Psychologist logo
Graphic of two heads with quadrant of their brains in black and white
Brain, Developmental, Social and behavioural

More intelligent people are quicker to learn (and unlearn) social stereotypes

People with superior cognitive abilities may be equipped to efficiently learn and use stereotypes about social groups.

16 August 2017

By Emma Young

Smart people tend to perform better at work, earn more money, be physically healthier, and be less likely to subscribe to authoritarian beliefs. But a new paper reveals that a key aspect of intelligence – a strong "pattern-matching" ability, which helps someone readily learn a language, understand how another person is feeling or spot a stock market trend to exploit – has a darker side: it also makes that person more likely to learn and apply social stereotypes.

Previous studies exploring how a person's cognitive abilities may affect their attitudes to other people have produced mixed results. But this might be because the questions asked in these studies were too broad.

In the new study, published in the Journal of Experimental Biology, David Lick, Adam Alter and Jonathan Freeman at New York University decided to home in on social stereotyping. "Because pattern detection is a core component of human intelligence, people with superior cognitive abilities may be equipped to efficiently learn and use stereotypes about social groups," they theorised.

To explore this, they conducted a total of six online studies involving 1,257 people recruited via Amazon's Mechanical Turk survey website. In the first two studies, volunteers saw pictures of aliens that varied on four dimensions (colour, face shape, eye size, ears), with most of the blue aliens  paired with an "unfriendly" behaviour (like "spat in another alien's face") and most of the yellow aliens paired with a friendly behaviour (like "gave another alien a bouquet of flowers"). The volunteers also completed items from Raven's Advanced Progressive Matrices, which assesses pattern-matching ability.

A subsequent memory test involved the same participants attempting to pair previously seen faces with their earlier behaviours, but there were also some new blue and yellow faces that actually they hadn't seen before.  Participants who were better at pattern-matching were more likely to attribute unfriendly behaviours to new blue aliens than to new yellow aliens – suggesting that they'd learned colour–behaviour stereotypes more readily, and applied them.

In studies three and four, volunteers were instead shown realistic pictures of male human faces. The displays were manipulated, so that most of the faces with a wide nose (for some participants) or a narrow nose (for others) were paired with negative behaviours – like "laughed and jeered at a homeless person". Most of the faces with the other nose type were paired with friendly behaviours – like "sent flowers to someone who was sick".

After viewing the faces, the volunteers played a trust game involving sharing money. They were led to believe this was an unrelated interlude in the study. Before the game began, they chose an avatar from a large group of faces to represent them online. They then played 12 rounds of what they believed was a real game, each time with a different partner who was represented by their own avatar.

In fact, the volunteers weren't playing with real partners, and the experimenters manipulated the "partners''" avatar photos, so that some had wider noses, and some had narrower noses (there were also female "partners" whose nose width did not systematically vary). The team found that volunteers who did better on the test of pattern detection gave less money to partners whose avatars had a nose width related, in the earlier trial, to unfriendly behaviour.

However, when these volunteers were given new information that contradicted the stereotype they had implicitly developed, the better pattern-detectors were also quicker to update their stereotype – to reverse their biases.

In a final experiment, the team used a real-world set of stereotypes, relating to traits they believe are often associated with men (such as being more authoritative) and with women (such as being more submissive). After counter-stereotype training – effectively being told that being authoritative is more associated with women rather than men, for example – good pattern-detectors showed a stronger decrease in stereotyping.

"To our knowledge, these findings are the first to systematically demonstrate that cognitive ability is associated with greater stereotyping," the researchers write, before adding, "people with superior pattern detection abilities appear to act as naive empiricists, both learning and updating their stereotypes based on incoming information."

While existing research tends to focus on the benefits of intelligence, these "findings join a small body of work guiding the field toward a more balanced understanding of the consequences of human aptitudes," they note. For example, it's also been suggested that superior, misguided, pattern-matching may play a role in obsessive-compulsive disorder (OCD).