Oxytocin

01 · What it is

The intranasal oxytocin literature is the most instructive failure of replication in this project, and it comes with a retracted meta-analysis whose retraction notice actually says why.

In 2015 a meta-analysis of placebo-controlled studies reported an effect of intranasal oxytocin on psychiatric symptoms ². It was retracted in 2018 after it was brought to the authors’ attention that there were several significant data analytic errors linked to the data entry and the software program used ^2,8. That is a rare thing: a retraction that states its grounds, and grounds that are honest error rather than misconduct.

The retraction is the smallest part of the story. Across a decade of randomised trials and meta-analyses, intranasal oxytocin has repeatedly failed to do what the early literature suggested. Trials in schizophrenia found it did not improve any aspect of symptoms across eight randomised controlled trials ³. Meta-analysis of clinical trials to improve social cognition in neurodevelopmental disorders yielded inconclusive results ⁶. A variance meta-analysis concluded that repeated administration does not improve most of the core symptoms of schizophrenia and autism spectrum disorder ¹⁷. And a 24-week placebo-controlled phase 2 trial in children and adolescents with autism was published in the New England Journal of Medicine ¹⁶ after experimental studies and small clinical trials had suggested benefit.

A single study captures the field’s problem better than any meta-analysis. In 2024, healthy participants were randomised to a three-day course of oxytocin nasal drops that were actually a sham treatment, delivered by a practitioner displaying either high or low warmth and empathy. The practitioner’s manner changed the reported response ²⁰.

What oxytocin actually is. An endogenous hypothalamic-pituitary nonapeptide, structurally similar to arginine-vasopressin, with broad physiological actions including regulation of caloric intake and metabolism ²¹. Its established medical uses are outside the scope of what this guide searched.

The compound is real, its receptor is real, and the specific claim that sniffing it improves social function is the one that has not held up.

02 · Evidence at a glance

Evidence grade

Mixed Evidence — abundant randomised human data whose conclusions conflict and largely trend negative

What it is

An endogenous hypothalamic-pituitary nonapeptide, structurally similar to arginine-vasopressin ²¹

Scope of this guide

Intranasal administration. Established obstetric and other clinical uses were not searched and are not covered

A retracted meta-analysis in the record

Yes — 2015, retracted 2018 for data-entry and software analytic errors ^2,8

Schizophrenia

Did not improve any aspect of symptoms across 8 RCTs ³; inconsistent findings ⁹; no beneficial effect on positive symptoms in main meta-analysis ¹³

Autism spectrum disorder

24-week placebo-controlled phase 2 trial in NEJM ¹⁶; multilevel meta-analysis ¹¹; repeated administration does not improve most core symptoms ¹⁷

Social cognition in neurodevelopmental disorders

Inconclusive ⁶

Where effects are reported

Emotion recognition in healthy individuals, particularly fear ⁵; cooperative behaviour, moderate effect ¹²; attachment insecurity ¹⁵

The methodological problem

Central and peripheral oxytocin concentrations correlate only under specific conditions ⁴

The expectancy problem

Practitioner warmth altered response to sham oxytocin nasal drops ²⁰

Adverse events, long-term ASD use

Systematically reviewed; no common adverse event statistically associated with treatment allocation ⁷

Literature size

~713 records on this search alone

03 · Mechanism of action

Getting it to the brain is the unsolved problem

Intranasal delivery is premised on the peptide reaching the central nervous system. A pre-registered systematic review and meta-analysis of correlations between central and peripheral oxytocin concentrations found coordination of central and peripheral release after stress and after intranasal administration — but the general relationship between the two compartments is the question, and it is the reason plasma measurements have limited value as a proxy for brain exposure ⁴.

Every behavioural claim about intranasal oxytocin depends on an unstated premise about how much reaches the brain. That premise has its own literature, and it is not settled.

What the receptor does, physiologically

Oxytocin and arginine-vasopressin are structurally similar hypothalamic-pituitary peptides with broad physiologic actions including regulation of caloric intake and metabolism ²¹. Endogenous oxytocin release is associated with anxiety and reduction in meal size ¹⁴.

Note the direction: endogenous oxytocin rises with anxiety. The popular framing of this molecule as a straightforward “bonding” or calming agent does not survive contact with its physiology.

