Prenatal testosterone does not explain sex differences in spatial ability


The most consistent sex differences in cognition are found for spatial ability, in which males, on average, outperform females. Utilizing a twin design, two studies have shown that females with male co-twins perform better than females with female co-twins on a mental rotation task. According to the Twin Testosterone Transfer hypothesis (TTT) this advantage is due to in-uterine transmission of testosterone from males to females. The present study tested the TTT across 14 different spatial ability measures, including mental rotation tasks, in a large sample of 19–21-year-old twins. Males performed significantly better than females on all spatial tasks, with effect sizes ranging from η2 = 0.02 to η2 = 0.16. Females with a male co-twin outperformed females with a female co-twin in two of the tasks. The effect sizes for both differences were negligible (η2 < 0.02). Contrary to the previous studies, our results gave no indication that prenatally transferred testosterone, from a male to a female twin, influences sex differences in spatial ability.


Sex differences are small to negligible in most cognitive traits1,2. However, some measures show differences between males and females3,4,5,6. The most consistent finding of cognitive sex differences comes from spatial ability, in which men, on average, consistently outperform women7,8. This finding has been replicated in large cross-cultural samples9,10. As spatial skills have shown positive correlations with academic and career success in the Science, Technology, Engineering and Mathematics (STEM) disciplines11, they may partly account for the current underrepresentation of women in these areas12,13.

Spatial ability can be described as the ability to produce, recall, store, and modify spatial relations among objects14 and to visualize the transformation of these relations due to changes, for example, in perspective15,16,17. Spatial ability is a component of general cognitive ability, alongside many others, such as working memory, verbal ability and processing speed18. However, spatial ability also involves aspects differentiable from general cognitive ability18,19. Conceptually, spatial ability is often described as having several, separate components, although their definitions overlap19. This may be partly due to the complex nature of the tasks, requiring many parallel cognitive processes, such as apprehending and encoding spatial forms20. However, two recent studies, using data from a large longitudinal twin sample in the United Kingdom (also used in this study), have shown evidence for a uni-factorial structure of spatial ability across a variety of different spatial measures, both phenotypically and genetically19,21. In both studies, the first order factor explained approximately 42 per cent of the variance across diverse spatial tasks.

Previous studies have shown that males outperform females in several spatial ability measures7. For example, studies on spatial navigation have showed large sex differences favoring men22,23. Also, on a mental rotation task (MRT), males outperform females by almost one standard deviation7. A MRT is an example of a widely-used measure of spatial visualization, which involves complex, multi-stage manipulations of spatial information24. This difference has been found consistently across several cultures6 and it has been documented in infants as young as 3–5 months25. However, one spatial task in which females perform better than males is object location memory. A meta-analysis found an overall effect size of d = 0.27 favoring females over males in a memory-dependent spatial task of remembering object locations6.

The factors that contribute to sex differences in spatial ability are still poorly understood. Both biological and environmental factors have been proposed to initiate and maintain the sex differences in spatial ability. Environmental explanations have highlighted the role of previous experiences and learning environments26,27, whereas biological investigations have concentrated on genetic and hormonal effects28,29,30,31. To date, behavioral genetic studies have only found small32, or non-existent33, sex differences in etiology of any cognitive abilities. In line with this, studies on the etiology of spatial ability found only negligible differences in genetic and environmental factors driving sex differences in spatial ability in males and females19, including in mental rotation28.

Hormonal effects, which are influenced by genes and environments, are a biological mechanism affecting spatial cognition34,35. The sex hormone testosterone, necessary for sexual development and sexual behavior, is also present in brain areas associated with cognitive abilities such as language and spatial ability36,37. Some evolutionary arguments suggest that testosterone is a factor maintaining spatial sex differences34. According to such accounts, the greater elaboration of the neurocognitive basis of spatial ability, especially in 3-dimensional environments, is due to navigating and tracking movement that had more evolutionary relevance for males than females34,35. In line with the evolutionary argument, naturally occurring testosterone levels vary between sexes: typical testosterone levels in clinical assessment, measured in blood, range between 0.5 to 2.4 nmol/L in females, and from 10 to 38 nmol/Lin males38.

Several studies have investigated the effect of testosterone on individual differences in spatial ability within sexes. Studies have shown that better spatial ability was associated with elevated testosterone levels, both due to natural fluctuations and extraneous administration39,40,41,42. For example, a study showed that females with higher levels of testosterone performed significantly better than females with lower testosterone levels on a visual maze task39. Other studies have reported the relationship between extraneous administration of testosterone and improvement in spatial ability task performance. The effect has been reported among older men40; female-to-male transsexuals41; and young women42. However, not all studies have supported the association between elevated testosterone levels and better spatial ability performance among females. One study found no within-sex associations between mental rotation task and endogenous, non-fluctuating testosterone when measured in blood43. Among males, the studies on the relationship between the level of testosterone and spatial ability performance have shown mixed results. Some studies have reported that high levels of testosterone are negatively associated with spatial ability among males44,45. One study tested and supported a curvilinear relationship between testosterone levels and spatial ability performance among young adults, suggesting that after exceeding an optimal level, additional testosterone may impair spatial performance46.

