The Game Theory of Evolution
Animals within a species competing for food, shelter or mating partners oftentimes display conflict that can be best described as a ‘limited war’ (Smith). This means that whereas in the normal notion of a war, opponents fight to kill or injure their rivals, in this limited war ideology, opponents seek to prove their dominance over opponents through less dangerous methods. This is beneficial to animal species which partake since competition for these resources does not lead to dangerous fighting which would diminish population and health of the species as a whole. Therefore the species can succeed and continue to grow even when it faces competition from its own members.
This “limited war” is clearly beneficial to the whole species, but Smith and Price wanted to determine whether or not it is also beneficial to the individual members of a species. They set up a simulation of animals which react in various ways in conflict to see who would fare the best in these situation. They made the simulation into an action- reaction- reaction- etc. between two players which choose from three possible moves. The three options which they can choose are:
C, which is ‘conventional tactics’ unlikely to cause serious injury
D, or ‘dangerous tactics’ which are likely to cause injury and.
R, or ‘retreat’, which ends the round. the one which does not retreat is the winner, (even if it is seriously injured)
Next, there are 5 types of animals, which are the same species but just called different things based on how they act to the conflict.
First there is ‘Mouse’, which never plays D. it immediately retreats if it receives an action of D by an opponent (Smith).
Second, ‘Hawk’, always plays D and continues until he is seriously injured or opponent retreats (Smith).
Third is the ‘Bully’ which always plays D if it has the first move, or in reaction to C, while it plays C in response to D (Smith).
Fourth we have ‘Retaliator’, which plays C if it has the first move and plays C in response to C, but will eventually retreat after a preassigned amount of rounds. If opponent plays D, the retaliator will generally play D as well (Smith).
Fifth and last there is the ‘Prober-Retaliator’ which is like the retaliator except it has a small probability of defecting to D at each round (Smith).
After running the simulation, the researchers used the results to make a payoff matrix to understand how well off each strategy is against all of the other strategies, to determine if it is evolutionarily stable. It was found that depending on the frequency of other reaction types, it was either the retaliator or the prober-retaliator which fared the best overall (Smith). Based on the averages, researchers determined that these strategies would be most likely to be evolutionarily stable (Smith). Furthermore this strategy uses general cooperation by using conventional tactics for the most part, which supports Smith and Price’s hypothesis that acting in a less dangerous manner is beneficial to the individual, not just the species (Smith).
This shows us that the concepts of game theory can be applied to real situations in nature. The researchers did note that they made many oversimplifications of how animals behave in conflict which impact some of the nuances of their results. However it is sufficient to show the benefit of this general concept of ‘limited warfare’.
Smith, J. Maynard, and G. R. Price. “The Logic of Animal Conflict.” Nature, vol. 246, no. 5427, 1973, pp. 15–18., doi:10.1038/246015a0.