Where effects have been found

The 2003 study that helped launch the field found that oxytocin combined with social support suppressed cortisol and subjective responses to psychosocial stress, with oxytocin appearing to enhance the buffering effect of social support ¹. Meta-analysis finds a single dose improves recognition of basic emotions, particularly fear, and increases expression of positive emotions among healthy individuals ⁵. Hormone-administration meta-analysis finds a moderate positive effect on cooperative behaviours — against a large negative effect for vasopressin ¹².

These are real, replicated findings in healthy volunteers on laboratory tasks. They are also not the same thing as clinical benefit in a disorder, and the gap between those two is where this literature came apart.

Where it has not

Across eight randomised controlled trials in schizophrenia, oxytocin did not improve any aspect of symptoms ³. A later meta-analysis and dose-response meta-analysis found no beneficial effect on positive symptoms in its main analysis ¹³. Efficacy findings in schizophrenia have been inconsistent ⁹. In neurodevelopmental disorders, trials to improve social cognition have yielded inconclusive results ⁶. A meta-analysis of variance in schizophrenia and autism concluded that repeated administration does not improve most core symptoms ¹⁷.

Four independent syntheses, different methods, same direction.

04 · Key research findings

The foundational stress study, 2003. Participants randomised to intranasal oxytocin 24 IU or placebo, with or without social support from their best friend before a psychosocial stressor. Oxytocin appeared to enhance the buffering effect of social support on cortisol and subjective stress responses ¹.

Elegant, influential, and about an interaction with social context rather than a drug effect on its own — a nuance the subsequent popular framing lost.

The retracted meta-analysis, 2015 and 2018. A meta-analysis of placebo-controlled studies of intranasal oxytocin on psychiatric symptoms ², retracted in 2018 after the authors were made aware of several significant data analytic errors linked to data entry and the software program used ⁸.

Fourth retraction encountered in this project and the first whose notice states grounds. Worth crediting: the authors corrected the record.

The schizophrenia syntheses, 2017–2021. A multivariate Bayesian meta-analysis of eight RCTs (n = 238) finding oxytocin did not improve any aspect of symptoms, in a field where evidence across eight RCTs and three prior meta-analyses had been inconsistent ³. A meta-analysis of adjunctive intranasal oxytocin in schizophrenia ⁹. A systematic review, meta-analysis and dose-response meta-analysis in negative symptoms, finding no beneficial effect on positive symptoms in the main analysis ¹³.

The delivery question, 2017. A pre-registered systematic search and meta-analysis of correlations between central and peripheral oxytocin concentrations ⁴.

Pre-registered, which in this literature is worth noting explicitly.

Emotion processing, 2017. Meta-analysis finding a single dose significantly improved recognition of basic emotions, particularly fear, and increased expression of positive emotions among healthy individuals, while not significantly influencing other measures examined ⁵.

Social cognition in neurodevelopmental disorders, 2018. Meta-analysis noting that clinical trials to improve social cognition have yielded inconclusive results ⁶.

Long-term safety in autism, 2018. A systematic review and meta-analysis of reported adverse events of long-term intranasal oxytocin in autism spectrum disorder. No common adverse event was statistically associated with treatment allocation, all p-values above 0.1. Five severe adverse events were reported, including aggression — one in placebo, two on oxytocin ⁷.

The most useful safety document in the file, and its numbers are small enough that they should be read as reassurance about common events rather than as exclusion of rare ones.

Autism, 2021. A multilevel meta-analysis assessing oxytocin’s efficacy in autism spectrum disorders ¹¹, and a 24-week placebo-controlled phase 2 trial of intranasal oxytocin in children and adolescents with autism spectrum disorder, published in the New England Journal of Medicine, which opens by noting that experimental studies and small clinical trials had suggested benefit ¹⁶.

This guide has read the abstract only and does not report the trial’s result. Its design — 24 weeks, placebo-controlled, in a top-tier journal — makes it the most important single study in the file, and the honest position is that this guide has not read it properly.

