Applications of agent based models to complex social systems: coordinated human behaviour from the bottom Up

This work investigates how complex collective behaviors emerge in social systems by employing agent‐based modeling to bridge micro‐level interactions and macro‐level phenomena. Three interrelated studies form the core of this research. The first study explores the spontaneous formation of investor clusters in financial markets, demonstrating that adaptive learning and imitation—without any preset behavioral biases—can drive the emergence of distinct trading typologies. In the second study, a novel framework incorporating subjective reputation mechanisms and gossip is developed to assess how decentralized information exchange fosters cooperation. The results reveal that local reputation assessments not only enhance cooperative behavior but also lead to the formation of centralized hubs within social networks. Finally, the third study examines cooperation in pre‐state societies, using simulation models inspired by the Aksum civilization. This investigation shows that while short-term cooperation can arise in the absence of coercive authority, long-term stability may ultimately depend on the emergence of centralized power structures. Collectively, these findings provide new insights into the evolutionary dynamics of complex social systems and underscore the versatility of agent‐based modeling as a tool for understanding emergent collective phenomena.