Collaborative Research: Cognitive and Physiological Effects of Stereotypes on Problem Solving: Causal Mechanisms and Prescriptive Measures

Williams College, Williamstown MA

Investigators

Abstract

The failure of large groups of students to perform to their potential.such as females in math, or African Americans across a variety of academic disciplines.is one of the great tragedies in our educational system. The significance of this problem has inspired numerous investigations of its scope, nature, and potential causes. A recent breakthrough points to the power that social context can have in creating, perpetuating, or eliminating the underperformance that has hindered these groups. Steele, Aronson, Spencer, and their colleagues have demonstrated the crucial role of stereotype threat, a situational phenomenon that occurs when individuals who are targets of stereotypes alleging intellectual inferiority are reminded of the possibility of confirming these stereotypes (e.g., Spencer, Steele, & Quinn, 1999; Steele, 1997). For example, Spencer et al. (1999) showed that high-achieving female college students performed significantly worse than males on a standardized math test when the stereotype about their math ability was made salient (.Males have performed better than females on this test in the past..). However, this gender gap was eliminated simply by changing the words used for introducing the test (.Males and females have performed equally well on this test in the past..). A counterintuitive finding is that only individuals who are highly identified with success and achievement in given stereotyped domains are the ones who show deficits under threat, in contrast to stigmatized individuals who are not achievement oriented in these domains (Steele, 1997). Thus, it is the people in the vanguard of their group who are the most vulnerable to situations in which stereotypes about their ability become salient. The detrimental effects of stereotype threat extend to African Americans (e.g., Steele & Aronson, 1995), Latinos (e.g., Aronson & Salinas, 1997), and students of low socioeconomic status (e.g., Croizet & Claire, 1998), who are concerned with academic success. We hypothesize that the priming of a stereotype results in heightened physiological arousal. According to the classic Yerkes-Dodson (1908) law of physiological arousal, performance is optimal at intermediate levels of arousal and decreases when arousal is either low or high, resulting in an inverted-U shaped function. More recent neurophysiological research has corroborated and expanded on the classic findings of the Yerkes-Dodson law in animals and in humans. For example, Lupien and McEwen (1997) provided evidence for an inverted-U shape relationship between the level of the corticosteroid, cortisol, and cognitive processes in a variety of animal and human studies. We hypothesize that high-achieving individuals in stigmatized domains approach a problem-solving task in the given domains with an optimal level of arousal for performing well. Stereotype threat may interfere with these individuals. performance by leading to arousal that exceeds an optimal level, causing performance deficits. By trying to uncover the role of physiological arousal in mediating stereotype threat and the cognitive nature of the resultant deficits, we are hoping to contribute to understanding the theoretical underpinnings of stereotype threat as well as to help stigmatized high-achieving students to overcome its effects. In the psychometric literature, Gallagher (1992, 1998) has shown that females performed more poorly than males on math-SAT problems that required the use of meaningful mental short cuts and estimation (versus problems that required standard calculation). There is reason to believe that cortisol reactivity may be implicated in defaulting to rote algorithmic versus meaningful problem solving (Forget, Lacroix, Somma, & Cohen, 2000). We will test interventions for combating stereotype threat by reducing arousal, improving the use of meaningful problem-solving strategies, and creating a practice of self-affirmation. In sum, the proposed research is designed to examine: (1) the role of physiological arousal as a mediator of stereotype threat; (2) whether reasoning becomes more algorithmic and less meaningful under stereotype threat; and (3) ways in which the detrimental effects of stereotype threat can be mitigated or eliminated.