Skip to main content

We Learn and Make Connections Better When Information Comes from People We Like

The way we’re “wired” to learn may divide us

Illustration of two parts of a brain, with a thumbs up on one side and a thumbs down on the other side

Thomas Fuchs

The human brain tends to play favorites. Its prejudices, well demonstrated by psychological studies, include the “halo effect”: if we like a certain quality in a person, we’re more likely to perceive their unrelated traits positively as well. There’s also “affinity bias,” which refers to how we gravitate toward people with backgrounds or characteristics similar to our own.

Now a study shows how cognitive biases could profoundly affect our most basic learning and memory processes. “What we show is not that people are biased; that we already kind of know,” says Inês Bramão, a psychologist at Sweden’s Lund University and co-author of the new study, published in Communications Psychology. “We give an explanation of why people are biased. The fundamental mechanism may be that we are more likely to expand our knowledge based on information provided by people we like.” Such bias could help explain how people develop strongly polarized views.

Study participants first chose “teammates” and “opponents” from among images of random faces based on their like or dislike of the faces. Then they created imaginary personas for each chosen face, giving characteristics and identities they liked to teammates and ones they disliked to opponents.

On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.

Next, participants viewed images of each face set in a landscape or other background alongside a common household object, as if the person were “showing” the participant the object. Later, the participants tried to match up objects that had shared the same background—this time, without the faces displayed. This tested their ability to learn new information through a process called memory integration: linking memories of multiple past events to make new inferences. The participants did significantly better when linking objects that had initially been “presented” by a persona they liked, which the researchers say indicates a fundamental bias in how we associate previously learned information with a new, partially related event.

The study authors suggest this finding helps to show how people’s opinions can become intensely polarized and increasingly extreme. If we tend to build understanding based mostly on what we learn from a limited set of liked individuals—largely because of their similarities to us—these beliefs can remain unchallenged, leading to narrowing viewpoints.

Psychologist Charles Stone of the City University of New York says that this study is just the beginning and that further research could move beyond images to test learning with real-world events. “This could have important implications for how people make inferences and connect dots about their beliefs that then match their worldviews,” he says. “There’s a lot of fodder moving forward.”

Kate Graham-Shaw is a journalist based in New York City. She covers international news for Japanese media and also covers health and science topics as a freelancer.

More by Kate Graham-Shaw
Scientific American Magazine Vol 330 Issue 6This article was originally published with the title “Learning Favoritism” in Scientific American Magazine Vol. 330 No. 6 (), p. 15