If you’ve read my blog, you’ve probably realized that I’m not one of those people who likes to go around stating unproblematically that transness is “inborn” or due to “nature”. If you’ve ever read my Twitter account, this becomes even more clear that I dislike ‘innate-ness’ explanations.
So when there’s a person going around reddit posting some “masterpost” of how trans identity is biological, it’s sure to arouse some suspicion in me. In my previous post, I addressed the political, scientific and sociological aspects of “brainsex” within trans politics. But the “neuro”-origin myths about being trans are not the only ones that exist: There are also the hormonal ones & the genetic ones. “Biotroofs” are present in all sorts of manners, trying to inscribe identity, lived experience & group differences to various aspects of the body.
Not One, But Two?
One of the most common appeals is to mythical “genes”. Milton Diamond’s paper (forgive the constant misgendering) is especially famous for this, as he examined trans identity among twins. The problem with his study is that not only did he calculate heritability coefficients by using the MZ (monozygotic – identical twins) and DZ (dizygotic – fraternal twins) data that he collected, but he uses a scientific methodology that has been torn to pieces.
For those not acquainted with twin studies, they are a methodology originally created by Francis Galton to ‘determine’ the relative contributions of ‘genetics’ and ‘environment’ to his measures of intelligence. Essentially, they compare the similarity among MZ twins (their concordance rate) and their similarity among DZ twins (their concordance rate) and use that to estimate the genetic contribution. The typical model for this study involves the analysis of variances. Variance, V, of the population is presumed to be a combination of the variances in genes & environments; V=G+E. The way to calculate “G” and “E” was like this: we assume that the variances of twin phenotypes is simply the addition of genes and environment. Monozygotic twins, by virtue of sharing all their genes, had their correlation . Dizygotic twins had the equation because they shared only half of their genes. Solving the set of simultaneously equations yields , so .
This is the original model that psychologists and eugenicists used in the early 1900s, but it has since been the subject of numerous critiques and some advancements. Notably, it has been noted that the “environment” is typically divided into two parts: shared environment (anything that makes twins more similar) and non-shared environment (anything that makes twins less similar). Researchers have also noted the interaction of genes and environment: a gene that “codes” for a particular phenotype may only do so in the presence of an environmental influence. This is typically denoted by GxE and has been the subject of a burgeoning literature, alongside statistical problems. It has been centered within the discussion of the controversial MAOA gene, allegedly* demonstrating that the gene only affects mental health outcomes in the presence of abusive environments.
While twin studies have been one of the most-used paradigms within the ‘heritability study’ scheme because they claim to be able to partition genes and environment, they have been subject to extensive critiques due to their inability to adequately control for confounds, as well as their uselessness in isolating the specific origins of a particular phenotype. Contrary to popular belief, the effects of genes and environment are not separable. As Lickliter so succinctly describes the heritability study methods, it is a fallacy of partitioning. Or as my friend always puts it “development doesn’t work that way“.
Beyond the conceptual disputes over what a heritability statistic means, there are some relevant biases that twin studies face. Most notably is known as the “equal environment assumption”. The equal environment assumption (or EEA) states that MZ twins and DZ twins share equal correlations in environment (i.e. that the environment will be no more similar for MZ twins than DZ twins). We have substantial reasons to believe that this is untrue, however. As a 2001 paper pointed out, one can potentially explain the entire difference in concordance rates between monozygotic twins and dizygotic twins by environmental similarity.
If we again review Diamond’s ‘transsexual twin’ study, we can note a few things. First: that he strangely refused to calculate the heritability statistic (from the naive traditional twin model), which from his Table 3, would be for ‘males’ and for ‘females’, from the bibliographic search. From Table 4, the survey search demonstrates for ‘males’, for ‘females’. And from Table 5, the overall aggregated data implies for ‘males’ and . Interestingly, when we go to calculate the “shared environment” contribution to the phenotype, we get a model violation: , which for ‘males’ is and for ‘females’ is . The ACE model described above does not permit for negative contributions, meaning that the computation of the statistic is meaningless. Regardless, if we calculate the ‘non-shared portion’ () we get for ‘males’ and for ‘females’. This would seemingly imply a very large non-shared environment contribution to the trait, but alas the statistic is meaningless. But as many recognize (note that this primer is uncritical of twins reared apart studies that have numerous problems), finding a negative estimate for would imply that either the EEA (equal environments assumption) or NNE (no non-additive effects) are false.
