Know Your Enemy: Applying Cognitive Modeling in Security Domain

Game Theory -based decision aids have been successfully employed in real-world policing, anti-terrorism, and wildlife conservation efforts (Tambe, Jiang, An, & Jain, 2013). Cognitive modeling, in concert with model tracing and dynamic parameter fitting techniques, may be used to improve the performance of such decision aids by predicting individual attacker behavior in repeated security games. We present three simulations, showing that (1) cognitive modeling can aid in greatly improving decision-aid performance in the security domain; and (2) despite the fact that individual attackers will differ in initial preferences and in how they learn, model parameters can be adjusted dynamically to make useful predictions for each attacker.