Sexual response and reproductive control

Established sex differences and their scale

Humans have two reproductive sexes, organized around the production of large ova or small sperm. They have different typical gonads, genital anatomy, endocrine organization, reproductive roles, and reproductive functions. Variations in sex development can alter chromosomes, gonads, hormone production or response, internal ducts, or external anatomy; they do not create an additional gamete class (Bhargava et al., 2021; Lehtonen et al., 2016).

Brain comparisons require more specific language because each result belongs to a measurement, sample, and model. Analyses of overlapping UK Biobank imaging data report sex differences in regional volumes, cortical surface areas, thicknesses, and other structural measures, including differences that remain after modelling total brain size or allometric scaling; their direction and magnitude vary by region and measure (Ritchie et al., 2018; Williams et al., 2021). The 5,216-participant and 40,028-participant analyses come from the same population resource and are not independent replications. The companion exhibit on brain structure, scaling, and variability examines these distinctions directly.

An unedited early-access 2026 analysis of 978 Human Connectome Project participants, 455 male and 523 female, across seven fMRI tasks found widespread, largely task-specific functional associations of small-to-moderate magnitude that were stable across repeated split-half analyses within that cohort and spatially unaligned with structural volume differences (Liu et al., 2026). Only anatomy was checked in an independent UK Biobank cohort. Sexual orientation was not reported as an analysis variable. The publisher still labels the article an unedited early-access version and warns that it may contain production errors, so its claims should be rechecked against the edited version when available.

Hemispheric asymmetry illustrates why evidence strength and effect magnitude must be kept separate. A critical review found slightly stronger average lateralisation in males across many measures, but the meta-analytic estimates reviewed were small, while large-sample structural work mapped regional sex associations within several distributed asymmetry patterns rather than one global asymmetry (Hirnstein et al., 2019; Saltoun et al., 2023). A separate meta-analysis found no reliable male-female difference in language lateralisation measured through planum-temporale asymmetry, dichotic listening, or functional imaging (Sommer et al., 2008). Robust evidence can therefore describe a small or null effect; robust does not mean large. Distributional overlap limits what a group mean can establish about an individual, but it does not erase a measured mean difference.

Desire and motivation

Sexual desire and motivation emerge from distributed interactions among attention, incentive value, salience, interoception, endocrine state, learning, and context. Reviews and meta-analyses implicate the ventral striatum, amygdala, insula, cingulate cortex, and hypothalamus as non-exclusive contributors, alongside sensory and association cortices (Bittoni and Kiesner, 2023; Stoleru et al., 2012).

These sources study different things. Mitricheva and colleagues pooled 61 visual-stimulus fMRI studies and 1,850 people. The source descriptions included heterosexual, homosexual, bisexual, and transsexual samples, but the latter two subsets were too sparse for reliable subgroup analysis. Poeppl and colleagues restricted their key quantitative comparison to healthy heterosexual participants and analysed experiments and activation foci rather than one deduplicated participant cohort. They reported some sex-related differences; stimulus modality, analysis choices, and sample composition are plausible contributors to the disagreement (Mitricheva et al., 2019; Poeppl et al., 2016). A direct methodological letter and reply then disputed experiment selection, matching, and false-positive control (Poeppl et al., 2020; Mitricheva et al., 2020). Sexual orientation is a distinct sampling variable, not a proxy for sex. Activation shows association under a paradigm, not that a region is necessary for desire. No highlighted mesh is a desire centre.

Arousal and pleasure

Genital and autonomic response, subjective arousal, desire, pleasure, and consent are distinct. They can covary, but none can be inferred from another or from a scan. Human imaging reviews describe changing somatosensory, thalamic, insular, cingulate, hypothalamic, autonomic, and evaluative participation across paradigms rather than a single arousal circuit (Georgiadis and Kringelbach, 2012; Stoleru et al., 2012).

