Game Theory May Explain How Personalities Evolved In Humans And Other Social Species

Game theory can be used to predict the choices of individuals based on the behaviour of other individuals.

It appears that watching the behaviour of others makes people more socially aware, which in turn exagerates the personality traits of both individuals. This has the impact of making some people more cooperative as they became aware of the effect their decisions are having on their own reputations. On the other hand, some individuals become less cooperative and even actively exploit more trusting individuals. This evolutionary self-perpetuating variation in personalities, begets more variation, so exaggerating the differences between the cooperative trusting and exploitative selfish individuals.

John McNamara, Philip Stephens, Sasha Dall, and Alasdair Houston have recently published evidence for the above in their paper - Evolution of trust and trustworthiness: social awareness favours personality differences - in Proceedings of The Royal Society B (February 2009). They modelled an infinite population of actors playing an asymmetric trust and cooperation game - see diagram - to produce their findings.



Decision tree for the trust and cooperation game

Abstract
Interest in the evolution and maintenance of personality is burgeoning. Individuals of diverse animal species differ in their aggressiveness, fearfulness, sociability and activity. Strong trade-offs, mutation–selection balance, spatio-temporal fluctuations in selection, frequency dependence and good-genes mate choice are invoked to explain heritable personality variation, yet for continuous behavioural traits, it remains unclear which selective force is likely to maintain distinct polymorphisms. Using a model of trust and cooperation, we show how allowing individuals to monitor each other's cooperative tendencies, at a cost, can select for heritable polymorphisms in trustworthiness. This variation, in turn, favours costly ‘social awareness’ in some individuals. Feedback of this sort can explain the individual differences in trust and trustworthiness so often documented by economists in experimental public goods games across a range of cultures. Our work adds to growing evidence that evolutionary game theorists can no longer afford to ignore the importance of real world inter-individual variation in their models.

The full paper can be downloaded as Full Text (Free) or Full Text (PDF) Free.

doi: 10.1098/rspb.2008.1182 Proc. R. Soc. B 22 February 2009 vol. 276 no. 1657 605-613