5 ms·

The author is seriously mistaken about game theory. The Nash Equilibrium requires every player to be maximizing their payoffs given what the other players are d

The author is seriously mistaken about game theory. The Nash Equilibrium requires every player to be maximizing their payoffs given what the other players are doing. But game theory itself does not say what the nature of the payoffs are: there is no problem in modelling altruistic players who attach a high payoff to outcomes which are good for everyone and low payoffs to outcomes which benefit only themselves. The Pareto Optimal is not an outcome which is best for both, since the whole point of this concept is to avoid giving a definition of "best for both". Rather it is an outcome from which no player can be made better off without making some other player worse off. So, for example, for two greedy people sharing a cake, any division of the cake, including one which gives the entire cake to one person, is Pareto optimal. Anyone seriously interested in understanding what game theory really says and avoiding superficial mistakes in its interpretation should read at least the first chapter of Binmore's 'Playing for Real'.

robbrown451 12y ago
I think when you allow for altruistic players, the core insights you get from game theory tend to vanish. I think the author had that part mostly right, actually.
- yomritoyj 12y ago
  But if you really are a selfish player, and you have somehow persuaded the other player to trust you and play "deny" then you should definitely play "confess" to maximize your payoff. With selfish players there is no way to move to ("deny","deny") in a one-shot Prisoners' Dilemma. Cultural change cannot help unless it amounts to a change in payoffs. [The story would have been different in a coordination game with multiple Pareto-ranked equilibria where a change in beliefs can move you from a bad equilibrium to a good equilibrium.]
- Symmetry 12y ago
  In the case of a perfectly selfless person playing with a selfish person yes, that is a place where game theory ceases to be useful. But real altruists are imperfect, and have somewhat different notions of utility and planning horizons. Some games cease to be interesting but others turn from pure conflict to interesting games. In fact, most of the examples I remember from The Strategy of Conflict are among groups that generally mutually altruistic, like parent and child or soldiers fighting on the same side in a war. Of course, games played among roughly altruistic players tend to have much better outcomes...
dropit_sphere 12y ago
> there is no problem in modelling altruistic players who attach a high payoff to outcomes which are good for everyone and low payoffs to outcomes which benefit only themselves. Isn't this something of a contradiction in terms? But say it's not. Aren't such players likely to die out in a lot of ecologies? Deriving utility from helping others won't stop you from starving.
- dragonwriter 12y ago
  > Aren't such players likely to die out in a lot of ecologies? Deriving utility from helping others won't stop you from starving. It might, as it gives other (selfish) actors a reason to prefer you to survive over a selfish competitor (other than themself.)
- yomritoyj 12y ago
  It is not necessarily a contradiction as long as you accept the existence of two different ways of ranking outcomes: one for desirability and one for decision-making. When a cake is being shared I may want to eat the whole cake alone, but I may still offer you half the cake out of a sense of fairness. For the predictions of game theory it is the decision-making ranking and not the desirability one which enters into payoffs. If you deny the meaningfulness of talking about different ways of ranking then of course any notion of altruism becomes self-contradictory: "He gains enjoyment by starving himself to feed the poor, that's why he does it." But in practice we distinguish between masochists and saints. Helping others may help you survive if selfish people are ostracized by others (as seem to happen in human societies) or if a good turn today entitles you to a good turn tomorrow. But the evolutionary theory of altruism is not something I understand very well. My only point was that contrary to what the OP suggests, the formal techniques of game theory such as Nash equilibrium are agnostic with respect to nature of payoffs. And both casual observation and systematic experiments in economics show that many humans act out of a sense of fairness and expect fairness from others even when interacting with strangers.

robbrown451 12y ago

I think when you allow for altruistic players, the core insights you get from game theory tend to vanish. I think the author had that part mostly right, actually.

yomritoyj 12y ago
But if you really are a selfish player, and you have somehow persuaded the other player to trust you and play "deny" then you should definitely play "confess" to maximize your payoff. With selfish players there is no way to move to ("deny","deny") in a one-shot Prisoners' Dilemma. Cultural change cannot help unless it amounts to a change in payoffs. [The story would have been different in a coordination game with multiple Pareto-ranked equilibria where a change in beliefs can move you from a bad equilibrium to a good equilibrium.]
Symmetry 12y ago
In the case of a perfectly selfless person playing with a selfish person yes, that is a place where game theory ceases to be useful. But real altruists are imperfect, and have somewhat different notions of utility and planning horizons. Some games cease to be interesting but others turn from pure conflict to interesting games. In fact, most of the examples I remember from The Strategy of Conflict are among groups that generally mutually altruistic, like parent and child or soldiers fighting on the same side in a war. Of course, games played among roughly altruistic players tend to have much better outcomes...

dropit_sphere 12y ago

> there is no problem in modelling altruistic players who attach a high payoff to outcomes which are good for everyone and low payoffs to outcomes which benefit only themselves. Isn't this something of a contradiction in terms? But say it's not. Aren't such players likely to die out in a lot of ecologies? Deriving utility from helping others won't stop you from starving.

dragonwriter 12y ago
> Aren't such players likely to die out in a lot of ecologies? Deriving utility from helping others won't stop you from starving. It might, as it gives other (selfish) actors a reason to prefer you to survive over a selfish competitor (other than themself.)
yomritoyj 12y ago
It is not necessarily a contradiction as long as you accept the existence of two different ways of ranking outcomes: one for desirability and one for decision-making. When a cake is being shared I may want to eat the whole cake alone, but I may still offer you half the cake out of a sense of fairness. For the predictions of game theory it is the decision-making ranking and not the desirability one which enters into payoffs. If you deny the meaningfulness of talking about different ways of ranking then of course any notion of altruism becomes self-contradictory: "He gains enjoyment by starving himself to feed the poor, that's why he does it." But in practice we distinguish between masochists and saints. Helping others may help you survive if selfish people are ostracized by others (as seem to happen in human societies) or if a good turn today entitles you to a good turn tomorrow. But the evolutionary theory of altruism is not something I understand very well. My only point was that contrary to what the OP suggests, the formal techniques of game theory such as Nash equilibrium are agnostic with respect to nature of payoffs. And both casual observation and systematic experiments in economics show that many humans act out of a sense of fairness and expect fairness from others even when interacting with strangers.