Year: 2012
Type: other
Abstract: We examine distribution of strategies in demographic Donor-Recipient game where patterns of move-play are restricted to simple structure. Players are initially randomly distributed in square lattice of cells. In each period, players move locally to random cell in neighbors or globally to random unoccupied cell in the whole lattice, and play multiple games against local neighbors or against randomly selected global players. We restrict patters of move (play) to local or global; local (global) means with high probability the player moves (plays) locally (globally). If wealth (accumulated payoff) of player becomes negative or his age becomes greater than his lifetime, he dies. If his wealth becomes greater than some amount and there is unoccupied cell in neighbors, he has an offspring. Players consist of TFT-like strategies; TFT uses Defect if the state is 0, Cooperate if it is 1 and the smallest state number that prescribes Cooperate is regarded as threshold. We allow up to three states and assume players may change their threshold at most once depending on their experience. We show, by Agent-Based Simulation, for example, Cooperative strategies evolutionarily tend to move and play locally, Defective do not, and AllC and AllD are abundant unless all strategies initially play locally.
Authors: Tsuneyuki Namekata, Yoko Namekata
Language: English
Subfield: Sociology and Political Science
Field: Social Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W2163006772