Colorblind? Researchers found white subjects shrink from using relevant racial descriptors when looking at cards like these.

The Price of Words Unspoken

After Barack Obama's landmark speech on race on 18 March, it was hard to tell what got more media attention: What the Democratic presidential candidate said or that he had said it at all. Regardless, many pundits agreed that as an African-American, Obama could discuss race in ways few white people would dare. That's because most white Americans today have learned not to talk about race for fear of seeming racist, says Samuel Sommers, a social psychologist at Tufts University in Medford, Massachusetts. Two new studies back up this idea, and the research shows that deliberately avoiding race when it's clearly relevant may impair decision-making skills.

Humans are hard-wired to notice race. The average person registers the race of another human face in less than 100 milliseconds, according to past studies. This instantaneous perception clashes sharply with the American cultural taboo against using race to identify someone. Watch people at a party trying to describe another person, says Michael Norton, a marketing researcher at Harvard Business School. "They'll launch into these long explanations until someone in the group might eventually say, 'Oh, you mean the Asian guy?'"

To measure the impact of such verbal gymnastics on cognition, Norton, Sommers, and colleagues employed a modified version of the children's game "Guess Who?" Cards depicting people of different genders and races are laid face-up on a table, and one player mentally selects a card. The second player has to figure out as quickly as possible which card the first player picked using as few yes-or-no questions as possible, such as "Is your person blonde?"

Past studies have suggested that children internalize social taboos about discussing race at about age 10. The researchers compared the performances of 51 children that were 8 to 9 years old with a similarly sized group of 10- to 11-year-olds. Both groups were equally composed of girls and boys, and the participants were predominantly white. In this game, asking about race was perfectly legitimate, because it could help the child pick the target card faster, Sommers says. Whereas almost 77% of the younger children asked about race, only 37% of the older children did. Consequently, the younger group guessed the target card after an average of 7.4 questions, but the older students averaged 8.3 questions, the team reports online this week in Developmental Psychology.

The researchers conducted similar experiments with about 100 men and women to see if the race of the playing partner would affect whether the subject brought up race. All of the volunteers were white, and in one test, half the subjects were paired with a black partner. These subjects asked about the race of the person on the target card 66.7% of the time, compared with 88.8% in subjects with white partners. This suggests that white subjects working with a black partner were less likely to mention race, the researchers report in a second paper, published online this week in the Journal of Personality and Social Psychology.

This avoidance of discussing race appears to impair cognition. Adults who avoided mentioning race during the "Guess Who?" game subsequently performed worse on a color-matching test, a standard test of how quickly the mind processes information, than volunteers who did mention race. "Trying to come up with alternate ways of identifying people has drained them of valuable cognitive resources," Sommers says.

Taken as a whole, the studies suggest that efforts to appear race-blind can be counterproductive in some situations, says Victoria Plaut, a social psychologist at the University of Georgia, Athens, who was not involved in the study. "This probably means that egalitarian whites think they're doing the right thing by avoiding race," she says, but they may not be.