- Email: [email protected]
Gender difference does not mean genetic difference: Externalizing improves performance in mental rotation
Learning and Individual Differences 22 (2012) 20–24
Contents lists available at SciVerse ScienceDirect
Learning and Individual Differences journal homepage: www.elsevier.com/locate/lindif
Gender difference does not mean genetic difference: Externalizing improves performance in mental rotation Angelica Moè ⁎ Department of General Psychology, University of Padua, Italy
a r t i c l e
i n f o
Article history: Received 17 March 2011 Received in revised form 15 October 2011 Accepted 8 November 2011 Keywords: Gender differences Genetics Stereotype threat Mental rotation Regulatory focus
a b s t r a c t The fear of underperforming owing to stereotype threat affects women's performance in tasks such as mathematics, chess, and spatial reasoning. The present research considered mental rotation and explored effects on performance and on regulatory focus of instructions pointing to different explanations for gender differences. Two hundred and one participants were asked to perform the Mental Rotation Test (MRT) and were told that men perform better than women. Then they were divided into four sub-groups and provided with no additional information (control condition) or one of three explanations: (a) genetic factors, (b) widely-held stereotype, or (c) time limit. A decrease in performance was predicted for the genetic instruction and an increase for the two alternative explanations based on externalizing. Results showed that both women and men are harmed by the genetic explanation and relieved by both the stereotype and the time limit explanations. Explanations stressing genetics and time limit as factors affecting performance favor prevention focus. © 2011 Elsevier Inc. All rights reserved.
1. Introduction Individual differences based on gender or gender roles have been often studied. Many researchers have studied the differences between the performances of men and women in completing various cognitive tasks. Their results confirmed that there are many gender-related stereotypes or beliefs and few true differences. One difference did show up in a test of mental rotation, the ability to rotate mental representations of two- and three-dimensional objects, a task that men perform better than women due to a wide range of factors. Research has demonstrated that one's performance is not dependent on ability alone but on situational factors, such as gender role beliefs and test instructions. Imagine that in the instructions you are told that there is a gender difference for a task you are asked to perform. What would you think? You may consider genetic factors to be the reason for the difference. Indeed, there is a tendency to attribute differences in ascribed identities (e.g., gender or ethnicity) to biologically-rooted reasons (Prentice & Miller, 2006). Research, however, has widely demonstrated that many other explanations may be valid. One is based on the fear of confirming a widely-held stereotype (Aronson, Quinn, & Spencer, 1998), known as stereotype threat. Reminding a common-held stereotype causes performance decrements in minority members (e.g., women told that men perform better in math, elderly told having poor memory). ⁎ Department of General Psychology Via Venezia, 8 Padua, Italy. Tel.: + 39 049 8276689; fax: + 39 049 8276600. E-mail address: [email protected]. 1041-6080/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.lindif.2011.11.001
Common ways to raise stereotype threat are priming gender in some preliminary questions (Steele & Aronson, 1995) or instructing that gender differences occur in the test at hand (Moè, 2009). Effects due to stereotype threat activation have been documented with mental rotation, an ability in which men have been shown to score up to one standard deviation higher than women (Voyer, Voyer, & Bryden, 1995), due to genetic explanations, such as hemispheric specialization (e.g., Jordan, Wuestenberg, Heinze, Peters, & Jaencke, 2002), and prenatal brain organization (e.g. Burton, Henninger, & Hafetz, 2005), and also to strategic, motivational, experiential, and culturebased factors, including stereotype activation (Moè & Pazzaglia, 2006). Men prefer holistic strategies (e.g., Hugdahl, Thomsen, & Ersland, 2006) that are more effective than analytic (Linn & Petersen, 1985), and they have more functional motivational beliefs than women, for example that men have a higher spatial perception ability (Moè & Pazzaglia, 2006). Due to cultural factors, people consider spatial ability as a predominantly masculine characteristic (Devlin, 2001), and beginning in childhood, boys are encouraged more strongly than girls to engage in activities that promote the development of spatial skills, such as computer games, math and science courses, participating in basket-ball, volleyball or being music performance majors and this experience with spatial activities increases performance in mental rotation (Cherney & Neff, 2004; Ginn & Pickens, 2005). Flaberty (2005) found a positive correlation between the number of masculine activities performed (e.g., carpentry, making objects, building train/car set) and mental rotation scores, irrespective of gender. Terlecki and Newcombe (2005) found that men use computers and videogames more than women and that this kind of spatial experience mediates the gender difference in mental
A. Moè / Learning and Individual Differences 22 (2012) 20–24
rotation performance. Mental rotation performance can be improved in both genders through experience and practice with spatial tasks and through videogame training and repeated testing (Newcombe, Mathason, & Terlecki, 2002; Terlecki, Newcombe, & Little, 2008). People with low baseline spatial abilities need longer interventions and may take some time to begin to show spatial gains, but there is evidence of transfer and long-term maintenance for both genders. This suggests the opportunity to encourage women to acquire spatial experience through appropriate training in order to better perform spatial tasks. In addition, experience aids in developing a higher self-perception of one's ability, which is harmed due to stereotypical threats or having no practice performing a task. Many methods have been proposed for alleviating the stereotype threat effects: priming a positive stereotype (Marx & Roman, 2002), self-affirming (Martens, Johns, Greenberg, & Schimel, 2006), teaching (Aronson, Fried, & Good, 2002) or having (Moè, Meneghetti, & Cadinu, 2009) an incremental theory, remembering positive ascribed identities (McGlone & Aronson, 2006), giving reminders of individuals who have succeeded in the stereotyped task (e.g., women who excelled in masculine professions: McIntyre, Paulson, & Lord, 2003), and explaining that performance depends on effort (Moè & Pazzaglia, 2010) or that the gender gap is due to a stereotype and not to lack of ability (Johns, Schmader, & Martens, 2005). All these results stress the role played by instructions. Among them research has outlined the importance of those priming what the test measures (spatial ability or empathy) in conjunction with gender-role beliefs. Women perform better when instructed that the test (actually a spatial one) measures empathy and they perceive themselves as a feminine person or measure spatial ability and they perceive themselves as a masculine person (Massa, Mayer, & Bohon, 2005). In this study instructions presenting the test as a measure of spatial abilities were provided. Previous research has found that both men and women are affected by instructions stressing that one gender performs better/worse than the other on the task at hand (Moè, 2009; Moè & Pazzaglia, 2006). Unfortunately, nothing was said in the instructions about the reasons for this gender difference, while previous research has shown that the participant's knowledge or the pre-existent causal belief about the reason for underperforming plays a crucial role regarding performance, at least for women. When prevented from believing that what matters are genetic factors, women have been shown not to be affected by the stereotype activation (Dar-Nimrod & Heine, 2006). This is a very intriguing result found only with studies involving mathematics, and it should be confirmed with mental rotation and deserves deeper understanding of the underlying mechanisms. One possibility is facilitating by externalization the reason for underperforming, i.e. by suggesting an external (e.g., stereotypes held by others, situational constraints) rather than internal (e.g., lack of ability) causal attribution for a poor performance. If this is true, alternative instructions favoring externalization should produce the same effects on performance. Among them, those based on commonly-held stereotypes and on time limits were examined. The first aim of this research was to test the hypothesis that an instruction priming a participant that a stereotype is the source of underperformance helps women to increase mental rotation performance. Effects of this instruction were compared with instructions based on genetics. The prediction was that women instructed that men have an advantage that is dependent on a commonly-held stereotype perform better than women instructed that genetic factors are the main source of the observed gender differences in performance. The second aim was to examine whether the advantage for women instructed on stereotype effects depend on providing an external reason (stereotypes does this) for their underperformance. To this end, a third instruction based on externalizing the reason for underperforming was provided. Participants were told that a time restriction negatively affects women's performance. Indeed, this is a
