# Oxytocin increases trust in humans

- **DOI:** [10.1038/nature03701](https://doi.org/10.1038/nature03701)
- **Rating:** Sound (3.4 out of 5)
- **Verdict:** A pioneering randomized trust experiment isolates a social-risk effect of oxytocin, but its headline claim exceeds the robustness of the primary analysis.
- **Framework version:** v5.8
- **Pipeline version:** v0.4
- **Evaluation date:** 2026-08-24
- **Evaluation page:** https://www.nabu.science/paper/10.1038%2Fnature03701

## Key claims

1. intranasal administration of oxytocin causes a substantial increase in trusting behaviour
   - Support: Tentative
   - Type: Primary finding
   - Basis: the narrow randomized task effect relies on a modest sample and a one-sided primary test
   - Location: Conclusion
2. oxytocin specifically affects an individual’s willingness to accept social risks arising through interpersonal interactions
   - Support: Supported
   - Type: Secondary
   - Basis: the matched risk task supports social specificity, though separate task samples temper the inference
   - Location: Abstract; Results, risk experiment
3. oxytocin does not increase the general inclination to behave prosocially
   - Support: Tentative
   - Type: Secondary
   - Basis: a nonsignificant trustee comparison in modest samples cannot establish the absence of a general prosocial effect
   - Location: Results, trustee-behaviour analysis
4. investors given oxytocin show more trusting behaviour but do not hold significantly different beliefs about the trustworthiness of others
   - Support: Tentative
   - Type: Secondary
   - Basis: the belief result is a modest-sample null used to exclude an alternative mechanism
   - Location: Results, belief-elicitation analysis

## Strengths

- **Matched control separates social risk** The risk experiment preserves the transfer choices and payoff-risk distribution while replacing the trustee with a random mechanism, directly probing whether the treatment effect is specific to interpersonal risk.
  - Dimension: methodological rigour
  - Location: Results, risk experiment; Figure 2; Table 1
- **Multiple alternative mechanisms are tested** Mood and calmness, trustee back-transfers, and investor beliefs are reported alongside the primary result, exposing several competing explanations to empirical checks.
  - Dimension: contribution
  - Location: Results, mood analyses and Figure 3
- **Focused integration of relevant literatures** The study connects animal evidence on oxytocin and social attachment with established trust-game and betrayal-aversion research, creating a clear rationale for the human experiment.
  - Dimension: positioning
  - Location: Introduction; references 6–8 and 19–25

## Limitations

- **Primary inference depends on one-sided testing** The headline trust comparison reports P=0.029 using a one-sided Mann–Whitney test, while most other comparisons are two-sided; the corresponding two-sided signal would not meet the conventional 0.05 threshold.
  - Dimension: methodological rigour
  - Location: Results, primary trust-game comparison
- **Population claims exceed sampled participants** The title and abstract refer to trust “in humans,” but the study recruited healthy male Zurich students and does not trace that restriction into its conclusions.
  - Dimension: positioning
  - Location: Title; Abstract; Methods, Subjects
- **Null mechanisms receive definitive wording** Nonsignificant trustee and belief comparisons are used to reject general prosociality and belief-based mechanisms, although the modest samples provide limited power for establishing absence of effects.
  - Dimension: contribution
  - Location: Results, trustee behaviour and belief analyses; Figure 3

## Reliability checks

Caveats (4 of 4 checks).

- **Internal coherence:** Noted. The central figure required a post-publication correction, but the corrigendum states that the omitted observations were included in every statistical test. The disclosed one-sided primary test is analytically consequential but does not constitute an impossible or internally contradictory result.
- **Research conduct:** Clean. Ethics approval, written informed consent, random assignment, double blinding, funding, and absence of competing financial interests are declared. Missing registration and data or code availability statements are treated as era-appropriate reporting gaps rather than conduct concerns.
- **Reference integrity:** Clean. 25 of 25 checkable references verified
  - 30 references in manuscript
  - 4 have no canonical index record — counted, but not index-checkable
  - 1 could not be checked — not the same as no match found
  - 3 references confirmed by manual review
  - References not verified automatically are resolved manually within 24 hours.

## Post-publication record

- **Status:** Clean
- **Summary:** No retraction or expression of concern found in Crossref (includes Retraction Watch).
- **Last checked:** 2026-10-17
