The objects and people children play with as early as toddlerhood may provide clues to their eventual sexual orientation, reveals the largest study of its kind. The investigation, which tracked more than 4500 kids over the first 15 years of their lives, seeks to answer one of the most controversial questions in the social sciences, but experts are mixed on the findings.
“Within its paradigm, it’s one of the better studies I’ve seen,” says Anne Fausto-Sterling, professor emerita of biology and gender studies at Brown University. The fact that it looks at development over time and relies on parents’ observations is a big improvement over previous studies that attempted to answer similar questions based on respondents’ own, often unreliable, memories, she says. “That being said … they’re still not answering questions of how these preferences for toys or different kinds of behaviors develop in the first place.”
The new study builds largely on research done in the 1970s by American sex and gender researcher Richard Green, who spent decades investigating sexuality. He was influential in the development of the term “gender identity disorder” to describe stress and confusion over one’s sex and gender, though the term—and Green’s work more broadly—has come under fire from many psychologists and social scientists today who say it’s wrong to label someone’s gender and sexuality “disordered.”
In the decades since, other studies have reported that whether a child plays along traditional gender lines can predict their later sexual orientation. But these have largely been criticized for their small sample sizes, for drawing from children who exhibit what the authors call “extreme” gender nonconformity, and for various other methodological shortcomings.
Seeking to improve on this earlier research, Melissa Hines, a psychologist at the University of Cambridge in the United Kingdom, turned to data from the Avon Longitudinal Study of Parents and Children. The study includes thousands of British children born in the 1990s. Parents observed and reported various aspects of their children’s behavior, which Hines and her Cambridge colleague, Gu Li, analyzed for what they call male-typical or female-typical play.
An example of stereotypical male-typical play, as defined by the study, would include playing with toy trucks, “rough-and-tumble” wrestling, and playing with other boys. Female-typical play, on the other hand, would include dolls, playing house, and playing with other girls.
Hines and Li looked at parental reporting of children’s play at ages 2.5, 3.5, and 4.75 years old, and arranged them on a scale of one to 100, with lower scores meaning more female-typical play and higher scores more male-typical play. They then compared those results to the participants’ self-reported responses as teenagers to a series of internet-administered questions about their sexuality.
Beginning with the 3.5-year-old age group, the team found that children who engaged mostly in “gender-conforming” play (boys who played with trucks and girls who played with dolls, as an example) were likely to report being heterosexual at age 15, whereas the teenagers who reported being gay, lesbian, or not strictly heterosexual were more likely to engage in “gender-nonconforming” play. The same pattern held true when they expanded the teenagers’ choices to a five-point spectrum ranging from 100% heterosexual to 100% homosexual.
Teens who described themselves as lesbian scored on average about 10 points higher on the gender-play scale at age 4.75 (meaning more stereotypically male play) than their heterosexual peers, and teens who described themselves as gay men scored about 10 points lower on the scale than their peers, the researchers report in Developmental Psychology. Questions of transgender identity were not addressed in the study.
“I think it's remarkable that childhood gender-typed behavior measured as early as age 3.5 years is associated with sexual orientation 12 years later,” wrote Li in an email. “The findings help us to understand variability in sexual orientation and could have implications for understanding the origins of this variability.”
The paper “is just a well-done study in terms of getting around some of the problems that have plagued the field,” says Simon LeVay, a retired neuroscientist whose 1991 paper in Science sparked interest in brain differences associated with sexual identity. “It shows that something is going on really early in life and points away from things like role modeling and adolescent experiences as reasons for becoming gay.”
Others dispute the paper’s methods and significance. Parents’ own beliefs and biases about gender almost certainly influence how they described their children’s gendered play, which could skew their reporting, says Patrick Ryan Grzanka, a psychologist who studies sexuality and multicultural issues at the University of Tennessee in Knoxville. But more worrisome to him are the cultural assumptions underlying the study itself. The authors appear to regard gender nonconformity as the primary marker of gayness, which doesn’t align with current research suggesting that your individual preferences for either stereotypically male or female behaviors and traits has little to do with your sexual orientation, he says.
Grzanka is also dismayed that the paper fails to critique the history of similar research that investigated whether childhood behaviors lined up with eventual sexual orientation. It wasn’t long ago that such research was used to stigmatize and pathologize gender-nonconforming children, he says. “I think it’s important to ask why we’re so invested in this purported link [between gender conformity and sexuality] in the first place.”
*Correction, 13 March, 12:19 p.m.: An earlier version of this incorrectly asserted that Richard Green and Melissa Hines are life partners.