21
true explanation. The gender gap in performance increases with a time restriction (Scali, Brownlow, & Hicks, 2000). The prediction was that this instruction would lead to the same increased performance effect caused by the stereotype instruction. The third aim was to examine effects of instructions on the regulatory focus. Higgins (1999) distinguished between two regulatory focuses or modalities of approaching a test: prevention versus promotion. Prevention is characterized by the fear of failing and it leads to avoiding risks, or being cautious. Persons characterized by a preventive focus tend to provide no answer when they are uncertain. Promotion sustains hope to succeed and it leads to approaching the task even with the risk of making mistakes or not to provide the right answer. It is expected that thinking that genetic factors affect performance can create apprehension and favor a greater prevention focus. Believing that time restriction affects performance can increase a feeling of pressure that could result in a more prevention-oriented focus. The stereotype threat explanation of gender differences should favor a promotion focus because it alleviates the fear of inability. Effects of the three instructions about genetics, stereotype and, the time limit were compared with a control condition in which participants are told that men perform better on the test at hand. This instruction was preferred to one in which nothing is said about gender differences, because it is more explicit. Not all the participants would recognize and endorse the stereotype about women underscoring on the test to the same extent and this differential level of endorsement could cause differential performance within the control group. The fourth aim was to explore effects for men, not considered by the research of Dar-Nimrod and Heine (2006). In previous research considering mental rotation (Moè, 2009; Moè & Pazzaglia, 2006), the fact that the instructions included the words ‘men perform better’ caused men to perform better due to an effect called ‘stereotype lift’ (Walton & Cohen, 2003). However no additional information was provided about the causes for men's higher performance in mental rotation compared to women. It could be expected that stressing genetic causes could cause decreased performance due to the apprehension not to be able to confirm the stereotype suggested by the instructions, an effect called “choking under pressure” (Beilock & Carr, 2001). Effects on performance and on the regulatory focus of the four instructions on genetics, stereotype effects, time limit and control were then compared in order to assess these hypotheses and to confirm that, in women, attributing gender differences to factors different from genetics cause increased performance and, for the stereotype threat explanation, a more promotion-oriented focus. 2. Method 2.1. Participants Ninety-five female and one hundred and six male high school students participated in the research on a voluntary basis. Their mean age was 15.50 years (SD = 0.92, range 14–18 years old). They were divided into four groups, each following different instructions: genetic (n = 56, 26 females and 30 males), stereotype (n = 48, 23 females and 25 males), time limit (n = 51, 20 females and 31 males), and control (n = 46, 26 females and 20 males). Data collection was done in Italy considering students attending different high schools, namely professional (n = 52), technical (n = 93), and liceo (n = 56). None stressed topics such as geometry, technical drawing, or spatial reasoning that previous research has shown enhanced mental rotation performance (e.g., Baenninger & Newcombe, 1995; Kirby & Boulter, 1999). 2.2. Materials Mental rotation ability was assessed by using a modified version (Moè & Pazzaglia, 2006) of the Mental Rotation Test (MRT) originally
22
A. Moè / Learning and Individual Differences 22 (2012) 20–24
devised by Vandenberg and Kuse (1978). Respondents were asked to identify two figures out of four that were identical to the target — a 3D object made up of assembled cubes, but rotated in space. There were two halves of equal difficulty, ten items per half, one administered before and the other after the experimental manipulation. For scoring, one point was assigned for each response item correctly identified (maximum score 20 for each of the two halves). The regulatory focus was measured after mental rotation ability through a self-assessment questionnaire (Promotion Focus Questionnaire) adapted from Maass, D'Ettole, and Cadinu (2008). It assessed the stated tendency to afford the task (promotion) vs. being cautious (prevention) and it presented six opposite attitudes toward the task (e.g., “In this task, it is important to answer only if certain of the correctness” vs. “In this task, it is important to answers all of the questions, with the hope of providing the right answer”). Participants were asked to assign a value ranging from 1 (closer to the prevention focus) to 7 (closer to the promotion focus). A total score of promotion score was obtained by summing up the raw scores assigned by the participants (maximum theoretical score 42). Alpha value for this adapted version was 0.79, M = 23.71, SD = 6.29, range 8–40.
2.3. Procedure Participants were randomly assigned to one of the four groups and tested in collective sessions of mixed gender sized 14 to 23, corresponding to the classes. Past research has shown that women are more affected by gender stereotypes in the presence of men (Inzlicht & Ben-Zeev, 2000): a greater number of men correspond to a greater decrease in women's performance, whereas men are not affected by the gender stereotype, even when they are in the minority. However, in order to create a more ecological situation, the mixedgender testing session was preferred over the same-gender. All materials were on paper. First, participants were presented with the three practice items of the MRT. Then, they were told to complete the first half of the test in 4 min. Previous researchers allotted either 6 or 8 min to perform the whole test. In some studies the test was given in two halves in order to place instructions in the middle, and in this case, researchers always set 4 min to complete each half. The same was done here. A start signal given by the experimenter followed. After a stop signal at the end of the assigned time, the experimenter read aloud the instructions written on a sheet that the participants could also read. All of the groups were instructed in the following manner: “This test measures spatial abilities. These are very important in everyday life, i.e., for finding a route on a map, orienting in a new environment, describing a road to a friend. Research has shown that men perform better than women in this test and obtain higher scores.” Then, different instructions were issued to the different groups, except for the control group (no additional information). The genetic explanation read, “Research has shown that male superiority is caused by biological and genetic factors.” The stereotype explanation read, “This superiority is caused by a gender stereotype, i.e., by a common belief in male superiority in spatial tasks, and has nothing to do with lack of ability.” The time limit explanation read, “Research has shown that women are generally more cautious than men and require more time to answer. Hence, their poor performance is owed to time limit and has nothing to do with lack of ability.” The participants were then asked to complete the second half of the test in 4 min, beginning after the start signal and ending at the stop signal. They were then asked to fill out the Promotion Focus Questionnaire. At the end of the allotted time, they were asked the following two questions for a manipulation check: (1) “In the instructions, were you told that in this test (a) women perform better, (b) men perform better, or (c) there is no gender difference?” and (2) “Is this caused by (a) a stereotype, (b) genetic factors, or (c) time
