Hostname: page-component-7b7567c8c-2n9c9 Total loading time: 0 Render date: 2026-11-15T14:22:15.664Z Has data issue: false hasContentIssue false
  • English
  • Français

MATERNAL ACTIVITY DURING PREGNANCY AND SEXUALLY DIMORPHIC TRAITS IN OFFSPRING

Published online by Cambridge University Press:  03 April 2017

Lee Ellis,
Russell Eisenman and
Anthony Hoskin
Lee Ellis*
Affiliation:
Minot State University, Minot, ND, USA
Russell Eisenman
Affiliation:
University of Texas Rio Grande Valley, Edinburg, TX, USA
Anthony Hoskin
Affiliation:
Idaho State University, Pocatello, ID, USA
*
1 Corresponding author. Email: lee.ellis@hotmail.com
Rights & Permissions [Opens in a new window]

Summary

Studies have found positive correlations between prenatal exposure to testosterone and masculinization of offspring traits, particularly among females. The present study sought to determine if physical or sexual activity by the mother during pregnancy was related to masculinized/defeminized offspring traits in adulthood. Data were obtained from a large sample of North American college students (offspring) and their mothers. Information about maternal activity levels during pregnancy were reported retrospectively by each mother. The offspring provided self-ratings of various sexually dimorphic traits. Several significant correlations were found. By and large, as maternal physical activity increased, feminine mannerisms decreased and masculine mannerisms increased in the offspring, particularly for females. Maternal physical activity was also associated with increased upper- and lower-body strength and especially with adult height among offspring. Sexual activity by the mother was only associated with upper-body strength and adult height, particularly of the female offspring. Several sexually dimorphic physical traits in offspring are associated with maternal activity levels during pregnancy. Prenatal testosterone is almost certainly involved. The associations could either reflect genetic influences (given that prenatal testosterone is highly heritable) or an effect of maternal testosterone being transferred to the developing fetus. More research is needed to assess the relative merit of these two possibilities.

Information

Type
Research Article
Information
Journal of Biosocial Science , Volume 50 , Issue 1 , January 2018, pp. 114 - 123
Copyright
Copyright © Cambridge University Press, 2017 

Introduction

Research has shown that throughout life males are, or at least seek to be, more physically and sexually active than females (reviewed by Ellis et al., Reference Ellis, Hershberger, Field, Wersinger, Pellis and Geary2008, pp. 603–605, 630–601, 718–724; Sallis et al., Reference Sallis, Prochaska and Taylor2000). At least two studies have indicated that one of the main causes of these sex differences is exposure to prenatal testosterone (PT) (Eliakim et al., Reference Eliakim, Brasel, Mohan, Wong and Cooper1998; Hönekopp et al., Reference Hönekopp, Manning and Müller2006).

According to twin studies in which 2D:4D finger ratios were used to estimate PT exposure, the heritability of PT is quite high, i.e. 66% (Paul et al., Reference Paul, Kato, Cherkas, Andrew and Spector2006) or even 81% (Voracek & Dressler, Reference Voracek and Dressler2007). To the extent that PT is heritable, parents and offspring should resemble one another for traits that are influenced by PT. These traits include masculine/feminine mannerisms, masculine/feminine body appearance, physical strength, voice timbre and strength of sex drive (Wells, Reference Wells2007; Hodges-Simeon et al., Reference Hodges-Simeon, Gaulin and Puts2011). Therefore, one would expect to find physical and sexual activity by mothers during pregnancy to be positively correlated with masculinization/defeminization of these sexually dimorphic traits in offspring. This expectation would be stronger in the case of female rather than male offspring since males receive many of the genes for sexually dimorphic traits from the Y chromosome (which only fathers, not mothers, transmit to their male offspring).

