Collaborating with Style: Using an Agent-Based Model to Simulate Cognitive Style Diversity in Problem Solving Teams

Abstract

Collaborative problem solving can be successful or counterproductive. The performance of collaborative teams depends not only on team members' abilities, but also on their cognitive styles. Cognitive style measures differences in problem-solving behavior: how people generate solutions, manage structure, and interact. While teamwork and problem solving have been studied separately, their interactions are less understood. This paper introduces the KAI Agent-Based Organizational Optimization Model (KABOOM), the first model to simulate cognitive style in collaborative problem solving. KABOOM simulates the performance of teams of agents with heterogeneous cognitive styles on two contextualized design problems. Results demonstrate that, depending on the problem, certain cognitive styles may be more effective than others. Also, intentionally aligning agents' cognitive styles with their roles can improve team performance. These experiments demonstrate that KABOOM is a useful tool for studying the effects of cognitive style on collaborative problem solving.