ABSTRACT
With an increasing shift towards asymmetric warfare, soldiers must be trained to enhance their social and behavior skills. Models/agents based on Adaptive Control of Thought-Rational (ACT-R) cognitive architecture have been widely used in military simulation platforms to educate and track trainee performance in various tactical domains. However, these models are inept in emulating the full range of human behavior in socio-cognitive tasks. Through this project, we have been successful in building an enhanced ACT-R (E+) model by incorporating an emotion (E) mechanism that produces better performance mapping compared to conventional ACT-R (E-) model for simple socio-cognitive games of Prisoner’s dilemma (PD) and Intergroup Prisoner’s Dilemma with Intragroup Power Dynamics – (IPD^2). By extending this fundamental work to incorporate various facets that affect decision-making in a social environment, the E+ model/agents could be used to understand and forecast the outcomes of real-world socio-cultural interactions. This study entails developing high-fidelity cognitive models/agents that can be employed for training of soldiers to improve their decision-making skills in scenarios that involve socio-cultural interactions with civilians or allied forces.