There is a second reason to expect physical and sexual activity by mothers during pregnancy to be positively correlated with masculinization/defeminization of these sexually dimorphic traits in offspring. It involves noting that mothers may transfer some of the testosterone that their ovaries and adrenal glands produce to their developing fetus via the placenta connecting the mother’s and the fetus’ blood systems (Gitau et al., Reference Gitau, Adams, Fisk and Glover2005; Palomba et al., Reference Palomba, Marotta, Di Cello, Russo, Falbo, Orio and Sala2012). If this transfer occurs, women who are most physically and sexually active would be likely to transfer more testosterone than women who are minimally active. In this regard, a positive correlation has been found between circulating testosterone among mothers while pregnant and fetal testosterone levels (Gitau et al., Reference Gitau, Adams, Fisk and Glover2005; Sarkar et al., Reference Sarkar, Bergman, Fisk, O’Connor and Glover2007; Palomba et al., Reference Palomba, Marotta, Di Cello, Russo, Falbo, Orio and Sala2012). Furthermore, maternal circulating testosterone during pregnancy was recently reported to be inversely correlated with 2D:4D ratios in their offspring, indicating greater fetal exposure to PT (Barona et al., Reference Barona, Kothari, Skuse and Micali2015).

The purpose of the present study was to determine if some of the most basic masculine/feminine offspring traits are correlated with maternal activity during pregnancy. While the study’s research design is not able to decipher which of the two explanations just presented has greater support, it will at least help to determine if either of them deserve further exploration. The specific hypothesis is that mothers who are most physically and sexually active during pregnancy will have offspring who are more masculine/less feminine regarding multiple sexually dimorphic traits.

Methods

The sample

Over 11,000 students attending 22 universities – 20 in the United States and two in Canada – completed a ten-page questionnaire pertaining to their physical characteristics, personalities, preferences and past experiences. Only a few items on the questionnaire were analysed for the present study.

Except for students who were adopted or whose mothers were no longer living, all of the respondents were asked to invite their mothers to complete a separate companion questionnaire. The mothers’ nine-page questionnaire contained questions that primarily had to do with their pregnancy history with the student who completed the student questionnaire (henceforth, the students will be referred to as the offspring).

Matching the mother’s and offspring’s questionnaires

Because all questionnaires were completed anonymously, matching of each mother–offspring questionnaire set first involved the use of a common code number on each set. However, the mother’s and offspring’s questionnaires were nearly always returned to the data collection centre independently (and in separate envelopes), so special efforts were undertaken to ensure that each mother–offspring set was properly matched. This was done by having asked the sex and the birth date of the offspring on both questionnaires. Only when the code numbers along with the sex and the birth dates for the offspring matched was a mother–offspring questionnaire confidently declared.

Demographics

The 3330 mothers who completed their questionnaire ranged in age from 38 to 77 years, with a mean of 48.14 (SD=7.43). Their average years of education was about one and a half years beyond high school, i.e. 13.69 (SD=2.45) years.

For the offspring, there were 994 males and 2336 females. The greater number of female offspring partially reflects the fact that more females than males completed a questionnaire. Also, female offspring were more successful than male offspring in getting their mothers to fill out and return the mothers’ companion questionnaires. The average age of the offspring was 22.55 (SD=5.73). From an ethnic standpoint, they were 85% white, 4% black, 2% Native American, 2% Asian/Pacific Islander, 1% Hispanic, with the remaining 6% providing no response that could be subsumed under one of these five categories.

Independent variables

Two types of maternal activity were measured: physical activity and sexual activity. To determine the first, mothers were asked to report as best as their recall allowed how physically active they were during each month of pregnancy with the offspring who had completed the offspring questionnaire. Their estimates were made using a 10-point scale from 0 (meaning ‘not at all active’) to 9 (meaning ‘extremely active’).

To measure sexual activity throughout pregnancy, mothers were asked to recall the extent to which they were sexually active during each month of pregnancy. In this regard, mothers provided their estimates using the same 0–9 rating scale as they used in estimating overall physical activity. The averages for both forms of activity throughout pregnancy were also calculated. While the use of questionnaires for measuring physical activity certainly has reliability limitations (Shephard, Reference Shephard2003), this study was designed to compensate for the low reliability by utilizing very large samples.