Second: there is another assumption that is almost certain to be violated: random selection & attrition, that is, ascertainment bias. When ‘gathering’ the twin data, especially for such rare identities like being trans, there is a gigantic problem with how data is collected. Because trans people are so rare, the methods that researchers use to identify them are already a huge issue in regular trans research, but in twin research, this problem is compounded: many times. Because twins that are both trans are much more noteworthy (and perhaps, because of the shared identity, the twins could have maintained a closer relationship during adulthood), it is certain that twins that are both trans (i.e. concordant) are more likely to be found by researchers & respond to surveys. This violation of ascertainment bias has been shown to upwardly bias estimates of heritability. Even more, the lack of zygosity testing (i.e. that more similar twins are more likely to consider themselves MZ when they are in actuality DZ) and the circular assumption of zygosity from narratives (i.e. presuming that similarity is a result of genetics and then concluded as such) are two more mechanisms by which the correlations of monozygotic twins are overestimated and dizygotic twins are underestimated.
And, of course, this circles right back to the underlying incoherence of the twin model. Development doesn’t work that way! Peter Taylor has demonstrated at length how underlying developmental heterogeneity invalidates the assumptions of heritability studies, while a review of the research in criminology has sparked a call for their abandonment. What is clear is that the research into the ‘genes’ of trans identity is inherently complex, politically fraught & not going to come to an end any time soon.
It’s Still In The Genes!
Despite the launch of attacks upon twin study methods I’ve alluded to, a careful reader might note that there have been reports of ‘transsexual genes‘ in the news, perhaps rebuking my critical analysis of the twins that Diamond has presented. A closer look, however, will note that it’s part of another failed tool from the hereditarian toolkit: the “candidate gene”. After decades of ‘heritability studies’ purportedly demonstrated that every trait in existence was, in part, genetic, the wave of genomics ushered in a new era of psychologist, geneticist and behavioral scientist alike, all hoping to ‘find the genes’. Their methodology was to identify a ‘locus’ involved in a particular body process. For depression, it was the 5-HTTLPR gene. For various personality traits, it was the DRD4 gene. For aggression, researchers presumed that MAOA gene may be involved. The entire ‘candidate gene’ shtick was nothing but a house of cards, with a recent psychiatric study making waves within genetics communities and the general public alike. Similar null results have been shown for the MAOA gene, and a review of the entire candidate gene literature found nothing but inconsistency and internal contradictions. The minuscule sample sizes (leading to both random and systematic error) for testing effects along with researcher degrees of freedom, p-hacking and publication bias have produced a research environment where non-existent genetic effects have been able to be touted as having large influences.
The paper in question is not much better. Along with the small sample size (N=112, 258), the association detected was only barely significantly at p=0.04. Right under the 0.05 margin. You’ll note that once classified into subgroups of ‘long’ and ‘short’, the difference between the cis & trans groups is entirely insignificant:
The AR genotype, being X-linked, is hemizygous, and thus the comparison undertaken was between short and long genotypes. An independent samples t-test revealed no significant association for the AR gene when sub-classified (p > .05).
I noticed that they had an outlier: a trans individual with 36 CAG repeats, so I did a quick and dirty recomputation without. Following the methodology in the studies I will discuss afterwards, I didn’t calculate the base pairs (as this could artificially inflate significance and because I have yet to hear back from the authors on how to calculate these figures & even more, the following literature doesn’t do so), but rather just the number of repeats (note that this is a rough reconstruction from their Figure 1A).
I performed a basic one-tailed Student’s t-test for the sake of ease and got a t-value of 1.31906 and a p-value of .094307. Thus, the number of repeats between the groups was not significant. My suspicion is that the difference found was an artifact of converting the number of repeats into the number of base pairs. See Appendix B for input data & more information about the calculation.
