Stereotype Susceptibility Narrows the Gender Gap in Imagined Self-Rotation Performance
Wraga, Maryjane, Duncan, Lauren, Jacobs, Emily C., Helt, Molly, Church, Jessica, Psychonomic Bulletin & Review
Three studies examined the impact of stereotype messages on men's and women's performance of a mental rotation task involving imagined self-rotations. Experiment 1 established baseline differences between men and women; women made 12% more errors than did men. Experiment 2 found that exposure to a positive stereotype message enhanced women's performance in comparison with that of another group of women who received neutral information. In Experiment 3, men who were exposed to the same stereotype message emphasizing a female advantage made more errors than did male controls, and the magnitude of error was similar to that for women from Experiment 1. The results suggest that the gender gap in mental rotation performance is partially caused by experiential factors, particularly those induced by sociocultural stereotypes.
The remarks of former Harvard President Lawrence Summers have reignited the debate on the nature of intelligence and its consequences for academic achievement in men and women (Ripley, 2005). Although traditional gender gaps in cognitive performance have diminished over past decades, one of the few exceptions to this trend involves mental rotation, a skill of spatial reasoning that is critical to success in academic fields such as mathematics and science (Feingold, 1988; Masters & Sanders, 1993). Mental rotation tasks consistently yield the largest and most reliable gender differences of any cognitive task, with men's performance surpassing that of women by approximately 1 SD (Campos, Pérez-Fabello, & GomezJuncal, 2004; Linn & Petersen, 1985; Masters & Sanders, 1993; Voyer, Voyer, & Bryden, 1995).
A predominant explanation for the performance gap in mental rotation involves biological factors, such as differential hormonal levels (Geary, 1995; Kimura & Hampson, 1994). However, studies exploring the impact of stereotypes on cognitive performance suggest that social factors might also play a role. For example, in comparison with men, women perform worse on math tests when they are reminded of gender differences in mathematical abilities (Spencer, Steele, &Quinn, 1999). Steele and colleagues (Steele, 1997; Steele, Spencer, & Aronson, 2002) posited that stereotype threat-the fear of confirming a negative stereotype about a group to which one belongs-is responsible for such decrements in cognitive performance. Social psychologists have demonstrated the effects of stereotype threat on numerous other stigmatized groups, including African Americans (Steele & Aronson, 1995), Asian Americans (Shih, Pittinsky, & Ambady, 1999), and the elderly (Levy, 1996).
The stereotype-threat literature has yielded at least two other noteworthy findings. The first is that the cognitive performance of dominant groups, such as White men, may also be susceptible to stereotype messages (see, e.g., Aronson et al., 1999; Brown & Josephs, 1999; Leyens, Désert, Croizet, & Darcis, 2000). Aronson et al. found that White men who were exposed to the stereotype that Asian men outperform White men in mathematics performed significantly worse on a math exam than did a control group of White men for whom the stereotype was not mentioned. Leyens et al. found selective deficits in men's cognitive performance after exposure to the stereotype that men are relatively inefficient at processing affective information. Men made significantly more errors than did women on a subsequent lexical decision task involving affective processing; however, no differences were found between groups for nonaffective judgments, such as distinguishing between words and nonwords. Thus, men's susceptibility to stereotype messages emphasizing women's abilities was specific to the task reflecting the stereotype.
Another important finding of the stereotype-threat literature is that stereotype messages also may have a positive impact on cognitive performance (see, e.g., Johns, Schmader, & Martens, 2005; Shih et al. …