Dependent variables

Ten dependent variables were derived from the questionnaires completed by the offspring of each of the mothers. Offspring self-reported their (a) feminine mannerisms, (b) feminine body appearance, (c) masculine mannerisms, (d) masculine body appearance, (e) overall physical strength, (f) upper-body strength, (g) lower-body strength, (h) depth of voice (voice timbre) and (i) strength of sex drive. In all of these cases, the responses were allowed to range from 1 (representing the lowest possible score) to 100 (representing the highest possible score). To measure height, all answers provided by respondents were converted to inches.

Statistics

To confirm that all ten of the dependent variables were sexually dimorphic, t-tests were used. Pearson correlation was used to test the hypotheses that maternal physical and sexual activity during pregnancy independently affect sexually dimorphic traits in offspring. Because the hypotheses are directional, they could have been tested with one-tailed tests of significance. However, to be more conservative, two-tailed tests were used throughout.

Results

Table 1 shows that the expected gender differences in all ten of the dependent variables were present. Specifically, females were higher for feminine mannerisms and feminine body appearance (the first two traits) and males were higher for all eight of the remaining sexually dimorphic traits.

Table 1

Gender differences in sexually dimorphic traits of offspring

Regarding the hypothesis that maternal physical activity during pregnancy is associated with sexually dimorphic traits in the offspring, pertinent evidence appears in Table 2. This presents the correlations between the amount of physical activity during each month of pregnancy (as reported by the mother) and the ten sexually dimorphic traits (as reported by the offspring). It can be seen that while most of coefficients are not statistically significant, many are. The bulk of the significant correlations can be described as follows:

Table 2

Correlations between physical activity by the mother during each month of pregnancy and sexually dimorphic traits among offspring as adults

*p<0.05; **p<0.01; ***p<0.001 (two-tailed test).

First, feminine mannerisms were significantly diminished, especially for male offspring, during the first half of pregnancy. In other words, mothers who were most physically active in the early months of pregnancy had offspring of both sexes who rated their feminine mannerisms as being significantly lower.

Second, the most active mothers had female offspring with elevated self-ratings of masculine mannerisms. This pattern held true throughout pregnancy, but comparable effects for male offspring were not found.

Third, while overall strength was not significantly affected, upper-body strength in females was positively associated with maternal physical activity during pregnancy. Upper-body strength in male offspring, on the other hand, was not significantly affected by maternal activity levels.

Fourth, as with upper-body strength, lower-body strength also appears to be significantly impacted by maternal activity, albeit to a somewhat lesser degree, in females but not in males. As maternal activity increased so too did the lower-body strength of the female offspring.

Fifth, the sexually dimorphic trait of adult height was affected in both male and female offspring by maternal physical activity throughout pregnancy. Mothers who reported being most active had offspring of both sexes who were relatively tall when compared with offspring of mothers who were least active.

Turning attention to maternal sexual activity rather than physical activity, Table 3 presents the correlations between the mother’s responses and the offspring’s ten sexually dimorphic traits. Only two traits – upper-body strength and adult height – were significantly affected. Specifically, mothers who reported being most sexually active throughout pregnancy had female offspring who rated their upper-body strength as being significantly higher than did females whose mothers were least sexually active. This pattern held throughout pregnancy, but no comparable effects were found for male offspring.

Table 3

Correlations between sexual activity by the mother during each month of pregnancy and sexually dimorphic traits among offspring as adults

*p<0.05; **p<0.01; ***p<0.001 (two-tailed test)

In the case of adult height, as with maternal physical activity, both male and female offspring appear to have been affected by maternal sexual activity during pregnancy. That is to say that the most sexually active mothers had relatively tall female offspring. For male offspring, the correlations between maternal sexual activity and adult height was limited to the latter third of pregnancy.