Fascinatingly, there are not only one, but two subsequent papers disconfirming the link. A 2014 paper in Spain tested all of the purported candidate genes (ERβ, AR, and CYP19A1) in the largest sample size yet (N=915) and found no relationship. The study found marginal significance for repeat length and the CYP19A1 gene, finally completing the “significant” result for the trio. But as all previous and subsequent research has not found any association, it is certainly a spurious result. A 2009 study also tested the purported candidate genes for both trans men and trans women with relatively high sample sizes & failed to replicate the AR & ERβ results from previous research. They also found no associations for any of the testosterone/estrogen-related genes they tested (increasing the total number of candidate genes from previous research). We should also note that other studies purporting to link CAG repeats to reproductive/sex-related phenotypes have come up with contradictions and publication bias. I think this qualifies as a robust falsification of the hypothetical aetiology, at least until the gene people break out GWA and start making more bullshit developmentally-ignorant associations.
From Phrenology to Fingerology
Next-up in the never-ending train of purported biological influences on trans identity are the fingers. One might question the relevance of one’s digits to the seemingly neurologically grounded (at least according to the trans-essentialists!) trans identity, but researchers have ‘shown’ (to use the word lightly since this thesis has been challenged numerous times and the relationship to various gender attributes rarely replicates) that the ratio of one’s second finger to their fourth finger is associated with the prenatal level of testosterone, at least purportedly. This result has been thrown up as “evidence” that trans people’s identities are determinisically caused by prenatal hormone exposure, despite the data being so mixed. Let’s take a look.
After a 2017 study (which was both of little value and performed by a doctor that is known to be harmful to trans people) tested the widely reported, sometimes replicated and sometimes not, relationship between 2D:4D and trans identity, there was a comment on the paper. One was from a set of researchers I trust highly, who performed a meta-analysis on the disparate literature on the topic. They found that trans women had 2D:4D ratios that were significantly more “feminized” in their right-hand, while the difference in the left-hand was not significant (although it’s p-value was marginally above the significance level). Trans men did not have “masculinized” ratios on either of their hands (although the g effect sizes were trending in the ‘correct’ direction). Despite the claims of other papers, it did not seem that measurement method affected the results in the meta-analysis. The heterogeneity among studies, then, seems to result from sampling characteristics (perhaps things like ascertainment bias). Most notably, the results indicated that even in the case that the small positive among trans women’s right-hand was not driven by things like developmental heterogeneity (a correlation produced by the fact that the link between 2D:4D and prenatal androgen exposure is weak in the first place), publication bias, ascertainment bias or other sampling biases, this still indicates an overlap of somewhere from 92-98%. If we frame this in terms of diagnostic accuracy, it would only give 51.7-55.3%, only marginally above the random guess of 50%.
I can already hear people wondering that if I don’t think that transness is biologically inborn, then where do I think that transness arises from? To be honest, I don’t think it’s a very interesting or politically relevant question. Because trans identity is historically and socially contingent and the irrelevance of a ‘biological’/’genetic’ grounding to a trans-affirmative politics, whether or not transness (as formulated by medical gatekeepers) is ‘biological’ just seems like a non-sequitur to every issue facing trans people. As Canguilhem has demonstrated in The Normal and the Pathological, to investigate the cause of something is to pathologize it by casting it as abnormal and need of investigation. Why is it that we have never searched for the cause of cisgenderism, of heterosexuality, of gender-normative behavior? It is precisely because these are norms, the ideal upon which non-cisheteronormative individuals are compared to: to create difference.
Despite the many affirmations that biological explanations are favorable to trans (and LGBQA) people, it’s unclear why this form of biologism would convince anyone. As any cursory engagement with philosophy would tell us, one cannot derive an ought (a moral/normative/ethical/evaluative statement) from a set of is statements (descriptive/empirical statements) alone. Due to this fact, it is quite easy for an anti-trans individual to reject any narrative affirming trans people based solely on vacuous appeals to biological and psychiatric authority (which, of course, ends up reinscribing transmisic oppression). All they have to do is point out that any purported biological influences are not inconsistent with the right-wing’s favorite “mental illness model” (warning: transmisia & homomisia). They could also just deny that a biological grounding of the ever-illusory ‘gender identity’ has any bearing on their privileging of ‘sex’ based on ontological or normative grounds. This realization that a trans-affirmationist politics cannot be based solely on empirical results helps us question the relevance of these findings overall: if the efficacy of the research in debates inevitably reduces to an ontological, ethical and metaphysical debate, why not just start there in the first place?
