July 29, 2005
Fear Factor May Fuel Racial Divides
Integration is best way to unite different groups, study suggests
Overcoming fear of members of another race may not be as easy as some might hope, a new study suggests.
The research, which included the use of photos of people and mild electric shocks administered to study participants, revealed the same responses for both black and white Americans.
The only factor that helped diminish the fear was experience with interracial dating, the study found.
"This provides support for something that's happened to this country for quite a while -- efforts to integrate," said Elizabeth Phelps, senior author of the study and a professor of psychology and neural science at New York University. "Those efforts may be one of the best ways to combat these natural biases that are not about conscious attitudes."
Mahzarin Banaji, a study co-author and a professor of psychology at Harvard University, said via e-mail: "Avoiding knowledge about disagreeable aspects of our nature will only lead us to underestimate the work needed to overcome our spontaneous fear of other groups -- a fear that makes little sense in our increasingly global social interactions.
"It's not good business practice and it fundamentally sits in opposition to our aspiration to treat each person fairly and equally," Banaji added.
The study results appear in the July 29 issue of the journal Science.
The researchers noted that similar fear responses have been noted before, albeit in the wild. For instance, humans and other primates tend to hold on to fears of snakes and spiders longer than fear of birds and butterflies.
"There are certain categories of stimuli that are more readily associated with aversive outcomes than other categories and that, once acquired, are harder to get rid of," Phelps explained. "There's a bias in our fear-learning mechanism for certain classes of stimuli."
Phelps and her colleagues wanted to see if this pattern persisted in social groups defined by race. They relied on a variation of a method devised by the Russian scientist Ivan Pavlov a century ago; in his experiments, dogs learned to salivate simply by hearing the ringing of a bell.
In Pavlov's case, the stimulus -- the bell -- was paired with something positive -- food.
In the new study, the stimulus -- photographs -- was paired with something negative -- a mild electrical shock.
"It's learning by association," Phelps said. "It's a very simple type of learning."
The photos consisted of pictures of two black male faces and two white male faces. All wore neutral expressions.
In the first phase of the experiment, each participant received a small electric shock while looking at one of the black faces and one of the white faces. The severity of the shock was chosen by each participant to be uncomfortable but not painful.
In the second phase, the participants looked at the same faces, but without the shock.
Fear responses were measured through changes in the sweat glands.
Not surprisingly, all the participants acquired a fear response to the images that were associated with a shock.
But when the shocks were taken away, the fear response to the face from the participant's own race lessened while the fear response to the face from the other race persisted.
"It was the same whether it was black or white. It was an in-group, out-group effect, not a race-specific effect," Phelps said.
"If you are looking at someone in a race group not your own and that particular individual is now linked with an aversive consequence, that learning is going to be stronger and harder to get rid of than if it's someone of your own race," she continued.
When the participants' attitudes and beliefs about race and contact with members of other groups were probed, the only factor that helped diminish the fear was their experience with interracial dating.
"About 20 percent of white and 50 percent of black participants had dated someone of another race," Phelps said. "If we eliminated all these subjects, the magnitude of the effect was equivalent," she pointed out.
"It's hard to know what to do with findings like this," Phelps added. "We can be aware of the fact that there may be biases in how we respond and how we learn about individuals, based on their social group identification relative to our own. The best we can do is use that information to try to let those types of biases influence choices that have some substance."