Discussion

Research seems to be closing in on a variety of ways that both genetic and prenatal environmental factors can have at least subtle influences on human sexual differentiation. For instance, the 2D:4D finger ratio not only differs on average between males and females due to androgen-regulating genetic influences (Hönekopp & Watson, Reference Hönekopp and Watson2010), but also appears to vary somewhat according to month of birth (James, Reference James2014; Szwed et al., Reference Szwed, Kosinska and Manning2017).

Over the years, studies have shown that physical activity by the mother during pregnancy is associated with a variety of offspring traits, including miscarriage rates, birth weight and childhood diabetes (Rao et al., Reference Rao, Kanade, Margetts, Yajnik, Lubree, Rege and Fall2003; Oken et al., Reference Oken, Ning, Rifas-Shiman, Radesky, Rich-Edwards and Gillman2006; Hegaard et al., Reference Hegaard, Pedersen, Bruun Nielsen and Damm2007; Owe et al., Reference Owe, Nystad and Bø2009; Melzer et al., Reference Melzer, Schutz, Boulvain and Kayser2010; Qi, et al., Reference Qi, Liu, Zhang, Wang, Yang, Gong and Du2013). To the authors’ knowledge, this is the first study to explore the possibility that maternal activity during pregnancy might be associated with sexually dimorphic traits in offspring.

This study’s results clearly support the hypothesis that sexually dimorphic traits in offspring are altered by maternal activity during pregnancy. In particular, the more physically active the mother reported she had been during pregnancy, the more her offspring reported exhibiting a variety of masculinized/defeminized traits in adulthood. This was especially true for maternal physical activity and for female offspring. Nevertheless, the effects are minor in the sense that they would never be detectable except when large groups of mothers and offspring are sampled.

As noted in the introduction, there are two ways to explain why physical and sexual activity by mothers during pregnancy would be associated with sexually dimorphic traits among offspring. The first is a rather straightforward genetic explanation. It would assert that because PT is substantially heritable (Paul et al., Reference Paul, Kato, Cherkas, Andrew and Spector2006; Voracek & Dressler, Reference Voracek and Dressler2007), one would expect mothers and offspring to resemble one another regarding testosterone production. Presumably, this would be truer for mothers and their female offspring than for mothers and their male offspring.

The other way to account for an association between maternal activity levels and offspring masculinized/defeminized traits would involve the possibility that mothers transfer testosterone to offspring during pregnancy (Gitau et al., Reference Gitau, Adams, Fisk and Glover2005; Palomba et al., Reference Palomba, Marotta, Di Cello, Russo, Falbo, Orio and Sala2012). In effect, the PT being produced by the offspring’s testes (for males) or ovaries (by females) and adrenals (for both sexes) is being augmented by circulating testosterone from the mother’s ovaries and adrenals. The transfer would presumably be by way of the fetus’ placental connection to the mother’s blood supply. In this latter case, one would again expect to find more physically and sexually active mothers to be producing more testosterone than more sedentary mothers, and to thereby transfer more testosterone to their fetuses.

Recently, a study reported evidence that seems to bear on the hypothesis that maternal testosterone is being transferred to the offspring in sufficient quantities to affect fetal sexual differentiation (Ventura et al., Reference Ventura, Gomes, Pita, Neto and Taylor2013). It found a weak but significant correlation between maternal plasma testosterone levels and the 2D:4D finger length ratio of both male and female newborn offspring. In addition, it reported that the 2D:4D finger ratios of mothers who had sons were lower (i.e. more masculinized) than for mothers of daughters. Through a series of analyses, the Ventura et al. (Reference Ventura, Gomes, Pita, Neto and Taylor2013) study concluded that its results ‘are suggestive that there is no significant transport of testosterone between the mother and the fetal environment’ (p. 112). If so, the only remaining explanation for why the present study found masculinized offspring traits correlated with high activity levels in their mothers is the genetic explanation.

