Gender Identity

Abstract

Gender identity is a person's internal sense of being male, female, both, or neither, a self-representation that is established early in childhood and is highly resistant to later change. The concept was separated from biological sex in mid-twentieth-century sexology and given its enduring form by the psychoanalyst Robert Stoller, and it became a central object of cognitive-developmental psychology through Lawrence Kohlberg's account of gender constancy and the gender-schema theories of Sandra Bem and of Carol Martin and Charles Halverson. This article treats gender identity as a cognitive construct: it traces how children acquire the categories of gender and apply them to themselves, sets out the cognitive-developmental, gender-schema, social-cognitive, and biological accounts of that acquisition, and reviews the recent longitudinal work on transgender children that has tested those accounts against a population the classic theories did not anticipate.

Keywords: gender identity, gender constancy, gender schema, sex typing, gender development

Gender is among the first social categories a child uses to organize the world, and the developmental question of how a private sense of one's own gender arises has occupied cognitive psychology since the 1960s. The distinction on which the field rests is between sex, the biological attributes of male and female, and gender identity, the person's own sense of their gender, a distinction drawn out of clinical work on intersex conditions by John Money and colleagues (Money et al., 1955) and sharpened by Stoller, who coined the term to name the stable inner conviction of one's gender (Stoller, 1964). Kohlberg then reframed the question as a cognitive one: rather than gender-typed behaviour being stamped in by reward or imitation, he argued that the child first constructs an understanding of gender as a stable, constant category and then actively works to behave in line with it (Kohlberg, 1966). The modern literature retains that cognitive emphasis while integrating schematic, social-learning, and biological contributions, and it has been tested against longitudinal studies of transgender children whose identities diverge from their assigned sex (Martin et al., 2002; Gülgöz et al., 2019).

What Gender Identity Is

Gender identity is the internal, subjective sense of being male, female, some combination, or neither. It is distinguished from three neighbouring concepts that are easily conflated with it. Sex refers to the biological attributes — chromosomes, gonads, hormones, and anatomy — on which a person is assigned to a category at birth. Gender role refers to the behaviours, presentations, and social expectations a culture attaches to a gender, a term introduced by Money and colleagues to name the outward, public counterpart of the inner identity (Money et al., 1955). Sexual orientation refers to the pattern of a person's attractions, which is independent of gender identity. Stoller fixed the modern usage, treating gender identity as a stable inner conviction established in the first years of life and largely settled before the biological events of puberty (Stoller, 1964).

For cognitive psychology the interest is developmental and representational: gender identity is a component of the self-concept, the organized cognitive representation a person holds of who they are, and the central questions are how a child comes to possess the category gender, how the child locates the self within it, and how that self-categorization then shapes what the child attends to, remembers, and does. These are questions about concept formation, categorization, and self-representation, which is why gender development became a testing ground for cognitive-developmental theory rather than remaining a topic in clinical sexology alone (Kohlberg, 1966; Martin et al., 2002). A recurring theme of the recent literature is that these categories are neither as binary nor as fixed across people as the classic theories assumed, and that the science is better served by treating gender as multi-dimensional (Hyde et al., 2019).

Types of Gender Identity

The Medical Subject Headings classify gender identity under the broader heading of psychological identification, and the descriptor has three narrower descriptors, listed in Table 1. These are the categories under which the indexing literature is organized, not a theory of how gender identity is structured in the mind: MeSH is a bibliographic indexing classification, built to make papers findable, so the list records how the field files its literature rather than a claim about carving gender at its joints. The subtypes are not mutually exclusive dimensions of a single scale — masculinity and femininity, in particular, are treated in modern measurement as separable rather than as opposite ends of one continuum, a point central to Bem's work on androgyny (Bem, 1974).

