Non-cooperative two-person games in biology: A classification (Q1077339)

Evolutionary equilibrium concepts developed by theoretical biologists are classified within classical game theory which assumes rational players. It will turn out that these notions, although they are based on a highly non-teleological model involving neither heigher rationality nor even consciousness, mostly resemble some solutions in non-cooperative two- person game theory which are based on the idea that players do behave rationally, but whose thoughts and acts are sometimes disturbed by errors. The first section deals with some helpful results in bimatrix games needed for the mentioned classification and considers the biologically important case of symmetry in payoff. Many refinements of the Nash equilibrium concept are mentioned. The second section considers the idea of ''games'' between members of one or two populations which appeared first in connection with the evolution of the sex ratio as given by Fisher. Hamilton and Stewart were the first to use explicit game theoretic terminology. The author studies the ESS evolutionary stable strategy introduced by Maynard Smith and Price. In the final section he dynamizes the considered model by means of difference or differential equation systems, tools used frequently in many other fields of theoretical biology, such as population genetics or ecology. He follows Taylor-Jonker investigating here the asexual game dynamics continuous in time. Following Zeeman he finds that the same dynamical equilibrium with additional properties is a regular equilibrium point. Some relative important results are further considered.