Prenatal testosterone may also influence cognitive development: it affects brain functions and neural structure during early prenatal development47,48. One line of evidence for the association between elevated prenatal testosterone levels and increased spatial ability in females comes from clinical samples. Congenital Adrenal Hyperplasia (CAH) is a genetic condition that elevates fetal testosterone levels. A meta-analysis of studies on the association between CAH and spatial ability found that females with CAH perform better on spatial tasks in comparison to control groups49. However, the evidence for the role of prenatal testosterone in spatial ability is mixed. Some studies have reported null results when the exposure on prenatal testosterone was measured as a 2D:4D ratio50,51. Additionally, one study found no difference in mental rotation performance between CAH females and the control group45.

Levels of prenatal testosterone can be measured in amniotic fluid. One study found an association between higher testosterone levels in amniotic fluid and better mental rotation ability for girls at age 752. However, the sample and effect sizes in amniotic fluid studies are small, calling for further investigations to have confidence in the results53. Additionally, studies utilizing twin samples have found evidence for a beneficial effect for females of having a male twin. Several studies have explored the Twin Testosterone Transfer Hypothesis36,54,55. According to the TTT, having a male co-twin improves females’ spatial ability due to the transmission of prenatal testosterone during gestation. Two studies have supported the TTT hypothesis, showing that females with a male co-twin (Fm) performed better in MRTs than females with a female co-twin (Ff)54,55. The first study, based on a single 3D MRT on a sample of 804 twins, showed that females with a male co-twin outperformed females with a female co-twin (d = 0.30)54. The second study, based on a sample comparing 100 females from fraternal same-sex and 100 females from opposite-sex twin pairs, replicated the results with similar effect size (d = 0.38)55, giving further support for the TTT hypothesis.

The evidence for TTT from twin samples does not provide a definitive conclusion regarding the etiology of sex differences. Differences in spatial ability between females with a male co-twin and females with a female co-twin could be due to postnatal environmental influences, namely growing up with a brother. For example, some play behaviors have been shown to improve spatial ability26,56,57. A review has concluded that playing video games can improve spatial cognition56. A study demonstrated that playing ten hours of a video game, requiring spatial skills, significantly improved females’ mental rotation ability (η2 = 0.39)26. Additionally, the positive effects of training on mental rotation have been shown to endure for several months and the improvement was more long-lasting among females57. To address the question of the effect of increased participation in spatial activities due to having a brother (not due the transmission of prenatal testosterone), two studies have employed samples of females with non-twin brothers of similar age55,58. Both studies found no advantage in mental rotation performance for females with brothers (of similar age) over females with no brothers55,58. These results gave indirect support to TTT, suggesting that performance in mental rotation is not influenced by the sex of the sibling via environmental pathways.

The evidence for TTT in spatial ability is currently limited only to mental rotation. The effect of prenatal testosterone on other spatial measures is unclear. A recent study suggested that TTT is not applicable to a range of measures of verbal and non-verbal abilities, in a sample aged between 2 to 16 years2. However, the study did not include any spatial ability measures. Additionally, no evidence for TTT was found for mathematical achievement in an adolescent sample59. As such, previous research associated prenatal testosterone only with mental rotation performance. The present study fills the gap in the literature by exploring systematically the TTT in relation to fourteen spatial measures (including three tasks involving mental rotation). To achieve this aim, we utilize a large representative sample of twins that provides the statistical power to detect even small effects. We hypothesize that (i) males will perform better than females in all 14 spatial tasks; and that (ii) females with a male co-twin will outperform females with a female co-twin on all tasks.


Sex differences in spatial ability

Males outperformed females on all thirteen spatial ability measures (raw scores are available from the authors). To enable meaningful comparisons in sex differences across the measures, the reported values were standardised for the whole sample (males and females combined) with a group mean of 0 and standard deviation of 1. For all measures, males’ average scores were positive and females’ average score were negative, reflecting the overall lower performance of females. The results from the individual tests are plotted in Fig. 1. As shown, the 95% confidence intervals did not overlap between males and females in any of the 14 measures. Overall, the effect sizes were small to moderate (η2 = 0.02–0.16). The differences for the age-corrected means between males and females maintained their significance after the alpha levels were adjusted to account for the family-wise error rate using the Bonferroni correction (see Supplementary Material S1 for the correlations between all the study measures).

Figure 1

Standardized spatial ability mean scores with 95% confidence intervals, separately for males and females. Note. The means for each task are based on one randomly selected member from each twin pair. All differences were significant at p < 0.01. Effects were considered significant only if they replicated in both halves of the twin sample.

The average mean difference for the overall spatial ability measure, Spatial Ability Total, was significant between males (m = 0.50, sd = 0.92) and females (m = −0.32, sd = 0.90). However, the distributions were largely overlapping. For example, for the Spatial Ability Total, the scores ranged for males between −2.4 and 2.1, and for females between −3.1 and 2.0.

