Hisashi Ohtsuki
(Department of Biology, Faculty of Sciences, Kyushu University)

06/01/24, 13:30 at Room 1203 (2nd floor of building 1 of the Faculty of Sciences)

I study the evolution of cooperation in social relationships. In particular I study indirect reciprocity. In this mechanism the donor of help cannot receive reciprocation back from the beneficiary but from a third party. This is achieved through social information such as reputation of individuals. The conditional strategy prescribes cooperation with good persons but not with bad persons makes it possible for cooperation to be stably sustained in a population. My talk consists of three parts. First I investigate the evolvability of "discriminator" strategy by Nowak and Sigmund (1998). By using adaptive dynamics I study a continuum of reactive strategies and check whether or not the discriminator strategy is an ESS (evolutionarily stable strategy). As a result I find that it cannot be an evolutionary outcome, because discriminators lack the concept of justified punishment. Second I study the definition of goodness. There are a huge number of possible rules that judge the goodness of the action of donors. I call that moral rule "reputation dynamics" and study what reputation dynamics can realize highly cooperative society. Out of 4096 possible combinations of reputation dynamics and behavioral strategies, I find eight ESS pairs which maintain the highest level of cooperation in a society. I call those "leading eight". I show that the leading eight are considerably robust against cost-benefit ratio of donation or error rates. Third, I explore the nature of the leading eight in depth and try to give an intuitive explanation behind those morals. Also I reveal the relationship between reputation dynamics and behavioral strategies in the leading eight

Back: Japanese / English