Now that I’ve avoided presenting my model of gender identity long enough, I’ll release the pressure. For a long time, I just ignored any ‘model’ of gender identity since I didn’t (and still really don’t) believe it’s a coherent, unitary or separable construct. Despite my persistent concerns, it may be better to adopt a tenuous model: one that can be deployed only for those who are insistent that we have to have some idea of where gender identity comes from. One of my favorite authors is Anne Fausto-Sterling, a biologist, feminist & gender studies scholar (she is also known as a leading expert on the development of gender identity!). She is most renowned for her oft-cited book Sexing the Body. In her 2008 work, Sex/Gender: Biology in a Social World, she analyzes the evidence as to what biological factors are associated with gender identity development. She concludes that chromosomes, gonads, reproductive organs, genitalia, prenatal & pubertal hormones do not have much, if any, evidence for a causal influence on gender identity. After reviewing the psychological literature on gender identity formation in early childhood (as well as the literature on intersex individuals), she concludes that gender identity is the result of a complex developmental process involving early postnatal gendered experiences and individual embodiment. I highly recommend reading the entire chapter (chapter 5: Am I A Boy Or A Girl? —The Emergence of Gender Identity). Since I follow her on Twitter, I noticed when she published a new article this year titled: Gender/Sex, Sexual Orientation, and Identity Are in the Body: How Did They Get There? Most of the article is focused on applying a phenomenological perspective to gender/sex, sexual orientation and identity, and in doing so, she develops a more thorough theory of embodied development: how sex/gender and sexual orientation arise and become a part of the body.
Because the focus of this article was on the alleged biological causes of trans identity, I didn’t focus on several other of the points that the reddit user drewiepoodle that sparked this ‘anti-genetics’ (to quote some famous hereditarian redditors) tirade made in their copypasta.
First up is the monkey myth:
A growing body of research is showing how biology influences gender expression, sexual orientation and gender identity — characteristics that can also fall outside of strict, socially defined categories. Toy-preference tests, a popular gauge of gender expression, have long shown that cis boys and cis girls will typically gravitate to toys that are stereotypically associated with their gender (cars and guns for boys, for instance, or plush toys for girls). While one might argue that this could be the by-product of a child’s environment — parental influence at play or an internalization of societal norms — Melissa Hines, a former UCLA researcher and current professor of psychology at the University of Cambridge, in England, has shown otherwise. In 2008, she demonstrated that monkeys showed the same sex-based toy preferences as humans — absent societal influence.
The “monkey studies” have been weaponized against the “gender feminists” (to borrow Christina Hoff Sommers’ term) who point out the pivotal role of socialization and gender roles in the development and expression of gendered preferences for various social features. Here the user is referencing Melissa Hines (who is, funnily enough, a critic of the brain organization thesis in many ways) and her team’s study on vervet monkeys. The study tested whether male or female vervet monkeys would have sex-typed preferences for toys, presumably because monkeys are free from social influence. They tested 6 toys, 2 ‘masculine’ (police car & ball), 2 ‘neutral’ (picture book & stuffed dog) and 2 ‘feminine’ (doll & cooking pan). Interesting, the toy of a ‘ball’ has previously been classified a ‘neutral’ toy in previously studies testing sex differentiation in toy preferences. Overall, male monkeys spent the most time with the dog (neutral) and equal amounts of time with the ball, police car (masculine) and cooking pan (feminine). The female monkeys spent the most time with the cooking pan (feminine), dog (neutral), then doll (feminine) and then less time with the police car and ball (masculine). While there was an overall statistical difference between male and female monkeys in their choices of ‘masculine’ vs ‘feminine’ toys, it doesn’t exactly explain why there were such heterogeneous results: why neutral toys dominated and why there were ‘cross-sex’ results for some toys. This is only compounded by looking at the other monkey study on toys, which is reviewed in Gina Rippon’s The Gendered Brain (pages 194-195), Rebecca Jordan-Young’s Brain Storm (pages 234-236), and Cordelia Fine’s Delusions of Gender (pages 123-129).