Cognitive-Developmental Theory

Kohlberg's 1966 essay turned gender development from a problem of conditioning into a problem of cognition. Where learning-theory accounts held that gender-typed behaviour is installed by reinforcement and imitation, Kohlberg argued the causal arrow runs the other way: the child first constructs an understanding of gender, and that understanding then motivates the child to acquire the behaviours that fit it (Kohlberg, 1966). On this account the acquisition proceeds through three cognitive milestones. Gender identity is the ability to label the self and others by gender, achieved around age two or three. Gender stability is the understanding that gender is stable over time — that a boy will become a man — achieved around age three or four. Gender constancy is the understanding that gender is invariant across superficial changes of appearance and activity, so that putting on a dress does not make a boy a girl; this is the most cognitively demanding step and is typically reached between about five and seven.

The developmental logic mirrors Piaget's account of conservation: just as a child comes to understand that the amount of liquid is conserved across a change in the shape of its container, the child comes to understand that gender is conserved across a change in hair, clothing, or play. The claim that carried the most theoretical weight was motivational — that once a child grasps gender constancy, the child actively values and seeks out same-gender models and activities, so that constancy drives sex typing rather than following it. Subsequent research complicated this strong claim: children display marked gender-typed preferences well before achieving full constancy, which is one of the observations that motivated the schema theories (Martin et al., 2002). The demonstration below lets the reader move through the three stages and see the task a child at each stage can and cannot pass.

Gender-Schema Theory

Gender-schema theory recast gender development as a special case of schematic information processing. A schema is an organized knowledge structure that guides how information is attended to, encoded, and retrieved, and Bem's insight was that the sexes are among the most pervasive categories a culture makes available, so that children readily build a gender schema and then use it to interpret the social world and themselves (Bem, 1981). In Bem's formulation, sex typing is the readiness to process information — including information about the self — in terms of the gender schema, and individuals differ in how strongly the schema organizes their self-concept. Bem had earlier developed the measurement side of this view, arguing against treating masculinity and femininity as opposite poles of a single scale and introducing the concept of psychological androgyny for people who endorse both masculine and feminine attributes (Bem, 1974).

Martin and Halverson gave the developmental version of the theory its most explicit processing model. On their account a child who can reliably label the sexes forms an in-group/out-group schema — a simple grouping of objects, activities, and traits as for one's own sex or the other — and a more detailed own-sex schema of information relevant to the self (Martin & Halverson, 1981). The model makes sharp predictions about memory: children attend to and remember schema-consistent information better than schema-inconsistent information, and, crucially, they distort inconsistent information to fit the schema, misremembering a picture of a boy cooking as a girl cooking. This distortion is the theory's signature, because it shows the schema operating not as a passive filter but as an active reconstruction of experience. The Worked Example below quantifies the memory asymmetry the model predicts, and the demonstration lets the reader vary the recall and distortion rates and watch the recalled set diverge from what was actually presented.

Biological and Social Influences

The cognitive theories describe how gender categories are represented and applied but do not by themselves explain why identity so often, though not always, aligns with natal sex, and here the biological literature is informative. The most studied biological factor is prenatal androgen exposure. Girls with congenital adrenal hyperplasia, who are exposed to atypically high androgen levels before birth, show on average more male-typical play preferences and, in a minority, more male-typical identity outcomes than unaffected girls, evidence that early hormones bias the development of gendered behaviour (Hines, 2011). The effects are real but partial and probabilistic rather than determining: androgen exposure shifts distributions of behaviour, it does not fix identity, and its influence is best described as one contribution to a complex developmental process (Berenbaum, 2018).

Social-cognitive theory supplies the complementary account of how the social environment shapes gender. Bussey and Bandura's model treats gender development as governed by three interacting modes of influence — modelling, enactive experience (the consequences of one's own gendered actions), and direct tuition — integrated through the child's growing capacity for self-regulation, so that children come to guide their own behaviour by internalized gender standards rather than by external reward alone (Bussey & Bandura, 1999). The contemporary consensus is integrative: cognitive-developmental, schematic, social-cognitive, and biological accounts each capture part of the process, and the comprehensive reviews treat them as complementary levels of explanation rather than as rival theories to be decided between (Martin et al., 2002). A further strand of recent theory argues that the binary framing shared by many of these accounts is itself a simplification that the evidence no longer supports, and calls for models that treat gender as multi-dimensional and partly independent of natal sex (Hyde et al., 2019).

