ReducingStereotypeThreat.org

Jamieson & Harkins, 2007

This paper reports four experiments testing the notion that one consequence of stereotype threat is increased motivation to perform well. Increased motivation to perform should facilitate the dominant response on a given task, and dominant responses, if correct, should improve task performance. If the dominant response is incorrect, however, then performance should be harmed under greater motivation. To test these hypotheses, students completed a visual attention task (an antisaccade task) in which an individual must inhibit the tendency to look at an irrelevant, peripheral stimulus while directing vision to determine the orientation of a target stimulus. In Experiment 1, in which the target display duration was set to 150 ms, undergraduate men and women completed this task after it had been described as "linked to math ability" that either had "been shown to produce gender differences" (stereotype threat for women) or "not been shown to produce gender differences" (control). Women under stereotype threat rated the task as more difficult, and women in the stereotype threat condition were less accurate than students in all other conditions in reporting the orientation of the target stimuli. In Experiment 2, in which only women performed, the target display duration was increased to 250 ms which provided threat participants with the opportunity for correcting for the dominant response. At this duration, women in the stereotype threat condition reported target orientation more quickly than controls with no cost in accuracy. Experiment 3 tracked eye movements and showed that women under stereotype threat were more likely to look at the irrelevant cue (demonstrate the dominant response) but were also better able to quickly correct for this error than women in the control condition. On trials where individuals successfully avoided looking at the irrelevant cue, eye glances toward the target were initiated more quickly in the stereotype threat compared with the control condition. Experiment 4 crossed the stereotype threat manipulation with a manipulation of cognitive load. Results under low cognitive load replicated previous findings at the 250 ms display duration: women under stereotype threat responded more quickly than women in the control condition with no cost in accuracy. However, under high cognitive load, women under stereotype threat performed more poorly than women in the control condition. Thus, when working memory was impaired, the performance advantage of threat participants was not only eliminated, but reversed.

The authors argue that these findings are consistent with the mere effort account, but not with a working memory account, which they contend would also predict more looking toward the cue (a result of impaired capacity to inhibit a dominant response), but would predict slower, not faster, attempts to initiate corrective saccades and would not predict faster response times when the threatened individuals’ eyes reached the target area. These results are suggestive that stereotype threat can produce increased motivation to perform on the threat-relevant task and that performance on relatively simple tasks can be facilitated by stereotype threat. Of course, increased motivation might not always benefit task performance, and other factors such as task difficulty play a role in determining the consequences of increased effort. In addition, other demonstrated consequences of stereotype threat (e.g., increased distraction, lowered expectations) might also play roles along with motivation and effort in determining task performance.