- an integrative framework for modelling frequency-dependent evolution -
The theory of adaptive dynamics provides a framework for modelling evolution by natural selection in complex ecological systems, where fitness depends on the frequencies of the interacting phenotypes. This framework integrates and extends concepts and techniques from evolutionary game theory, with special emphasis on dynamical phenomena such as the origin and divergence of new lineages by evolutionary branching. The general theory of adaptive dynamics yields algorithms that can readily be applied to analyse concrete ecological settings.
The most intriguing phenomena discovered is evolutionary branching, whereby a population that initially contains a single phenotype splits up into two lineages with increasingly different phenotypes. Evolutionary branching occurs at particular phenotypes (called branching points) that are attractors of monomorphic evolution but are not evolutionarily stable in the sense that are not immune to invasion by mutants. Near such a phenotype, coexistence of very similar types (such as the original and its mutant) is possible. Once the population has become dimorphic, selection is disruptive and consequently the coexisting phenotypes undergo divergent coevolution. Evolutionary branching is the clonal reminiscent of non-allopatric speciation, and ecological systems where evolutionary branching occurs provide the selective environment for speciation.
A great strength of adaptive dynamics is its capability to incorporate ecological complexity and to model long-term evolution as driven by ecological interactions. The theory of the ESS has been enormously successful in analysing evolution in complex ecological and behavioural interactions. Adaptive dynamics is an extension of the ESS-theory that preserves its wide applicability to diverse ecological systems. Applications already include models exploring the evolution of competitive and predator-prey systems, mutualism, multiple habitats, temporally fluctuating environments, host-parasite systems, sex allocation, seed ecology, dispersal strategies, etc.
From the mathematical point of view, adaptive dynamics deals with a novel type of stochastic dynamical systems, with the unorthodox property of changing the number of state variables (evolving phenotypes) by extinction and by evolutionary branching.
Genetic complexity is incorporated at various levels ranging from simple clonal inheritance in many phenotypic models to single-locus (or few-locus) diploid models and to the most realistic multilocus individual-based simulations. The genetically explicit models support the conclusions from the analytically more tractable phenotypic approach. Most notably, several studies show that assortative mating can evolve in the selective environment present during evolutionary branching, which thus gives the first step towards the formation of new, reproductively isolated species in sympatry.
At present, the theory of adaptive dynamics is best developed for one-dimensional phenotypes and for monomorphic populations. Intense research concentrates on multiple-trait evolution, on coevolving multispecies communities, and on ecological systems with complex population dynamics including multiple attractors and chaos. A bifurcation theory of adaptive dynamics is currently being developed. New insights may be expected from the analysis of genetic models, especially in sympatric speciation and in the evolution of genetic systems.
For further information, see a list of papers on adaptive dynamics.
Other links: Adaptive dynamics and the evolution of biodiversity (PDF, in Hungarian)