SYDNEY, 4 July 2006 - Many hands won't necessarily make light work. In fact, the reverse could be true, according to Harvard University researchers.
Arne Traulsen and Martin Nowak of Harvard's Program for Evolutionary Dynamics developed a mathematical model for the evolution of cooperation among groups and they said that the size of the group determines how co-operative its members will be. And the bigger the group, the less likely its members are to pitch in.
"If everybody is paying for tickets on the subway, it's best for you as an individual not to do it, because you can get a free ride on the contributions of others," said Traulsen. "But if no one pays, then the group can't travel – the group is better off if everyone pays."
According to group selection theory, genes can spread in a population because of the benefits they provide to groups rather than individuals. But some biologists dismiss this idea, arguing that natural selection can act only on individuals.
Traulsen and Nowak agree that natural selection only operates on individuals, but view group selection as an emergent property of individual fitness and group structure.
"A lot of models presume that there is some group fitness or something like that, and that's not necessarily natural," said Traulsen. "We start from a system which has only individual selection, and group selection arises naturally out of that."
The model classifies individuals into two categories: "cooperators" pay a cost that ensures other group members receive a benefit; while "defectors" pay no cost and provide no benefit to others.
Individuals are assigned fitness levels, and reproduce in proportion. Since defectors benefit from any cooperators present in their group, defectors have higher fitness than cooperators in a mixed group. In homogenous groups, however, cooperators have higher fitness.
When groups reach a certain critical size, individuals are distributed randomly into two daughter groups, or a single individual is eliminated to keep the group at its maximum size.
The model predicts that cooperative behaviour will be favoured by higher benefit-to-cost ratios, and by smaller group sizes and larger numbers of groups.
The researchers have found a real-world example to support their model's findings. Plasmids – viruses that infect bacterial cells – seem to follow the predicted patterns, particularly as they operate as isolated groups.
"Plasmids can cooperate and produce an enzyme that's of benefit to the whole cell, or they can not cooperate, and benefit individually by using the enzyme made by others," says Nowak. "In the long run, cells containing cooperators are better off."
While Troulsen and Nowak acknowledge the limited real-world application of a model that assumes isolated, finite populations, they stress the importance of group selection in the natural world.
"Group selection is an important organising principle that permeates evolutionary processes," they wrote in their paper in the Proceedings of the National Academy of Sciences, "from the emergence of the first cells to ... the economics of nations."