limit?” Finally all the sheets were collected and participants debriefed and thanked. 3. Analysis and results 3.1. Manipulation check Understanding of instructions was verified. No participant failed to recognize the right instructions (men perform better and the explanation provided) and so all were considered in the analysis. 3.2. Effects of instructions on performance Given that increased or decreased performance was expected, in the analysis the mean improvement was considered as a dependent variable rather than the absolute mean score. Hence, a differential score was computed by subtracting the mean accuracy score before instructions from the mean accuracy score afterwards (M = 0.31, SD= 2.73, range −7 to +7). Then, a 4 groups (genetic, stereotype, time limit, control) × 2 (gender) MANOVA was computed on the differential score, showing a significant effect of groups, F (3, 201) = 5.67, p = 0.001 (see Table 1). Tukey-a post hoc comparisons were run in order to compare the four groups' mean values. They showed that the genetic group scored similarly to the control group and lower than the stereotype and the time limit groups, which did not differ from each other. The same results were found running an ANCOVA 4 (groups) × 2 (gender) on the mean accuracy score after instructions having performance before instructions as covariate: groups significantly differed, F(3, 192) = 6.43, p b 0.001. Men scored higher than women, F(1, 192) = 7.50, p = 0.007. The effect of the covariate was significant as well, F(1, 192) = 162.70, pb 0.00, but this was not due to a difference in mean accuracy scores before instructions. Actually an ANOVA 4 (groups)×2 (gender) run on pre-test scores showed no effect due to groups, whose means and standard deviations are the following: genetic M=12.38, SD=3.84, stereotype M=12.00, SD=3.33, time limit M=12.55, SD=3.52 and control M=12.24, SD=3.78. Neither the main effect of gender [F(1, 201) = 0.02, p > 0.05] nor the interaction [F(3, 201) = 0.14, p > 0.05] yielded significant effects. This does not mean that there was no gender difference, but that the effects of instructions were the same for men and women. Actually, both differences between genders were significant: before instructions, t(199) = 6.30, p b 0.001 [women M = 10.75, SD = 3.32, men M = 13.69, SD = 3.29], and after instructions, t(199) = 6.54, p b 0.001 [women M = 10.99, SD = 3.49, men M = 14.07, SD = 3.17]. 3.3. Effects of instructions on the regulatory focus A 4 groups (genetic, stereotype, time limit, control) × 2 (gender) MANOVA was computed on the promotion regulatory focus score obtained by the Promotion Focus Questionnaire (the higher the score, the greater a person's promotion focus, and lower scores indicate a prevention focus). One female participant assigned to the “time limit” group failed to complete the questionnaire. Hence, the whole sample was made up of 200 rather than 201 participants for
Table 1 Means and SDs of increase (or decrease) in mental rotation performance after instructions on genetic, stereotype, or time limits. Groups
Genetic Stereotype Time limit Control
Males
Females
Total
M
SD
M
SD
M
SD
− 0.60 0.92 1.06 0.10
2.99 2.74 3.03 1.16
− 0.92 1.26 0.95 − 0.04
3.06 2.24 2.42 2.60
− 0.75 1.08 1.02 0.02
3.00 2.49 2.78 2.08
A. Moè / Learning and Individual Differences 22 (2012) 20–24
this analysis. A significant main effect of groups was found: F(3, 200) = 7.06, p b 0.001 (see Table 2). In addition, neither the main effect of gender [F(1, 200) = 0.06, p > 0.05] nor the interaction [F(3, 200) = 0.82, p > 0.05] yielded a statistically significant result, thus suggesting that the effects of instructions are the same for participants of both genders. Tukey-a post hoc comparisons showed that the genetic group scored similarly to the time limit group and lower than both the stereotype and the control groups, which did not differ from each other. 4. Discussion Research has largely confirmed that there is a gender difference that favors men in performing the mental rotation test, and scholars have provided various explanations. The present research did not examined this issue further, but explored the effectiveness of an instructional strategy intended to enable women to be less affected by the awareness to belong to the group who under-perform. Previous research has shown that instructions stressing the belief that performance depends on effort (controllable and malleable cause) rather than on ability (uncontrollable and fixed cause) increases performance in mental rotation (Moè & Pazzaglia, 2010). Here, beliefs in controllable and malleable causes were primed by instructing participants that either gender stereotypes or time limit (rather than genetic factors) are the source of the observed gender differences. Increased performance was expected following this kind of instructions, and a decrease was assumed after instructions based on genetics, which stress the belief that performance depends on uncontrollable and fixed causes. Results confirmed that performance was affected negatively by instructions stressing genetic factors as the source of gender differences in performance and positively by instructions stressing the role of common stereotypes or time limit, which suggests externalization of the source of gender difference in performance. The regulatory focus became more prevention-oriented following instructions stressing time limit and genetic factors as the reasons for the observed gender differences and more promotion-oriented for instruction stressing that a common-held stereotype sustains the gender gap in performance. The difficulty of the test could have played a crucial role for this effect to emerge. Research examining the mechanisms implied in favoring a more prevention-oriented focus showed that it emerges mainly with unpractised items and highly demanding problems (Beilock, Kulp, Holt, & Carr, 2004). The same is true for the size of the gender difference, that it increased in face of difficult items or a difficult situation, namely reduced time. When given unlimited time to finish the task, women perform as well as men (Scali et al., 2000), but this result is controversial. For instance, Peters (2005) found that when the standard time allowed is doubled, women solve more items, but the same is true for men. The level of difficulty of the items can explain this result. When additional time is provided, the extra items solved are the difficult ones, and research has shown that when the difficult increases, the difference between men's and women's performance increases (Collins & Kimura, 1997). No significant difference in performance was found between instructions positing stereotype threat or time limit as the reasons for
Table 2 Means and SDs in the regulatory focus (higher scores means promotion, lower prevention) after instructions on genetic, stereotype, or time limits. Groups
Genetic Stereotype Time limit Control
Males
Females
Total
M
SD
M
SD
M
SD
20.67 25.40 22.06 27.10
5.67 6.63 7.23 3.75
22.38 25.78 22.79 25.12
5.64 5.93 4.97 6.80
21.46 25.58 22.34 25.98
5.67 6.24 6.42 5.71
23
women underscoring. This confirms, as hypothesized, that externalizing (rather than ascribing to personal factors) is the reason for a potential failure that helps women to perform better because it prevents ascribing the gender difference to genetics. However, giving a reminder about stereotype rather than advocating time limit appears to favor both performance and a promotion regulatory focus. Future research could consider another control condition: having no instruction about gender differences. Doing this was not considered here because it would create the possibility that the differential knowledge and endorsement of the stereotype by the control group would affect performance. However, this condition could better assess what happens in natural situations, namely when nothing is overtly stated about commonly-held stereotypes. No gender individual difference was found in the effects of instructions. Men were harmed by the genetic instructions to the same extent as women. However, the underling mechanisms could differ between genders. The awareness of belonging to a group expected to perform well could have created apprehension in men, thus favoring a cautious way of solving the test questions and decreasing performance levels due to choking under pressure. In contrast, a stereotype lift effect occurred for the other instructions. Differently, women, following the genetic explanation, experienced a setback due to the occurrence of a stereotype threat effect, while when they did not received instructions making them think that genetic factors are what affects performance, they were relieved from fearing underperform and they performed better. It is noteworthy that the control group received instructions that are usually provided in order to activate the stereotype threat. They were told that men score higher than women with no additional explanation. Actually, they scored the same as the genetic group. This testifies that, unless alternative instructions are provided, people may assume that gender difference equals genetic difference. Future research could explicitly assess this through a post-session survey. 5. Conclusions Women's mental rotation performance is affected negatively by various factors, among them gender stereotypes, which have been proven to be an important source of individual differences. Results of this study evidenced that performance is not harmed by the knowledge of the stereotype (the stereotype group increased performance), but by accepting it. It does not matter if the genetic explanation is really true or to what extent it is true. What makes the difference is the belief that failures or difficulties are dependent on genetic reasons. In fact, instructions stressing alternative explanations caused increased performance. This is the effect of externalizing, namely of attributing the performance on factors more malleable and over which a person has greater control than those primed by a genetic instruction. Both stereotype and time limit instructions have the same effect of increased performance. Future research could confirm the attributional hypothesis by assessing participants' causal attributions. In addition, the belief about the cause of gender differences in performance could be directly assessed in order to confirm that people hold a genetic explanation for gender differences if nothing is said. As previously outlined, the inclusion of a group having no instruction about what the test measures could help a researcher to understand what happens in natural situations. A future study could confirm these effects due to causal instructions by having the test presented as measuring an ability that is not stereotyped or positively stereotyped, e.g. empathy. Gender-role beliefs could be included in order to have a clearer pattern as observed in the research of Massa et al. (2005). Finally, this instructional strategy (saying that stereotype or time limit, rather than genetics, causes gender differences) could be added or compared with those based on training the participants
