In a social and political context marked by increasing polarisation around issues of gender and sexuality, Benjamin deMayo and colleagues conducted a large-scale longitudinal study to better understand how young people’s gender identities and sexual orientations evolve over the course of development. This research, carried out with more than 900 North American children and adolescents between 2013 and 2024 (with follow-up spanning average ages 8 to 14), constitutes the largest quantitative survey to date on these identity trajectories. It is situated within a historical moment in which LGBTI+ youth are increasingly visible – but also increasingly contested – in the public sphere. The 166-page study has just been published open access in the journal Monographs of the Society for Research in Child Development.
Three groups to compare trajectories
The core of the sample consists of trans children who undertook a “binary” social transition (change of name, pronouns, clothing, hairstyle) before the age of 12 (on average at 6.5 years), all of whom were supported by their parents (N = 317). Two comparison groups were included: their cisgender siblings (N = 218), and a control group of cisgender children with no family ties to the trans children, but matched by age and gender (N = 377). This design makes it possible not only to assess the developmental trajectories of trans children, but also to compare them with those of children growing up in family contexts that are more or less informed about, or exposed to, gender diversity. To strengthen the reliability of the data, the researchers collected self-reports from children and adolescents at each wave of data collection, while also asking parents about their child’s gender identity. This dual source of information makes it possible to check the consistency of accounts, identify potential discrepancies, and ground the analyses in a concrete family context. The researchers developed precise coding systems to analyse gender identity and sexual orientation at each visit, taking into account the evolution of language and categories over time.
Predominant identity stability across all groups
This study by deMayo and colleagues builds on the original sample of the Trans Youth Project, previously analysed by Kristina Olson and colleagues (2022), and addresses several limitations identified in that earlier work by extending the maximum follow-up period by four years, diversifying measures of gender identity, and adding comparison groups. The results indicate that identity stability is the norm, regardless of group membership. More than 80 % of participants maintained the same gender identity throughout their participation in the study, whether they were initially identified as cisgender or transgender. These figures directly challenge the hypothesis that early trans identities are particularly volatile. Parental support, when present from early childhood, appears to strengthen the coherence of developmental trajectories and enables trans children to develop a stable identity on a par with their cisgender peers.
Figure 7 in the report presents the stability of gender identity across the three groups studied: trans children who had undergone a social transition, their cisgender siblings, and cisgender children in the control group. It shows that the vast majority of participants – between 80 % and 90 % depending on the group – maintained the same gender identity throughout the study, with no statistically significant differences between the three groups.
Gender identity stability is high even among trans children (only 4 % detransitioned toward their sex assigned at birth between a mean age of 8 and a mean age of 15), which contradicts the idea that these identities are more unstable than those of cisgender children. The rare identity changes observed mostly involved shifts toward non-binary identities. These findings confirm that stability is the norm, regardless of trans or cis status, when the family environment is supportive.
Moreover, this figure shows that the siblings of trans children overwhelmingly remain cisgender throughout their development, even though they grow up in the same family environment as the trans child. This directly challenges the commonly held but erroneous belief that parents are in some way responsible for their child’s gender nonconformity. On the contrary, as most parents reported to the researchers, their child’s gender nonconformity came as a surprise: parents had dressed the child from an early age in clothing stereotypical of their assigned gender, and the child’s bedroom had been decorated in similarly stereotypical ways. Both of these elements were objectively confirmed by researchers, who were unaware of the child’s sex assigned at birth, through photographs from that period.
Identity changes and non-binary identities
Despite the predominance of stability, the study shows that 11.9 % of children who were initially cisgender experienced a change in their gender identity over the follow-up period. This proportion, far higher than what classical models of developmental psychology had anticipated, invites a re-examination of fixed conceptions of gender identity. Traditional models, inherited from 20th-century cognitive psychology, assume that children develop a stable and binary understanding of gender around the age of three, which then remains unchanged. Because most research has focused on children considered “typical,” it has rarely considered – or even measured – the possibility of gender identity change beyond early childhood.
These identity changes, although in the minority, mostly involve shifts toward non-binary identities, signalling a growing visibility of this identity modality among younger generations. In the 2010s, few children used gender-neutral pronouns or identified as non-binary; today, these identities are better known, more socially accepted, and more frequently claimed by young people.
The authors emphasise that these identity variations are not associated with specific ages in childhood or adolescence. No critical period was identified in which changes were particularly frequent. This suggests that the capacity to question or redefine one’s gender can occur at different points in development and does not correspond to a typical adolescent crisis or a puberty-related effect.
