When does evolution optimize?

作者：Metz J A J^*; Mylius S D; Diekmann O

来源：Evolutionary Ecology Research, 2008, 10(5): 629-654.

• eco-evolutionary feedback
• environmental dimension
• evolutionary optimization
• invasion fitness
• life-history theory
• r-optimization
• R(0)-optimization

Aim: To elucidate the role of the eco-evolutionary feedback loop in determining evolutionarily stable life histories, with particular reference to the methodological status of the optimization procedures of classical evolutionary ecology.
Key assumption: The fitness p of a type depends both on its strategy X and on the environmental E, p = p(X, E), where E comprises everything, biotic and abiotic, outside an individual that may influence its population dynamically relevant behaviour. Through the community dynamics, this environment is determined (up to non-evolving drivers) by the resident stratergy X(r): E = E(aur)(X(r)).
Procedures: Use the indicated notation to derive necessary and sufficient conditions for the existence of an evolutionary optimization principle, and for the reduction of such a principle to straightforward r- or R(0)-maximization. Develop quick tests to diagnose whether an eco-evolutionary model supports an optimization principle.
Results: It is necessary and sufficient for the existence of an optimzation principle that the stratergy affects fitness in an effectively monotone one-dimensional manner, or equivalently, that the environment affects fitness in an effectively monotone one-dimensional manner. In particular, there should exist functions psi of the strategies and phi of the environments such that sign[p(X,E)] = sign[psi(X) + phi(E)]. Pairwise invasibility plots of an eco-evolutionary model that supports an optimization principle have a special, easily recognizable shape. Natural selection just maximizes r, or R(0), if and only if r(X, E) can be written as alpha(r(X, E(0)), E) or R(0)(X, E) can be written as exp[alpha(In[R(0)(X, E(0))], E)], with alpha increasing in its first argument, and E(0) fixed, but otherwise arbitrary.
Conclusion: A pure optimization approach holds water only when the eco-evolutionary feedbacks are of a particularly simple kind.

• 出版日期2008-7
• 单位国际应用系统分析学会(IIASA)