| Stereotype threat (i.e., the concern that is experienced when one feels "at risk of confirming, as self-characteristic, a negative stereotype about one's group" (Steele & Aronson, 1995, abstract)) has been shown to negatively impact performance. A working memory interference account for this effect suggests that cognitive resources that could be devoted to task performance are instead expended on processing the information resulting from the activation of the negative stereotype. It is this reduction in working memory capacity that produces the performance debilitation. More specifically, Beilock et al. (2007) suggest that these worries occupy the phonological loop of Baddeley's (1986; 2000; Baddeley & Logie, 1999) multicomponent model of working memory. According to Trbovich and LeFevre (2003), participants solving mental arithmetic problems in a horizontal format rely on the phonological loop to keep track of intermediate values. Thus, Beilock et al. suggests that when threatened participants are presented with horizontal mental arithmetic problems that involve borrowing (e.g., 57 - 28), performance should be debilitated. To test this hypothesis, females completed horizontal or vertical modular arithmetic problems (subtracting the second number from the first, dividing by the mod, and then reporting, true or false, whether the result was a whole number) under threat and no threat. Consistent with the prediction, the performance of threatened females was debilitated on horizontal problems.;Although Beilock et al.'s (2007) findings support the working memory account, they do not rule out the possibility of a motivational explanation: the mere effort account (Harkins, 2006). The mere effort account argues that stereotype threat motivates stigmatized participants to want to perform well, potentiating the prepotent response on the given task (Jamieson & Harkins, 2007; 2009). If the prepotent response is correct, stereotype threat facilitates performance for stigmatized individuals. If the prepotent response is incorrect and participants do not know, or lack the knowledge or time required for correction, performance is debilitated.;In the current studies, I examined how working memory and mere effort contribute to stereotype threat's effects on horizontal arithmetic. In Experiment 1, replicating Beilock et al. (2007) in a paradigm sensitive to motivational effects, I found that threatened females did worse on horizontal modular arithmetic problems, but, consistent with a motivational account, they performed better on vertical problems than non-threatened females. Experiment 2 showed that threat facilitated division performance, suggesting that division alone does not account for the debilitation in modular arithmetic. Additionally, Beilock et al.'s account itself focuses on the subtraction component of the modular arithmetic problems. Focusing on subtraction, Experiments 3 and 4 identified a possible prepotent response for horizontal subtraction, termed the method of adjustment (e.g., adjust the second number to the nearest 10, subtract the two numbers, and then add the adjustment). Experiment 3 showed that participants prefer to work horizontal mental arithmetic problems from left-to-right and report using the method of adjustment. Consistent with the mere effort account, Experiment 4 found that stereotype threat potentiated the prepotent approach, the method of adjustment. Experiment 5 pitted the mere effort account against the working memory account. Working memory predicts debilitation effects on horizontal subtraction problems when participants need to use the phonological loop (i.e., entering answers from left-to-right), whereas mere effort predicts that the potentiated use of the method of adjustment should facilitate performance when answers must be entered from left-to-right. Results supported the mere effort account. Overall, these experiments support a motivational, mere effort, account of stereotype threat's effects on horizontal subtraction performance. |
