cally consuming resources shared by others (such as certain yeast cells) have evolved metabolisms that use the fast-but-wasteful approach to producing ATP. In multicellular organisms, though, cells behave more cooperatively with their neighbors, evolving reaction pathways that produce ATP more efficiently.

Intriguingly, cancer cells seem to violate the cooperation strategy and behave more selfishly (in terms of using inefficient ATP-producing processes). Game theory hasn’t exactly cured cancer yet, but insights into such properties of cancer cells may contribute to progress in fighting it.

On a higher evolutionary level, a combination of network math and game theory may be able to explain more advanced forms of human cooperative behavior. Evolutionary game theory’s assault on the cooperation problem—how altruistic behavior can evolve in societies of seemingly selfish individuals—has relied mainly on playing the Prisoner’s Dilemma game under a variety of circumstances. In some versions of the game, the players (or agents) may encounter anybody else in the population and then decide whether to defect or cooperate. In one version, though, the agents face such decisions only in interactions with their immediate neighbors (the game, in other words, is “spatially structured”). It appears that cooperation is more likely to evolve in games with spatial constraints, at least when the game is the Prisoner’s Dilemma.

But perhaps the Prisoner’s Dilemma does not always capture the essence of real life very accurately. Life might sometimes more closely resemble a different kind of game. One candidate is the “snowdrift” game, in which the best strategic choice differs from the classic Prisoner’s Dilemma. In a Prisoner’s Dilemma, each player earns the highest payoff by defecting, regardless of what the other player does. In the snowdrift game, your best move is to defect only if your opponent cooperates. If the opponent defects, you are better off cooperating.12 As it turns out, spatial constraints also influence the evolution of cooperation in the snowdrift game, but in a different way—inhibiting cooperation rather than enhancing it. That is a perplexing finding, calling into question game theory’s validity for studying the cooperation issue.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement