This study shows how larger group size can enhance punishing behavior in social dilemmas and hence support higher levels of cooperation. Unlike existing approaches that focus on uncoordinated punishment, I view punishment to be a coordinated activity ...
This study shows how larger group size can enhance punishing behavior in social dilemmas and hence support higher levels of cooperation. Unlike existing approaches that focus on uncoordinated punishment, I view punishment to be a coordinated activity which may be resisted by those being punished and study the conflict situation induced by punishment. By developing the conflict model describing such situation and combining it with the n-player prisoner dilemma game, I show that coordinated punishment is still viable in the large population and hence the concentration effect induced by the coordinated punishment enhances cooperation in the large group. I show that, surprsingly, even when the degree of coordination decays as the group size increases, as long as the rate of decaying is relatively slow, the large group advantage still obtain. The coordinated punishment, thus, implies that punishers can ``divide and conquer'' defectors more effectively as the size of the group increases. The conflict model in the paper is also related to Lanchester's equations - equations which describe the time evolution of the strengths of two competing armies. I generalize these equations to obtain a conflict model of punishers and defectors in the public good games.