2003/01/14 by Daniel Eriksson, Henrik Jeldtoft Jensen
Biochemistry, Genetics and Molecular Biology · Physics and Astronomy · Social Sciences · #Evolution and Genetic Dynamics #Evolutionary Game Theory and Cooperation #Gene Regulatory Network Analysis #cond-mat.stat-mech #q-bio.PE
paper · pdf · doi:10.1088/1742-5468/2004/09/p09001
published as JStat P09001 (2004) · 4 pages, 5 figures
arxiv created 2003/01/14 · openalex publication_date 2004/09/03 · arxiv updated 2009/11/30 · openalex created_date 2016/06/24 · openalex updated_date 2026/07/30
We study the action of Darwinian selection on the phenotype using a collection of agents playing local games with no pure Nash equilibria. After each cycle, the agent who has lost the most games is replaced by a new agent with a new random strategy and game partner. The network generated exhibits a very broad power-law-like distribution of lifetimes of the agents. The model is interesting in the sense that surviving agents are forced to keep adapting to their environments, no matter what the evolution of the network dynamics, while the specific tasks and strategies of agents are not pre-specified. Hence both the tasks and strategies of agents are self-organized properties of the network. While the underlying games are unstable, supporting only single winners, it is found that surviving agents are not extraordinarily successful, but rather tend to perform similarly well as their environments. Yet agents display a broad range of different behaviours (strategies).