The twin testosterone transfer hypothesis

To test the Twin Testosterone Transfer hypothesis on all spatial ability measures, comparisons were conducted between four groups: Mm (males with a male co-twin), Mf (males with a female co-twin), Fm (females with a male co-twin) and Ff (females with a female co-twin). The results showed significant group differences for all measures. To investigate the TTT in detail, post hoc comparisons with Bonferroni correction were carried out between females with a male co-twin (Fm) and females with a female co-twin (Ff). Two of the measures, 2D Bricks and Elithorn Mazes, showed significant average differences between the two female groups. In both measures, Fm group outperformed Ff group. The effect sizes for the differences were very small: Bricks 2D (η2 < 0.01); Elithorn Mazes (η2 < 0.02). The means with 95% confidence intervals for the twin groups in 2D Bricks and Elithorn Mazes tasks are plotted in Figs 2 and 3.

Figure 2

Mean Bricks 2D test scores (with 95% confidence intervals) for males and females from same-sex and opposite-sex twin pairs. Note. Mm = Males with a male co-twin; Mf = Males with a female co-twin; Fm = Females with a male co-twin; Ff = Females with a female co-twin. All the tests were standardized for the whole sample (males and females combined) with a mean of 0. Analyses were then run on these standardized values for males and females separately. For all measures, males’ average scores were positive and females’ average scores were negative.

Figure 3

Mean Elithorn Mazes test scores (with 95% confidence intervals) for males and females from same-sex and opposite-sex twin pairs. Note. Mm = Males with a male co-twin; Mf = Males with a female co-twin; Fm = Females with a male co-twin; Ff = Females with a female co-twin. All the tests were standardized for the whole sample (males and females combined) with a mean of 0. Analyses were then run on these standardized values for males and females separately. For all measures, males’ average scores were positive and females’ average scores were negative.

The mean scores, standard deviations, F-values and effect sizes for the four twin groups (Mm, Mf, Fm, Ff) on all spatial ability measures are shown in Table 1. To visually assess the group differences, the mean scores are plotted in Fig. 4.

Table 1 The age adjusted mean scores, standard deviations, sample sizes, F-values and effect sizes for the four twin groups (Mm, Mf, Fm and Ff) after randomly selecting one twin per pair.
Figure 4

Standardized mean scores for 14 spatial ability scores, separately for four twin groups, based on the sex of the twin and co-twin. Note. Mm = Males with a male co-twin; Mf = Males with a female co-twin; Fm = Females with a male co-twin; Ff = Females with a female co-twin. All the tests were standardized for the whole sample (males and females combined) with a mean of 0. Analyses were then run on these standardized values for males and females separately. For all measures, males’ average scores were positive and females’ average scores were negative.”


The first aim of the study was to investigate whether previously found sex differences for spatial ability are present across all aspects of spatial ability. As hypothesized, our findings showed that males performed significantly better than females on all examined tasks. Effect sizes ranged from small to moderate, consistent with the previous research which has shown that men outperform women on spatial ability tasks with advantage of up to one standard deviation7. Although sex differences in spatial ability were robust, individual differences within sexes explained far more variance in spatial ability than differences between sexes.

The consistently better performance of males over females in all spatial ability measures provides further indirect support for viewing spatial ability as a unitary construct: sex differences in spatial ability maybe maintained by the general spatial ability factor19,21. Future studies could explore the role of spatial anxiety in sex differences in spatial ability. A recent study found a small, but significant sex difference: females demonstrated higher spatial anxiety, on average, when facing spatial tasks60.

The present study also investigated systematically the degree to which sex differences in spatial ability could be explained by differences in prenatal testosterone. We tested this hypothesis by examining the potential effects of prenatal testosterone transmission between opposite-sex twins. The hypothesis that the influence of prenatal testosterone would be detected in all spatial tasks, was not supported by our data. Only in 2 of the 14 measures, the average scores of Fm twin group were significantly higher in comparison to the Ff group. The effect sizes of the differences were negligible. One of the tasks with a significant difference between the two female groups was 2D Bricks, which is a measure of mental rotation ability. However, there were no significant mean differences between the female groups in the two other mental rotation tasks (3D Bricks and Shape Rotation). Put together, these results are not in line with the previous research, which showed evidence for the superiority of females from opposite-sex twin pairs over females from the same-sex twin pairs in mental rotation54,55.

Several factors may account for the differences in results between the current study and previous studies that provided evidence for TTT on mental rotation. Whereas the two previous studies, conducted in Finland54 and Germany55, utilized convenience samples, the sample of this study is part of a representative UK twin sample. Additionally, the larger number of participants in the present study increased the power of detecting the potential effect. There were also differences in the used measures and data collection methods. The two previous studies only employed a single mental rotation task and data were collected using paper-and-pencil method54,55, whereas the current study utilized 14 different computerized spatial measures. Previous research has shown indications that sex differences in spatial ability may vary as a function of the presented stimuli and data collection method. The strength of the sex differences in spatial tasks have been shown to decrease when the used stimuli was presented as real, 3D models instead of 3D images61; and when data was collected in virtual environment instead of pencil-and-paper62.