24
A. Moè / Learning and Individual Differences 22 (2012) 20–24
with spatial tasks and teaching them the best strategies to use, as done in the research of Newcombe et al. (2002) and Terlecki et al. (2008). Long term effects of the instructions provided could be assessed as well. Overall, the results obtained outlined the importance of instructions in testing because they can foster gender differences in an individual's performance. What is said can cause either an increase or a decrease in performance independently of a person's baseline ability. Actually, men scored higher than women in the mental rotation test used, but the effects of instructions were the same. No individual difference occurred in the effects of instructions. Actually, the same effects on performance have been found in both genders. This is a very interesting result that tells us that women can improve and that even though they start from different points, they can take the same advantage of instructions as men. This shed light on practical applications. Women could perform better in a task where they usually function less well than men (such as mental rotation and, speculatively, every spatial task) if they were told that “commonly-held stereotypes” (better) or “time limits” are the factors that hamper their performance. On the contrary, suggesting or reminding them of genetic reasons could further decrease their performance. Additionally, saying that a “time limit” is the crucial factor could decrease their motivation toward the task and over time their performance because sustaining a prevention regulatory focus makes participants attempt fewer items and hence to achieve a lower score. Using instructions that favor more malleable causal attributions and externalizations should be encouraged for men in order to prevent the observed decrease in their performance that follows giving them genetic instructions The take-home message is that in a task where research has shown women underscoring greatly compared to men and the occurrence of genetic difference as a cause of this gender gap, teaching that a stereotype affects performance is a good way to free women from the fear of underperforming and help them to dare to do their best. Acknowledgments This research has been supported by a Liripac Grant. I wish to thank Dr. Francesco Martello for his help with data collection. I am very grateful to the two anonymous reviewers for their insightful comments. References Aronson, J., Fried, C. B., & Good, C. (2002). Reducing the effects of stereotype threat on African American college students by shaping theories of intelligence. Journal of Experimental Social Psychology, 38, 113–125. Aronson, J., Quinn, D. M., & Spencer, S. J. (1998). Stereotype threat and the academic underperformance of minorities and women. In J. K. Swim, & C. Stangor (Eds.), Prejudice: The target's perspective (pp. 83–103). San Diego, CA: Academic Press. Baenninger, M., & Newcombe, N. (1995). Environmental input to the development of sex-related differences in spatial and mathematical ability. Learning and Individual Differences, 7, 363–379. Beilock, S. L., & Carr, T. H. (2001). On the fragility of skilled performance: What governs choking under pressure? Journal of Experimental Psychology. General, 130, 701–725. Beilock, S. L., Kulp, C. A., Holt, L. E., & Carr, T. H. (2004). More on the fragility of performance: Choking under pressure in mathematical problem solving. Journal of Experimental Psychology. General, 133, 584–600. Burton, L., Henninger, D., & Hafetz, J. (2005). Gender differences in mental rotation, verbal fluency and SAT scores to finger length ratios as hormonal indexes. Developmental Neuropsychology, 28, 493–505. Cherney, I. D., & Neff, N. L. (2004). Role of strategies and prior exposure in mental rotation. Perceptual and Motor Skills, 98, 1269–1282. Collins, D. W., & Kimura, D. (1997). A large sex difference on a two-dimensional mental rotation task. Behavioral Neuroscience, 111, 845–849.
Dar-Nimrod, I., & Heine, S. J. (2006). Exposure to scientific theories affects women's math performance. Science, 314, 435. Devlin, A. S. (2001). Mind and maze. Spatial cognition and environmental behavior. Westport, CT: Praeger Publishers. Flaberty, M. (2005). Gender differences in mental rotation ability in three cultures: Ireland, Ecuador and Japan. Psychologia, 48, 31–38. Ginn, S. R., & Pickens, S. J. (2005). Relationships between spatial activities and scores on the mental rotation test as a function of sex. Perceptual and Motor Skills, 100, 877–881. Higgins, T. E. (1999). Promotion and prevention as a motivational duality: Implications for evaluative processes. In S. Chaiken, & Y. Trope (Eds.), Dual-process theories in social psychology (pp. 503–525). New York, NY, US: Guilford Press. Hugdahl, K., Thomsen, T., & Ersland, L. (2006). Sex differences in visuo-spatial processing: An fMRI study of mental rotation. Neuropsychologia, 44, 1575–1583. Inzlicht, M., & Ben-Zeev, T. (2000). A threatening intellectual environment: Why females are susceptible to experiencing problem-solving deficits in the presence of males. Psychological Science, 11, 365–371. Johns, M., Schmader, T., & Martens, A. (2005). Knowing is half the battle. Teaching stereotype threat as a means of improving women's math performance. Psychological Science, 16, 175–179. Jordan, K., Wuestenberg, T., Heinze, H. J., Peters, M., & Jaencke, L. (2002). Women and men exhibit different cortical activation patterns during mental rotation tasks. Neuropsychologia, 40, 2397–2408. Kirby, J. R., & Boulter, D. R. (1999). Spatial ability and transformational geometry. European Journal of Psychology of Education, 14, 283–294. Linn, M. C., & Petersen, A. C. (1985). Emergence and characterization of sex differences in spatial ability: A meta-analysis. Child Development, 56, 1479–1498. Maass, A., D'Ettole, C., & Cadinu, M. (2008). Checkmate? The role of gender stereotypes in the ultimate intellectual sport. European Journal of Social Psychology, 38(2), 231–245. Martens, A., Johns, M., Greenberg, J., & Schimel, J. (2006). Combating stereotype threat: The effect of self-affirmation on women's intellectual performance. Journal of Experimental Social Psychology, 42, 236–243. Marx, D. M., & Roman, J. S. (2002). Female role models: Protecting women's math performance. Personality and Social Psychology Bulletin, 28, 1183–1193. Massa, L. J., Mayer, R. E., & Bohon, L. M. (2005). Individual differences in gender role beliefs influence spatial ability test performance. Learning and Individual Differences, 15, 99–111. McGlone, M. S., & Aronson, J. (2006). Stereotype threat, identity salience, and spatial reasoning. Journal of Applied Developmental Psychology, 27, 486–493. McIntyre, R. B., Paulson, R. M., & Lord, C. G. (2003). Alleviating women's mathematics stereotype threat through salience of group achievements. Journal of Experimental Social Psychology, 39, 83–90. Moè, A. (2009). Are males always better than females in mental rotation ? Exploring a gender belief explanation. Learning and Individual Differences, 19(1), 21–27. Moè, A., Meneghetti, C., & Cadinu, M. (2009). Women and mental rotation: Incremental theory and spatial strategy use enhance performance. Personality and Individual Differences, 46(2), 187–191. Moè, A., & Pazzaglia, F. (2006). Following the instructions! Effects of gender beliefs in mental rotation. Learning and Individual Differences, 16, 369–377. Moè, A., & Pazzaglia, F. (2010). Beyond genetics in mental rotation test performance. The power of effort attribution. Learning and Individual Differences, 20, 464–468. Newcombe, N. S., Mathason, L., & Terlecki, M. (2002). Maximization of spatial competence: More important than finding the cause of sex differences. In A. McGillicuddyDe Lisi, & R. De Lisi (Eds.), Biology, society and behavior: The development of sex differences in cognition (pp. 183–206). Westport, CT: Ablex Publishing. Peters, M. (2005). Sex differences and the factor of time in solving Vandenberg and Kuse mental rotation problems. Brain and Cognition, 57, 176–184. Prentice, D. A., & Miller, D. T. (2006). Essentializing differences between women and men. Psychological Science, 17(2), 129–135. Scali, R. M., Brownlow, S., & Hicks, J. L. (2000). Gender differences in spatial task performance as a function of speed or accuracy orientation. Sex Roles, 43, 359–376. Steele, C. M., & Aronson, J. (1995). Stereotype threat and the intellectual performance of African Americans. Journal of Personality and Social Psychology, 69, 797–811. Terlecki, M. S., & Newcombe, N. S. (2005). How important is the digital divide? The relation of computer and videogame usage to gender differences in Mental Rotation ability. Sex Roles, 53, 433–441. Terlecki, M. S., Newcombe, N. S., & Little, M. (2008). Durable and generalized effects of spatial experience on Mental Rotation: Gender differences in growth patterns. Applied Cognitive Psychology, 22, 996–1013. Vandenberg, S. G., & Kuse, A. R. (1978). Mental rotations: A group test of threedimensional spatial visualization. Perceptual and Motor Skills, 47, 599–604. Voyer, D., Voyer, S., & Bryden, M. P. (1995). Magnitude of sex differences in spatial abilities: A meta-analysis and consideration of critical variables. Psychological Bulletin, 117, 250–270. Walton, G. M., & Cohen, G. L. (2003). Stereotype lift. Journal of Experimental Social Psychology, 39(5), 456–467.
