This page is part of François Rousset's minimal web site. Access the main page here.

Social evolution

Individuals compete as much as the cooperate with their neighbors, and the evolutionary outcome of such interactions can be understood in terms of analytical partitions that identify forces common to different evolutionary scenarios. For example, Hamilton’s much-discussed inclusive fitness concept is a partition of the fitness of bearers of an allele in terms of the effects of each individual’s phenotype on his or her fitness, and of the effects of other individuals’ phenotype on this fitness. Beyond solving the substantive scientific issues in the group selection debate a long time ago, this approach leads to a number of efficient approximations for analyzing complex models.

In my work, I have investigated how isolation by distance, complex demographies, multilocus interactions, or disruptive selection can be understood in terms of such partitions. Some introductions to this work are

 Lehmann and Rousset (2010) which reviews how demography affects selection processes from an ecological perspective (in more technical terms, it reviews applications of the weak-selection version of inclusive fitness in one-locus models under various demographic scenarios) ;

 Rousset 2006 investigates the population genetic basis of inclusive fitness fitness argument and of diffusion models under isolation by distance ;

 Rousset and Roze (2007) elucidate the conditions for the evolution of genetic kin recognition, and illustrate techniques for analyzing multilocus models ;

 Ajar (2003) discusses disruptive selection and evolutionary "branching". (What "Ajar" has to do with this ? Read Romain Gary’s Vie et mort d’Emile Ajar)

This work has been applied, often through collaborations, to questions about dispersal evolution, sex ratio evolution, and cooperation syndromes. For example, see Lehmann et al. (2007) on debates about cooperation among humans. Leturque and Rousset (2004) shows that intersexual resource competition may explain sex ratio biases in mammals, building on the still little recognized fact that the sex ratio should not necessarily be biased toward the sex that disperses more.

In more details, such works have included :

 deriving simple perturbation results for fixation probabilities of mutants in nontrivial ecological scenarios (Rousset and Billiard (2000)). Lehmann et al (2007) usefully comment on what has been done with this idea ;

 a no-nonsense, simple but general way of defining of relatedness coefficients used in kin selection theory (Rousset and Billiard 2000 again, and Rousset 2002) ;

 showing how what is often described as "non-additive interactions" is easily handled by weak-selection inclusive fitness techniques. See Roze and Rousset (2003) for the first development and Ohtsuki (2010) for a more recent discussion.

 the development of multilocus models in spatially structured population : see Roze and Rousset (2008) for some relatively general formulation and Roze and Rousset (2009) for a quantitative model of dispersal evolution in response to heterosis. For perturbation expansions of multilocus fixation probabilities see e.g. Lehmann and Rousset (2009) and Lehmann et al (2009) ;

 deriving selection gradients in ecological scenarios with demographic feedback (Rousset and Ronce 2004). See in particular an application by Lehmann et al. 2006 to "group selection" scenarios.