A Modeling Vaccine Hesitancy in a Dynamic World: Integrating Misinformation, Behavioral Shocks, and Policy Interventions in a Game-Theoretic Framework

Abstract

When we think about vaccination, it's not just a medical choice—it's a deeply human one, woven with perceptions, risks, and the subtle pull of collective behavior. Building on empirical measles data, this work extends a game-theoretic model of the public goods dilemma to capture the complexities of real-world hesitancy. We start with a baseline replication, grounding individual utilities in observed disease trends, but then layer in misinformation spread, sudden hesitancy shocks, and policy levers like subsidies and mandates. Using an agent-based simulation coupled with an SIR epidemiological framework, we explore how these factors interplay to shape coverage, incidence, and effective reproduction number (R_eff).
Key findings reveal the fragility of herd immunity: under a hesitancy shock—simulating a surge in doubt or misinformation—mean vaccination coverage drops to 6.4%, peak incidence spikes to around 59,433 cases, and final R_eff settles at 0.43, far below control thresholds. Sensitivity analysis via Sobol indices highlights R_0 as the dominant driver of final R_eff, with first-order sensitivity near 1.0, underscoring transmission's outsized role amid behavioral noise. Policy frontiers map trade-offs: combining subsidies (up to 0.5) and mandate penalties (up to 0.5) can push R_eff down to 0.18 while boosting welfare to 0.08, but intensity matters—overly aggressive mandates risk backlash. This model advances prior work by incorporating dynamic feedback loops and actionable policies, offering a compass for policymakers. It shows that while free-riding persists, targeted interventions can tip the balance toward resilience, even in uncertain times. Future extensions
could integrate network effects or evolving variants for deeper insights.