Contents lists available at SciVerse ScienceDirect
Learning and Individual Differences journal homepage: www.elsevier.com/locate/lindif
Gender difference does not mean genetic difference: Externalizing improves performance in mental rotation Angelica Moè ⁎ Department of General Psychology, University of Padua, Italy
a r t i c l e
i n f o
Article history: Received 17 March 2011 Received in revised form 15 October 2011 Accepted 8 November 2011 Keywords: Gender differences Genetics Stereotype threat Mental rotation Regulatory focus
a b s t r a c t The fear of underperforming owing to stereotype threat affects women's performance in tasks such as mathematics, chess, and spatial reasoning. The present research considered mental rotation and explored effects on performance and on regulatory focus of instructions pointing to different explanations for gender differences. Two hundred and one participants were asked to perform the Mental Rotation Test (MRT) and were told that men perform better than women. Then they were divided into four sub-groups and provided with no additional information (control condition) or one of three explanations: (a) genetic factors, (b) widely-held stereotype, or (c) time limit. A decrease in performance was predicted for the genetic instruction and an increase for the two alternative explanations based on externalizing. Results showed that both women and men are harmed by the genetic explanation and relieved by both the stereotype and the time limit explanations. Explanations stressing genetics and time limit as factors affecting performance favor prevention focus. © 2011 Elsevier Inc. All rights reserved.
1. Introduction Individual differences based on gender or gender roles have been often studied. Many researchers have studied the differences between the performances of men and women in completing various cognitive tasks. Their results confirmed that there are many gender-related stereotypes or beliefs and few true differences. One difference did show up in a test of mental rotation, the ability to rotate mental representations of two- and three-dimensional objects, a task that men perform better than women due to a wide range of factors. Research has demonstrated that one's performance is not dependent on ability alone but on situational factors, such as gender role beliefs and test instructions. Imagine that in the instructions you are told that there is a gender difference for a task you are asked to perform. What would you think? You may consider genetic factors to be the reason for the difference. Indeed, there is a tendency to attribute differences in ascribed identities (e.g., gender or ethnicity) to biologically-rooted reasons (Prentice & Miller, 2006). Research, however, has widely demonstrated that many other explanations may be valid. One is based on the fear of confirming a widely-held stereotype (Aronson, Quinn, & Spencer, 1998), known as stereotype threat. Reminding a common-held stereotype causes performance decrements in minority members (e.g., women told that men perform better in math, elderly told having poor memory). ⁎ Department of General Psychology Via Venezia, 8 Padua, Italy. Tel.: + 39 049 8276689; fax: + 39 049 8276600. E-mail address: [email protected]. 1041-6080/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.lindif.2011.11.001
Common ways to raise stereotype threat are priming gender in some preliminary questions (Steele & Aronson, 1995) or instructing that gender differences occur in the test at hand (Moè, 2009). Effects due to stereotype threat activation have been documented with mental rotation, an ability in which men have been shown to score up to one standard deviation higher than women (Voyer, Voyer, & Bryden, 1995), due to genetic explanations, such as hemispheric specialization (e.g., Jordan, Wuestenberg, Heinze, Peters, & Jaencke, 2002), and prenatal brain organization (e.g. Burton, Henninger, & Hafetz, 2005), and also to strategic, motivational, experiential, and culturebased factors, including stereotype activation (Moè & Pazzaglia, 2006). Men prefer holistic strategies (e.g., Hugdahl, Thomsen, & Ersland, 2006) that are more effective than analytic (Linn & Petersen, 1985), and they have more functional motivational beliefs than women, for example that men have a higher spatial perception ability (Moè & Pazzaglia, 2006). Due to cultural factors, people consider spatial ability as a predominantly masculine characteristic (Devlin, 2001), and beginning in childhood, boys are encouraged more strongly than girls to engage in activities that promote the development of spatial skills, such as computer games, math and science courses, participating in basket-ball, volleyball or being music performance majors and this experience with spatial activities increases performance in mental rotation (Cherney & Neff, 2004; Ginn & Pickens, 2005). Flaberty (2005) found a positive correlation between the number of masculine activities performed (e.g., carpentry, making objects, building train/car set) and mental rotation scores, irrespective of gender. Terlecki and Newcombe (2005) found that men use computers and videogames more than women and that this kind of spatial experience mediates the gender difference in mental
A. Moè / Learning and Individual Differences 22 (2012) 20–24
rotation performance. Mental rotation performance can be improved in both genders through experience and practice with spatial tasks and through videogame training and repeated testing (Newcombe, Mathason, & Terlecki, 2002; Terlecki, Newcombe, & Little, 2008). People with low baseline spatial abilities need longer interventions and may take some time to begin to show spatial gains, but there is evidence of transfer and long-term maintenance for both genders. This suggests the opportunity to encourage women to acquire spatial experience through appropriate training in order to better perform spatial tasks. In addition, experience aids in developing a higher self-perception of one's ability, which is harmed due to stereotypical threats or having no practice performing a task. Many methods have been proposed for alleviating the stereotype threat effects: priming a positive stereotype (Marx & Roman, 2002), self-affirming (Martens, Johns, Greenberg, & Schimel, 2006), teaching (Aronson, Fried, & Good, 2002) or having (Moè, Meneghetti, & Cadinu, 2009) an incremental theory, remembering positive ascribed identities (McGlone & Aronson, 2006), giving reminders of individuals who have succeeded in the stereotyped task (e.g., women who excelled in masculine professions: McIntyre, Paulson, & Lord, 2003), and explaining that performance depends on effort (Moè & Pazzaglia, 2010) or that the gender gap is due to a stereotype and not to lack of ability (Johns, Schmader, & Martens, 2005). All these results stress the role played by instructions. Among them research has outlined the importance of those priming what the test measures (spatial ability or empathy) in conjunction with gender-role beliefs. Women perform better when instructed that the test (actually a spatial one) measures empathy and they perceive themselves as a feminine person or measure spatial ability and they perceive themselves as a masculine person (Massa, Mayer, & Bohon, 2005). In this study instructions presenting the test as a measure of spatial abilities were provided. Previous research has found that both men and women are affected by instructions stressing that one gender performs better/worse than the other on the task at hand (Moè, 2009; Moè & Pazzaglia, 2006). Unfortunately, nothing was said in the instructions about the reasons for this gender difference, while previous research has shown that the participant's knowledge or the pre-existent causal belief about the reason for underperforming plays a crucial role regarding performance, at least for women. When prevented from believing that what matters are genetic factors, women have been shown not to be affected by the stereotype activation (Dar-Nimrod & Heine, 2006). This is a very intriguing result found only with studies involving mathematics, and it should be confirmed with mental rotation and deserves deeper understanding of the underlying mechanisms. One possibility is facilitating by externalization the reason for underperforming, i.e. by suggesting an external (e.g., stereotypes held by others, situational constraints) rather than internal (e.g., lack of ability) causal attribution for a poor performance. If this is true, alternative instructions favoring externalization should produce the same effects on performance. Among them, those based on commonly-held stereotypes and on time limits were examined. The first aim of this research was to test the hypothesis that an instruction priming a participant that a stereotype is the source of underperformance helps women to increase mental rotation performance. Effects of this instruction were compared with instructions based on genetics. The prediction was that women instructed that men have an advantage that is dependent on a commonly-held stereotype perform better than women instructed that genetic factors are the main source of the observed gender differences in performance. The second aim was to examine whether the advantage for women instructed on stereotype effects depend on providing an external reason (stereotypes does this) for their underperformance. To this end, a third instruction based on externalizing the reason for underperforming was provided. Participants were told that a time restriction negatively affects women's performance. Indeed, this is a
21
