The Interdisciplines website for research in philosophy, cognitive science and social science is currently running an online conference on adaptation and representation. Discussion has just closed on an interesting paper by Ryan McKay and Daniel Dennett on the evolution of misbelief, and a new paper by Robert Brandon on biological adaptation and function has just gone live. The conference is due to last through to July and there is an exciting line-up of papers to come. Readers are invited to comment on live papers, and the discussions are archived on the site. It’s an excellent project.
Why do we play games? Is there an evolutionary story for this?
David, a good friend of mine, asked me this a while back and I thought it might be interesting to begin a discussion on this here. To stimulate this I will ask my own questions. Is anyone aware of a literature on the evolution of play? Are there obvious ways to test between different ideas? I also think, more generally, that this is a nice context within which to explore some familiar problems involved in putting together accurate evolutionary explanations.
These were/are my first thoughts on why we play games:
1. Play is practice. It is a form of trial-and-error learning with the advantage that the error signal is less costly than that of the equivalent adult activity. For example, a kitten that plays with inanimate objects or with small prey can do so with a minimal error-associated injury risk. The maturing cat can therefore gain competence (to reduce the risk of error) before hunting larger prey that can bite back. This suggests that play is an adaptive trait and we might predict its occurrence in the early stages of learning a new skill. If adults were to play hunting games or, if juveniles always hunted and devoured small nutritive prey, it might be hard to justify this account.
2. Play is a by-product. Play occurs because motivational systems that make us enjoy adult activities can be stimulated out of context. For example, cats are motivated to hunt and might engage in play hunting because they are motivated by conspecifics or mere movement in a risk-free context.
3. Play is a kind of contest. Some or all of what we think of as games might be means to address real conflicts. Conflicts are often settled by display and ritualised behaviour rather than outright fighting because both weaker and stronger individuals in a conflict benefit from retreating and not fighting, respectively (these conclusions have long been established from game theoretic mathematical models of fighting tactics). Kin selection might increase play between family members where the costs of losing are shared in part by both party’s genes (this is hard to test though because the probabilities of meeting have to be factored in).
Explanations 1 and 2 are not mutually exclusive. In particular number 2 does not exclude number 1 and the fact that play behaviour often requires restraint (soft biting and retracted claws in our cats) suggests an adapted (if not adaptive) element. Both explanations predict that in a species where there is a division of labour between the sexes we should expect a sex difference in play behaviours. A rather amusing example of this, for those who posit cultural stereotyping, can be seen in vervet monkeys (Alexander and Hines (2002) Sex differences in response to children’s toys in nonhuman primates (Cercopithecus aethiops sabaeus). 23:467-79; see below for link).
But maybe a more radical adaptationist hypothesis of play is possible: has anyone any ideas? And do we need more elaborate theories to explain the evolution of abstract games with fixed rules in human cultures? Is it correct to say that chess or go evolved because they were consciously designed, by individuals, to foster an aptitude for war strategies amongst players? When were the rules written?
For link to vervet monkey study mentioned above see:
After a lengthy hiatus, Evolving Ideas is relaunching this month as a forum for the discussion of evolutionary issues generally, with an interdisciplinary perspective. The owners and main contributors are Ben Dickins, Tom Dickins, and Keith Frankish – respectively a geneticist, an evolutionary psychologist, and a philosopher of psychology. We shall be also recruiting new contributors soon, and we aim to maintain a regular stream of postings. Please check back regularly and feel free to add your comments.