Not only do the actual results raise questions about what the actual preferences among monkeys is, it raises questions about monkeys themselves. People often assume that, no, animals don’t have complex social structure. The assumption is that since an animal exhibits a sex difference, the corresponding sex difference in humans is not only natural, but intractable. (Un)fortunately, this is a naive way of looking at development. Just like humans, many primate species have complex social structures with varying hierarchies. It might be anathema to some, but it has been shown that primates and other animals alike have forms of social learning, what some term “socialization”. Some research has even indicated that primates have a form of gender roles and perhaps even gender identity. This research has come alongside with a review of our anthropological concepts of culture and the realization that primates (and other animals!) indeed do have what can be termed ‘culture’. Because there are varying developmental experiences and trajectories among male and female primates, the fact that a difference emerged (however self-contradictory) is not evidence for an innate sex difference, it hints towards a complex developmental origin. In fact, a friend of mine has noted that the only thing that these ‘monkey studies’ can show is the particular troop dynamics.
Next up, a few lessons to anyone (including the subject of my critique) who may read this. Behavioral traits are, if genetic at all, polygenic, meaning originating from numerous genes. In fact, nearly all traits are polygenic, with the exception with a minuscule number of allele substitutions which produce strong changes in phenotype (sickle cell disease, for example). The claim that:
We have no idea what the details are (a gene, multiple genes, etc?) but we have pretty strong data that it’s something durable and biological.
indicates an underlying misunderstanding of the way that not only scientific research (and norms) work, but of how developmental genetics works. If there are going to be any genes that influence trans identity, it isn’t going to be a gene or multiple genes, it’s going to be numerous. Estimates of polygenicity (a numerical quantification of how many genes are involved in the development of a phenotype) for traits such as Alzheimer’s disease and schizophrenia have estimated numbers of genes into the thousands, simply for a single trait. And again, to the extent that genes are purported to exist, they are very likely to be mediated, moderated and interact with environmental factors.
Of the final extraneous points I wanted to make is the discussion of intersex people & their gender identities;
Also, the attempts by the medical establishment to surgically change body parts of intersex children based on what seemed easiest surgically was not always in line with the person’s actual gender. The thinking back then(and even today) was that gender identity was not biological. When the data was carefully collected, a majority of kids treated this way have a gender identity at odds with their surgically created body parts and upbringing(socialized as male/female). This is proof that we cannot change the gender identity someone already has innately.
Rebecca Jordan-Young has reviewed the literature (in the aforementioned book Brain Storm) on a variety of research progammes investigating sex differences in various traits, mostly related to the purported causal link between hormones (testosterone, estrogen), brain development and the aforementioned social traits. One of those programmes of research has focused on the development of gender identity in intersex individuals. She has thoroughly demonstrated that the typical consensus that intersex individuals will ‘revert’ to their ‘biological gender identity’ is not back-up by the data (this still doesn’t mean we ought to mutilate intersex children).
Alexander, G. M., & Hines, M. (2002). Sex differences in response to children’s toys in nonhuman primates (Cercopithecus aethiops sabaeus). Evolution and Human Behavior, 23(6), 467–479. doi:10.1016/s1090-5138(02)00107-1
Alia-Klein, N., Kriplani, A., Pradhan, K., Ma, J. Y., Logan, J., Williams, B., … Fowler, J. S. (2008). The MAO-A genotype does not modulate resting brain metabolism in adults. Psychiatry Research: Neuroimaging, 164(1), 73–76.doi:10.1016/j.pscychresns.2007.12.010
Ancelin ML, Ryan J. (2017). 5-HTTLPR × stress hypothesis: Is the debate over? Mol Psychiatry.doi: 10.1038/mp.2017.195
Berenbaum, S. A., Bryk, K. K., Nowak, N., Quigley, C. A., & Moffat, S. (2009). Fingers as a Marker of Prenatal Androgen Exposure. Endocrinology, 150(11), 5119–5124.doi:10.1210/en.2009-0774
Border R, Johnson EC, Evans LM, et al (2019). No support for historical candidate gene or candidate gene-by-interaction hypotheses for major depression across multiple large samples. Am J Psychiatry, 176:376–387.doi:10.1176/appi.ajp.2018.18070881
Burt, C. H., & Simons, R. L. (2014). Pulling Back the Curtain on Heritability Studies: Biosocial Criminology in the Postgenomic Era. Criminology, 52(2), 223–262. doi:10.1111/1745-9125.12036
Canguilhem, G. (1989). The normal and the pathological. New York: Zone Books. Originally published as Le normal et le pathologique (Paris: Les Presses Universitaires de France, 1966).
