Game theory and human evolution: A critique of some recent interpretations of experimental games
Found through Cornell Summons
Blog 2
Game theory and human evolution: A critique of some recent interpretations of experimental games
This research paper studies and critiques past evolutionary theories in relation to game theory. One such critique is based on the fact that some animals, such as humans, would arbitrarily in certain one-time situations choose to harm another person while receiving a negative payoff themselves (aka revenge).
Evolutionarily stable strategy explains some behaviors that evolved through selection, which is apparently closely linked to the concept of Nash equilibrium.
However, to more accurately explain behavior, it is helpful to use aspects of psychological and social science into the game theory involved in explaining evolutionary behaviors.
The instance where an animal or person is willing punish others at the expense of themselves is called strong reciprocity. This typically occurs when someone violates the norms of a social group or type. Experiments have also shown that in an ultimatum game where a proposer proposes a certain amount of money for a responder to accept, sometimes the responder would choose instead to not accept and get 0 payoff, whereas had the offer been accepted, both players would have gotten a positive payoff. This occurs when two interacting people are strangers interacting anonymously.
One proposed explanation is that inequity causes a certain kind of negative payoff, and that people strive to minimize that in favor of everyone simply getting a positive monetary payoff. These non-maximization scenarios of payoffs in groups are called “altruism”. “Punishment” goes to defectors from the norms of a group, and altruism decreases when defectors decrease. This is called “cultural group selection”.
This selection involves conformist transmission, where people try to behave the same way, and prestige-biased transmission, where successful people are imitated.
Another theory to explain non-maximizing behaviors is the mismatch hypothesis. For example, evolutionarily in the past, venomous snakes and spiders were truly dangerous to humans. However, today, they kill less than 20 people per year compared to tens of thousands of deaths due to automobile and electronic accidents. Yet, humans don’t run away from cars, but most are afraid of snakes and spiders. Thus, this explains many weird negative human behaviors as a result of our evolutionary past, where the absence of “retaliation” often meant a much more negative payoff compared a successful retaliation.
In their conclusion, the researchers believe that experimental game results do show that most people genuinely care about others. However, they do not believe that this altruism is a group-selected norm due to lack of evidence.