The postcentral-gyrus mesh is a gross proxy, and the paracentral-lobule mesh is a procedural proxy, for a finer somatotopic genital representation that this atlas cannot separate. Their highlight marks approximate anatomical territory, not a dedicated genital-arousal structure and never a measure of pleasure or consent.

Orgasm

Orgasm is not the same event as ejaculation: either can occur without the other, and ejaculation adds reproductive-tract and spinal motor processes that an orgasm contrast does not automatically measure. Orgasm imaging instead describes a temporally changing pattern of sensory, motor, interoceptive, reward, cortical, cerebellar, and brainstem participation (Georgiadis and Kringelbach, 2012).

The evidence is limited by unusually small samples and paradigms that are sensitive to head motion, variable latency, self-report timing, stimulation method, and the chosen baseline. The review by Stoleru and colleagues covered 73 original studies and emphasized a literature dominated by healthy young heterosexual men. Wise and colleagues recruited 14 women; ten completed an analysed orgasm, ages 29 to 74, seven exclusively heterosexual and three reporting some bisexual experience. Five analysed first orgasms were self-induced and five partner-induced. That study found activity rising toward orgasm with no detected deactivation (Stoleru et al., 2012; Wise et al., 2017). Its postcentral and precentral results appeared only at a relaxed exploratory cluster-forming threshold after the standard-threshold immediate-pre-orgasm contrast was null. Dedicated brainstem maps used still more permissive thresholds, although a broader pons focus also appeared in a standard whole-brain contrast. The calibrated conclusion is that prefrontal activation and deactivation results conflict across small, different paradigms, not that the prefrontal cortex simply switches off.

Reproductive endocrine control

The hypothalamic-pituitary-gonadal axis regulates reproductive endocrine function and must be kept separate from desire, arousal, pleasure, and orgasm. Its established sequence is (Endotext, 2024 update):

1. The hypothalamic KNDy network integrates endocrine and other inputs through kisspeptin and neurokinin-B signalling and regulates GnRH neurons.
2. Pulsatile GnRH enters hypophyseal portal vessels and signals to gonadotrophs in the anterior pituitary.
3. The anterior pituitary releases LH and FSH, which signal to the gonads and regulate steroid production and gamete maturation.
4. Gonadal steroids and inhibin feed back at hypothalamic and pituitary levels; the timing and sign of feedback differ between typical ovarian and testicular reproductive physiology.

Peripheral kisspeptin also modulated extra-hypothalamic limbic and paralimbic responses to sexual and bonding images in a randomized, double-blind, placebo-controlled crossover study. Investigators recruited 31 healthy heterosexual men, excluded two for motion greater than 2 mm, and analysed 29 with a mean age of 25. The experiment does not directly localize a hypothalamic effect or turn the activated regions into a universal human mechanism (Comninos et al., 2017). Human postmortem evidence from 38 HIV-negative adults, 18 male and 20 female, ages 17 to 65, found sex-related variation in the microscopic INAH3 within anterior hypothalamic territory while finding no such volume difference in the other INAH nuclei. Sexual orientation was not studied, and the present preoptic mesh is much coarser (Byne et al., 2000).

The pituitary, hypophyseal portal vessels, and gonads lie outside this brain atlas’s physical scope. Individual GnRH neurons and kisspeptin/neurokinin-B/dynorphin, or KNDy-like, cells lie outside its spatial resolution. The highlighted hypothalamic meshes therefore show gross territory, not the endocrine pathway itself.

Comparative rodent evidence

Rodent AVPV, MPOA, VMH, medial-amygdala, BNST, PAG, and brainstem studies provide comparative evidence about cell types, hormone sensitivity, and species-specific mating patterns. Those causal rodent mechanisms are not established human male and female circuits and are not represented by the gross atlas meshes; Endotext likewise distinguishes rodent from larger-mammal and human organization (Endotext, 2024 update).

Reviewed status: Owner review was completed by Taig Mac Carthy on 21 July 2026 for publication. Reviewed status records editorial approval; it is not clinical endorsement.