Development in Early Childhood

The empirical timetable of gender development has been mapped in detail, and it begins earlier than the classic theories assumed. Infants discriminate male from female faces and voices in the first year, and the ability to label the sexes — the entry point for schema formation — emerges in the second year: in a longitudinal study, children who began producing gender labels earlier went on to show gender-typed play sooner, exactly the sequence gender-schema theory predicts (Zosuls et al., 2009). Gender-typed toy and playmate preferences are robust by the third year, gender stability follows, and full gender constancy is typically in place by the early school years, giving the ordered progression the figure summarizes.

The developmental picture has been sharpened by longitudinal research on transgender children, which tests whether the ordinary developmental account applies when identity and natal sex diverge. Studies of socially transitioned transgender preschool and school-age children find that their gender identity, and their gender-typed preferences and implicit associations, closely resemble those of cisgender children of the same identified gender rather than of their natal sex (Fast & Olson, 2018). A large cross-sectional comparison reached the same conclusion across several measures, reporting that transgender children's gender development looks much like that of cisgender children matched on identified gender (Gülgöz et al., 2019). These findings are consistent with the view that gender identity is a genuine early self-categorization rather than a product of natal sex acting through it, and they extend the cognitive theories of gender development to a population those theories did not originally consider.

Figure 1

The Ordered Emergence of Gender Development Milestones

Worked Example

Consider the memory asymmetry that Martin and Halverson's schematic-processing model predicts, using the kind of design their studies employed (Martin & Halverson, 1981). Suppose a child is shown 16 pictures of people performing activities, of which 8 are schema-consistent (a girl playing with a doll, a boy with a truck) and 8 are schema-inconsistent (a boy with a doll, a girl with a truck). The model holds that schema-consistent pictures are recalled at a higher rate than inconsistent ones; take representative rates of 0.75 for consistent and 0.45 for inconsistent items. The child is then expected to correctly recall 8 × 0.75 = 6 of the consistent pictures and 8 × 0.45 = 3.6, or about 4, of the inconsistent pictures.

The model's signature prediction is not merely that inconsistent items are recalled less often but that some are distorted to fit the schema. Suppose that of the inconsistent pictures a child does recall, a fraction of 0.5 are misremembered with the actors' roles reversed so that the scene becomes schema-consistent. Then of the roughly 4 inconsistent pictures recalled, about 4 × 0.5 = 2 are recalled in distorted, schema-conforming form and only about 2 are recalled accurately. The child therefore leaves the task reporting a world more gender-stereotyped than the one actually presented — 6 consistent scenes remembered accurately, 2 inconsistent scenes remembered accurately, and 2 inconsistent scenes actively converted into consistent ones. This is the sense in which the gender schema is an active reconstruction rather than a passive filter, and it is the arithmetic the demonstration above reproduces as its recall and distortion rates are varied.

Discussion

The trajectory of this field is one of successive theories each preserving the cognitive insight of the last while widening its scope. Kohlberg's contribution was to make gender development a cognitive achievement — something the child constructs and then acts upon — rather than something done to the child by reinforcement (Kohlberg, 1966). Gender-schema theory kept that constructive emphasis but relocated the engine from the single milestone of constancy to the everyday operation of a schema that begins working as soon as the child can label the sexes, which better fit the observation that gender-typed behaviour precedes full constancy (Bem, 1981; Martin & Halverson, 1981). Social-cognitive and biological accounts then supplied the mechanisms the cognitive theories left implicit — how the environment transmits gender and how prenatal biology biases its development — so that the mature view is layered rather than monolithic (Bussey & Bandura, 1999; Hines, 2011).