The associations revealed in this study are particularly compelling given that mother’s and offspring’s questionnaires were completed without any cross-consultation (at least as far as the authors were aware). This research design all but eliminates any possibility that the measurement of the independent variables (reported by the mothers) could have had any influence on the measurement of the dependent variables (reported by the offspring). An additional strength of this study was the large sample size, leaving little doubt that the relationships observed are real.

Study limitations primarily involve the fact that it was based entirely on self-reports and that the data from the mothers required them to recall events that occurred 20-odd years in the past. Nevertheless, studies have shown that even after two decades, women are able to remember numerous events and specific information about their pregnancies that match medical records quite well, i.e. in the 80–95% accuracy range (Yawn et al., Reference Yawn, Suman and Jacobsen1998; Buka et al., Reference Buka, Goldstein, Spartos and Tsuang2004). Because errors due to maternal recall are most likely to be random rather than systematic, it is reasonable to assume that the findings provide a conservative estimate of the relationships that actually exist between the independent and dependent variables.

Acknowledgments

Data collection for this study was partially supported by funds from Minot State University and the Eugene Garfield Foundation.

References

Barona, M., Kothari, R., Skuse, D. & Micali, N. (2015) Social communication and emotion difficulties and second to fourth digit ratio in a large community-based sample. Molecular Autism 6, doi: 10.1186/s13229-13015-10063-13227.Google Scholar
Buka, S. L., Goldstein, J. M., Spartos, E. & Tsuang, M. T. (2004) The retrospective measurement of prenatal and perinatal events: accuracy of maternal recall. Schizophrenia Research 71, 417–426.CrossRefGoogle ScholarPubMed
Eliakim, A., Brasel, J. A., Mohan, S., Wong, W. L. T. & Cooper, D. M. (1998) Increased physical activity and the growth hormone-IGF-I axis in adolescent males. American Journal of Physiology-Regulatory, Integrative and Comparative Physiology 275, R308–R314.Google Scholar
Ellis, L., Hershberger, S., Field, E., Wersinger, S., Pellis, S., Geary, D. et al. (2008) Sex Differences: Summarizing More Than a Century of Scientific Research. Psychology Press, New York.Google Scholar
Gitau, R., Adams, D., Fisk, N. & Glover, V. (2005) Fetal plasma testosterone correlates positively with cortisol. Archives of Disease in Childhood – Fetal and Neonatal Edition 90, F166–F169.CrossRefGoogle ScholarPubMed
Hegaard, H. K., Pedersen, B. K., Bruun Nielsen, B. & Damm, P. (2007) Leisure time physical activity during pregnancy and impact on gestational diabetes mellitus, pre‐eclampsia, preterm delivery and birth weight: a review. Acta Obstetricia et Gynecologica Scandinavica 86, 1290–1296.Google Scholar
Hodges-Simeon, C. R., Gaulin, S. J. & Puts, D. A. (2011) Voice correlates of mating success in men: examining “contests” versus “mate choice” modes of sexual selection. Archives of Sexual Behavior 40, 551–557.CrossRefGoogle ScholarPubMed
Hönekopp, J., Manning, J. T. & Müller, C. (2006) Digit ratio (2D:4D) and physical fitness in males and females: evidence for effects of prenatal androgens on sexually selected traits. Hormones and Behavior 49, 545–549.Google Scholar
Hönekopp, J. & Watson, S. (2010) Meta‐analysis of digit ratio 2D: 4D shows greater sex difference in the right hand. American Journal of Human Biology 22, 619–630.CrossRefGoogle ScholarPubMed
