A generative model of trust formation and generations

Mehmet Çelik Özdeş

This thesis introduces an agent based simulation of a multi-generation population with a dynamic trust network in a world with limited resources. This population is assumed to be under evolutionary pressure incurring costs of life and individuals must act wisely in order to survive and reproduce. Over their lifetimes individuals encounter strangers, assess their reliability and build lasting connections. They retain the memory of all past interactions and re-evaluate their trust in others and in themselves after all actions. Individuals employ various strategies to decide their course of action and take advice from their trust networks. We use this generative model to investigate the emergence and consequences of generational differences under various conditions of the world.