04 August 2009

Two heads are better than one

Colchester Campus

If you are one of those people who firmly believe that making decisions by committee is never a good idea, you may have to eat your words.

A group of scientists from the Universities of Essex, Leeds and Bielefeld (Germany) have successfully demonstrated that groups of people can reach decisions more accurately than individuals. Their work will be published in the September issue of the journal Animal Behaviour.

The research shows how groups of volunteers navigate closer to a prescribed target than individuals, when given only limited information of where to go.

Dr Edward Codling, who is jointly appointed between the Departments of Mathematical Sciences and Biological Sciences at the University of Essex, designed the experiment to test ‘the many wrongs principle’. First suggested in 1964 as a way to describe the movement of animal groups, the principle states that individual error can be overcome by staying within a group, so groups can navigate more precisely than singletons. As a result, the group as a whole stays on target, and does so more accurately than any given individual.

Dr Codling and his colleagues (Jolyon Faria and Jens Krause from Leeds, and Fritz Trillmich from Bielefeld) found that when they provided only very low levels of information about the target, and hence uncertainty was high, large groups navigated the most accurately - as predicted by the principle. It did take groups longer to reach the target than individuals, however, suggesting that greater accuracy comes at the cost of slower decision-making.

The team hopes their new findings might lead to a better understanding of how many animals manage to navigate so well. Dr Coding said: ‘The principle has been demonstrated using computer simulations, and has some support from studies on bird flocks, but before our experiments, no one had tested whether humans would show a similar kind of improved group-based accuracy. This may help explain how some animals can navigate so accurately, and why animals often navigate in groups.’

The researchers recruited schoolchildren and colleagues to take part in their experiments, which involved them moving towards one of 16 numbered targets placed around the walls of a circular arena. Members of the group were given a shortlist of targets, only one of which was correct. Each group member had a different list of targets, but all lists included the correct target’s number. Group members – who were forbidden from communicating directly with one another - were then told to head toward whichever target they liked, as long as they stayed within arm’s reach of everyone else in their group.

The work is published online in the journal Animal Behaviour, and was funded by the Biotechnology and Biological Sciences Research Council (BBSRC).

...more news releases