Abstract

This dissertation investigates Multi Agent Decision Making (MADM) within Complex Societies (CS), systems of interdependent agents whose collective behaviors emerge from local interactions. Such emergent dynamics are nonlinear and difficult to predict, posing significant challenges towards understanding how individual decisions scale to societal outcomes. Existing approaches often fall short of capturing the combined attributes of CS, including asymmetry, decentralization, and adaptivity. To address this gap, we introduce new models and analyses of agent strategies that explain how cooperation and stability can arise in CS. Structured around three interrelated projects, this dissertation enhances our comprehension of MADM in CS. The research centers on the Junior High Games (JHG), a strategic network-based environment designed to simulate the essential features of complex social interaction. The first project establishes the JHG as a representative test-bed for simulating MADM within CS and presents baseline algorithms that leverage network topology to study collective action. The second project incorporates empirical human data, modeling human behavior in the JHG and demonstrating how artificial agents might successfully interact within human societies. The final project examines evolutionary stability and productivity of societies of agents, exploring the interdependence of agent interactions by analyzing which strategies are evolutionarily stable as well as identifying which strategies contribute to societal productivity. Collectively, this dissertation contributes a unified framework for studying MADM in CS. Through introducing the JHG, modeling human strategies within it, and analyzing the evolutionary stability of cooperation, it demonstrates how local interactions scale into collective action and stability. The findings show that community awareness and higher-order reasoning are critical for sustaining cooperation, while also providing tools for comparing artificial and human strategies in shared environments. These contributions advance theory in multi-agent systems and complex networks and lay groundwork for applying these insights to real-world domains where cooperation is fragile, ranging from political coalitions to economic networks and human-AI collaboration.

Degree

PhD

College and Department

Computer Science; Computational, Mathematical, and Physical Sciences

Rights

https://lib.byu.edu/about/copyright/