Cooperation, food intake and attachment, 2021. Moderate positive effect on cooperative behaviours ¹²; effects on food intake and craving ¹⁴; and a meta-analysis finding that intranasal but not peripheral oxytocin regulates behaviours of attachment insecurity ¹⁵. Also a study finding intranasal oxytocin decreased self-oriented learning ¹⁰.

The variance meta-analysis, 2022. Examining heterogeneity in response to repeated intranasal oxytocin in schizophrenia and autism, and concluding that repeated administration does not improve most core symptoms ¹⁷.

A meta-analysis of variance asks whether some people respond even when average effects are null. Its answer here was not encouraging.

Trauma memory, 2022. A randomised, double-blind, placebo-controlled study in which 220 healthy women received a single dose of intranasal oxytocin 24 IU or placebo before a trauma film paradigm soliciting intrusive memories ¹⁸.

Well-powered by this field’s standards, and in healthy volunteers on an experimental paradigm rather than in patients.

The sham-treatment study, 2024. Eighty-four healthy participants randomised to a three-day course of oxytocin nasal drops that were in fact sham treatment, delivered in conjunction with high or low practitioner warmth and empathy. Practitioner warmth attenuated the nocebo effect and enhanced the placebo effect ²⁰.

This does not test oxytocin. It tests the setting oxytocin is given in, and it demonstrates that the setting alone moves reported outcomes. Any open-label or weakly blinded oxytocin result should be read against it.

Metabolic physiology, 2025. Arginine-vasopressin secretory dynamics in relation to food intake and an eight-week course of intranasal oxytocin in adults with obesity ²¹.

05 · Evidence overview

06 · Safety profile

Human data. This is one of the better-characterised safety pictures in the project. A systematic review and meta-analysis of reported adverse events during long-term intranasal oxytocin in autism spectrum disorder found no common adverse event statistically associated with treatment allocation, all p-values above 0.1. Five severe adverse events were reported across the pooled studies, including aggression, distributed one in placebo and two on oxytocin ⁷. Randomised trials in healthy volunteers have used single 24 IU doses ^1,18 and multi-week courses have been run in clinical populations ^16,21.

Common adverse events appear no more frequent than placebo. The severe-event numbers are too small to support any conclusion in either direction, and the guide states them rather than rounding them into reassurance.

What is genuinely unknown. How much intranasally administered oxytocin reaches the brain, and in whom — the central–peripheral concentration relationship remains an active research question ⁴. Whether any of the healthy-volunteer laboratory effects ^1,5,12 correspond to anything a person would notice outside a laboratory. What repeated administration does over months, beyond the trial durations studied.

Two further gaps. Vasopressin cross-reactivity. Oxytocin and arginine-vasopressin are structurally similar ²¹, and vasopressin administration has a large negative effect on cooperative behaviour where oxytocin has a moderate positive one ¹² — two similar molecules with opposing effects on the same endpoint is a reason to care about receptor selectivity at any given dose, and nothing located here characterises it. And the expectancy problem is a safety issue as well as an efficacy one: if practitioner manner alone moves reported outcomes with sham drops ²⁰, then self-reported benefit from an unblinded product carries very little information.

Nothing is published on the identity or purity of any material sold under this name outside a pharmacy supply chain.

07 · US regulatory status

Current as of 7 September 2026. Oxytocin is an established medicine with recognised clinical uses. This guide searched intranasal administration only, and did not examine oxytocin’s approved indications, formulations, labelling or regulatory records in any jurisdiction. It therefore makes no statement about approval status, and that limitation is logged.

What can be said within the scope searched: intranasal oxytocin has been studied extensively in randomised trials for psychiatric and neurodevelopmental indications ^{3,6,9,13,16,17}, and this guide located no approved intranasal indication among the sources reviewed.

It is not a controlled substance in the United States.

Under the World Anti-Doping Code, oxytocin does not appear as a named prohibited substance in the classes reviewed for this guide. Competitors should consult the current Prohibited List directly rather than rely on secondary summaries, including this one.