Charney, E. (2008). Genes and Ideologies. Perspectives on Politics, 6(02).doi:10.1017/s1537592708080626
Charney, E., & English, W. (2012). Candidate Genes and Political Behavior. American Political Science Review, 106(01), 1–34.doi:10.1017/s0003055411000554
Cirulli, E. T., & Goldstein, D. B. (2007). In vitro assays fail to predict in vivo effects of regulatory polymorphisms. Human Molecular Genetics, 16(16), 1931–1939. doi:10.1093/hmg/ddm140
Daw, J., Guo, G., & Harris, K. M. (2015). Nurture net of nature: Re-evaluating the role of shared environments in academic achievement and verbal intelligence. Social Science Research, 52, 422–439.doi:10.1016/j.ssresearch.2015.02.011
Diamond, M. (2013). Transsexuality Among Twins: Identity Concordance, Transition, Rearing, and Orientation. International Journal of Transgenderism, 14(1), 24–38.doi:10.1080/15532739.2013.750222
Fausto-Sterling, Anne (2000). Sexing the Body: Gender Politics and the Construction
of Sexuality. New York: Basic Books.
Fausto-Sterling, A. (2012). Sex/gender: Biology in a social world. New York: Routledge.
Favé, M.-J., Lamaze, F. C., Soave, D., Hodgkinson, A., Gauvin, H., Bruat, V., … Awadalla, P. (2018). Gene-by-environment interactions in urban populations modulate risk phenotypes. Nature Communications, 9(1).doi:10.1038/s41467-018-03202-2
Fernández, R., Esteva, I., Gómez-Gil, E., Rumbo, T., Almaraz, M. C., Roda, E., … Pásaro, E. (2014). Association Study of ERβ, AR, and CYP19A1 Genes and MtF Transsexualism. The Journal of Sexual Medicine, 11(12), 2986–2994. doi:10.1111/jsm.12673
Fine, Cordelia (2010). Delusions of Gender: How Our Minds, Society, and Neurosexism Create Difference, New York: Norton.
Fosse, R., Joseph, J., & Richardson, K. (2015). A Critical Assessment of the Equal-Environment Assumption of the Twin Method for Schizophrenia. Frontiers in Psychiatry, 6.doi:10.3389/fpsyt.2015.00062
Goldman, N., Glei, D. A., Lin, Y.-H., & Weinstein, M. (2010). The serotonin transporter polymorphism (5-HTTLPR): allelic variation and links with depressive symptoms. Depression and Anxiety, 27(3), 260–269. doi:10.1002/da.20660
Gottlieb, G. (1997). Synthesizing nature-nurture : prenatal roots of instinctive behavior. Hillsdale (N.J.): Erlbaum.
Guo, S.-W. (2001). Does Higher Concordance in Monozygotic Twins Than in Dizygotic Twins Suggest a Genetic Component? Human Heredity, 51(3), 121–132.doi:10.1159/000053333
Hare, L., Bernard, P., Sánchez, F. J., Baird, P. N., Vilain, E., Kennedy, T., & Harley, V. R. (2009). Androgen Receptor Repeat Length Polymorphism Associated with Male-to-Female Transsexualism. Biological Psychiatry, 65(1), 93–96.doi:10.1016/j.biopsych.2008.08.033
Jordan-Young, R. M. (2010). Brainstorm: The flaws in the science of sex differences. Cambridge, MA: Harvard University Press
Joseph, J. (1998). The Equal Environment Assumption of the Classical Twin Method: A Critical Analysis. The Journal of Mind and Behavior, 19(3), 325-358.
Joseph, J. (2015). The trouble with twin studies: A reassessment of twin research in the social and behavioral sciences. New York, NY, US: Routledge/Taylor & Francis Group.
Juhasz, G., Gonda, X., Hullam, G., Eszlari, N., Kovacs, D., Lazary, J., … Bagdy, G. (2015). Variability in the Effect of 5-HTTLPR on Depression in a Large European Population: The Role of Age, Symptom Profile, Type and Intensity of Life Stressors. PLOS ONE, 10(3), e0116316.doi:10.1371/journal.pone.0116316
Keller, E. F. (2010). The mirage of a space between nature and nurture. Durham, NC: Duke University Press.