James, W. H. (2014) An update on the hypothesis that one cause of autism is high intrauterine levels of testosterone of maternal origin. Journal of Theoretical Biology 355, 33–39.CrossRefGoogle ScholarPubMed
Melzer, K., Schutz, Y., Boulvain, M. & Kayser, B. (2010) Physical activity and pregnancy. Sports Medicine 40, 493–507.Google Scholar
Oken, E., Ning, Y., Rifas-Shiman, S. L., Radesky, J. S., Rich-Edwards, J. W. & Gillman, M. W. (2006) Associations of physical activity and inactivity before and during pregnancy with glucose tolerance. Obstetrics and Gynecology 108, 1200–1207.Google Scholar
Owe, K. M., Nystad, W. & Bø, K. (2009) Association between regular exercise and excessive newborn birth weight. Obstetrics & Gynecology 114, 770–776.Google Scholar
Palomba, S., Marotta, R., Di Cello, A., Russo, T., Falbo, A., Orio, F. & Sala, G. B. L. (2012) Pervasive developmental disorders in children of hyperandrogenic women with polycystic ovary syndrome: a longitudinal case-control study. Clinical Endocrinology 77, 898–904.Google Scholar
Paul, S. N., Kato, B. S., Cherkas, L. F., Andrew, T. & Spector, T. D. (2006) Heritability of the second to fourth digit ratio (2d:4d): a twin study. Twin Research and Human Genetics 9, 215–219.CrossRefGoogle Scholar
Qi, L., Liu, J., Zhang, Y., Wang, J., Yang, M., Gong, T. & Du, Y. (2013) Risk factors for non‐syndromic oral clefts: a matched case-control study in Hubei Province, China. Oral Diseases, doi: 10.1111/odi.12200.Google Scholar
Rao, S., Kanade, A., Margetts, B., Yajnik, C., Lubree, H., Rege, S. & Fall, C. (2003) Maternal activity in relation to birth size in rural India. The Pune Maternal Nutrition Study. European Journal of Clinical Nutrition 57, 531–542.CrossRefGoogle ScholarPubMed
Sallis, J. F., Prochaska, J. J. & Taylor, W. C. (2000) A review of correlates of physical activity of children and adolescents. Medicine and Science in Sports and Exercise 32, 963–975.Google Scholar
Sarkar, P., Bergman, K., Fisk, N., O’Connor, T. & Glover, V. (2007) Amniotic fluid testosterone: relationship with cortisol and gestational age. Clinical Endocrinology 67, 743–747.Google Scholar
Shephard, R. J. (2003) Limits to the measurement of habitual physical activity by questionnaires. British Journal of Sports Medicine 37, 197–206.CrossRefGoogle Scholar
Szwed, A., Kosinska, M. & Manning, J. T. (2017) Digit ratio (2D:4D) and month of birth: a link to the solstitial-melatonin-testosterone effect. Early Human Development 104, 23–26.Google Scholar
Ventura, T., Gomes, M., Pita, A., Neto, M. & Taylor, A. (2013) Digit ratio (2D:4D) in newborns: influences of prenatal testosterone and maternal environment. Early Human Development 89, 107–112.Google Scholar
Voracek, M. & Dressler, S. G. (2007) Digit ratio (2D: 4D) in twins: heritability estimates and evidence for a masculinized trait expression in women from opposite-sex pairs. Psychological Reports 100, 115–126.Google Scholar
Wells, J. C. (2007) Sexual dimorphism of body composition. Best Practice & Research: Clinical Endocrinology & Metabolism 21, 415–430.Google Scholar
Yawn, B. P., Suman, V. J. & Jacobsen, S. J. (1998) Maternal recall of distant pregnancy events. Journal of Clinical Epidemiology 51, 399–405.Google Scholar
Figure 0

Table 1Gender differences in sexually dimorphic traits of offspring

Figure 1

Table 2Correlations between physical activity by the mother during each month of pregnancy and sexually dimorphic traits among offspring as adults

Figure 2

Table 3Correlations between sexual activity by the mother during each month of pregnancy and sexually dimorphic traits among offspring as adults