08 · Limitations of the evidence

1. A meta-analysis in this literature was retracted for analytic error ^2,8. The record was corrected, which is to the authors’ credit, and it means any synthesis published before 2018 should be checked against what superseded it.
2. The clinical evidence trends null where it matters most. No improvement in any aspect of schizophrenia symptoms across eight RCTs ³; inconclusive results for social cognition in neurodevelopmental disorders ⁶; no beneficial effect on positive symptoms in the main analysis of a dose-response meta-analysis ¹³; and no improvement in most core symptoms of schizophrenia or autism with repeated administration ¹⁷.
3. The gap between laboratory effects and clinical benefit is the central problem. Emotion recognition improves in healthy volunteers ⁵ and cooperative behaviour shifts moderately ¹², while symptom scores in patients do not move ^3,17. Both sets of findings can be true, and only one of them is what a person would be buying the compound for.
4. Central delivery is assumed rather than demonstrated ⁴. Every behavioural claim rests on it.
5. Expectancy effects are demonstrably large in this specific context. Practitioner warmth altered response to sham oxytocin nasal drops ²⁰, which sets a floor on how much any unblinded report can be trusted.
6. This guide did not read the most important trial. The 24-week placebo-controlled phase 2 trial in autism ¹⁶ was read as an abstract only, and its result is deliberately not reported here. Anyone relying on this guide for that question should read the paper.
7. Severe adverse event numbers are too small to interpret ⁷. Five events across pooled studies supports no conclusion about rare harms.
8. The scope is intranasal only. Oxytocin’s established medical uses were not searched, so this guide is silent on them by design rather than by oversight.
9. Fewer than twenty-five of roughly 713 records were examined, weighted toward meta-analyses and methodological critiques. A different selection would emphasise different things.
10. This guide has read abstracts, not full texts, for every source cited.
11. Nothing is known about material sold under this name outside a pharmacy supply chain.

09 · References

1. Heinrichs M, Baumgartner T, Kirschbaum C, Ehlert U. Social support and oxytocin interact to suppress cortisol and subjective responses to psychosocial stress. Biol Psychiatry. 2003 Dec 15;54(12):1389–1398.

   PMID 14675803 ↗

2. Hofmann SG, Fang A, Brager DN. RETRACTED: Effect of intranasal oxytocin administration on psychiatric symptoms: a meta-analysis of placebo-controlled studies. Psychiatry Res. 2015 Aug 30;228(3):708–714. Retracted — see reference 8..

   PMID 26094200 ↗

3. Williams DR, Bürkner PC. Effects of intranasal oxytocin on symptoms of schizophrenia: a multivariate Bayesian meta-analysis. Psychoneuroendocrinology. 2017 Jan;75:141–151. Review.

   PMID 27825069 ↗

4. Valstad M, Alvares GA, Egknud M, Matziorinis AM, Andreassen OA, Westlye LT, Quintana DS. The correlation between central and peripheral oxytocin concentrations: a systematic review and meta-analysis. Neurosci Biobehav Rev. 2017 Jul;78:117–124.

   PMID 28442403 ↗

5. Leppanen J, Ng KW, Tchanturia K, Treasure J. Meta-analysis of the effects of intranasal oxytocin on interpretation and expression of emotions. Neurosci Biobehav Rev. 2017 Jul;78:125–144. Review.

   PMID 28467893 ↗

6. Keech B, Crowe S, Hocking DR. Intranasal oxytocin, social cognition and neurodevelopmental disorders: a meta-analysis. Psychoneuroendocrinology. 2018 Jan;87:9–19. Review.

   PMID 29032324 ↗

7. Cai Q, Feng L, Yap KZ. Systematic review and meta-analysis of reported adverse events of long-term intranasal oxytocin treatment for autism spectrum disorder. Psychiatry Clin Neurosci. 2018 Mar;72(3):140–151.

   PMID 29232031 ↗

8. Retraction notice to “Effect of intranasal oxytocin administration on psychiatric symptoms: a meta-analysis of placebo-controlled studies”. Psychiatry Res. 2018 May;263:299.

   PMID 29678249 ↗

9. Zheng W, Zhu XM, Zhang QE, Yang XH, Cai DB, Li L, Li XB, Ng CH, Ungvari GS, Ning YP, Xiang YT. Adjunctive intranasal oxytocin for schizophrenia: a meta-analysis of randomized, double-blind, placebo-controlled trials. Schizophr Res. 2019 Apr;206:13–20. Review.

