I don’t think cooperate/defect are good action labels because it doesn’t seem much like a standard prisoner’s dilemma. It is not quite a game of Chicken either, but it is closer to Chicken than PD.
The mirror/mirror outcome in Ganch is like getting in a car crash in Chicken, because it is the worst for everyone and anyone who swerves unilaterally will make both happier with the outcome, and that outcome potentially functions as a threat to use against the other player to get your way if a conflict comes up.
Ganch is unlike Chicken in that Chicken’s swerve/swerve outcome gives an honorable tie, and the other person playing a different move causes you to lose outright, which is something you’d like to prevent if you can.
Contrast this with the carve/carve outcome in Ganch, which is worse than having the carving role in a carve/mirror outcome, because playing carve against mirror at least the room looks OK but playing carve against carve leaves you with an artistic mess and both of you look bad. So in Ganch (unlike Chicken) the players have the same worst and second worst outcomes and these outcomes are in a diagonal relationship to each other.
Basically, Ganch is a kind of coordination game and the closest “famously named” coordination game I know to Ganch is the 2x2 Battle Of The Sexes.
This name goes back at least as far as Duncan and Raiffa’s 1957 book “Games and Decisions: Introduction and Critical Survey”, reviewed here.
The basic dynamic in BotS is that you are playing a coordination game, but not a coordination game with perfect alignment of goals. It is generally silly to play coordination games without talking first, but in this variant the conversations are more delicate than otherwise because there’s some of self-interest and meta-game-fairness issues that come up when picking which of the two “basically acceptable” Nash equilibriums to aim for.
I don’t think cooperate/defect are good action labels because it doesn’t seem much like a standard prisoner’s dilemma. It is not quite a game of Chicken either, but it is closer to Chicken than PD.
The mirror/mirror outcome in Ganch is like getting in a car crash in Chicken, because it is the worst for everyone and anyone who swerves unilaterally will make both happier with the outcome, and that outcome potentially functions as a threat to use against the other player to get your way if a conflict comes up.
Ganch is unlike Chicken in that Chicken’s swerve/swerve outcome gives an honorable tie, and the other person playing a different move causes you to lose outright, which is something you’d like to prevent if you can.
Contrast this with the carve/carve outcome in Ganch, which is worse than having the carving role in a carve/mirror outcome, because playing carve against mirror at least the room looks OK but playing carve against carve leaves you with an artistic mess and both of you look bad. So in Ganch (unlike Chicken) the players have the same worst and second worst outcomes and these outcomes are in a diagonal relationship to each other.
Basically, Ganch is a kind of coordination game and the closest “famously named” coordination game I know to Ganch is the 2x2 Battle Of The Sexes.
This name goes back at least as far as Duncan and Raiffa’s 1957 book “Games and Decisions: Introduction and Critical Survey”, reviewed here.
The basic dynamic in BotS is that you are playing a coordination game, but not a coordination game with perfect alignment of goals. It is generally silly to play coordination games without talking first, but in this variant the conversations are more delicate than otherwise because there’s some of self-interest and meta-game-fairness issues that come up when picking which of the two “basically acceptable” Nash equilibriums to aim for.
And yet carve/carve is what they were expected to do and be judged for...
They set up the problem really badly, then. What they asked for would not have looked good at all.
Perhaps the loser had to redo his part after the winner’s. Oh wait...:)