Evolutionary Change and Darwinian Demons

Abstract

Models of adaptive evolution often have the property that change is guided by, but not fully determined by fitness. In a given situation many different mutant phenotypes may have a fitness advantage over the residents, and are thus potential invaders, implying that the mutational process plays an important role in deciding which particular invasion will take place. By introducing an imaginary 'Darwinian demon' in charge of mutations, one can examine the maximal role that mutation could play in determining evolutionary change. Taking into account pleiotropic mutations and shifting fitness landscapes, it seems likely that a Darwinian demon could exert considerable influence and most likely would be able to produce any viable form of organism. This kind of perspective can be helpful in clarifying concepts of evolutionary stability.

Joint evolution of altruistic cooperation and dispersal in a metapopulation of small local populations

“…A singular strategy s * is an evolutionary attractor (Eshel, 1983) if the repeated invasion of nearby mutant strategies into resident strategies will lead to the convergence of resident strategies towards s * . For scalar strategies, the condition for such convergence is D ′ (s * ) < 0, whereas the situation with vector-valued strategies is more complicated (Christiansen, 1991;Marrow et al, 1996;Matessi and Di Pasquale, 1996;Geritz et al, 1998;Leimar, 2001;Meszéna et al, 2001). Because cooperation and dispersal are physiologically rather different traits, we find it realistic to assume that these traits evolve independently.…”

Social and individual learning of helping in humans and other species

“…Where the two perspectives differ, however, is in their relative emphasis on inherited versus individually learned mappings: the literature on other-regarding preferences tends to treat individuals as irreversibly committed by genetic or cultural helping variants to express a fixed pattern of behaviors. The self-regarding literature, by contrast, suggests a highly plastic behavior map, which can be continually adjusted in a way that yields the actors the highest payoff in each new game [15], probably the closest we can get to a darwinian demon [87]. Rather than focusing on other-or self-regarding preferences of individuals, therefore, we believe that the debate over the nature of human helping would be more usefully focused on the effects and relative importance of behavioral plasticity, whether socially or individually learned.…”

Function-valued adaptive dynamics and optimal control theory

“…This canonical equation can be used to study the transient dynamics and convergence towards evolutionarily singular strategies (see, e.g., Heino et al, 2008). Already for vector-valued strategies, conditions for such convergence are more complicated than for scalar strategies (Christiansen, 1991;Marrow et al, 1996;Matessi and Di Pasquale, 1996;Geritz et al, 1998;Leimar, 2001;Meszéna et al, 2001), usually requiring dynamical analysis of the kind the canonical equation allows. In some simple cases, the equilibria of the canonical equation can be solved analytically, and singular strategies thus can be obtained.…”