true explanation. The gender gap in performance increases with a time restriction (Scali, Brownlow, & Hicks, 2000). The prediction was that this instruction would lead to the same increased performance effect caused by the stereotype instruction. The third aim was to examine effects of instructions on the regulatory focus. Higgins (1999) distinguished between two regulatory focuses or modalities of approaching a test: prevention versus promotion. Prevention is characterized by the fear of failing and it leads to avoiding risks, or being cautious. Persons characterized by a preventive focus tend to provide no answer when they are uncertain. Promotion sustains hope to succeed and it leads to approaching the task even with the risk of making mistakes or not to provide the right answer. It is expected that thinking that genetic factors affect performance can create apprehension and favor a greater prevention focus. Believing that time restriction affects performance can increase a feeling of pressure that could result in a more prevention-oriented focus. The stereotype threat explanation of gender differences should favor a promotion focus because it alleviates the fear of inability. Effects of the three instructions about genetics, stereotype and, the time limit were compared with a control condition in which participants are told that men perform better on the test at hand. This instruction was preferred to one in which nothing is said about gender differences, because it is more explicit. Not all the participants would recognize and endorse the stereotype about women underscoring on the test to the same extent and this differential level of endorsement could cause differential performance within the control group. The fourth aim was to explore effects for men, not considered by the research of Dar-Nimrod and Heine (2006). In previous research considering mental rotation (Moè, 2009; Moè & Pazzaglia, 2006), the fact that the instructions included the words ‘men perform better’ caused men to perform better due to an effect called ‘stereotype lift’ (Walton & Cohen, 2003). However no additional information was provided about the causes for men's higher performance in mental rotation compared to women. It could be expected that stressing genetic causes could cause decreased performance due to the apprehension not to be able to confirm the stereotype suggested by the instructions, an effect called “choking under pressure” (Beilock & Carr, 2001). Effects on performance and on the regulatory focus of the four instructions on genetics, stereotype effects, time limit and control were then compared in order to assess these hypotheses and to confirm that, in women, attributing gender differences to factors different from genetics cause increased performance and, for the stereotype threat explanation, a more promotion-oriented focus. 2. Method 2.1. Participants Ninety-five female and one hundred and six male high school students participated in the research on a voluntary basis. Their mean age was 15.50 years (SD = 0.92, range 14–18 years old). They were divided into four groups, each following different instructions: genetic (n = 56, 26 females and 30 males), stereotype (n = 48, 23 females and 25 males), time limit (n = 51, 20 females and 31 males), and control (n = 46, 26 females and 20 males). Data collection was done in Italy considering students attending different high schools, namely professional (n = 52), technical (n = 93), and liceo (n = 56). None stressed topics such as geometry, technical drawing, or spatial reasoning that previous research has shown enhanced mental rotation performance (e.g., Baenninger & Newcombe, 1995; Kirby & Boulter, 1999). 2.2. Materials Mental rotation ability was assessed by using a modified version (Moè & Pazzaglia, 2006) of the Mental Rotation Test (MRT) originally
22
A. Moè / Learning and Individual Differences 22 (2012) 20–24
devised by Vandenberg and Kuse (1978). Respondents were asked to identify two figures out of four that were identical to the target — a 3D object made up of assembled cubes, but rotated in space. There were two halves of equal difficulty, ten items per half, one administered before and the other after the experimental manipulation. For scoring, one point was assigned for each response item correctly identified (maximum score 20 for each of the two halves). The regulatory focus was measured after mental rotation ability through a self-assessment questionnaire (Promotion Focus Questionnaire) adapted from Maass, D'Ettole, and Cadinu (2008). It assessed the stated tendency to afford the task (promotion) vs. being cautious (prevention) and it presented six opposite attitudes toward the task (e.g., “In this task, it is important to answer only if certain of the correctness” vs. “In this task, it is important to answers all of the questions, with the hope of providing the right answer”). Participants were asked to assign a value ranging from 1 (closer to the prevention focus) to 7 (closer to the promotion focus). A total score of promotion score was obtained by summing up the raw scores assigned by the participants (maximum theoretical score 42). Alpha value for this adapted version was 0.79, M = 23.71, SD = 6.29, range 8–40.
2.3. Procedure Participants were randomly assigned to one of the four groups and tested in collective sessions of mixed gender sized 14 to 23, corresponding to the classes. Past research has shown that women are more affected by gender stereotypes in the presence of men (Inzlicht & Ben-Zeev, 2000): a greater number of men correspond to a greater decrease in women's performance, whereas men are not affected by the gender stereotype, even when they are in the minority. However, in order to create a more ecological situation, the mixedgender testing session was preferred over the same-gender. All materials were on paper. First, participants were presented with the three practice items of the MRT. Then, they were told to complete the first half of the test in 4 min. Previous researchers allotted either 6 or 8 min to perform the whole test. In some studies the test was given in two halves in order to place instructions in the middle, and in this case, researchers always set 4 min to complete each half. The same was done here. A start signal given by the experimenter followed. After a stop signal at the end of the assigned time, the experimenter read aloud the instructions written on a sheet that the participants could also read. All of the groups were instructed in the following manner: “This test measures spatial abilities. These are very important in everyday life, i.e., for finding a route on a map, orienting in a new environment, describing a road to a friend. Research has shown that men perform better than women in this test and obtain higher scores.” Then, different instructions were issued to the different groups, except for the control group (no additional information). The genetic explanation read, “Research has shown that male superiority is caused by biological and genetic factors.” The stereotype explanation read, “This superiority is caused by a gender stereotype, i.e., by a common belief in male superiority in spatial tasks, and has nothing to do with lack of ability.” The time limit explanation read, “Research has shown that women are generally more cautious than men and require more time to answer. Hence, their poor performance is owed to time limit and has nothing to do with lack of ability.” The participants were then asked to complete the second half of the test in 4 min, beginning after the start signal and ending at the stop signal. They were then asked to fill out the Promotion Focus Questionnaire. At the end of the allotted time, they were asked the following two questions for a manipulation check: (1) “In the instructions, were you told that in this test (a) women perform better, (b) men perform better, or (c) there is no gender difference?” and (2) “Is this caused by (a) a stereotype, (b) genetic factors, or (c) time
limit?” Finally all the sheets were collected and participants debriefed and thanked. 3. Analysis and results 3.1. Manipulation check Understanding of instructions was verified. No participant failed to recognize the right instructions (men perform better and the explanation provided) and so all were considered in the analysis. 3.2. Effects of instructions on performance Given that increased or decreased performance was expected, in the analysis the mean improvement was considered as a dependent variable rather than the absolute mean score. Hence, a differential score was computed by subtracting the mean accuracy score before instructions from the mean accuracy score afterwards (M = 0.31, SD= 2.73, range −7 to +7). Then, a 4 groups (genetic, stereotype, time limit, control) × 2 (gender) MANOVA was computed on the differential score, showing a significant effect of groups, F (3, 201) = 5.67, p = 0.001 (see Table 1). Tukey-a post hoc comparisons were run in order to compare the four groups' mean values. They showed that the genetic group scored similarly to the control group and lower than the stereotype and the time limit groups, which did not differ from each other. The same results were found running an ANCOVA 4 (groups) × 2 (gender) on the mean accuracy score after instructions having performance before instructions as covariate: groups significantly differed, F(3, 192) = 6.43, p b 0.001. Men scored higher than women, F(1, 192) = 7.50, p = 0.007. The effect of the covariate was significant as well, F(1, 192) = 162.70, pb 0.00, but this was not due to a difference in mean accuracy scores before instructions. Actually an ANOVA 4 (groups)×2 (gender) run on pre-test scores showed no effect due to groups, whose means and standard deviations are the following: genetic M=12.38, SD=3.84, stereotype M=12.00, SD=3.33, time limit M=12.55, SD=3.52 and control M=12.24, SD=3.78. Neither the main effect of gender [F(1, 201) = 0.02, p > 0.05] nor the interaction [F(3, 201) = 0.14, p > 0.05] yielded significant effects. This does not mean that there was no gender difference, but that the effects of instructions were the same for men and women. Actually, both differences between genders were significant: before instructions, t(199) = 6.30, p b 0.001 [women M = 10.75, SD = 3.32, men M = 13.69, SD = 3.29], and after instructions, t(199) = 6.54, p b 0.001 [women M = 10.99, SD = 3.49, men M = 14.07, SD = 3.17]. 3.3. Effects of instructions on the regulatory focus A 4 groups (genetic, stereotype, time limit, control) × 2 (gender) MANOVA was computed on the promotion regulatory focus score obtained by the Promotion Focus Questionnaire (the higher the score, the greater a person's promotion focus, and lower scores indicate a prevention focus). One female participant assigned to the “time limit” group failed to complete the questionnaire. Hence, the whole sample was made up of 200 rather than 201 participants for
Table 1 Means and SDs of increase (or decrease) in mental rotation performance after instructions on genetic, stereotype, or time limits. Groups
Genetic Stereotype Time limit Control
Males
Females
Total
M
SD
M
SD
M
SD
− 0.60 0.92 1.06 0.10
2.99 2.74 3.03 1.16
− 0.92 1.26 0.95 − 0.04
3.06 2.24 2.42 2.60
− 0.75 1.08 1.02 0.02
3.00 2.49 2.78 2.08
A. Moè / Learning and Individual Differences 22 (2012) 20–24