Overall, given the strengths of this study (larger, representative sample; 14 spatial measures), the role of prenatal testosterone in spatial ability can be called into question. As indicated by our results, the effect of prenatally transferred testosterone, from male to female fetus, may be too subtle to influence the development of neurocognitive functions associated with spatial ability. Alternatively, it is plausible that the influence of prenatally transmitted testosterone may be only evident for females whose male co-twin produces high levels of testosterone.

Another way to investigate the relationship between the influence of testosterone and cognitive sex differences is to measure gender as a behavioral measure, namely as sex-typed behavior. A recent study has reported the mediating influence of sex roles between sex and spatial ability performance; the results suggested that sex role identity may be more informative in explaining spatial ability than biological sex alone63. Since prenatal testosterone levels have been associated with sex role identity64, as well as with interest in male-typed activities52 it may be beneficial to investigate the role of perceived sex roles in future studies. Perceived sex role identity could be studied not just in relation to testosterone levels, but also in relation to learning and activities that enhance spatial ability (e.g., if sex role identity mediates the relationship between testosterone levels and engagement in spatial activities).

Understanding the causes of sex differences in spatial ability is essential as they may account for today’s underrepresentation of women in STEM professions. The results of this study add to the consistent finding of males’ better performance in spatial ability. However, the role of prenatal testosterone in spatial ability was not supported.



The Twins Early Development Study (TEDS) sample was used in this study. TEDS is a large, longitudinal twin sample that includes more than 13,000 twin pairs born between 1994 and 199665, representative of the population of England and Wales. A subsample of twins, aged 19 to 21, completed a range of spatial ability tests in two separate collection waves. Data from participants with severe medical conditions, or whose mothers had experienced perinatal complications, were excluded from the analyses. After the exclusions, the sample size included 2,928 individuals. Sample sizes varied between the tasks due to different completion rates for different measures. The combined mean age for the participants in the present study was 19.6 years (SD = 0.48).

The data were derived from two spatial ability batteries: Bricks and King’s Challenge test batteries19,21. The measures in the Bricks test battery were partly based on the classic mental rotation task and designed to investigate mental rotation and spatial visualisation separately, in both 2D and 3D21. The 10 King’s Challenge tasks were selected from the pool of 27 spatial measures that resulted from the literature research19. The selection was based on the psychometric properties of the measures19. Cronbach’s alphas were calculated for all the items of each measure to evaluate their internal consistency/reliability (see Table S3 in the Supplementary Material).

Bricks is a test battery comprising 6 separate tasks (see Supplementary Material S6 for sample stimuli). Some of the tasks are based on the classic mental rotation task66. Three of the tasks are 2D versions and three tasks are 3D versions. Two tasks measure visualization (2D and 3D), two tasks measure rotation (2D and 3D) and two tasks assess the combination of visualization and rotation (2D and 3D). Each test includes 12 items, out of which the nine psychometrically best performing items were scored. Due to their short length, the Bricks subtests are not recommended to be used individually21. In line with this, three composite Bricks scores, based on the mean performance, were used in the present study: a 2D composite (the three 2D tasks), a 3D composite (the three 3D tasks) and the overall total Bricks score (all six tasks). Reliabilities for the three composite scores ranged from α = 0.70–0.84.

The King’s Challenge is a battery of spatial tests, in which the measures are administered within a game-like narrative to encourage participation19. It assesses several aspects of spatial ability, not just mental rotation, and is therefore complimentary to the Bricks test battery. King’s Challenge test battery includes the following ten tests: 1) Cross-Sections, 2) Elithorn Mazes, 3) Mechanical Reasoning, 4) Shape Rotation, 5) Perspective-Taking, 6) Mazes, 7) 2D Drawing, 8) 3D Drawing, 9) Paper-Folding, and 10) Pattern Assembly. Demonstration of the gamified test is available from In the present study, the reliability of two of the tests was lower than recommended α = 0.70 (Mechanical Reasoning α = 0.54; Mazes α = 0.55). However, in the preliminary testing stage18 both tests showed good test-retest reliabilities (Mechanical Reasoning r = 0.69, n = 46, p < 0.001; Mazes r = 0.74, n = 42, p < 0.001); and therefore they were included in the test battery. The reliabilities for the remaining tests in the present study were α = 0.78–0.92.

To investigate group differences in overall spatial ability, a Spatial Ability Total measure was created. This measure, based on the Principal Component Analysis, was derived from the primary component loadings of each of the spatial ability measures in this study. Spatial Ability Total explained 46% of the total variance among all the study measures. The values for each spatial measure were assigned with the regression method. The Spatial Ability Total is the combination of standardized scores for each participant based on the scores on each measure weighted by the factor loadings. The sample size for Spatial Ability Total was smaller in comparison to the other measures; Spatial Ability Total was calculated only for participants with complete data (who responded to all 13 spatial measures).