Kendler, K. S., & Eaves, L. J. (1989). The Estimation of Probandwise Concordance in Twins: The Effect of Unequal Ascertainment. Acta Geneticae Medicae et Gemellologiae: Twin Research, 38(3-4), 253–270.doi:10.1017/s000156600000266x
Koehler, N., Simmons, L. W., & Rhodes, G. (2004). How well does second-to-fourth-digit ratio in hands correlate with other indications of masculinity in males? Proceedings of the Royal Society B: Biological Sciences, 271(Suppl_5), S296–S298.doi:10.1098/rsbl.2004.0163
Kohn, G. M. (2019). How social systems persist: Learning to build a social network in an uncertain world. Preprint.
Leadbeater, E., & Chittka, L. (2007). Social Learning in Insects — From Miniature Brains to Consensus Building. Current Biology, 17(16), R703–R713. doi:10.1016/j.cub.2007.06.012
Leinung, M., & Wu, C. (2017). The Biologic Basis of Transgender Identity: 2D:4D Finger Length Ratios Implicate A Role For Prenatal Androgen Activity. Endocrine Practice, 23(6), 669–671.doi:10.4158/ep161528.or
Lewontin, R. C. (2006). The analysis of variance and the analysis of causes. International Journal of Epidemiology, 35(3), 520–525.doi:10.1093/ije/dyl062
Lickliter, R. (2009). The Fallacy of Partitioning: Epigenetics’ Validation of the Organism-Environment System. Ecological Psychology, 21(2), 138–146.doi:10.1080/10407410902877157
Lippa, R. A. (2006). Finger lengths, 2D:4D ratios, and their relation to gender-related personality traits and the Big Five. Biological Psychology, 71(1), 116–121.doi:10.1016/j.biopsycho.2005.02.004
Maier, R. M., Visscher, P. M., Robinson, M. R., & Wray, N. R. (2017). Embracing polygenicity: a review of methods and tools for psychiatric genetics research. Psychological Medicine, 48(07), 1055–1067.doi:10.1017/s0033291717002318
Manning, J. T. (2017). Prenatal Sex Steroids And Transgender Identity: Is There A Link With Digit Ratio? Endocrine Practice, 23(6), 738–740.doi:10.4158/ep171843.co
McGrew W. C. (1998). Culture in Nonhuman Primates? Annual Review of Anthropology 27:1, 301-328.doi:10.2307/223373
Medland, S. E., Zayats, T., Glaser, B., Nyholt, D. R., Gordon, S. D., Wright, M. J., … Evans, D. M. (2010). A Variant in LIN28B Is Associated with 2D:4D Finger-Length Ratio, a Putative Retrospective Biomarker of Prenatal Testosterone Exposure. The American Journal of Human Genetics, 86(4), 519–525.doi:10.1016/j.ajhg.2010.02.017
Moore, D. S., & Shenk, D. (2016). The heritability fallacy. Wiley Interdisciplinary Reviews: Cognitive Science, 8(1-2), e1400.doi:10.1002/wcs.1400
Nikulina, V., Widom, C. S., & Brzustowicz, L. M. (2012). Child Abuse and Neglect, MAOA, and Mental Health Outcomes: A Prospective Examination. Biological Psychiatry, 71(4), 350–357. doi:10.1016/j.biopsych.2011.09.008
Ottman R. (1996). Gene-environment interaction: definitions and study designs. Preventive medicine, 25(6), 764–770.
Perry, S. (2016). Cultural Primatology. The International Encyclopedia of Primatology, 1–5.doi:10.1002/9781119179313.wbprim0201
Ptacek, R., Kuzelova, H., & Stefano, G. B. (2011). Dopamine D4 receptor gene DRD4 and its association with psychiatric disorders. Medical Science Monitor, 17(9), RA215–RA220.doi:10.12659/msm.881925
Richardson, K., & Norgate, S. (2005). The equal environments assumption of classical twin studies may not hold. British Journal of Educational Psychology, 75(3), 339–350.doi:10.1348/000709904×24690
Rippon, Gina (2019). Gendered Brain: the new neuroscience that shatters the myth of the female brain. London: The Bodley Head Ltd.