   PMID 30573406 ↗

10. Liao Z, Huang L, Luo S. Intranasal oxytocin decreases self-oriented learning. Psychopharmacology (Berl). 2021 Feb;238(2):461–474.

    PMID 33156402 ↗

11. Huang Y, Huang X, Ebstein RP, Yu R. Intranasal oxytocin in the treatment of autism spectrum disorders: a multilevel meta-analysis. Neurosci Biobehav Rev. 2021 Mar;122:18–27.

    PMID 33400920 ↗

12. Yang X, Wang W, Wang XT, Wang YW. A meta-analysis of hormone administration effects on cooperative behaviours: oxytocin, vasopressin, and testosterone. Neurosci Biobehav Rev. 2021 Jul;126:430–443. Review.

    PMID 33819546 ↗

13. Sabe M, Zhao N, Crippa A, Strauss GP, Kaiser S. Intranasal oxytocin for negative symptoms of schizophrenia: systematic review, meta-analysis, and dose-response meta-analysis of randomized controlled trials. Int J Neuropsychopharmacol. 2021 Aug 20;24(8):601–614.

    PMID 33890987 ↗

14. Chen CY, Chiang YC, Kuo TC, Tam KW, Loh EW. Effects of intranasal oxytocin in food intake and craving: a meta-analysis of clinical trials. Clin Nutr. 2021 Oct;40(10):5407–5416.

    PMID 34600216 ↗

15. Zhang K, Fan Y, Yu R, Tian Y, Liu J, Gong P. Intranasal oxytocin administration but not peripheral oxytocin regulates behaviors of attachment insecurity: a meta-analysis. Psychoneuroendocrinology. 2021 Oct;132:105369. Review.

    PMID 34340132 ↗

16. Sikich L, Kolevzon A, King BH, McDougle CJ, Sanders KB, Kim SJ, Spanos M, Chandrasekhar T, Trelles MDP, Rockhill CM, Palumbo ML, Witters Cundiff A, Montgomery A, Siper P, Minjarez M, Nowinski LA, Marler S, Shuffrey LC, Alderman C, Weissman J, Zappone B, Mullett JE, Crosson H, Hong N, Siecinski SK, Giamberardino SN, Luo S, She L, Bhapkar M, Dean R, Scheer A, Johnson JL, Gregory SG, Veenstra-VanderWeele J. Intranasal oxytocin in children and adolescents with autism spectrum disorder. N Engl J Med. 2021 Oct 14;385(16):1462–1473.

    PMID 34644471 ↗

17. Martins D, Paduraru M, Paloyelis Y. Heterogeneity in response to repeated intranasal oxytocin in schizophrenia and autism spectrum disorders: a meta-analysis of variance. Br J Pharmacol. 2022 Apr;179(8):1525–1543. Review.

    PMID 33739447 ↗

18. Schultebraucks K, Maslahati T, Wingenfeld K, Hellmann-Regen J, Kraft J, Kownatzki M, Behnia B, Ripke S, Otte C, Roepke S. Intranasal oxytocin administration impacts the acquisition and consolidation of trauma-associated memories: a double-blind randomized placebo-controlled experimental study in healthy women. Neuropsychopharmacology. 2022 Apr;47(5):1046–1054.

    PMID 34887528 ↗

19. Lopes S, Osório FL. Effects of intranasal oxytocin on pain perception among human subjects: a systematic literature review and meta-analysis. Horm Behav. 2023 Jan;147:105282.

    PMID 36463692 ↗

20. Barnes K, Wang R, Faasse K. Practitioner warmth and empathy attenuates the nocebo effect and enhances the placebo effect. Appl Psychol Health Well Being. 2024 May;16(2):421–441.

    PMID 37793644 ↗

21. Aulinas A, Galbiati F, Wronski ML, Carter CS, Nazarloo S, Holman K, Golden ER, Muhammed M, Hadaway N, Asanza E, Smith SE, Plessow F, Lawson EA. Arginine-vasopressin dynamics in relation to food intake and 8-week intranasal oxytocin treatment in adults with obesity. J Clin Endocrinol Metab. 2025 Nov 18;110(12):e3972–e3979.

    PMID 40314146 ↗