When people make decisions that seem irrational, a common explanation is that they seek evidence confirming what they already believe. But confirmation bias may not be the only factor that skews the way we gather information, according to a new PLOS Biology study led by Laurence Hunt of University College London.
“I’m particularly interested in whether some of the simpler explanations accounting for such biases in animal behavior might also explain some of the biases that we see in humans, which have often been thought of in more ‘cognitive’ terms,” Hunt says.
One of these simpler explanations is reward-driven behavior, which is known to affect how animals and people learn as well as what they pay attention to. To test whether this primitive behavior might also influence how people gather information, Hunt and colleagues collected data from more than 30,000 people who played a gambling card game via a smartphone app.
The game set tasks such as picking the row of cards with the highest or lowest sum, and players spent points for the opportunity to turn over cards – that is, gather information – to help them choose. Altogether the dataset included more than 3 million decisions. To see if the players’ decisions were biased, the researchers compared their choices with those predicted by a model of the optimal ways to gather information.
The researchers found that players often made choices that were not optimal, confirming that the possibility of a reward can bias the way people gather information. But while irrational in the context of the game, the researchers think reward-driven biases in information gathering are likely to be adaptive in other contexts.
When choosing what to eat, for example, “it’s generally helpful to find out more information about the item that’s associated with a rewarding outcome but ignore items that you don’t think will lead to reward,” Hunt says. “This is the same kind of behavior that we see in our subjects. It’s also, I think, linked to the kinds of decisions that many of our evolutionary ancestors made when searching for food in the wild.”
Reference: Hunt LT, Rutledge RB, Malalasekera WMN, Kennerley SW, Dolan RJ (2016) Approach-Induced Biases in Human Information Sampling. PLoS Biol 14(11): e2000638. doi:10.1371/journal.pbio.2000638
Image Credit: _DJ_, Flickr