Game Theory: Are People Really as Selfish as We Think?
The majority of research in Economics rests on the assumption that human beings, by nature, are selfish and are motivated to act selfishly in all aspects of everyday life. Like most areas of Economics, game theory relies on this assumption, as the Nash equilibrium occurs when each player maximizes his own payoff in relation to the other player’s strategy. However, according to a paper published by Southern Illinois University Edwardsville, when games are designed such that altruism can influence their decisions, people do not act purely out of self-interest: rather than simply playing according to a strategy that maximizes a player’s own payoff, a player will aim for a strategy that results in a positive payoff for the other player as well.
One game that the paper mentions that illustrates this is called the “Ultimatum” game. In this game, two players—the paper labels them “A” and “B”—are shown $10, and A has the option of offering a portion of that $10 to B, which B can accept or reject. If B accepts, then each receives the amount that A proposes; if B rejects, then they both walk away with nothing. Under the assumption that people act purely out of self-interest, then one would assume that B would accept any offer greater than $0, as accepting this will always allow B to leave with a greater payoff than if B rejects the offer. Thus, since A knows that B would accept any offer greater than $0, A’s best response is to offer the smallest amount greater than $0. However, according to the paper, studies conducted at universities found that A’s mean offer ranges between 42% and 48% of the $10. Under the assumption that A is acting purely out of his own self-interest, there should be no reason for him to offer that much; therefore, he must somehow also benefit from B receiving a substantial portion of the $10.
What is interesting about this paper is that it allows us to rethink Nash equilibria in certain games with the assumption that people do not operate purely out of self-interest. One of these games that we can rethink is Prisoner’s Dilemma. As discussed in class, the Nash equilibrium of a Prisoner’s Dilemma game is that both players defect rather than cooperate and receive a negative payoff. However, this strategy is only optimal if each player truly has no regard for the other player’s payoff, because the optimal strategy results in the other player receiving the most negative payoff possible. If people really are as altruistic as the paper suggests, then it would be reasonable to assume that the optimality of defecting would be reduced, and that it would be common to see situations where both players cooperate rather than defect.
Reference: http://www.siue.edu/~evailat/gt-behavior.htm