Mathematical cognitive modelling enhanced by deep learning for understanding human decision making in complex sequential tasks

Ananthakrishnan, Gayathri; Ghanimi, Hayder M. A.; Babu, Kolluru Suresh; Kata, Anilambica; Sravanthi, Gunaganti; Smerat, Aseel; Sengan, Sudhakar; Dadheech, Pankaj

doi:10.47974/JIM-2573

Open Access Research Article

Mathematical cognitive modelling enhanced by deep learning for understanding human decision making in complex sequential tasks

^*Gayathri AnanthakrishnanCorresponding authorgayathri.a@vit.ac.inDepartment of Information TechnologySchool of Computer Science Engineering and Information SystemsVellore Institute of TechnologyVellore, Tamil Nadu, 632014, IndiaView full profile → , Hayder M. A. Ghanimihayder.alghanami@uowa.edu.iqAffiliation 1Department of Information TechnologyCollege of ScienceUniversity of Warith Al-AnbiyaaKarbala, 56001, IraqAffiliation 2Department of Computer ScienceCollege of Computer Science and Information TechnologyUniversity of KerbalaKarbala, 56001, IraqView full profile → , Kolluru Suresh Babukollurusuresh@gmail.comDepartment of Computer Science and EngineeringVasireddy Venkatadri Institute of TechnologyGuntur, Andhra Pradesh, 522508, IndiaView full profile → , Anilambica Katahranilambica@gmail.comSymbiosis International (Deemed University)Symbiosis Institute of Business ManagementHyderabad, Telangana, 509217, IndiaView full profile → , Gunaganti Sravanthisravanthi887@gmail.comDepartment of Computer Science and EngineeringMalla Reddy College of Engineering & TechnologyHyderabad, Telangana, 500100, IndiaView full profile → , Aseel Smeratsmerat.2020@gmail.comHourani Centre for Applied Scientific ResearchAl-Ahliyya Amman UniversityAmman, 19328, JordanView full profile → , Sudhakar Sengansudhasengan@gmail.comDepartment of Computer Science and EngineeringPSN College of Engineering & Technology (Autonomous)Tirunelveli, Tamil Nadu, 627152, IndiaView full profile → , Pankaj Dadheechpankajdadheech777@gmail.comDepartment of Computer Science and EngineeringSwami Keshvanand Institute of Technology, Management & Gramothan (SKIT)Jaipur, Rajasthan, 302017, IndiaView full profile →

* Corresponding author · click or hover a name for details

pp. 1033–1044Vol. 29Issue 4April 2026DOI: 10.47974/JIM-2573XML

Received:: 01 Sep 2025
Published Online:: 09 Apr 2026
Article type:: Research Article
Language:: EN
Article no.:: JIM-2573
Pages:: 1033–1044

Abstract

Conventional mathematical models and data-driven deep learning (DL) methods are both tested for Human Decision-Making (HDM) in Complex Sequential Tasks (CST). These methods include complicated cognitive functions. Considering the use of adaptive timing mechanisms, the present research introduces a hybrid model that includes statistical cognitive dynamic models with Bidirectional Long Short-Term Memory (BiLSTM) attention networks. While DL serves as the basis for collecting temporal behavioural patterns, the mathematical model is vital for recording value changes, cognitive load, and decision selection by its application of classical Boltzmann logic. The mathematical model of contraction mapping serves as the basis for the invention of convergence guarantees. The recommended model achieves 84.7% prediction accuracy and improved cognitive state alignment (CSD=0.187) when tested on the Iowa Gambling Task dataset, involving 617 participants. The proposed model improves the accuracy of mathematical models by 12.4% and the validity of DL by 3.5%. This research has enabled the interpretation of artificial intelligence (AI) for human-centred decision support systems.

Keywords

Cognitive modelling Deep learning Human decision-making Sequential decision processes BiLSTM Attention mechanisms Hybrid systems Iowa gambling task Cognitive state alignment Adaptive gating

