Experimental Evolution: Concepts, Methods, and Applications of Selection Experiments

19. Theoretical and Experimental Approaches to the Evolution of Altruism and the Levels of Selection

585 19 THEORETICAL AND EXPERIMENTAL APPROACHES TO THE EVOLUTION OF ALTRUISM AND THE LEVELS OF SELECTION Benjamin Kerr All mankind . . . is one volume; when one man dies, one chapter is not torn out of the book, but translated into a better language. JOHN DONNE, MEDITATION XVII THEORETICAL BACKGROUND A Simple Trait Group Model Defining Altruism Statistical Association and the Evolution of Altruism The Fitness Structure and Context Formation Statistical Equations and the Levels of Selection EXPERIMENTAL RESEARCH Artificial Selection Experiments Quasi-Natural Selection Experiments Natural Selection Experiments DISCUSSION Experimental Evolution: Concepts, Methods, and Applications of Selection Experiments, edited by Theodore Garland, Jr., and Michael R. Rose. Copyright © by the Regents of the University of California. All rights of reproduction in any form reserved. One of the central themes in Donne’s meditation is the interconnectivity between human beings. Our lives are not stand-alone chapters from an edited volume (like the one you are reading), but more like chapters from an elaborate novel, each setting the stage for chapters to come while simultaneously depending on chapters already read. Biological systems, from subcellular biochemical networks to multispecies food webs, display striking forms of interconnectivity in their parts. How does the theory of biological evolution by natural selection handle this interconnectivity? The simplest description of natural selection starts by ignoring interdependence. For simplicity, individuals in a population are assumed to affect neither each other’s fitness nor the form of their environment . As an example of the logic, consider the giant anteater. A standard story of natural selection would maintain that anteaters with longer, stickier tongues have been selected because they are able to gather more ants and termites. After all, it is these food resources that improve survival and the production of (long-sticky-tongued) progeny. However, tongue length of one anteater is assumed not to affect the fitness of another, and tongue stickiness is assumed not to influence the behavior or morphology of the insect prey. In effect, each anteater is treated as “an island, entire of itself” (Donne 1624/1839). Under the “organism-as-island” incarnation of natural selection, there is an extreme premium placed on personal fitness. An individual with a phenotype that best solves current environmental challenges “cashes in” by earning the highest personal fitness. Within this perspective, what are we to make of the existence of individuals that appear to sacrifice personal fitness to improve the fitness of others? Of course, the answer is that such behavior is fundamentally backward and should be swiftly eradicated by the action of natural selection. Individuals that exhibit restraint and self-sacrifice for the benefit of others are foolishly eroding the precious commodity of personal fitness and should be undone by selfish counterparts. Nonetheless, the biological world is filled with examples of altruistic leanings, from slime mold cells that sacrifice themselves to form the somatic stalks of fruiting bodies (on which reproductive spores sit) to the nonreproductive workers of eusocial insect colonies. A second reaction to the existence of altruists is that the organism-as-island version of natural selection is misleading when considering such behaviors, as compared with simpler behaviors, such as locomotion in an activity wheel (Rhodes and Kawecki this volume ; Swallow et al. this volume). By its very nature, altruism is a social activity. The fitness of an individual depends on the behavior of others. Social organisms do not simply solve the challenges of an external environment—they are the selective environment (or at least part of it). The interconnectivity ignored by the organism-as-island model is now front and center. Indeed, how organisms are connected turns out to be the critical issue. Most explanations for the evolution of altruism depend on altruists disproportionately finding themselves in the company of other altruists. For instance, organisms may interact with relatives (kin selection), organisms may condition their own altruistic behavior on the cooperative behavior of partners (reciprocal altruism), or organisms may exclude 586 • A P P L I C A T I O N S non-cooperating members from their pool of interactors (policing). If altruists are able to associate preferentially with other altruists, then the personal fitness cost can be viewed as the price of admission to a beneficial social milieu. Sometimes cooperative interactions between the members of a group lead to a type of functionality and cohesiveness at the group level. For instance, eusocial insect colonies have been called “superorganisms” to underline the degree...