Two open problems define the current debate. The first is the adequacy of the binary framing that most classic theories assume; the accumulating evidence for gender diversity and for the partial independence of gender's several dimensions has led to calls to rebuild the theories on a non-binary footing (Hyde et al., 2019). The second is the developmental status of transgender identity, where longitudinal work showing that transgender children's gender cognition tracks their identified rather than their natal gender has become an important test case for any general theory of gender development (Fast & Olson, 2018; Gülgöz et al., 2019). The field's progress has been less in choosing a single winning theory than in assembling a multi-level account that can accommodate both the typical case and its instructive exceptions.

Current Directions

The most active current line of research is the longitudinal study of transgender and gender-diverse children, which has, in a little over a decade, supplied evidence the classic theories could only speculate about. Projects following socially transitioned children have shown that their measured gender identity and gender-typed cognition align with their identified gender across explicit and implicit measures, and the accumulating cohorts are beginning to yield the developmental and outcome data the field long lacked (Fast & Olson, 2018; Gülgöz et al., 2019). A second direction is the ongoing effort to move measurement and theory beyond the gender binary, treating masculinity, femininity, and identity as separable dimensions and asking how each develops, an agenda that reaches back to Bem's separation of the masculine and feminine scales but is now pursued with contemporary methods and larger, more diverse samples (Bem, 1974; Hyde et al., 2019). A third is the continued refinement of the biological picture, in which prenatal androgen effects are being characterized as probabilistic contributions interacting with cognitive and social processes rather than as direct determinants of identity (Berenbaum, 2018).

Common Misconceptions

Gender identity is the same thing as biological sex.

They are distinct constructs. Gender identity is the internal sense of one's gender, separated from the biological attributes of sex in the foundational work of Money and Stoller, and the two can diverge (Stoller, 1964; Money et al., 1955).

Children only become gender-typed once they understand gender constancy.

Marked gender-typed preferences appear well before full constancy is achieved. This is why gender-schema theory locates the onset of sex typing at gender labelling, not at constancy (Martin & Halverson, 1981; Zosuls et al., 2009).

Masculinity and femininity are opposite ends of one scale.

Bem's work established that they are separable dimensions: a person can score high on both, the pattern she termed psychological androgyny, so a single masculine-to-feminine continuum misrepresents the data (Bem, 1974).

Glossary

Androgyny.

In Bem's usage, a personality profile that combines high masculine and high feminine attributes, treated as evidence that the two are separable dimensions rather than opposite poles.

Cisgender.

Describing a person whose gender identity matches the sex they were assigned at birth; the complement of transgender.

Congenital adrenal hyperplasia.

A genetic condition producing elevated prenatal androgen exposure, studied for its effects on the development of gender-typed behaviour.

Gender constancy.

The understanding, in Kohlberg's theory, that a person's gender is invariant across superficial changes of appearance or activity; typically achieved between about five and seven years.

Gender identity.

A person's internal sense of being male, female, both, or neither; established early, highly resistant to change, and distinct from biological sex.

Gender role.

The behaviours, presentations, and social expectations a culture associates with a given gender; the outward counterpart of the inner identity.

Gender schema.

An organized knowledge structure about the sexes that guides attention, encoding, and retrieval of gender-relevant information, and that can distort schema-inconsistent input.

Gender stability.

The understanding that gender is constant over time — that a boy becomes a man — the second of Kohlberg's three milestones, reached around three to four years.

Gender typing.

The process by which children acquire the preferences, behaviours, and self-concept a culture associates with their gender; also called sex typing.

In-group/out-group schema.

In Martin and Halverson's model, a simple classification of objects and activities as belonging to one's own sex or the other, formed once a child can label the sexes.

Prenatal androgen exposure.

The influence of male-typical hormones before birth, which shifts the distribution of gender-typed behaviour probabilistically rather than determining identity.

Self-concept.

The organized cognitive representation a person holds of who they are, of which gender identity is one component.

Sex.

The biological attributes — chromosomes, gonads, hormones, and anatomy — on which a person is assigned to a category at birth; distinguished throughout from gender identity.