Samek, D. R., Bailey, J., Hill, K. G., Wilson, S., Lee, S., Keyes, M. A., … McGue, M. (2016). A Test-Replicate Approach to Candidate Gene Research on Addiction and Externalizing Disorders: A Collaboration Across Five Longitudinal Studies. Behavior Genetics, 46(5), 608–626. doi:10.1007/s10519-016-9800-8
Schönemann, P. H. (1997). On models and muddles of heritability. Genetica, 99(2/3), 97–108.doi:10.1023/a:1018358504373
Schuppli, C., & van Schaik, C. P. (2019). Animal cultures: how we’ve only seen the tip of the iceberg. Evolutionary Human Sciences, 1. doi:10.1017/ehs.2019.1
Suhay, E., & Kalmoe, N. P. (2010). The equal environment assumption in twin studies of
political traits: Social confounds and suggested remedies. Unpublished manuscript.
Taylor, P. (2006). Heritability and Heterogeneity: The Limited Relevance of Heritability in Investigating Genetic and Environmental Factors. Biological Theory, 1(2), 150–164.doi:10.1162/biot.2006.1.2.150
Taylor, P. (2008). The Under-recognized Implications of Heterogeneity: Opportunities for Fresh Views on Scientific, Philosophical, and Social Debates about Heritability. History and Philosophy of the Life Sciences, 30(3/4), 431-456.doi:10.2307/23334458
Taylor, P. (2008). Underlying heterogeneity: a problem for biological, philosophical, and other analyses of heritability? Biology & Philosophy, 23(4), 587–589.doi:10.1007/s10539-008-9114-1
Turkheimer, E. (2000). Three Laws of Behavior Genetics and What They Mean. Current Directions in Psychological Science, 9(5), 160–164.doi:10.1111/1467-8721.00084
Ujike, H., Otani, K., Nakatsuka, M., Ishii, K., Sasaki, A., Oishi, T., … Kuroda, S. (2009). Association study of gender identity disorder and sex hormone-related genes. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 33(7), 1241–1244.doi:10.1016/j.pnpbp.2009.07.008
Voracek, M., Kaden, A., Kossmeier, M., Pietschnig, J., & Tran, U. S. (2018). Meta-Analysis Shows Associations of Digit Ratio (2D:4D) and Transgender Identity Are Small At Best. Endocrine Practice, 24(4), 386–390.doi:10.4158/ep-2017-0024
Whiten, A. (2000). Primate Culture and Social Learning. Cognitive Science, 24(3), 477–508. doi:10.1207/s15516709cog2403_6
Whiten, A., & van de Waal, E. (2018). The pervasive role of social learning in primate lifetime development. Behavioral Ecology and Sociobiology, 72(5). doi:10.1007/s00265-018-2489-3
Wilson, G. D. (1983). Finger-length as an index of assertiveness in women. Personality and Individual Differences, 4(1), 111–112. doi:10.1016/0191-8869(83)90061-2
Xiao, F., Lan, A., Lin, Z., Song, J., Zhang, Y., Li, J., … Yang, X. (2016). Impact of CAG repeat length in the androgen receptor gene on male infertility – a meta-analysis. Reproductive BioMedicine Online, 33(1), 39–49.doi:10.1016/j.rbmo.2016.03.012
Because I was lazy and my computer doesn’t like to install software these days, I just used an online t-test calculator.
For Treatment 1, I input: 16, 16, 17, 17, 17, 17, 17, 17, 17, 17, 17, 18, 18, 18, 18, 18, 18, 18, 18, 18, 19, 19, 19, 19, 19, 19, 19, 19, 19, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 22, 22, 22, 22, 22, 22, 22, 22, 22, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 24, 24, 24, 24, 24, 24, 24, 25, 25, 26, 26, 26, 26, 26, 27, 28, 28, 28, 28, 28, 36, 36
And for Treatment 2, I input: 12, 13, 14, 14, 15, 16, 16, 16, 17, 17, 17, 17, 17, 18, 18, 18, 18, 18, 18, 18, 18, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 22, 22, 22, 22, 22, 22, 22, 22, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 24, 24, 24, 24, 24, 24, 24, 25, 25, 26, 26, 26, 26, 27, 27, 28, 29, 30, 32
Note that these don’t actually reflect the actual numbers of the sample as these were reconstructed from frequency data. Because I didn’t want to go through the mess of multiplying the frequencies by the sample sizes for the trans & control groups, I just essentially ‘normalized’ the Ns to ~100 (the actual sizes were both slightly above 100 because of errors in estimating the frequency from the graph).