MAO – Predicting Self-Initiated Preventive Behavior Against Epidemics with an Agent-Based Relative Agreement Model
Liang Mao
Article first published online: 31 Oct 2015 Journal of Artificial Societies and Social Simulation
DOI: 10.18564/jasss.2892
ABSTRACT: Human self-initiated behavior against epidemics is recognized to have significant impacts on disease spread. A few epidemic models have incorporated self-initiated behavior, and most of them are based on a classic population-based approach, which assumes a homogeneous population and a perfect mixing pattern, thus failing to capture heterogeneity among individuals, such as their responsive behavior to diseases. This article proposes an agent-based model that combines epidemic simulation with a relative agreement model for individual self-initiated behavior. This model explicitly represents discrete individuals, their contact structure, and most importantly, their progressive decision making processes, thus characterizing individualized response to disease risks. The model simulation and sensitivity analysis show the existence of critical points (threshold values) in the model parameter space to control influenza epidemic including minimum values for the initially positive population size, the communication rate, and the attitude uncertainty. These threshold effects shed insights on preventive strategy design to deal with the current circumstances that new vaccines are often insufficient to combat emerging communicable diseases.
Read the full publication at Journal of Artificial Societies and Social Simulation