Adaptive Agent-Based Models with Real-Time Social Feedback Loops for Epidemic Control

Hotton et al. (2025) and Pando et al. (2023) focus on agent-based modeling and the role of social determinants in disease spread, but most models use static or pre-programmed behavioral rules. This research introduces adaptive agent-based models where individual decisions (e.g., to isolate, vaccinate, or ignore guidelines) change in response to real-time local infection rates, peer behavior, and global news or misinformation trends. Network connections may also rewire dynamically as agents avoid or seek contacts based on perceived risk. The model would be validated against empirical data from recent outbreaks with strong behavioral feedbacks (e.g., COVID-19 waves). This direction is novel because it tightly integrates behavioral adaptation and network evolution, providing a more realistic simulation platform for testing intervention strategies that rely on public compliance and real-time information dissemination.

References:

1. Incorporating social determinants of health into agent-based models of HIV transmission: methodological challenges and future directions. Anna Hotton, Pedro Nascimento de Lima, A. Fadikar, Nicholson T. Collier, Aditya S. Khanna, D. Motley, E. Tatara, Sara Rimer, Ellen Almirol, Harold A. Pollack, John A Schneider, R. Lempert, J. Ozik (2025). Frontiers in Epidemiology.
2. A social network analysis model approach to understand tuberculosis transmission in remote rural Madagascar. Christine Pando, A. Hazel, Lai Yu Tsang, K. Razafindrina, Andry Andriamiadanarivo, Roger Mario Rabetombosoa, Ideal Ambinintsoa, Gouri B. Sadananda, Peter M. Small, A. Knoblauch, N. Rakotosamimanana, S. Grandjean Lapierre (2023). BMC Public Health.