News: News Archives
German Researchers Find New Insight on Social Punishment, Human Cooperation
FRANKFURT, GermanyPeople are willing to inflict financial pain to achieve collective gain, according to a new study in the 7 April 2006 issue of the journal Science.
In an experiment with volunteers at the University of Erfurt in Erfurt, Germany, researchers have found that groups in which individual members have the option to punish freeloaders were more popular than groups without this option.
Though two thirds of the study participants initially chose the group in which members could not punish others, many abandoned this non-punishing group after seeing the greater financial rewards that come with cooperation that is maintained when individuals choose to punish freeloaders.
This new evidence for a competitive advantage for groups in which individuals can punish freeloaders may bring scientists closer to a general theory of human cooperation, the authors say.
"We were the first researchers who were able to show that the sanctioning institution is the undisputed winner in a competition between a sanction-free and punishing institution," Science author Bernd Irlenbusch from the London School of Economics in London, UK, said at a news conference Thursday in Frankfurt. "The important question is how can we encourage people to impose these social sanctions...to provide for the public good which could affect the whole world community"on issues like the environment, for example.
Whether in small societies or in a global context, there must be shared norms to create incentives for environmental stewardshipand disincentives for the sort of harmful behavior that causes global warming, added Science author Bettina Rockenbach from the University of Erfurt in Erfurt, Germany. "If you don't have clear norms it's very hard to implement a common cooperation in achieving this goal," she said.
Listen to a Science podcast that includes an interview with Bernd Irlenbusch, one of the authors of the study, here.
Understanding the circumstances under which people cooperate is of great interest because addressing some of the world's most pressing issues, such as global climate change, may require people to act in the best interest of the group, according to the author of a related "Perspective" article.
The researchers asked 84 students from a German university to either join a virtual group that does not financially punish freeloaders or a group that is nearly identical but allows members to punish freeloaders.
The differences in cooperation between the two groups can be likened to neighboring towns with different reactions to those who pollute, Irlenbusch explained before the news conference.
In a town where people notice and disapprove of those who pollute the environment, the threat of informal forms of punishment such as social exclusion or reputation damage can discourage people from polluting. If, in a neighboring town, people do not notice who the polluters are, then pollution is likely to be more of a problem.
Just as a town in which inhabitants punish polluters might be cleaner and thus more popular than a more polluted nearby town whose inhabitants do not punish polluters, the scientists found that a group that punishes or threatens to punish its freeloaders holds a competitive advantage over a nearly identical group that has no mechanism for punishing freeloaders.
In the laboratory, the participants were asked to join a punishing or a non-punishing group and received 20 "monetary units" to manage, with the instruction that the more money they amassed for themselves during the experiment, the more money they would be paid at the end of the experiment.
Players could choose to contribute money to a group project or deposit the money in a private account. All contributions to the group project were increased by about two thirds and divided equally among all players, regardless of their contributions.
After the players made their contributions and learned of the contributions of others, those in the punishing group had the option to punish non-cooperating group members. Punishing another person reduced the payoff of the punisher by one unit and the person being punished by three units.
At the end of the punishing period, all participants from both groups received detailed but anonymous payoff information about each of the other members of both institutions. At this point in the game, people could choose to stay in their own group for the next round, or switch to the other group.
At the start of the first round of the game, two-thirds of the study participants chose the punishment-free group.
After the first few rounds, members of the non-punishing group who initially behaved cooperatively by contributing to the group project scaled back their contributions after seeing that others were freeloading.
Many of these initially punishment-averse people switched over to the punishing group and immediately adopted the established culture of cooperation. The former freeloaders contributed significantly to the group account and punished freeloaderseven though they had to pay to punish.
"You can't explain this dramatic change in behavior by saying that people are just looking for the best payoff. People gave up payoff to follow the cooperative norm," Rockenbach explained before the news briefing.
"I was surprised so many of the freeloaders exerted punishment on others when they entered the sanctioning institution," said Irlenbusch.
The immediate switch from freeloader to cooperator and punisher is consistent with theories of cultural and institutional selection, the authors say. These theories suggest that people preferentially migrate to groups with higher payoffs and imitate the decisions prevalent in those groups.
"New members of the sanctioning institution punish because it is common to do so," said Rockenbach. She noted that this new work supports the idea that cultural selectionthe notion that culture can evolve in ways similar to Darwin's natural selectionplays a role in the establishment and maintenance of cooperation.
By forcing sanctioning and non-sanctioning institutions to compete head-to-head in an experimental setting, the authors also present empirical support for the idea that institutions with built-in sanctioning mechanisms can establish norms of cooperation and out-compete institutions lacking mechanisms for punishing freeloaders.
Within the sanctioning institution, the culture of cooperation remained, and even strengthened as streams of outsiders from a non-cooperative environment joined and quickly adopted their cooperative behavior, the scientists found.
When the game ended after 30 rounds, the non-punishing group was almost completely depopulated and the punishing group functioned at a high level of cooperation. There was not much actual punishment being applied during the later rounds of the game; apparently the threat of punishment induced cooperative behavior, the authors say.
The new Science study should spur researchers to devise experiments to look into questions that will provide further information on the development of cooperation within groups of people, writes "Perspective" author Joe Henrich from Emory University in Atlanta, Georgia, USA.
"What if switching institutions is costly, or information about the payoffs in the other institutions is poor?" Henrich wrote. "What happens if individuals cannot migrate between institutions, but instead can vote on adopting alternative institutional modifications?"
Oezguer Guererk and Bettina Rockenbach are at the University of Erfurt in Erfurt, Germany. Bernd Irlenbusch is at the London School of Economics in London, UK.
The news briefing was held at Die Industrie- und Handelskammer Frankfurt am Main (the Frankfurt am Main Chamber of Commerce and Industry).
Daniel B. Kane and Natasha Pinol
6 April 2006