The Neurobiology of Gender Incongruence: Beyond the Sexed-Brain Hypothesis

Research into the neurobiology of gender incongruence has progressed from small postmortem studies of sexually differentiated hypothalamic and extended-amygdala nuclei to structural magnetic resonance imaging (MRI), diffusion imaging, resting-state and task functional MRI, molecular imaging, longitudinal studies of gender-affirming hormone therapy (GAHT), developmental neuroimaging, and large multisite analyses. This literature demonstrates measurable group-level neural variation associated with transgender status and gender incongruence, but it does not support a simple model in which transgender individuals possess brains globally shifted from one binary sex endpoint toward the other. Early postmortem observations involving the bed nucleus of the stria terminalis (BSTc) and INAH3 reported sex-atypical characteristics in small transgender samples. Subsequent structural and diffusion studies identified differences involving cortical morphology, subcortical anatomy, and white-matter microstructure, but anatomical location, effect direction, treatment state, and apparent relation to cisgender reference groups vary across studies. Resting-state and task-based investigations provide recurring evidence involving systems participating in body representation, self-referential processing, salience, and multisensory integration. These observations motivate mechanistic investigation but do not establish that such neural differences precede or cause persistent gender incongruence. Molecular imaging remains sparse. In contrast, longitudinal endocrine-intervention studies provide comparatively strong evidence that adult neural structure, microstructure, functional activation, and selected neurochemical systems are hormone-responsive. Large-scale evidence is especially important for recalibrating interpretation. The ENIGMA Transgender Persons Working Group mega-analysis of 803 non-hormonally treated participants found region- and metric-dependent neuroanatomical differences rather than a uniform displacement along a male-female axis. Developmental evidence remains considerably weaker: pediatric and adolescent neuroimaging is dominated by cross-sectional designs, heterogeneous endocrine states, and small samples, leaving the temporal emergence of gender-incongruence-associated neural variation largely unresolved. This paper proposes a Multidimensional Neurodevelopmental Phenotype Model, in which measured neural phenotype is a time-dependent product of prior neural state, genomic variation, molecular and epigenetic regulation, effective steroid signaling, developmental environment, experience, medical intervention, and developmental stage. Eight methodological constraints are introduced: the Neural Reference-Class Constraint, Classifier–Identity Separation Principle, Neural Causal Direction Constraint, Neural Treatment-State Constraint, Developmental Temporal Precedence Constraint, Neural Phenotype Non-Equivalence Principle, Cohort-Independence Constraint, and Historical Phenotype Transportability Constraint. A Neural Null-Evidence Constraint is additionally proposed to ensure that etiological heterogeneity does not render neurobiological hypotheses unfalsifiable. The strongest present conclusion is neither neurological determinism nor neurological irrelevance. Persistent gender incongruence is associated with multidimensional neural variation, some measurable before GAHT and some demonstrably treatment-responsive. What remains unestablished is whether any particular neural state constitutes a necessary, sufficient, or general causal mechanism of persistent gender incongruence.