Evolutionary models for recombination and learning: Analytical and computational approaches

This thesis examines evolutionary models of recombination and learning. I use analytical and computational methods to explore factors that influence the evolution of recombination, and to characterize environmental conditions under which learning is most likely to evolve.; The evolution of a selectively neutral modifier of recombination is studied under different conditions of selection on a set of major genes. In a finite population a simulation study is carried out in which the phenotype is computed additively from the diploid genotype at twenty genes. The fitness is taken to be a function of the phenotype and I show that when this function is very jagged, low recombination has a strong advantage. When the function is smooth and of the disruptive-selection kind, high recombination may be favored in both finite and very large populations. In a numerical study of disruptive selection on two loci in a large population it is shown that the evolution of recombination depends on the initial frequencies at the selected loci and on the exact shape of selection and strength of the selection. In general, when the selection is disruptive and very strong, it is possible to find conditions under which higher recombination will be favored.; Next, I develop models for the evolution of learning in a population-genetic framework. I have studied a model in which different genotypes have different capabilities to form representations of a randomly changing environment. The stochastic properties of the environment are outlined. Then a haploid genetic model with two alleles, one of which allows a limited capability to represent the environment, is described. This is extended to a two-locus framework. The analysis concludes with a generalization to a diploid one-locus model with different genotypes conferring on their carriers different abilities to represent the environment. I discuss the results of a stochastic simulation study of this model and compare these with the local stability analyses of the mathematical model. In particular, it turns out that conditions favoring initial increase of learning may not guarantee its eventual fixation in the population.