Estimating Population Attitudes with CogSim

Introduction: Estimating the nonkinetic effects of irregular warfare (IW) actions on the population, such as those described in the U.S. Army counterinsurgency manual,1 is now incorporated into the analysis of IW war games. CogSim is an agent-based population attitude model developed at NRL as a prototype to predict coalition formation between several groups based on social science theories, namely the personal construct theory and social learning theory. In this approach, a cognitive agent adapts its belief system, represented in a casual belief map, in response to its social environment characterized by the belief systems of other agents with whom it interacts. The claim of this model is that group mental models will emerge from individual mental models through the learning and adaptation of causal belief maps.