Adaptive role switching promotes fairness in networked ultimatum game. Wu, T., Fu, F., Zhang, Y., & Wang, L.
Adaptive role switching promotes fairness in networked ultimatum game [pdf]Paper  abstract   bibtex   
In recent years, mechanisms favoring fair split in the ultimatum game have attracted growing interests because of its practical implications for international bargains. In this game, two players are randomly assigned two different roles respectively to split an offer: the proposer suggests how to split and the responder decides whether or not to accept it. Only when both agree is the offer successfully split; otherwise both get nothing. It is of importance and interest to break the symmetry in role assignment especially when the game is repeatedly played in a heterogeneous population. Here we consider an adaptive role assignment: whenever the split fails, the two players switch their roles probabilistically. The results show that this simple feedback mechanism proves much more effective at promoting fairness than other alternatives (where, for example, the role assignment is based on the number of neighbors). T he Ultimatum Game has been catching up with the Prisoner's Dilemma in characterizing and elucidating the issues surrounding the emergence and persistence of altruistic behavior. In a typical ultimatum game, two players are asked to divide a sum of money. Of the two players, one acts as the proposer and the other as the responder. The proposer proposes a portion of the money to give to the responder, whose reaction shall decide their own incomes. If the responder accepts the offered money, the remaining money flows into the proposer's account. Otherwise, they each get nothing. Obviously, a rational responder should accept any amount of the money, however small the amount is, as the opposite choice leaves him with zero income. The backward induction leads to the subgame perfect equilibrium prediction 1 that the proposer should claim almost all of the money and leave the remaining to the responder. However, deluging experimental findings are at odds with this purely analytical reasoning. These experiments evidence that most individuals do give 40 to 50% of the total money, and around half of their co-players reject offers below 30% 2–6

Downloads: 0