Agent-based modeling (ABM) is a modeling technique to analyze complex systems by simulating the interactions of individual agents with their surroundings. ABM enables the simulation of complex systems with emergent properties resulting from agent interaction. The modeling approach can be used to investigate a wide range of systems, such as biological, social, and economic systems. ABM can assist researchers in studying the effects of various policies, strategies, and interventions on complex systems and understanding how individual agent behavior affects system behavior.

The here presented study analyzes team cognition as an emergent property using an agent-based model. This is a mixed-method approach with group observations and laboratory experiments to provide qualitative and quantitative input, as well as an output validation of the model’s key aspects. The group observation generates information about the quantitative patterns of individual and team cognition. To analyze these patterns, we introduce the concept of a team cognition matrix and identify four types of empirical team cognition structures (consensus, majority, party-decision, and compromise). These types serve as the basis for developing a team decision algorithm that we subsequently implement in an agent-based model. In our output validation, we find that the simulated team results approximate the empirically observed team decisions. Overall, we contribute to the current literature by presenting an innovative research approach that opens the black box of team behavior beyond wellknown static attributes.

This presentation will primarily focus on this study, but also discuss how agentbased modeling can be successful in further mixed-method approaches such as with structural equation modeling. The presentation will also address how ABM may contribute to theory building and refer to the discussion around ABM and theory building and the difficulties that can arise.