How Do Terrorist Cells Self-Assemble? Insights from an Agent-Based Model

Abstract:     
In order to shed light on the process of radicalization (the adoption of violence for political ends), we constructed an agent-based computational simulation of a community in which radicalization occurs. In this simulated society we varied parameters such as community size, population homogeneity/diversity, and migration; examining how they affect the formation of self-starter radical cells. We have also looked at social venues, which we call "magnets", and modeled how they create new ties between agents. In agreement with case studies, we found that such venues radicalize even when they attract radicals and non-radicals alike. Additionally, we found that diversity reduces radicalization but population size has little effect. By understanding how radical cells form, the model facilitates better prediction of terrorist-cell formation and suggests possible counter-radicalization policies.

Keywords: terrorism, radicalization, simulation, agent-based model, home-grown terrorism, self-starter terrorism

Working Paper Series