed method. The proposed LGBM achieves the classification accuracy of 99.16% with high precision, sensitivity, and specificity. Feature importance analysis and ROC curve results confirmed the robustness, reliability of the model. The proposed method compared with traditional algorithms like Support Vector Machine (SVM), K-Nearest Neighbors, Random Forest, and XGBoost, hybrid SVM+XGBoost, gradient Boosting Decision Tree, and hybrid PCA+SVM. The proposed LGBM approach outperformed in terms of different object quality metrics. The results indicate that the proposed LGBM-based EEG lie detection framework is a highly accurate, reliable, and interpretable method for identifying deceptive behavior with smaller datasets.