this analysis. A significant main effect of groups was found: F(3, 200) = 7.06, p b 0.001 (see Table 2). In addition, neither the main effect of gender [F(1, 200) = 0.06, p > 0.05] nor the interaction [F(3, 200) = 0.82, p > 0.05] yielded a statistically significant result, thus suggesting that the effects of instructions are the same for participants of both genders. Tukey-a post hoc comparisons showed that the genetic group scored similarly to the time limit group and lower than both the stereotype and the control groups, which did not differ from each other. 4. Discussion Research has largely confirmed that there is a gender difference that favors men in performing the mental rotation test, and scholars have provided various explanations. The present research did not examined this issue further, but explored the effectiveness of an instructional strategy intended to enable women to be less affected by the awareness to belong to the group who under-perform. Previous research has shown that instructions stressing the belief that performance depends on effort (controllable and malleable cause) rather than on ability (uncontrollable and fixed cause) increases performance in mental rotation (Moè & Pazzaglia, 2010). Here, beliefs in controllable and malleable causes were primed by instructing participants that either gender stereotypes or time limit (rather than genetic factors) are the source of the observed gender differences. Increased performance was expected following this kind of instructions, and a decrease was assumed after instructions based on genetics, which stress the belief that performance depends on uncontrollable and fixed causes. Results confirmed that performance was affected negatively by instructions stressing genetic factors as the source of gender differences in performance and positively by instructions stressing the role of common stereotypes or time limit, which suggests externalization of the source of gender difference in performance. The regulatory focus became more prevention-oriented following instructions stressing time limit and genetic factors as the reasons for the observed gender differences and more promotion-oriented for instruction stressing that a common-held stereotype sustains the gender gap in performance. The difficulty of the test could have played a crucial role for this effect to emerge. Research examining the mechanisms implied in favoring a more prevention-oriented focus showed that it emerges mainly with unpractised items and highly demanding problems (Beilock, Kulp, Holt, & Carr, 2004). The same is true for the size of the gender difference, that it increased in face of difficult items or a difficult situation, namely reduced time. When given unlimited time to finish the task, women perform as well as men (Scali et al., 2000), but this result is controversial. For instance, Peters (2005) found that when the standard time allowed is doubled, women solve more items, but the same is true for men. The level of difficulty of the items can explain this result. When additional time is provided, the extra items solved are the difficult ones, and research has shown that when the difficult increases, the difference between men's and women's performance increases (Collins & Kimura, 1997). No significant difference in performance was found between instructions positing stereotype threat or time limit as the reasons for
Table 2 Means and SDs in the regulatory focus (higher scores means promotion, lower prevention) after instructions on genetic, stereotype, or time limits. Groups
Genetic Stereotype Time limit Control
Males
Females
Total
M
SD
M
SD
M
SD
20.67 25.40 22.06 27.10
5.67 6.63 7.23 3.75
22.38 25.78 22.79 25.12
5.64 5.93 4.97 6.80
21.46 25.58 22.34 25.98
5.67 6.24 6.42 5.71
23
women underscoring. This confirms, as hypothesized, that externalizing (rather than ascribing to personal factors) is the reason for a potential failure that helps women to perform better because it prevents ascribing the gender difference to genetics. However, giving a reminder about stereotype rather than advocating time limit appears to favor both performance and a promotion regulatory focus. Future research could consider another control condition: having no instruction about gender differences. Doing this was not considered here because it would create the possibility that the differential knowledge and endorsement of the stereotype by the control group would affect performance. However, this condition could better assess what happens in natural situations, namely when nothing is overtly stated about commonly-held stereotypes. No gender individual difference was found in the effects of instructions. Men were harmed by the genetic instructions to the same extent as women. However, the underling mechanisms could differ between genders. The awareness of belonging to a group expected to perform well could have created apprehension in men, thus favoring a cautious way of solving the test questions and decreasing performance levels due to choking under pressure. In contrast, a stereotype lift effect occurred for the other instructions. Differently, women, following the genetic explanation, experienced a setback due to the occurrence of a stereotype threat effect, while when they did not received instructions making them think that genetic factors are what affects performance, they were relieved from fearing underperform and they performed better. It is noteworthy that the control group received instructions that are usually provided in order to activate the stereotype threat. They were told that men score higher than women with no additional explanation. Actually, they scored the same as the genetic group. This testifies that, unless alternative instructions are provided, people may assume that gender difference equals genetic difference. Future research could explicitly assess this through a post-session survey. 5. Conclusions Women's mental rotation performance is affected negatively by various factors, among them gender stereotypes, which have been proven to be an important source of individual differences. Results of this study evidenced that performance is not harmed by the knowledge of the stereotype (the stereotype group increased performance), but by accepting it. It does not matter if the genetic explanation is really true or to what extent it is true. What makes the difference is the belief that failures or difficulties are dependent on genetic reasons. In fact, instructions stressing alternative explanations caused increased performance. This is the effect of externalizing, namely of attributing the performance on factors more malleable and over which a person has greater control than those primed by a genetic instruction. Both stereotype and time limit instructions have the same effect of increased performance. Future research could confirm the attributional hypothesis by assessing participants' causal attributions. In addition, the belief about the cause of gender differences in performance could be directly assessed in order to confirm that people hold a genetic explanation for gender differences if nothing is said. As previously outlined, the inclusion of a group having no instruction about what the test measures could help a researcher to understand what happens in natural situations. A future study could confirm these effects due to causal instructions by having the test presented as measuring an ability that is not stereotyped or positively stereotyped, e.g. empathy. Gender-role beliefs could be included in order to have a clearer pattern as observed in the research of Massa et al. (2005). Finally, this instructional strategy (saying that stereotype or time limit, rather than genetics, causes gender differences) could be added or compared with those based on training the participants
24
A. Moè / Learning and Individual Differences 22 (2012) 20–24
with spatial tasks and teaching them the best strategies to use, as done in the research of Newcombe et al. (2002) and Terlecki et al. (2008). Long term effects of the instructions provided could be assessed as well. Overall, the results obtained outlined the importance of instructions in testing because they can foster gender differences in an individual's performance. What is said can cause either an increase or a decrease in performance independently of a person's baseline ability. Actually, men scored higher than women in the mental rotation test used, but the effects of instructions were the same. No individual difference occurred in the effects of instructions. Actually, the same effects on performance have been found in both genders. This is a very interesting result that tells us that women can improve and that even though they start from different points, they can take the same advantage of instructions as men. This shed light on practical applications. Women could perform better in a task where they usually function less well than men (such as mental rotation and, speculatively, every spatial task) if they were told that “commonly-held stereotypes” (better) or “time limits” are the factors that hamper their performance. On the contrary, suggesting or reminding them of genetic reasons could further decrease their performance. Additionally, saying that a “time limit” is the crucial factor could decrease their motivation toward the task and over time their performance because sustaining a prevention regulatory focus makes participants attempt fewer items and hence to achieve a lower score. Using instructions that favor more malleable causal attributions and externalizations should be encouraged for men in order to prevent the observed decrease in their performance that follows giving them genetic instructions The take-home message is that in a task where research has shown women underscoring greatly compared to men and the occurrence of genetic difference as a cause of this gender gap, teaching that a stereotype affects performance is a good way to free women from the fear of underperforming and help them to dare to do their best. Acknowledgments This research has been supported by a Liripac Grant. I wish to thank Dr. Francesco Martello for his help with data collection. I am very grateful to the two anonymous reviewers for their insightful comments. References Aronson, J., Fried, C. B., & Good, C. (2002). Reducing the effects of stereotype threat on African American college students by shaping theories of intelligence. Journal of Experimental Social Psychology, 38, 113–125. Aronson, J., Quinn, D. M., & Spencer, S. J. (1998). Stereotype threat and the academic underperformance of minorities and women. In J. K. Swim, & C. Stangor (Eds.), Prejudice: The target's perspective (pp. 83–103). San Diego, CA: Academic Press. Baenninger, M., & Newcombe, N. (1995). Environmental input to the development of sex-related differences in spatial and mathematical ability. Learning and Individual Differences, 7, 363–379. Beilock, S. L., & Carr, T. H. (2001). On the fragility of skilled performance: What governs choking under pressure? Journal of Experimental Psychology. General, 130, 701–725. Beilock, S. L., Kulp, C. A., Holt, L. E., & Carr, T. H. (2004). More on the fragility of performance: Choking under pressure in mathematical problem solving. Journal of Experimental Psychology. General, 133, 584–600. Burton, L., Henninger, D., & Hafetz, J. (2005). Gender differences in mental rotation, verbal fluency and SAT scores to finger length ratios as hormonal indexes. Developmental Neuropsychology, 28, 493–505. Cherney, I. D., & Neff, N. L. (2004). Role of strategies and prior exposure in mental rotation. Perceptual and Motor Skills, 98, 1269–1282. Collins, D. W., & Kimura, D. (1997). A large sex difference on a two-dimensional mental rotation task. Behavioral Neuroscience, 111, 845–849.
