A new TRN article discusses a new quantitative model of agent competition. The new model has potential uses in many fields including the optimization of communication and control within robot collectives. Traditional game theory has limitations in describing complex interaction in collectives of intelligent agents. The new theory allows the agents to make choices based on incomplete information. Even though mistakes are made, each agent has access to the knowledge and experiences of its neighbors allowing the collective to evolve complex dynamics quickly. One of the more interesting results is that the a scalable leadership system of "hubs" within the collective evolves spontaneously. For a more in-depth look at the theory and research see the original research paper: Competition in Social Networks: Emergence of a Scale-free Leadership Structure and Collective Efficiency (PDF format). The research was done by M. Anghel, Zoltan Toroczkai, Kevin E. Bassler, and Gyorgy Korniss. It was funded by grants from the Department of Energy and National Science Foundation.