CAT 2017 Slot 2 VARC: What Animal Contests Owe to Game Theory

It is one of the odder achievements of twentieth-century biology that a mathematical framework invented to analyse poker strategy and Cold War nuclear standoffs turned out to explain, with striking precision, why animals behave the way they do toward one another in contests over mates, territory and food. Evolutionary game theory, developed chiefly in the 1970s, begins from a deceptively simple premise: an animal's best strategy in a contest with another member of its species is not fixed but depends on what strategy the rest of the population is playing, and natural selection, acting over many generations, will tend to favour whichever strategy performs best against the mix of strategies actually present, rather than against some hypothetical ideal opponent.

The clearest illustration is the so-called Hawk-Dove game, a model of contests over a resource, say, a territory or a mate, between individuals who can either fight aggressively ("Hawk") or display and retreat without physical contact ("Dove"). If every member of a population plays Dove, a single mutant Hawk does extremely well, since it wins every contest without ever risking injury, and Hawk strategies will spread. But if the population becomes entirely Hawk, individuals now pay a steep cost every time they meet, since every contest escalates into a costly fight, and a Dove mutant, who simply concedes without fighting, avoids these costs and can do comparatively better. Neither pure strategy, then, is what game theorists call an evolutionarily stable strategy, a strategy that, once common in a population, cannot be invaded by any rare alternative strategy; instead, the population settles at some stable mixture of Hawks and Doves, or equivalently, individuals who play Hawk some fraction of the time and Dove the rest, with the exact proportion set by the relative costs of fighting and the value of the resource being contested.

What makes this framework more than a mathematical curiosity is how well its predictions match observed animal behaviour across many species, from the ritualised, rarely injurious combat of red deer stags competing for mates, to the way certain fish species assess a rival's size before deciding whether to escalate a confrontation or retreat. Crucially, the theory does not require that animals consciously calculate payoffs; natural selection can produce the same outcome by simply favouring, over many generations, whatever behavioural tendencies happen to perform best given what other members of the species are doing, in exactly the way that market competition, without anyone designing it, can produce prices that reflect supply and demand. This is perhaps the framework's deepest lesson: strategic equilibrium can emerge from a process, natural selection, that has no strategist and no foresight at all, and that fact alone should give pause to any assumption that the appearance of strategic sophistication in nature implies conscious deliberation behind it.