Subject Classifications

Primary 91E10 and 68T07Secondary 37N40 and 62M45

References

[1] U. Rakhmanovich, M. Taslim, A. M. Raman, and A. A. A. Ugli, “The rise of behavioral economics: Leveraging AI to understand consumer decision-making,” in AIP Conference Proceedings, vol. 3306, no. 1, pp. 050007 (2025).
[2] I. Kotseruba and J. K. Tsotsos, “The Computational Evolution of Cognitive Architectures”. Oxford, U.K.: Oxford University Press (2025).
[3] K. Roy, “Process-grounded knowledge-infused learning and decision making,” Ph.D. dissertation, Univ. South Carolina, Columbia, SC, USA (2025).
[4] R. Mundlamuri, G. R. Gunnam, N. K. Mysari, and J. Pujuri, “The evolution of AI: From classical machine learning to modern large language models,” IEEE Access (2025).
[5] P. Peykani, F. Ramezanlou, C. Tanasescu, and S. Ghanidel, “Large language models: A structured taxonomy and review of challenges, limitations, solutions, and future directions,” Applied Sciences, vol. 15, no. 14 (2025).
[6] Z. Zhang, X. Ding, X. Liang, Y. Zhou, B. Qin, and T. Liu, “Brain and cognitive science inspired deep learning: A comprehensive survey,” IEEE Transactions on Knowledge and Data Engineering (2025).
[7] L. E. Guerrero, L. F. Castillo, J. Arango-Lopez, and F. Moreira, “A systematic review of integrated information theory: A perspective from artificial intelligence and the cognitive sciences,” Neural Computing and Applications, vol. 37, no. 11, pp. 7575–7607 (2025).
[8] T. Gopalakrishnan, H. Z. Almngoshi, V. K. Koppula, K. Dinesh, V. K. Sonthi, S. Sengan, and P. Dadheech, “Fractal image compression using digital cone metric space,” Journal of Interdisciplinary Mathematics, vol. 28, no. 2, pp. 647–658 (2025), doi: 10.47974/JIM-2108.
[9] L. Deng, H. Tang, and K. Roy, “Editorial: Understanding and bridging the gap between neuromorphic computing and machine learning,” Frontiers in Computational Neuroscience, vol. 15, pp. 665662 (2021), doi: 10.3389/fncom.2021.665662.
[10] V. K. Koppula, K. Dinesh, V. K. Sonthi, S. Sengan, and P. Dadheech, “Fractal image compression using digital cone metric space,” Journal of Interdisciplinary Mathematics, vol. 28, no. 2, pp. 647–658 (2025), doi: 10.47974/JIM-2108.
[11] S. Siddamsetti, M. M. Qasim, N. Sravanthi, K. Dinesh, V. Velmurugan, S. Sengan, and P. Dadheech, “Wireless sensor networks with modular metric spaces,” Journal of Interdisciplinary Mathematics, vol. 28, no. 2, pp. 659–667 (2025), doi: 10.47974/JIM-2109.
[12] S. Siddamsetti, H. Z. Almngoshi, K. Dinesh, G. C. Prashant, M. A. Bennet, P. Dadheech, and S. Sengan, “Modular metric spaces: Some fixed-point theorems and application of secure dynamic routing for WSN,” Journal of Interdisciplinary Mathematics, vol. 27, no. 2, pp. 393–401 (2024), doi: 10.47974/JIM-1890.
[13] S. H. Nowfal, G. R. K. Rao, V. Velmurugan, S. Sengan, R. K. Bommisetti, and P. Dadheech, “Advancing viscoelastic material modelling: Tackling time-dependent behavior with fractional calculus,” Journal of Interdisciplinary Mathematics, vol. 27, no. 2, pp. 307–316 (2024), doi: 10.47974/JIM-1827.
[14] P. V. Sagar, M. Rajyalaxmi, A. V. V. S. Subbalakshmi, S. Sengan, R. K. Bommisetti, and P. Dadheech, “Utilizing stochastic differential equations and random forest for precision forecasting in stock market dynamics,” Journal of Interdisciplinary Mathematics, vol. 27, no. 2, pp. 285–298 (2024), doi: 10.47974/JIM-182ract.
[15] S. Siddamsetti, H. Z. Almngoshi, K. Dinesh, G. C. Prashant, M. A. Bennet, P. Dadheech, and S. Sengan, “Modular metric spaces: Some fixed-point theorems and application of secure dynamic routing for WSN,” Journal of Interdisciplinary Mathematics, vol. 27, no. 2, pp. 393–401 (2024), doi: 10.47974/JIM-1890.
[16] S. H. Nowfal, G. R. K. Rao, V. Velmurugan, S. Sengan, R. K. Bommisetti, and P. Dadheech, “Advancing viscoelastic material modelling: Tackling time-dependent behavior with fractional calculus,” Journal of Interdisciplinary Mathematics, vol. 27, no. 2, pp. 307–316 (2024), doi: 10.47974/JIM-1827.

Views: 326Downloads: 135Citations: 0