Dar-Nimrod, I., & Heine, S. J. (2006). Exposure to scientific theories affects women's math performance. Science, 314, 435. Devlin, A. S. (2001). Mind and maze. Spatial cognition and environmental behavior. Westport, CT: Praeger Publishers. Flaberty, M. (2005). Gender differences in mental rotation ability in three cultures: Ireland, Ecuador and Japan. Psychologia, 48, 31–38. Ginn, S. R., & Pickens, S. J. (2005). Relationships between spatial activities and scores on the mental rotation test as a function of sex. Perceptual and Motor Skills, 100, 877–881. Higgins, T. E. (1999). Promotion and prevention as a motivational duality: Implications for evaluative processes. In S. Chaiken, & Y. Trope (Eds.), Dual-process theories in social psychology (pp. 503–525). New York, NY, US: Guilford Press. Hugdahl, K., Thomsen, T., & Ersland, L. (2006). Sex differences in visuo-spatial processing: An fMRI study of mental rotation. Neuropsychologia, 44, 1575–1583. Inzlicht, M., & Ben-Zeev, T. (2000). A threatening intellectual environment: Why females are susceptible to experiencing problem-solving deficits in the presence of males. Psychological Science, 11, 365–371. Johns, M., Schmader, T., & Martens, A. (2005). Knowing is half the battle. Teaching stereotype threat as a means of improving women's math performance. Psychological Science, 16, 175–179. Jordan, K., Wuestenberg, T., Heinze, H. J., Peters, M., & Jaencke, L. (2002). Women and men exhibit different cortical activation patterns during mental rotation tasks. Neuropsychologia, 40, 2397–2408. Kirby, J. R., & Boulter, D. R. (1999). Spatial ability and transformational geometry. European Journal of Psychology of Education, 14, 283–294. Linn, M. C., & Petersen, A. C. (1985). Emergence and characterization of sex differences in spatial ability: A meta-analysis. Child Development, 56, 1479–1498. Maass, A., D'Ettole, C., & Cadinu, M. (2008). Checkmate? The role of gender stereotypes in the ultimate intellectual sport. European Journal of Social Psychology, 38(2), 231–245. Martens, A., Johns, M., Greenberg, J., & Schimel, J. (2006). Combating stereotype threat: The effect of self-affirmation on women's intellectual performance. Journal of Experimental Social Psychology, 42, 236–243. Marx, D. M., & Roman, J. S. (2002). Female role models: Protecting women's math performance. Personality and Social Psychology Bulletin, 28, 1183–1193. Massa, L. J., Mayer, R. E., & Bohon, L. M. (2005). Individual differences in gender role beliefs influence spatial ability test performance. Learning and Individual Differences, 15, 99–111. McGlone, M. S., & Aronson, J. (2006). Stereotype threat, identity salience, and spatial reasoning. Journal of Applied Developmental Psychology, 27, 486–493. McIntyre, R. B., Paulson, R. M., & Lord, C. G. (2003). Alleviating women's mathematics stereotype threat through salience of group achievements. Journal of Experimental Social Psychology, 39, 83–90. Moè, A. (2009). Are males always better than females in mental rotation ? Exploring a gender belief explanation. Learning and Individual Differences, 19(1), 21–27. Moè, A., Meneghetti, C., & Cadinu, M. (2009). Women and mental rotation: Incremental theory and spatial strategy use enhance performance. Personality and Individual Differences, 46(2), 187–191. Moè, A., & Pazzaglia, F. (2006). Following the instructions! Effects of gender beliefs in mental rotation. Learning and Individual Differences, 16, 369–377. Moè, A., & Pazzaglia, F. (2010). Beyond genetics in mental rotation test performance. The power of effort attribution. Learning and Individual Differences, 20, 464–468. Newcombe, N. S., Mathason, L., & Terlecki, M. (2002). Maximization of spatial competence: More important than finding the cause of sex differences. In A. McGillicuddyDe Lisi, & R. De Lisi (Eds.), Biology, society and behavior: The development of sex differences in cognition (pp. 183–206). Westport, CT: Ablex Publishing. Peters, M. (2005). Sex differences and the factor of time in solving Vandenberg and Kuse mental rotation problems. Brain and Cognition, 57, 176–184. Prentice, D. A., & Miller, D. T. (2006). Essentializing differences between women and men. Psychological Science, 17(2), 129–135. Scali, R. M., Brownlow, S., & Hicks, J. L. (2000). Gender differences in spatial task performance as a function of speed or accuracy orientation. Sex Roles, 43, 359–376. Steele, C. M., & Aronson, J. (1995). Stereotype threat and the intellectual performance of African Americans. Journal of Personality and Social Psychology, 69, 797–811. Terlecki, M. S., & Newcombe, N. S. (2005). How important is the digital divide? The relation of computer and videogame usage to gender differences in Mental Rotation ability. Sex Roles, 53, 433–441. Terlecki, M. S., Newcombe, N. S., & Little, M. (2008). Durable and generalized effects of spatial experience on Mental Rotation: Gender differences in growth patterns. Applied Cognitive Psychology, 22, 996–1013. Vandenberg, S. G., & Kuse, A. R. (1978). Mental rotations: A group test of threedimensional spatial visualization. Perceptual and Motor Skills, 47, 599–604. Voyer, D., Voyer, S., & Bryden, M. P. (1995). Magnitude of sex differences in spatial abilities: A meta-analysis and consideration of critical variables. Psychological Bulletin, 117, 250–270. Walton, G. M., & Cohen, G. L. (2003). Stereotype lift. Journal of Experimental Social Psychology, 39(5), 456–467.