The theory of the "wisdom of the crowd" has been used to explain everything from the overall accuracy of Wikipedia to the logic of democracy. And in general, that principle is true: Choices made by many are usually better than those made by a few or one.
But new research from Arizona State University and Uppsala University in Sweden adds a caveat to that notion, showing that while crowds might indeed be wise when it comes to making tough, close calls, they are actually worse than individuals at choosing between two options, one of which is vastly superior to the other. When the choice is easy, in other words, the crowd can actually be pretty dumb.
For this study, the researchers used ants, given their propensity for acting as a group. The bugs were tasked with something humans might relate well to: Apartment hunting. Specifically, they were supposed to move from their existing dwelling to a new "house" -- one of two hollowed-out chunks of balsa wood, one of which was darker than the other. And if you're an ant, darker is better.
When they were hunting as a group, the ants made their choice by checking out the two houses, reaching a quorum on one or the other, and recruiting other ants from the home nest to come over. When the difference between the two houses was very slight, it helped to have a big group scouting together: Both places were pretty nice, it seemed, so more ant-votes made it more likely that the "right" house would be chosen. Eventually, enough ants would pick the nicer house, and they would recruit the rest of the colony to move in.
Meanwhile, when the difference between the two houses was stark, the group of ants tended to make the wrong decision more often than an individual ant house-hunting by himself. If enough ants happened upon the inferior house first, there was a slight chance that several of them would accept it, thinking that perhaps it was not so bad -- after all, all their friends were there.
The lead researcher on the study, ASU's Takao Sasaki, provided an example using a hypothetical figure of 100 house-hunting ants, of whom 10 or so wrongly believe that the far-worse house is the better one:
"Out of 100 ants evaluating a poor nest, 10 of them start recruiting [other ants to move in.] These recruiters start bringing nestmates to the poor nest without visiting the other nests. The ants who rejected it eventually find the other nest and compare, but by that time, the 'rash decision-making' ants have already recruited many nestmates from the home to the poor nest, so the 'careful decision-making' ants may not contribute much to the colony's decision."
Before long, the entire ant-village would end up in the worse house.
Meanwhile, when the ants worked individually, they didn't make the same mistake as frequently because there was no one around to mislead them.
Here's a chart of how the decision-making of individuals overtook that of the group once the difference between the two options became more striking:
I asked Sasaki if the same idea holds for other creatures, like, say, humans?
Not entirely, he said, but he did point out a few clear parallels for certain circumstances.
"I went to buy something on Amazon, and I was supposed to compare options and features and cost," he said. "But what I did instead was just buy the most popular thing."
The study authors write that such quirks in group decision-making are "known in many animal groups, including humans, ﬁsh, and cockroaches. Studies in honey bees have shown that social interactions do not always improve collective foraging."
Personally, these ants reminded me of soccer riots, mob attacks, or even the decision to join terror groups. (Or, perhaps less dramatically, inexplicably crowded brunch places with low Yelp ratings.) We can sometimes be conned into making an extremely poor choice -- even when a much better alternative is clearly available -- simply because those around us have made rash decisions and we're following their lead.
Or, as Sasaki said, "We often get fooled by others."
We want to hear what you think about this article. Submit a letter to the editor or write to firstname.lastname@example.org.