Exploiting Social Capabilities in the Minority Game

Abstract

The minority game (MG) is an inductive binary-decision model that is able to study emergent behaviors in a population of agents who compete, through adaptation, for scarce resources. The original formulation of the game was inspired by the W.B. Arthur’s El Farol Bar problem: a fixed number of people have to independently decide, each week, whether to go to a bar having a limited capacity. People’s choices are only affected by the information about the number of visitors who attended the bar in the past weeks. Basic MG assumes that the information about the past game outcomes is publicly available, and it does not contemplate any communication among players. This article proposes the Dynamic Sociality Minority Game (DSMG). DSMG is an original variant of the classic MG where (1) information about the outcome of the previously played game step is only known to agents that really attended the bar the previous week, and (2) a dynamically established acquaintance network is introduced to propagate such information to nonattendant players. Specific settings of the game are identified in which DSMG is able to show a better coordination level among players with respect to the standard MG. Emergent properties of the DSMG along with players’ wellness are thoroughly analyzed through agent-based simulation of a simple road-traffic model.