Girls Can Reach the Top in 3D Mental Rotations as Well: Stereotype Falsification Helps

Abstract

Explaining cognitive sex differences is a great challenge that calls the contribution of many scientific fields like psychology, biology, or sociology. Scientifically, understanding sex differences leads to a better knowledge of some cognitive processes, and practically, it answers to a high social demand given the consequences it may have for the whole social organisation (education, employment, etc.). Multiplicity of approaches is needed because the relation between the factors explaining ability differences (cerebral organisation, hormones, social experience, motivation…) seems highly intricate. Furthermore, most of those factors are difficult to control or manipulate experimentally, especially the social ones. So, despite some inconsistencies, biological hypotheses became very popular since the development of neuroimagery techniques. After decades of investigation in psychology, nobody can deny the existence of cognitive sex differences. No other test than Vandenberg & Kuse’s (1978) mental rotation (MR) task better exhibits how strongly men outperform women in visuo-spatial abilities (Voyer, Voyer, & Bryden, 1995). Regularity and robustness of this effect support the conclusion that social factors have only a minor influence. However, only macro-social factors - like the long term influence of social environment - have only been considered in correlational study. We propose to consider the social context in which a task is performed as a situational variable, manipulable as any experimental factor. Concerning MR, we ask if the test situation emphasizes for women the risk to confirm a negative stereotype concerning their visuo-spatial ability. Women with good visuo-spatial abilities (i.e. at the top of the performance distribution) may be threaten by the possibility to confirm this stereotype (by failing at a difficult MR test). Such self-related threat can precisely interfere with their performance. Steele (1997) theorized this interference and called it stereotype threat (ST). In Spencer, Steele, & Quinn (1999), for example, women with high math ability performed less well (than equally qualified men) on difficult math tests both when they were told that the test produced gender differences and when that information was not given, but performed as well as men when told that no gender differences had been found. The very fact that falsifying the gender stereotype about math not only reduced the male advantage but eliminated it altogether runs counter to any biological account of gender differences in this domain. 3D MR is a basic and pretended “hardware” process which mediates 64% of the gender difference in mathematics (Casey, Nuttal, & Pezaris, 1997). The question that arises therefore is whether the MR process itself is sensible to ST? We proposed the Vandenberg & Kuse MR task to 800 middle school pupils in ecological (full classroom) situation. MR performance were assessed in two conditions : one reproduced the usual “standard” condition of MR testing, whereas a falsification sentence (“usually girls and boys show the same performance on this test”) was added to instructions in the critical falsification condition. Tail-ratio analyses showed that a greater proportion of girl reached the top of the performance distribution in this falsification condition. Further analyses revealed that falsification prevented the drop of girls’ performance observed between the two parts of the test in the standard condition. It appears that during the test time course, girls attempted less and less problems and hence achieved worse. This suggests that under ST, girls accumulated doubts on their ability to perform MR. This effect was particularly strong for pure mental rotation problems. Taken together, these new findings do not just offer further support for ST theory, they also lead toward paying constant attention to the social environment of cognition.