New research on decision making in fish shows that the bigger the school, the higher the chance the group makes the right choice, scientists report online November 13 in Current Biology.
Trusting group decisions is a cornerstone of modern human society. Trial juries, Wikipedia and even American Idol are based on trust in consensus decision making. And underwater in the stickleback fish world, things aren’t so different.
Instead of choosing presidents or favorite singers, one species of stickleback fish, Gasterosteus aculeatus, has a more pressing concern — picking the right fish to follow home. Based on earlier experiments, the study’s researchers had a pretty good idea about such stickleback preferences. Fat, evenly colored fish are regarded as healthy and strong, while scrawny fish mottled with black spots may be considered diseased.
Coauthor Ashley Ward, a biologist from the University of Sydney in Australia, says of these sticklebacks, “Fish like large leaders, well-fed leaders and unparasitized leaders.”
Because sticklebacks are so judgmental, Ward and his colleagues were able to set up a situation where the fish were forced to choose between a “good” choice and a “bad” choice. Researchers then measured what percentage of fish went for the tall, dark and handsome (and, the team presumed, successful) leader.
To create such choices, researchers made replica fish, digital images of all sorts of sticklebacks — fat, skinny, long, short, clear and spotted — that were taped to both sides of a glass slide. Two replica fish, one “good” and the other “bad,” were then tugged toward one of two homes in an aquarium, forcing the real fish to choose which replica fish to follow.
Slightly more than half of the live fish tested alone made the right choice, following replicas that appeared healthy. Adding another live fish to the mix didn’t make much of a difference in selection.
But something very interesting happened as more and more live fish were put into the tank. When the group grew to three or more fish, the group’s decision took precedent.
When the size of the live fish group was increased to eight, 80 percent chose to follow the better leader. These results show “that the group as a whole is much better at finding the fat fish,” says coauthor David Sumpter of Uppsala University in Sweden.
Researchers speculate that the wisdom of the crowd results from individual fish that may have bits of information that the group uses to make a better decision. Say, for example, a fish caught sight of a mottled spot on a potential leader that other members of the group may have missed. If in a group, the discerning fish’s information will benefit other members, too.
“The most interesting bit — the positive social feedback — was really not something we expected to see,” says Ward.
“This is a very interesting paper,” comments Stephen Pratt, a biologist from Arizona State University in Tempe who studies consensus making in ants. “It has a different take on the advantages of living in a group.”
But before you jump off a bridge just because Billy did, understand that these group smarts come with a caveat. In a small number of the team’s experiments, all of the fish in a group chose the “bad” leader. “There’s always a danger of amplifying a bad effect,” says Pratt.
This notion of amplified mistakes could apply to studies done in humans that have examined how bad decisions, like unfortunate fashion trends and economic bubbles, snowball.
To navigate the tricky boundary between individual and group desires, the research team devised a mathematical equation that describes when a stickleback goes with its gut or goes with the crowd. The researchers explain that in this quorum rule, as the number of fish in the group rose, so too did the group’s influence on the decision.
“The quorum rule is a nice way to make compromises between using your own smarts or slavishly following the leader,” says Pratt.
