Game-theoretic behavioral adaptation in non-Markovian epidemic spreading on networks

Epidemic spreading is influenced by both temporal memory effects and adaptive human behavior. Although behavioral adaptation has been widely studied in Markovian epidemic models, its interaction with non-Markovian spreading dynamics remains poorly understood. we introduce AG-nSIS (Adaptive Game-coupled non-Markovian Susceptible–Infectious–Susceptible), a modeling framework that integrates adaptive game-theoretic decision-making with non-Markovian SIS dynamics on complex networks. AG-nSIS couples an evolutionary game for protective behavior with age-dependent infection and recovery processes, enabling us to derive a self-consistent condition for the steady-state level of protection. Through extensive numerical simulations, we show that adaptive behavior fundamentally reshapes non-Markovian epidemic spreading. Specifically, strategy-modulated transmission factors c critically regulate outbreak potential by tuning the effective transmission rate across different behavioral interactions. Moreover, the co-evolutionary game dynamics substantially curb both the final outbreak size and the endemic prevalence.