Preliminary analyses

Preliminary analyses showed significant differences in some of the 14 spatial measures between participants from monozygotic and dizygotic same-sex twin pairs (see Supplementary Material S3). However, the effect sizes of the differences were small and did not remain significant after correcting for the family-wise error rate. To increase power, the MZ and DZss twin groups were combined, separately for males and females. To test TTT, further analyses were conducted between four groups based on the sex of the participant and sex of the co-twin: Mm (males with a male co-twin), Mf (males with a female co-twin), Fm (females with a male co-twin) and Ff (females with a female co-twin).

Data collection

Data collection took place in spring 2015. The data collection was conducted in two waves, separately for Bricks and King’s Challenge batteries. All the tasks were computerized and completed online on the TEDS website by participants after they were sent login details via e-mail. The study was approved by an ethics committee at King’s College London and it was conducted following the approved guidelines. All participants gave their informed consent. The access to the TEDS data is conditional. The complete data access policy can be viewed at

Statistical analyses

The data was checked for normality and no data transformations were needed based on the skewness and kurtosis values. Different measures had different number of items, therefore the total scores for each test were standardized to enable comparisons between the tasks.

To control for the effect of age variation in test performances, comparisons were made using one-way ANCOVAs with age (in months) as a covariate. This study used multiple ANCOVAs to examine each spatial measure separately. This statistical technique was chosen over MANCOVA to explore the potential effect of TTT on spatial ability for each aspect of spatial ability, as previous studies suggested some partial independence of the measures. In a previous study, 42% of the variance across the 10 measures was explained by the first principle component19. In addition, the use of independent ANCOVAs enables comparisons with previous studies that used single spatial measures67.

Analyses to investigate sex differences (Hypothesis 1) were conducted by randomly selecting one individual from each twin pair (the twin group comparisons for the second half of the sample are reported in the Supplementary Material Figure S1 and Table S5). Random selection of one twin per pair created two independent samples, free from inflated inter-individual similarity observed in twins. This approach makes the sample comparable to other (non-twin) samples used in the literature. In addition, this approach allows for replication of the results in another sample (based on the other twin from each pair). If results are significant in one sample, but not in the replication sample – the significance may be a false positive, or the effect size is so small that it cannot be reliably demonstrated. For the comparisons between the twin groups (Hypothesis 2), the whole sample was used – in order to maximize power.

On some of the measures, the variances between the four twin groups (Mm, Mf, Fm, Ff) had significant differences, as shown by Levene’s test. Therefore, all group comparisons were re-run using non-parametric Kruskal-Wallis analysis. For four of the measures (Elithorn Mazes, Perspective Taking, 2D Drawing and Paper Folding), the variances of the four groups were found to be significantly different, and the groups differed in sample sizes. For this reason, we followed up the main analyses with the Kruskal-Wallis test to confirm the results. These further analyses replicated the findings from ANCOVA. Post hoc group comparisons showed that only in Elithorn Mazes there was a small, significant difference between females from same-sex and females from opposite-sex twin pairs.


  1. 1.

    Hyde, J. S. Sex and cognition: gender and cognitive functions. Current opinion in neurobiology 38, 53–56 (2016).

    Article  PubMed  CAS  Google Scholar 

  2. 2.

    Toivainen, T., Papageorgiou, K. A., Tosto, M. & Kovas, Y. Sex differences in non-verbal and verbal abilities in childhood and adolescence. Intelligence 64, 81–88 (2017).

    Article  Google Scholar 

  3. 3.

    Halpern, D. F., Beninger, A. S. & Straight, C. A. Sex differences in intelligence in The Cambridge Handbook of Intelligence (ed. Sternberg, R. J. & Kaufman S. B.) 253–272 (Cambridge University Press, 2011).

  4. 4.

    Stoet, G. & Geary, D. C. Sex differences in mathematics and reading achievement are inversely related: Within-and across-nation assessment of 10 years of PISA data. PloS one, 8, (2013).

  5. 5.

    Herlitz, A. & Lovén, J. Sex differences and the own-gender bias in face recognition: a meta-analytic review. Visual Cognition 21, 1306–1336 (2013).

    Article  Google Scholar 

  6. 6.

    Voyer, D., Postma, A., Brake, B. & Imperato-McGinley, J. Gender differences in object location memory: A meta-analysis. Psychonomic bulletin & review 14, 23–38 (2007).

    Article  Google Scholar 

  7. 7.

    Voyer, D., Voyer, S. & Bryden, M. P. Magnitude of sex differences in spatial abilities: A meta-analysis and consideration of critical variables. Psychological Bulletin 117, 250–270 (1995).

    Article  PubMed  MATH  CAS  Google Scholar 

  8. 8.

    Hromatko, I. & Tadinac, M. Testosterone levels influence spatial ability: further evidence for curvilinear relationship. Review of psychology 13, 27–34 (2006).

    Google Scholar 

  9. 9.

    Peters, M., Lehmann, W., Takahira, S., Takeuchi, Y. & Jordan, K. Mental rotation test performance in four cross-cultural samples (N = 3367): overall sex differences and the role of academic program in performance. Cortex 42, 1005–1014 (2006).

    Article  PubMed  Google Scholar 

  10. 10.

