African-American schoolchildren who completed a brief writing assignment designed to reaffirm their sense of self-worth received higher grades at the end of the semester than those given a control intervention, a new study finds. These better-performing children closed the grade gap with their white peers by 40 percent, apparently because the assignment interrupted the harmful effects of declining performance early in the semester.

Researchers have invoked a concept called stereotype threat to help explain why black students in the U.S. consistently perform worse in school than their white counterparts. In this view, members of a minority experience anxiety at the prospect of confirming negative stereotypes about their group, such as low intelligence. This anxiety impedes their performance in tasks that reflect on the stereotype, creating a downward spiral in which anxiety and poor performance feed on one another. In past experiments college students have been told that a test they are about to take is insensitive to race or gender, and such interventions can reduce group differences in test scores. "No one had looked at whether these kinds of processes could be manipulated in the real world and have long-term consequences," says social psychologist Geoffrey Cohen of the University of Colorado at Boulder.

Cohen and his colleagues studied a group of about 250 seventh graders, some black and some white, attending a suburban middle school in the Northeast. The children were randomly chosen to receive a 15-minute writing assignment that asked them to indicate their most important value and why it mattered to them, as a way to affirm their sense of integrity or worth. The remaining students wrote about their least important value and why it might matter to others.

Despite the intervention's brevity, the black children who received the affirmative assignment scored one fourth to one third of a grade point higher in that course than the black control group at the end of the term, and the difference showed up in other classes, too. In contrast, the assignment showed no effect on white students. "What's remarkable about this are the long-term effects on grades. It's just an eye-opener that a 15-minute intervention can have this effect," says Timothy Wilson of the University of Virginia, who was not involved in the study. Cohen agrees: "it's tapping at something that runs really deep in these kids." The intervention is described in the September 1 Science. Whether it would work in other school environments remains to be seen.