Evolving populations of agents with personalities in the minority game

Abstract

In our daily lives we often have to face binary decisions where we seek to take the minority's choice, e.g., in traffic scenarios where we have to choose between similar alternative routes. In previous papers, we have dealt with an agent coordination mechanism for binary decision models introduced in the literature recently: the El Farol Bar Problem. Extending this model, we have proposed personalities which model certain types of human behaviour, and we have simulated different populations of these personalities. Our previous work has given insights into the impact of commuters' behaviours and it addresses relegated issues in traditional traffic simulation. In this paper the simulation is extended to include evolutionary aspects. We evolve populations of agents with personalities in order to assess whether the personalities that did better in the original scenario (in particular the "wayward" agents) are evolutionarily stable. We find that personalities which are more flexible than the wayward agents do better in evolutionary terms. © Springer-Verlag 2000.