In June 23, 2003, the Supreme Court ruled on two affirmative action cases. Shortly before the rulings, NPR interviewed Justice Sandra Day O’Connor. Curiously, affirmative action never came up in the interview. Instead, the interviewer asked O’Connor about her role as the only woman on the court (later, she was joined by Justice Ruth Bader Ginsburg). Connor talked about how, when Ginsburg joined her, different things were expected from her. She no longer stressed as much about making the right choice “for women.” After O’Connor retired, and Ginsburg briefly became the only female justice, Ginsburg made comments suggesting that she felt “burdened with extra scrutiny.”
Sandra Day O’Connor was the first woman on the Supreme Court, and she reported feeling anxious and stigmatized as a result. By virtue of being the only woman on the court, O’Connor felt that she was being asked to speak on behalf of all women, and thus began questioning her own decisions more than she would have otherwise. O’Connor’s behavior mirrors that of the test-takers in some of Steele’s experiments—they feel burdened by the extra pressure of their minority status.
Essentially, Ginsburg and O’Connor were describing identity contingencies. With two female justices on the court, Ginsburg and O’Connor gained enough “critical mass” to circumvent the extra scrutiny of their peers. Psychologists have studied the concept of critical mass in many different settings. From their perspective, critical mass is defined as “the point at which there are enough minorities in a setting … that individual minorities no longer feel uncomfortable there because they are minorities.” For example, studies have shown that when the number of women in an orchestra approached forty percent, the women no longer feel as stressed and stigmatized—suggesting that forty percent is the “critical mass” for an orchestra.
In the 1990s, Steele researched the factors that lead to a person being affected by identity threats. Initially, he believed that psychological factors, such as confidence, were most relevant. But over time, he began to realize that often, the people who were most confident—such as O’Connor—suffered from identity threats the most. Steele began to understand that environmental factors were just as important in predicting people’s responses to identity threats. Steele assumed that environmental cues, some of them almost undetectable, have a discernible influence on people’s behavior.
By the 1990s, Steele has assembled a convincing case for the impact of environmental factors on the individual’s sense of stigmatization. However, Steele doesn’t discount the role of the individual’s personality: he’s found that the most driven, motivated people are often most susceptible to stereotyping cues. In this sense, stereotype threats arise from an intermingling of environmental and personal factors.
Steele and his research partners began to develop a theory of how environmental cues contribute to identity threats. It seemed that “identity cues” tended to influence people’s behavior when they accumulated over time. Individual identity cues, however, didn’t necessarily impact behavior. By “identity cues,” Steele means external stimuli that remind people of an identity stereotype. Many of these stimuli exist regardless of whether or not people are actively prejudiced. For example, when O’Connor was the only woman on the court, she wasn’t reacting to overt displays of prejudice; rather, she was reacting to the number of male justices itself. Steele posits that identity cues must gather a certain “critical mass” in order to influence people’s cognitive processes.
Steele has shown how minor, seemingly neutral verbal or visual cues can provoke feelings of stereotype anxiety in test-takers. However, here he begins to develop a more nuanced theory for how and why this happens. He suggests that a certain number of minor stereotyping cues, when put together, can trigger an anxious reaction that impedes the test-taker’s cognitive abilities.
At this point in his research, Steele had many new questions about identity contingencies, above all, “Can a few cues in a setting really undermine a person’s sense of belonging? Are people so attuned to the details of their social environments?” Working with three colleagues, Valerie Purdie-Vaughs, Paul Davies, and Mary Murphy, Steele developed experiments designed to measure the influence of environmental cues on behavior. Davies, working with Steve Spencer, ran an experiment in which students were asked to watch TV commercials, some of which depicted women in stereotypically gendered roles. Then they were asked to complete a test. The experiment suggested that women exposed to gender-neutral TV were likely to perform at the same level as the men, but when they watched women in stereotypical roles, they underperformed.
In this situation, a seemingly minor identity cue—a TV commercial with sexist content—provides enough of an identity threat to elicit strong feelings of inferiority and self-doubt in the female test-takers. In this case, the female test-takers’ lifelong experiences with sexism and misogyny have influenced them in such a way that even something as simple as a commercial is enough to interfere with their cognitive performance on the test.
Steele and Purdie-Vaughns developed their own experiment, in which black and white participants were given a newsletter, supposedly from a tech company, and then asked if they’d trust the company. Some copies of the newsletter included photographs showing a small number of black people at the company, while other copies showed larger numbers of black employees. Some copies specified that the company had a “color-blind” hiring policy, while other copies said the company valued diversity. White participants said they trusted the company no matter which version of the newsletter they read. Black participants were much more likely to say they trusted the company if they saw pictures of many black people, regardless of what the company’s diversity policy was. But they were also more likely to say they’d trust the company that valued diversity, even if there were fewer black people in the pictures.
This experiment provides more nuanced results than the experiment discussed in the previous section. Certain identity cues are powerful enough to provoke strong feelings in the test subjects, but other identity cues are less powerful, with the result that they can be masked or canceled out by other, contradictory identity cues. For example, the statement that the business values diversity, it would seem, is enough to neutralize the dearth of black people in the photographs from the business newsletter (at least at the time this study took place—“diversity” has recently become more of an empty buzzword that many minorities are, for good reason, suspicious of).
The results of the experiment might suggest that people with minority identities feel more comfortable when they see other minorities (i.e., the critical mass approach), but also when they understand that their teachers or employers value diversity. But the results further suggest that certain environmental cues (such as photographs of minorities) can “cancel out” others. Even if it’s impossible to control every detail of an environment, such as a classroom, psychologists’ research could be used to create enough positive cues that people feel relaxed and un-stigmatized.
The experiment sets the tone for the final third of Whistling Vivaldi, which is largely about ways of neutralizing stereotype cues. No environment can be controlled with total accuracy, but perhaps certain environments, such as classrooms, can be primed to limit the influence of stereotyping and thus maximize the potential of all students.
Mary Murphy proposed that the researchers study the physiological impact of identity cues in everyday situations (rather than during a test). With the help of James Gross, an emotional psychologist, Steele and Murphy developed experiments to measure the physiological impact of situational cues. Student participants watched a video advertising a science conference, and their heart rates were recorded. Some students watched a video showing more men than women at the conference, while others saw a video with equal numbers of men and women. Men who watched the videos showed little change in heart rate. Women, however, experienced an elevated heart rate while watching the video with more men than women, and also remembered more "incidental features," showing heightened vigilance.
Murphy, Gross, and Steele’s findings further reinforce the physiological basis for identity threats—and, therefore, the need to find ways to control environments and limit the influence of these identity threats. Without some kind of environmental control, Steele suggests, underrepresented groups, such as women, will continue to experience anxiety and underperform on tests, and perhaps in general.
Murphy, Steele, and Gross had shown that, under ordinary circumstances—something as banal as watching a commercial—women experienced an identity threat. This supported the hypothesis that 1) responses to identity threats were largely the result of environmental cues, rather than psychological disposition, and 2) it might be possible to neutralize these environmental cues with other cues.
After many years of research, Steele and his colleagues have uncovered a major source of the achievement gap problem (feelings of anxiety, caused by environmental cues and proven with physiological evidence) and, by the same token, proposed a possible solution (changing and neutralizing stereotype threats through different environmental cues).