Investigating Physiological and Self-Reported Mediators of Stereotype Lift Effects on a Motor Task

Abstract

Achievement gaps between social groups may result from stereotype threat effects but also from stereotype lift effects—the performance boost caused by the awareness that an outgroup is negatively stereotyped. We examined stereotype lift and threat effects in the motor domain and investigated their mediation by task involvement and self-confidence, measured by heart rate reactivity and self-reported indices. Males and females performed a balance task about which negative stereotypes about either males or females were given. No gender information was given in a control condition. Results showed no stereotype threat but a stereotype lift effect, participants performing significantly better after negative outgroup stereotypes were explicitly linked to performance on the balance task compared to the control condition. Concerning males, this effect was mediated by higher self-confidence and task involvement. The implications of these results for understanding the gender inequalities in the motor domain are discussed.

We would like to thank Mark Seery for his helpful comments on earlier drafts of this article and Hervé Dubouchaud for providing the physiological recording equipment.

Notes

¹Blood pressure (BP) indexes were also collected. However the meaning of BP is unclear, interpreted as an indicator of threat in some studies (e.g., Blascovich et al., 2001) and effort in others (e.g., Wright & Kirby, 2001). Moreover the analyses showed no effect of test frame manipulation on BP indexes, so they were dropped from this study.

Note: Pre-performance HR and mean HR during test are covariance-adjusted for mean baseline.

²Individual difference variables, like self-engagement in the performance domain (Stone, 2002; Stone et al., 1999), social anxiety or self-confidence, could be an explanation of the absence of ST. We examined whether the relationship between the activation of a negative ingroup stereotype and performance was moderated by pre-performance HR reactivity, anxiety or self-confidence, but these analyses did not reveal any significant interactions between the individual difference variables and the threat manipulation on the primary dependent measures. Investigating personal and contextual factors that determine threat responses when explicit negative stereotypes are linked to performance represents an important direction for future research.

Note: The correlations for males (n = 32) are below the diagonal, the correlations for females (n = 31) are above the diagonal. Physiological indices are the residuals of the reactivity scores regressed onto their baseline values. Test frame manipulation was coded as follows for males: 1: females inferior condition; 0: males inferior condition; –1: control condition, and as follows for females: 1: males inferior condition; 0: females inferior condition; –1: control condition.

^† p < .10

∗p < .05

∗∗p < .01

∗∗∗p < .001.

³When self-confidence and pre-performance HR reactivity were included in separate analyses, they both significantly predicted performance (β = .38, p = .03 and β = .44, p = .01, respectively), whereas the relationship between test frame manipulation and performance was not significant (β = .26, ns for both analyses).