Sexual orientation and identity exploration
Another major contribution of the study is its analysis of young people’s sexual orientations over time. More than 60 % of “binary” trans adolescents and around 33 % of cisgender adolescents reported a “queer” sexual orientation, meaning one that differs from heterosexuality. Among younger children (under 12), trans children were significantly more likely to report non-heterosexual patterns of attraction than cisgender children. This finding challenges the common assumption that childhood trans identity would prevent the development of a cisgender homosexual identity in adolescence. For example, among children who had already experienced romantic attraction, trans boys observed at an average age of 11 had already felt attraction to other boys in 34 % of cases, whereas 22 % of cisgender girls had already felt attraction to other girls (Table 23, p. 110). Another notable result concerns trans girls, only 32 % of whom reported being exclusively attracted to boys. This directly contradicts earlier theories (Blanchard, 1988) claiming that trans women who identified early as girls would be overwhelmingly and exclusively attracted to men.
With regard to non-binary youth, a strong majority reported attraction to more than one gender, with a high prevalence of bisexuality or pansexuality. These figures confirm that younger generations are redefining sexual norms, and that LGBTI+ identities are increasingly lived and claimed from adolescence onward.
This fluidity is also dynamic: about one third of participants experienced changes in their sexual orientation over the course of the study, with no significant difference in this evolution between trans and cisgender youth. This variability is not correlated with psychological instability, but rather appears to reflect a healthy phase of identity exploration. It confirms previous research showing that sexual orientation, far from being fixed, can evolve throughout adolescence and even beyond.
Can adolescent sexual orientation be predicted from childhood gender identity (more or less masculine or feminine) or gender-typed preferences (for peers, toys, clothing)? The researchers found that more feminine identities or preferences in boys significantly predicted a non-heterosexual orientation in adolescence, just as more masculine identities or preferences in girls predicted the same outcome. Crucially, this study demonstrates that the gender identities and preferences of trans boys and trans girls are neither more nor less predictive of their future sexual orientation than those of cisgender boys and girls.
A break with past pathologising approaches
Unlike traditional medical approaches that framed gender nonconformity as a disorder to be corrected, this study adopts a view of it as a normal variation of human development. DeMayo and colleagues recall that, for several decades, certain medical centres (notably those directed by Kenneth Zucker or Richard Green) attempted to modify the behaviour of gender-nonconforming children through conversion therapies or behavioural re-education. For example, the work of Rekers, Stoller or Coates proposed that boys displaying “effeminate” behaviours be encouraged – or even compelled – to adopt more masculine attitudes, in the hope of preventing a homosexual or transgender identity in adulthood. These interventions were based on a pathologising reading of gender nonconformity, which is now widely challenged.
The authors also criticise a belief that remains widespread in some medical publications: the idea that the majority of children who experience gender dysphoria in childhood will “desist” in adolescence, meaning that they will eventually identify with their sex assigned at birth. According to the authors, this belief rests on older studies conducted in clinics that did not socially recognise trans children and that grouped together, under the label of “gender dysphoria,” very different children (for example, boys who were simply gender-nonconforming, without marked and persistent identity incongruence).DeMayo and colleagues emphasise that these studies often confounded gender identity, gender expression, and sexual orientation, and were carried out in highly normative, and sometimes coercive, contexts. By contrast, their own sample – composed of socially recognised, affirmed trans children followed over many years – shows far higher gender identity stability than that reported in earlier research. The authors therefore denounce a misinterpretation of past data, which is frequently mobilised to justify restrictive policies or to delay the recognition of trans identities in young people.
A study conducted amid political backlash
A substantial part of the discussion focuses on the social consequences of these findings. As anti-trans legislation proliferates in the United States, the study is situated within a context of political regression. The authors explicitly cite the return to power of Donald Trump in 2024, accompanied by a series of executive orders banning the legal recognition of trans identities and threatening to withdraw federal funding from schools that acknowledge gender diversity. In 2023 alone, more than 600 anti-trans bills were introduced across the country, many of them aimed at banning gender-affirming medical care for minors. According to the authors, these measures are in direct contradiction with the available scientific evidence.
The diversity of identity trajectories documented in this study calls for abandoning linear models of development. The researchers emphasise that this variability is not a sign of disorder. On the contrary, it reflects an adaptive response to social contexts and cultural change. Young people today have access to language, communities and resources that previous generations did not, which transforms the ways they understand and articulate who they are. This work thus establishes itself as a key reference, demonstrating with robust data that openness to diverse identities is not a risk, but a resource for each individual’s healthy development.