Can Science Help People Unlearn Their Unconscious Biases?

Social events, sleep training and even meditation may offer ways for people to erase biases they probably didn’t know they held

The U.S. Supreme Court voted 5-4 to legalize marriage equality nationwide on June 26. Pete Marovich/Corbis

Last week's Supreme Court decisions offered many reasons for liberal-leaning people to celebrate: the justices upheld the constitutionality of the Affordable Care Act (better known as "Obamacare"), same-sex marriage and the Fair Housing Act. While equal treatment on paper does not always translate to equal treatment in reality, the latter two decisions in particular do strengthen the national conversation about bias and how people of different sexualities, genders, religions, races and ethnicities are treated by the government and their fellow citizens.

Those who championed the Supreme Court's decisions may see their opponents as bigoted or heartless—but even the most egalitarian among us may at times fall victim to unconscious biases that influence behavior in subtle ways.

These so-called implicit biases have been shown to have a broad array of downstream effects, ranging from hiring decisions to the quality of health care. Implicit biases are formed early in childhood—some studies show evidence in children as young as six years old—and they are reinforced through adulthood by social environments and the ubiquitous presence of mass media. Not to sound too much like "Avenue Q," but everyone, it seems, can be a little bit racist.

Today, a growing cadre of psychologists and cognitive scientists is working to unravel the implicit stereotypes and biases we hold against others. Their research suggests that these biases are not set in stone and can be unlearned— offering possible strategies to reduce their destructive impact.

“Correcting bias is actually more difficult than it seems. The first challenge is you have to be aware of your bias,” says Calvin Lai, a postdoctoral researcher at Harvard University. Awareness could reveal blind spots and “motivate individuals to seek out their biases and regulate them,” he says.

One well-established method of reducing intergroup prejudice is simply to interact with diverse groups of people. “In the 60 to 70 years of research on this idea of prejudice reduction, there has been this one idea that has stood firm as a gold standard, and that idea is intergroup contact,” says Lai. For example, a 2008 study suggests that people who have more frequent interactions with gay, lesbian or bisexual individuals show more favorable attitudes toward homosexual men (as measured by explicit and implicit tests) and also reported more motivation to eliminate their internal prejudices.

The same seems to apply to race. Psychologists have shown that the creation of a mixed-race group, such as a political party or an intramural sports team, can override pre-existing racial biases toward other group members. More interaction can also encourage people to develop individuation—the cognitive ability to see members of a racial group as unique individuals. Even being touched in a friendly manner by a member of another ethnicity has been shown to reduce implicit biases toward all members of that group, according to a 2014 paper.

And in a study released just last month, researchers at the University of Queensland showed that exposure to people of other ethnicities can impact whether we feel their pain. The team monitored the neural activity of 30 participants who watched videos that featured members of their own race or a foreign race experiencing a painful jab from a needle. While viewing foreign races in pain, participants who reported greater contact showed enhanced activation of the anterior cingulate cortex, a brain area that has been associated with empathy, when compared with those who reported less contact.

In some cases, the key is to surprise people by having them interact with someone who shatters their pre-existing stereotypes. “Research has shown that exposure to a female science professor by a female student is likely to change gender stereotypes about science and could potentially change career aspirations,” says Lai.

Increased accountability has also prompted some encouraging results. This strategy has gained some notoriety in light of recent debates about whether to place body cameras on police officers. “There has been a long tradition within psychological research showing that accountability is great for reducing all types of biases,” says Lai. “And although there has not been much direct research looking at the physical data of how effective they might be, there is promise in the idea of body cameras.

Even meditation has been explored as a means of reducing bias. An April 2015 study indicated that a mindfulness meditation audio recording can induce listeners to rely less on previously established associations, producing a reduction in implicit race and age biases.

But not all researchers are as optimistic that unconscious biases can be readily diminished, especially in the long run. "My understanding of the present state of research on reducing implicit biases is that there is no established method of achieving durable reductions of implicit biases that were formed in childhood," says Anthony Greenwald, a professor of psychology at University of Washington and co-creator of the Implicit Association Test

According to Lai, only a small proportion of studies have looked at the effects of bias reduction over time. And while numerous interventions have been shown to be effective in the short term, these effects tend to disappear soon after the experiment. “It is similar to assessing blood pressure,” says Lai. “If you make people run up some stairs, their blood pressure is temporarily elevated for a couple minutes. And in a similar way, most of the things that we have been looking at with implicit bias show similar short-term effects.”

Lai adds that many of the interventions that are effective in the controlled environment of a scientific experiment are far more difficult to implement in the real world. “Something as simple as encouraging more high quality intergroup contact is not something that is logistically easy to do, especially when people live in segregated neighborhoods,” he says.

Greenwald does point to some encouraging recent findings that reveal how sleep might help facilitate unlearning of implicit social biases. Psychologists at Northwestern University conducted a training experiment that associated counter-stereotypical visual stimuli (such as pairing women's faces with words linked to math and science) with unique audio cues. After training, the participants took a 90-minute nap, during which the unique sound was replayed subliminally. The result was a significant reduction in bias after training compared to participants who did not receive the audio cue.

Even after a week, participants still showed weaker unconscious stereotypes compared to baseline levels. The researchers believe that repetition of the audio sound helped to integrate counter-stereotypical information into a person's memory via a process called consolidation, which occurs during sleep. But while these results seem promising, Greenwald adds that he will remain skeptical "until the findings are convincingly reproduced."

As an additional strategy, Lai suggests trying some broader social interventions to prevent implicit bias from rearing its ugly head. “Rather than trying to change people at the individual level, we can think about trying to restructure the way that people make decisions,” says Lai. “For example, when you look at a name on a resume, you gain information about a person’s race and gender that can quickly leak in and influence your judgment of that candidate without you ever realizing it. So simple levers that take implicit bias out of the equation, like blinding resumes so you don't see a person’s name, can do wonders.”

Implicit biases are also likely to arise in situations when we are highly stressed, time-constrained and mentally or physically exhausted. For example, when faced with repetitive parole hearings that weigh heavy on cognitive load, judges are more likely to choose the default option of “no parole” just prior to lunchtime, when they are most drained. This principle could also apply in law enforcement or medicine, where police officers and doctors are most often exhausted and operating in high-stress environments. Counteracting fatigue in these types of jobs may help lower the odds of bias negatively affecting critical decisions.

"People aren't going to act on implicit bias if they don't have the opportunity to let implicit bias influence decision-making to begin with,” says Lai.

Get the latest Science stories in your inbox.