    Silverman, I., Choi, J. & Peters, M. The Hunter-Gatherer theory of sex differences in spatial abilities: Data from 40 countries. Archives of Sexual Behavior 36, 261–268 (2007).

    Article  PubMed  Google Scholar 

  11. 11.

    Shea, D. L., Lubinski, D. & Benbow, C. P. Importance of Assessing Spatial Ability in Intellectually Talented Young Adolescents: A 20-Year Longitudinal Study. J. Educ. Psychol. 93, 604–614 (2001).

    Article  Google Scholar 

  12. 12.

    Miller, D. I. & Halpern, D. F. The new science of cognitive sex differences. Trends in cognitive sciences 18, 37–45 (2014).

    Article  PubMed  Google Scholar 

  13. 13.

    Ceci, S. J., Williams, W. M. & Barnett, S. M. Women’s underrepresentation in science: sociocultural and biological considerations. Psychological bulletin 135, 218–261 (2009).

    Article  PubMed  Google Scholar 

  14. 14.

    Lohman, D. F. Spatial ability and g in Human abilities: Their nature and measurement (eds Dennis, I. & Tapsfield, P.) 97–116 (Lawrence Erlbaum Associates, 1996).

  15. 15.

    Mackintosh, N. J. IQ and human intelligence (Oxford University Press, 2011).

  16. 16.

    Wai, J., Lubinski, D. & Benbow, C. P. Spatial ability for STEM domains: Aligning over 50 years of cumulative psychological knowledge solidifies its importance. J. Educ. Psychol. 101, 817–835 (2009).

    Article  Google Scholar 

  17. 17.

    Newcombe, N. S. & Shipley, T. F. Thinking about spatial thinking: New typology, new assessments in Studying visual and spatial reasoning for design creativity (ed. Gero, J. S.) 179–192 (Springer Netherlands, 2015).

  18. 18.

    Carroll, J. B. Human cognitive abilities: A survey of factor-analytic studies (Cambridge University Press, 1993).

  19. 19.

    Rimfeld, K. et al. Phenotypic and genetic evidence for a unifactorial structure of spatial abilities. Proceedings of the National Academy of Sciences 114, 2777–2782 (2017).

    Article  CAS  Google Scholar 

  20. 20.

    Colom, R., Contreras, M. J., Shih, P. C. & Santacreu, J. The assessment of spatial ability with a single computerized test. European Journal of Psychological Assessment 19, 92–100 (2003).

    Article  Google Scholar 

  21. 21.

    Shakeshaft, N. G. et al. Rotation is visualisation, 3D is 2D: using a novel measure to investigate the genetics of spatial ability. Scientific Reports 6, 30545, (2016).

    ADS  Article  PubMed  PubMed Central  CAS  Google Scholar 

  22. 22.

    Moffat, S. D., Hampson, E. & Hatzipantelis, M. Navigation in a “virtual” maze: Sex differences and correlation with psychometric measures of spatial ability in humans. Evolution and Human Behavior 19, 73–87 (1998).

    Article  Google Scholar 

  23. 23.

    Astur, R. S., Ortiz, M. L. & Sutherland, R. J. A characterization of performance by men and women in a virtual Morris water task: A large and reliable sex difference. Behavioural brain research 93, 185–190 (1998).

    Article  PubMed  CAS  Google Scholar 

  24. 24.

    Maeda, Y. & Yoon, S. Y. A meta-analysis on gender differences in mental rotation ability measured by the Purdue spatial visualization tests: Visualization of rotations (PSVT: R). Educational Psychology Review 25, 69–94 (2013).

    Article  Google Scholar 

  25. 25.

    Moore, D. S. & Johnson, S. P. Mental rotation in human infants: A sex difference. Psychological Science 19, 1063–1066 (2008).

    Article  PubMed  PubMed Central  Google Scholar 

  26. 26.

    Feng, J., Spence, I. & Pratt, J. Playing an action video game reduces gender differences in spatial cognition. Psychological Science 18, 850–855 (2007).

    Article  PubMed  Google Scholar 

  27. 27.

    Ginn, S. R. & Pickens, S. J. Relationships between spatial activities and scores on the mental rotation test as a function of sex. Perceptual and Motor Skills 100, 877–881 (2005).

    Article  PubMed  Google Scholar 

  28. 28.

    Vuoksimaa, E. et al. Are there sex differences in the genetic and environmental effects on mental rotation ability? Twin Research and Human Genetics 13, 437–441 (2010).

    Article  PubMed  Google Scholar 

  29. 29.

    Kimura, D. & Hampson, E. Cognitive pattern in men and women is influenced by fluctuations in sex hormones. Current directions in psychological science 3, 57–61 (1994).

    Article  Google Scholar 

  30. 30.

    Phillips, K. & Silverman, I. Differences in the relationship of menstrual cycle phase to spatial performance on two-and three-dimensional tasks. Hormones and Behavior 32, 167–175 (1997).

    Article  PubMed  CAS  Google Scholar 

  31. 31.