Transgender.

An umbrella term for people whose gender identity differs from the sex they were assigned at birth.

Transsexualism.

The historical clinical term for a persistent cross-gender identity accompanied by a wish to alter the body to match it; a narrower descriptor under gender identity in MeSH.

Key Researchers

Sandra Lipsitz Bem (1944-2014). American psychologist at Cornell University who developed the Bem Sex-Role Inventory and gender-schema theory, reframing psychological androgyny and the cognitive processing of gender-typed information. Wikipedia - Wikidata

Sheri A. Berenbaum (contemporary). Developmental psychologist at Pennsylvania State University whose work on congenital adrenal hyperplasia established the role of early androgen exposure in gender development. ORCID - Faculty page

Melissa Hines (contemporary). Director of the Gender Development Research Centre at the University of Cambridge, whose reviews synthesize the evidence on prenatal hormone effects on gendered behaviour and the brain. Faculty page - Wikipedia - Wikidata

Lawrence Kohlberg (1927-1987). American psychologist at Harvard University who gave the first cognitive-developmental account of gender, proposing that children's understanding of gender as stable and constant drives gender-typed behaviour. Wikipedia - Wikidata

Carol Lynn Martin (contemporary). Developmental psychologist at Arizona State University who, with Halverson, built the schematic-processing model of sex typing and co-authored the definitive review of cognitive theories of gender development. ORCID - Faculty page

John Money (1921-2006). New Zealand-American psychologist at Johns Hopkins who introduced the term gender role and helped establish the distinction between biological sex and gender identity; his work remains controversial. Wikipedia - Wikidata

Kristina R. Olson (contemporary). Developmental psychologist at Princeton University who founded the TransYouth Project, the first large-scale longitudinal study of socially transitioned transgender children. Faculty page - Wikipedia - Wikidata

Robert J. Stoller (1924-1991). American psychoanalyst at UCLA who introduced the term gender identity and its separation from biological sex, providing the conceptual foundation of the field. Wikipedia - Wikidata

Frequently Asked Questions

What is gender identity? Gender identity is a person's internal sense of being male, female, both, or neither, established early in childhood and highly resistant to later change. It is distinct from biological sex and from the culturally defined gender role, a distinction fixed in the foundational work of Stoller (Stoller, 1964).

How is gender identity different from biological sex? Sex refers to the biological attributes of male and female, while gender identity is the person's own internal sense of their gender. The two usually align but can diverge, and the separation of the concepts was a foundational step in the field (Money et al., 1955).

What is gender constancy? Gender constancy is the understanding, in Kohlberg's cognitive-developmental theory, that a person's gender stays the same across superficial changes of dress or activity. It is the most demanding of his three milestones and is typically reached between five and seven years of age (Kohlberg, 1966).

What is gender-schema theory? Gender-schema theory holds that once children can label the sexes they form gender schemas, organized knowledge structures that guide attention, memory, and inference about gender and that can distort schema-inconsistent information (Bem, 1981; Martin & Halverson, 1981).

At what age does gender development begin? It begins in infancy: children discriminate male and female faces in the first year and begin labelling the sexes in the second, with earlier labelling predicting earlier gender-typed play (Zosuls et al., 2009).

Do prenatal hormones influence gender development? Prenatal androgen exposure shifts the distribution of gender-typed behaviour, as studies of girls with congenital adrenal hyperplasia show, but the effect is probabilistic and partial rather than a direct determinant of identity (Hines, 2011; Berenbaum, 2018).

How does the gender development of transgender children compare with that of cisgender children? Longitudinal studies of socially transitioned transgender children find their gender identity and gender-typed cognition closely resemble those of cisgender children of the same identified gender rather than of their natal sex (Fast & Olson, 2018; Gülgöz et al., 2019).

Are masculinity and femininity opposites? No. Bem's research established that they are separable dimensions rather than the two ends of one scale, so that a person can be high on both, a profile she called psychological androgyny (Bem, 1974).

References