In a paper published in the Proceedings of the National Academy of Sciences, researchers from the University of California, San Diego and Harvard say they have provided the first laboratory evidence that cooperative behavior is contagious.

 Researchers showed that when one person gave money to help others in a "public-goods game," where people had the opportunity to cooperate with each other, the recipients were more likely to give their own money away to other people in future games. This created a domino effect in which one person's generosity spread first to three people and then to the nine people that those three people interact with in the future, and then to still other individuals in subsequent waves of the experiment.

"Though the multiplier in the real world may be higher or lower than what we've found in the lab,"  UCSD political scientist James H. Fowler said, "personally it's very exciting to learn that kindness spreads to people I don't know or have never met. We have direct experience of giving and seeing people's immediate reactions, but we don't typically see how our generosity cascades through the social network to affect the lives of dozens or maybe hundreds of other people."

Participants were strangers to each other and never played twice with the same person, a study design that eliminates direct reciprocity and reputation management as possible causes.

The study is the first work to document experimentally Fowler and Christakis's earlier findings that social contagion travels in networks up to three degrees of separation, and the first to corroborate evidence from others' observational studies on the spread of cooperation.

The contagious effect in the study was symmetric; uncooperative behavior also spread, but there was nothing to suggest that it spread any more or any less robustly than cooperative behavior, Fowler said.

From a scientific perspective, these findings suggest the fascinating possibility that the process of contagion may have contributed to the evolution of cooperation: Groups with altruists in them will be more altruistic as a whole and more likely to survive than selfish groups.

"Our work over the past few years, examining the function of human social networks and their genetic origins, has led us to conclude that there is a deep and fundamental connection between social networks and goodness," said Harvard research Nicholas A Christakis. "The flow of good and desirable properties like ideas, love and kindness is required for human social networks to endure, and, in turn, networks are required for such properties to spread. Humans form social networks because the benefits of a connected life outweigh the costs."

Citation: James H. Fowlera, Nicholas A. Christakis, 'Cooperative behavior cascades in human social networks', PNAS, March 2010; doi:10.1073/pnas.0913149107