    Hausmann, M., Slabbekoorn, D., Van Goozen, S. H., Cohen-Kettenis, P. T. & Güntürkün, O. Sex hormones affect spatial abilities during the menstrual cycle. Behavioral neuroscience 114, 1245–1250 (2000).

    Article  PubMed  CAS  Google Scholar 

  32. 32.

    Galsworthy, M. J., Dionne, G., Dale, P. S. & Plomin, R. Sex differences in early verbal and non‐verbal cognitive development. Developmental Science 3, 206–215 (2000).

    Article  Google Scholar 

  33. 33.

    Kovas, Y. et al. The genetic and environmental origins of learning abilities and disabilities in the early school years. Monographs of the Society for research in Child Development 72, i–156 (2007).

    Google Scholar 

  34. 34.

    Geary, D. C. Sexual selection and sex differences in spatial cognition. Learning and Individual Differences 7, 289–301 (1995).

    Article  Google Scholar 

  35. 35.

    Geary, D. C. Male, female: The evolution of human sex differences (American Psychological Association, 1998).

  36. 36.

    Tapp, A. L., Maybery, M. T. & Whitehouse, A. J. O. Evaluating the twin testosterone hypothesis: A review of the empirical evidence. Hormones and Behavior 60, 713–722 (2011).

    Article  PubMed  CAS  Google Scholar 

  37. 37.

    Janowsky, J. S. Thinking with your gonads: testosterone and cognition. Trends in cognitive sciences 10, 77–82 (2006).

    ADS  Article  PubMed  Google Scholar 

  38. 38.

    Van Leeuwen, A. M. & Bladh, M. L. Textbook of Laboratory and Diagnostic Testing: Practical Application of Nursing Process at the Bedside. (FA Davis, 2016).

  39. 39.

    Burkitt, J., Widman, D. & Saucier, D. M. Evidence for the influence of testosterone in the performance of spatial navigation in a virtual water maze in women but not in men. HormBehav 51, 649–54 (2007).

    CAS  Google Scholar 

  40. 40.

    Cherrier, M. M. et al. Testosterone supplementation improves spatial and verbal memory in healthy older men. Neurology 57, 80–88 (2001).

    Article  PubMed  CAS  Google Scholar 

  41. 41.

    Slabbekoorn, D., van Goozen, S. H. M., Megens, J., Gooren, L. J. G. & Cohen-Kettenis, P. T. Activating effects of cross-sex hormones on cognitive functioning: A study of short-term and long-term hormone effects in transsexuals. Psychoneuroendocrinology 24, 423–447 (1999).

    Article  PubMed  CAS  Google Scholar 

  42. 42.

    Aleman, A., Bronk, E., Kessels, R. P., Koppeschaar, H. P. & Van Honk, J. A single administration of testosterone improves visuospatial ability in young women. Psychoneuroendocrinology 29, 612–617 (2004).

    Article  PubMed  CAS  Google Scholar 

  43. 43.

    Halari, R. et al. Sex differences and individual differences in cognitive performance and their relationship to endogenous gonadal hormones and gonadotropins. Behavioral neuroscience 119, 104–117 (2005).

    Article  PubMed  CAS  Google Scholar 

  44. 44.

    Gouchie, C. & Kimura, D. The relationship between testosterone levels and cognitive ability patterns. Psychoneuroendocrinology 16, 323–334 (1991).

    Article  PubMed  CAS  Google Scholar 

  45. 45.

    Hines, M. et al. Spatial abilities following prenatal androgen abnormality: targeting and mental rotations performance in individuals with congenital adrenal hyperplasia. Psychoneuroendocrinology 28, 1010–1026 (2003).

    Article  PubMed  CAS  Google Scholar 

  46. 46.

    Moffat, S. D. & Hampson, E. A curvilinear relationship between testosterone and spatial cognition in humans: possible influence of hand preference. Psychoneuroendocrinology 21, 323–337 (1996).

    Article  PubMed  CAS  Google Scholar 

  47. 47.

    Baron-Cohen, S., Knickmeyer, R. C. & Belmonte, M. K. Sex differences in the brain: implications for explaining autism. Science 310, 819–823 (2005).

    ADS  Article  PubMed  CAS  Google Scholar 

  48. 48.

    Williams, C. L. & Meck, W. H. The organizational effects of gonadal steroids on sexually dimorphic spatial ability. Psychoneuroendocrinology 16, 155–176 (1991).

    Article  PubMed  CAS  Google Scholar 

  49. 49.

    Puts, D. A., McDaniel, M. A., Jordan, C. L. & Breedlove, S. M. Spatial ability and prenatal androgens: meta-analyses of congenital adrenal hyperplasia and digit ratio (2D: 4D) studies. Archives of sexual behavior 37, 100–111 (2008).

    Article  PubMed  PubMed Central  Google Scholar 

  50. 50.

    Falter, C. M., Arroyo, M. & Davis, G. J. Testosterone: Activation or organization of spatial cognition? Biological psychology 73, 132–140 (2006).

    Article  PubMed  CAS  Google Scholar 

  51. 51.

    Rahman, Q., Wilson, G. D. & Abrahams, S. Biosocial factors, sexual orientation and neurocognitive functioning. Psychoneuroendocrinology 29, 867–881 (2004).

