May 04, 2010

Co-ordinated punishment and the evolution of co-operation

From the related public release:
Previous models of cooperation assumed that punishment of free-riders was uncoordinated and unconditional. One problem with these models was that the costs associated with punishment were often higher than the gains of cooperation. Thus, the cost of one group member's punishing a free-rider would be substantial and would not overweigh the gains achieved through increased cooperation.

Costs may be defined as loss of friendship or loss of relational closeness with other members of the group.

To address the problem, Boyd and his colleagues changed the assumptions built into previous cooperation/punishment models. First, they allowed for punishment to be coordinated among group members. In their model, group members could signal their willingness to punish someone who was not participating in the group, but punishment would only occur if it was coordinated. This meant the cost of punishing a free-rider would be distributed across members and would not be higher than the cost of gains achieved through increased cooperation.

Second, the researchers allowed for the cost of punishing a free-rider to decline as the number of punishers increased. Boyd explained that this new model was "catching up with common sense" because these two assumptions exist in reality.

Their model had three stages in which a large group of unrelated individuals interacted repeatedly. The first stage was a signaling stage where group members could signal their intent to punish. In the second stage, group members could choose to cooperate or not. The final stage was a punishment stage when group members could punish other group members.

The results of their model look a lot like what is seen in most human societies, where individuals meet and decide whether and how to punish group members who are not cooperating. This is coordinated punishment where group members signal their intent to punish, only punish when a threshold has been met and share the costs of punishing.

Boyd argues that even in societies without formal institutions for establishing rules and methods of punishment, group punishment appears to be effective at maintaining cooperation.
Here is an example of what the authors are talking about. Imagine you are living in the Old West and a gang has just robbed your neighbor's store. Clearly, if you went out to punish the perpetrators you would pay a high cost (they would be likely to kill you) and it wouldn't be successful. So, it doesn't make sense to punish individually. But, the sheriff could assemble a posse to go after the criminals. This would immediately reduce your risk (since you would be one target among many) and it would also increase your chance of success (as more punishers are more likely to achieve their goal).

Related posts:
Science, Vol. 328. no. 5978, pp. 617 - 620

Coordinated Punishment of Defectors Sustains Cooperation and Can Proliferate When Rare

Robert Boyd et al.


Because mutually beneficial cooperation may unravel unless most members of a group contribute, people often gang up on free-riders, punishing them when this is cost-effective in sustaining cooperation. In contrast, current models of the evolution of cooperation assume that punishment is uncoordinated and unconditional. These models have difficulty explaining the evolutionary emergence of punishment because rare unconditional punishers bear substantial costs and hence are eliminated. Moreover, in human behavioral experiments in which punishment is uncoordinated, the sum of costs to punishers and their targets often exceeds the benefits of the increased cooperation that results from the punishment of free-riders. As a result, cooperation sustained by punishment may actually reduce the average payoffs of group members in comparison with groups in which punishment of free-riders is not an option. Here, we present a model of coordinated punishment that is calibrated for ancestral human conditions and captures a further aspect of reality missing from both models and experiments: The total cost of punishing a free-rider declines as the number of punishers increases. We show that punishment can proliferate when rare, and when it does, it enhances group-average payoffs.


1 comment:

Andrew Oh-Willeke said...

Historically, one interesting variant is ancient Iceland, in which there was a collectively determination of eligibility for punishment, but actual infliction of the punishment via blood feuds, was a private function.

The set up for the Illiad also grapples quite expressly with the notion of how to secure collective participation in punishment of a wrongdoer (the oaths of the wooers and the subsuquent calling in of those oaths and related family obligations to support each other).