Feedbacks between human behavior, disease transmission, and vector ecology generate culture-linked dynamics in vector-borne diseases

Denis Tverskoi , Daniela Miteva, Bryce Morsky, Grzegorz A Rempala

Vector-borne diseases cause millions of deaths worldwide and are expanding into new regions and populations as climate change intensifies. Human behavior, shaped by social and cultural factors, is central to disease control, yet feedbacks among behavior, disease transmission, and vector ecology remain poorly understood. To address this gap, we develop a parsimonious malaria model coupling infection dynamics in human and mosquito populations with evolving protective behavior, perceived intervention effectiveness, and competition between insecticide-resistant and non-resistant mosquitoes. We show that the strength of social norms fundamentally reshapes disease dynamics. Weak norms can generate endemic waves of infection and protective behavior, whereas moderate norms can enable disease elimination but also produce culture-dependent tipping dynamics, including abrupt reversion to highly endemic states following ecological or economic change. Strong norms can lock in either beneficial or harmful behavioral states, sustaining high adoption when protective behavior is already established or promoted but otherwise stabilizing poor epidemiological outcomes. These complex socio-biological feedbacks can lead to counterintuitive results: the model suggests that under weak norms, untreated nets can yield lower infection prevalence than insecticide-treated nets. In addition, we identify dynamically changing behavior and perceived effectiveness as a standalone mechanism capable of generating the coexistence of insecticide-resistant and non-resistant mosquito populations. We complement the model with an exploratory analysis of the adoption of mosquito nets, malaria prevalence, and the changes in mosquito resistance in Liberia to examine whether some of the qualitative patterns suggested by the model are also visible in the available data. While the empirical analysis is not causal, the estimated associations are largely consistent with the theoretical predictions. Overall, our framework bridges game-theoretic and epidemiological approaches, providing a unified perspective for understanding vector-borne diseases as coupled human–biological systems and designing behaviorally informed interventions.