Title: Analysing EEG-based differential functional connectivity patterns during truth and lie responses

Authors: Sakshi Jethva; Jyoti Maheshwari

Addresses: School of Behavioural Forensics, National Forensic Sciences University, Gandhinagar, 382007, Gujarat, India ' Electronics and Communication Engineering, Nirma University, Ahmedabad, 382481, Gujarat, India

Abstract: With the growing application of machine learning and deep neural networks in biomedical signal processing, selecting the right features for model training has become crucial – especially in lie detection, where accurate classification has serious implications in forensics, law, national security, and research. This study emphasises the importance of feature selection by comparing functional connectivity (FC) networks during truth and lie conditions. Using a publicly available dataset, we applied multiple connectivity measures – correlation, phase locking value (PLV), phase lag index (PLI), coherence, and imaginary coherence (iCOH) – across frequency bands (global, delta, theta, alpha, beta, gamma). Results showed significantly higher FC during lying in global, delta, and theta bands, particularly in frontal-temporal regions, suggesting their relevance for deception detection. In contrast, alpha, beta, and gamma bands showed inconsistent FC patterns. These findings highlight the complex neural dynamics of lying and support the use of diverse connectivity measures to enhance the accuracy of lie detection systems.

Keywords: coherence; connectivity; correlation; deception; phase lag index; PLI; phase locking value; PLV.

DOI: 10.1504/IJBET.2025.149596

International Journal of Biomedical Engineering and Technology, 2025 Vol.49 No.2, pp.121 - 138

Received: 26 Jan 2025
Accepted: 06 Apr 2025
Published online: 07 Nov 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article