Authors : Dev Athwani; Stuti Srivastava; Dr. Gaurvi Shukla; Rinku Raheja

Volume/Issue : Volume 11 - 2026, Issue 4 - April

Google Scholar : https://tinyurl.com/mwbvy7sj

DOI : https://doi.org/10.38124/ijisrt/26apr912

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : Deception Detection is still a challenge in security, forensics and high stakes interviews. The conventional approaches such as polygraphs are inaccurate and can be easily tampered. The paper will analyze a multimodal artificial intelligence structure of detecting truthfulness, which involves three complementary modalities: vocal features, linguistic text pattern and facial micro-expression. Machine learning and deep learning are used in the methodology to detect minor and subconscious cues of deception that could be overlooked with single-modality analysis. The system processes acoustic, semantic and syntactic, and micro-expressions as well. Multimodal learning systems combine these cues to make them more robust and less ambiguous, in addition being more accurate. Very initial signs that can be obtained through the current literature and the test of prototypes prove that multimodal fusion is far better than unimodal methods in terms of reliability and usability. The possible uses include border control, fraud detection, law enforcement interrogation, recruitment screening, and digital communication systems in which authenticity seems paramount. The paper is an addition to the developing body of AI-based deception detection by offering a scalable, flexible, and ethically conscious framework.

Keywords : AI-Based Deception Detection; Micro-Expressions; Multimodal; Machine-Learning; Truthfulness Detection.

References :

1. Krishnamurthy, G., Majumder, N., Poria, S., & Cambria, E. (2018). A deep learning approach for multimodal deception detection. arXiv.
2. Benchmarking Cross-Domain Audio-Visual Deception Detection. (2024). arXiv.
3. 3D Facial Landmark-based Deception Detection. (2022). Journal of Kufa.
4. Multimodal Latent Emotion Recognition from Micro-expressions. (2025). ScienceDirect.
5. Deception Detection using Machine Learning and 3D Face Reconstruction. (2024). ScienceDirect.
6. MM-DFN: Multimodal Dynamic Fusion Network for Emotion Recognition in Conversations. (2022). arXiv.
7. Hu, X., et al. (2021). Multimodal fusion via deep graph convolution network. Semantic Scholar.
8. SMFNM: Semi-supervised multimodal fusion network with main modality consistency. (2023). ScienceDirect.
9. Multimodal Fusion via Hypergraph Autoencoder and Contrastive Learning. (2024). ScienceDirect.
10. Survey of Digital Forensic Methods for Multimodal Deepfake Detection on Social Media. (2024). University of Virginia.
11. Joshi, G., et al. (2025). Multimodal machine learning for deception detection using behavioural, verbal, and neurophysiological data. MIT Media Lab.
12. Multimodal machine learning for deception detection using text, audio, and vision. (2025). Nature.
13. Columbia University. (2020). Multimodal deception detection using automatically extracted features.
14. Multimodal Deception Detection Challenge. (2025). arXiv.
15. FusionNet. (2017). Fusing via fully-aware attention.
16. Burzo, M., et al. (2017). Multimodal deception detection. Morgan Claypool.
17. Zhang, J., et al. (2020). Multimodal deception detection using automatically extracted features. Interspeech.
18. MDPE: A Multimodal Deception Dataset with Personality and Emotional Characteristics. (2022). arXiv.
19. Jaiswal, M., et al. (2016). Multimodal analysis for deception detection. SenticNet.
20. Rao, K. V. S., et al. (2022). Artificial intelligence for deception detection: A multimodal review. ETJ.