    Article  PubMed  Google Scholar 

  52. 52.

    Grimshaw, G. M., Sitarenios, G. & Finegan, J. A. K. Mental rotation at 7 years: relations with prenatal testosterone levels and spatial play experiences. Brain and cognition 29, 85–100 (1995).

    Article  PubMed  CAS  Google Scholar 

  53. 53.

    Hines, M. Prenatal testosterone and gender-related behaviour. European Journal of Endocrinology 155, 115–121 (2006).

    Article  CAS  Google Scholar 

  54. 54.

    Vuoksimaa, E. et al. Having a male co-twin masculinizes mental rotation performance in females. Psychological Science 21, 1069–1071 (2010).

    Article  PubMed  PubMed Central  Google Scholar 

  55. 55.

    Heil, M., Kavšek, M., Rolke, B., Beste, C. & Jansen, P. Mental rotation in female fraternal twins: Evidence for intra-uterine hormone transfer? Biological Psychology 86, 90–93 (2010).

    Article  PubMed  Google Scholar 

  56. 56.

    Spence, I. & Feng, J. Video games and spatial cognition. Review of General Psychology 14, 92–104 (2010).

    Article  Google Scholar 

  57. 57.

    Terlecki, M. S., Newcombe, N. S. & Little, M. Durable and generalized effects of spatial experience on mental rotation: Gender differences in growth patterns. Applied Cognitive Psychology 22, 996–1013 (2008).

    Article  Google Scholar 

  58. 58.

    Frenken, H. et al. Siblings’s sex is linked to mental rotation performance in males but not in females. Intelligence 55, 38–43 (2016).

    Article  Google Scholar 

  59. 59.

    Ahrenfeldt, L., Petersen, I., Johnson, W. & Christensen, K. Academic performance of opposite-sex and same-sex twins in adolescence: a Danish national cohort study. Hormones and behavior 69, 123–131 (2015).

    Article  PubMed  PubMed Central  Google Scholar 

  60. 60.

    Malanchini, M. et al. The genetic and environmental aetiology of spatial, mathematics and general anxiety. Scientific Reports 7, 42218, (2017).

    ADS  Article  PubMed  PubMed Central  CAS  Google Scholar 

  61. 61.

    McWilliams, W., Hamilton, C. J. & Muncer, S. J. On mental rotation in three dimensions. Perceptual and Motor Skills 85, 297–298 (1997).

    Article  PubMed  CAS  Google Scholar 

  62. 62.

    Parsons, T. D. et al. Sex differences in mental rotation and spatial rotation in a virtual environment. Neuropsychologia 42, 555–562 (2004).

    Article  PubMed  Google Scholar 

  63. 63.

    Reilly, D., Neumann, D. L. & Andrews, G. Sex and sex-role differences in specific cognitive abilities. Intelligence 54, 147–158 (2016).

    Article  Google Scholar 

  64. 64.

    Berenbaum, S. A. & Beltz, A. M. How early hormones shape gender development. Current opinion in behavioral sciences 7, 53–60 (2016).

    Article  PubMed  PubMed Central  Google Scholar 

  65. 65.

    Haworth, C. M. A., Davis, O. S. P. & Plomin, R. Twins Early Development Study (TEDS): A genetically sensitive investigation of cognitive and behavioural development from childhood to young adulthood. Twin Research and Human Genetics 16, 117–125 (2013).

    Article  PubMed  Google Scholar 

  66. 66.

    Vandenberg, S. G. & Kuse, A. R. Mental rotations, a group test of three- dimensional spatial visualization. Percept Mot Skills 47, 599–604 (1978).

    Article  PubMed  CAS  Google Scholar 

  67. 67.

    Huberty, C. J. & Morris, J. D. Multivariate analysis versus multiple univariate analyses. Psychological bulletin 105, 302–308 (1989).

    Article  Google Scholar 

Download references


We gratefully acknowledge the on-going contribution of the participants in the Twins Early Development Study (TEDS) and their families. TEDS is supported by a program grant from the UK Medical Research Council (MR/M021475/1 and previously G0901245). TT is supported by the Economic and Social Research Council [grant number ES/J500124/1]. This research was supported by the Tomsk State University competitiveness improvement programme grant

Author information




T.T., G.P., K.P., M.M., K.R., N.S. and Y.K. designed the study. T.T., G.P. and K.P. conducted the analyses. T.T. and G.P. wrote the manuscript. Y.K. supervised all aspects of the study. All authors reviewed the manuscript.

Corresponding author

Correspondence to Yulia Kovas.

Ethics declarations

Competing Interests

The authors declare no competing interests.

Additional information

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Toivainen, T., Pannini, G., Papageorgiou, K.A. et al. Prenatal testosterone does not explain sex differences in spatial ability. Sci Rep 8, 13653 (2018).

Download citation


  • Spatial Vision
  • Prenatal Testosterone
  • Spatial Ability Measures
  • Mental Rotation
  • Mental Rotation Task (MRT)

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.