The Effects of Trust on Convention Emergence

In this paper, we consider the effects of trust and conventions on the evolution of a multi-agent system. To do so, we provide a simulation in which agents are able to use a trust mechanism to select interaction partners, and a learning mechanism that leads to convention emergence. We examine the impacts on the system of: different network topologies; the presence or absence of malicious agents; and the presence or absence of the trust and learning system. Our results indicate that while trust has a slight positive impact on the rate of convergence emergence, its main benefit arises when malicious agents are present in the system.