Moderate Contact between Sub-populations Promotes Evolved Assortativity Enabling Group Selection
Snowdon, James R.,
Watson, Richard A. (2011). Moderate Contact between Sub-populations Promotes Evolved Assortativity Enabling Group Selection. In: Kampis, G.,
Szathmáry, E. (Eds.) Advances in Artificial Life. Darwin Meets von Neumann. Lecture Notes in Computer Science, 5778/2011, () ( ed.). (pp. 45-52). : . Springer.
Group selection is easily observed when spatial group structure is imposed on a population. In fact, spatial structure is just a means of providing assortative interactions such that the benefits of cooperating are delivered to other cooperators more than to selfish individuals. In principle, assortative interactions could be supported by individually adapted traits without physical grouping. But this possibility seems to be ruled-out because any 'marker' that cooperators used for this purpose could be adopted by selfish individuals also. However, here we show that stable assortative marking can evolve when sub-populations at different evolutionarily stable strategies (ESSs) are brought into contact. Interestingly, if they are brought into contact too quickly, individual selection causes loss of behavioural diversity before assortative markers have a chance to evolve. But if they are brought into contact slowly, moderate initial mixing between sub-populations produces a pressure to evolve traits that facilitate assortative interactions. Once assortative interactions have become established, group competition between the two ESSs is facilitated without any spatial group structure. This process thus illustrates conditions where individual selection canalises groups that are initially spatially defined into stable groups that compete without the need for continued spatial separation.
+44 131 455 2718
Areas of Expertise
See all areas of expertise
The Bio-Inspired Algorithms group within the Centre for Algorithms, Visualisation and Evolving Systems is a large and thriving group with interests in nature-inspired computing that include Evolutionary Computing, Hyper-Heuristics, Artificial Immune Systems and Swarm Intelligence.