Poses, sprays, and the replication wave
Two TED-ready hormones. Two failed replications. The atlas keeps the original RCTs and the papers that took them down.
Folk claim
Stand like a CEO and your hormones will make you one. Spray oxytocin and strangers will trust you.
What the papers say
Carney, Cuddy, and Yap (2010) reported that two minutes of expansive posture raised testosterone, cut cortisol, and increased risk-taking (N=42). Ranehill et al. (2015) ran 200 people and found no hormone or risk effect. Simmons and Simonsohn (2017) p-curved the supportive set and found it consistent with selective reporting. Kosfeld et al. (2005) reported that intranasal oxytocin raised trust-game transfers. Nave, Camerer, and McCullough (2015) reviewed the follow-ups as inconclusive. Declerck et al. (2020), including Fehr, registered a high-powered replication with the original social-contact condition and found no OT effect on trusting behavior there.
A viral RCT that dies in a larger registered study stays on the atlas as smoke, with both papers linked. Replication is a stamp, not a vibe.
What the 2010 pose paper claimed
Carney, Cuddy, and Yap assigned 42 people to expansive or contractive poses for two minutes and reported a T rise, a cortisol drop, and more risk-taking. That is a real RCT. It is also N=42 on three correlated outcomes. The TED talk dropped the sample size and kept the endocrine story. The folk mechanism needs the hormone step. Without it, 'feeling powerful' is a mood item.
What the larger test and the p-curve took away
Ranehill, Dreber, Johannesson, Leiberg, Sul, and Weber (2015) used the same pose logic in 200 men and women and found no effect on T, cortisol, or risk. Simmons and Simonsohn (2017) p-curved the supportive postural-feedback set Carney et al. had cited and found a distribution you would expect if the average effect were zero and significant results were selected. A later reply argued for self-report effects in a wider set. That is a different claim than 'the pose moves androgens.' This atlas stamps the hormone and risk arrows smoke.
The spray, and Camerer on the right paper
Kosfeld, Heinrichs, Zak, Fischbacher, and Fehr (2005) is the Nature RCT behind 'oxytocin is the trust hormone.' Nave, Camerer, and McCullough (2015) is the review that should have ended the slogan — and the paper that actually has Camerer on it, unlike the ultimatum T trial the brief misattributed. Declerck, Boone, Pauwels, Vogt, and Fehr (2020) then registered a 677-person double-blind replication, including the original minimal-social-contact condition, and found no OT effect on trusting behavior there. Fehr signed both the discovery and the registered null. We keep both. We do not sell a spray.
Edges in this case
Randomized trial · High confidence
Expansive posture (power pose)→Testosterone
Two minutes in a stance does not identify an androgen change. The larger RCT is a null. The TED talk is not a protocol.
Randomized trial · High confidence
Expansive posture (power pose)→Economic risk tolerance
Expansive posture is not a Holt–Laury intervention. The 2010 risk result did not replicate.
Randomized trial · High confidence
Oxytocin→Interpersonal trust
The trust-hormone slogan is a 2005 RCT that did not survive its own authors' registered replication.
Sources
- D. R. Carney et al. (2010). Power Posing: Brief Nonverbal Displays Affect Neuroendocrine Levels and Risk Tolerance
- E. Ranehill et al. (2015). Assessing the Robustness of Power Posing: No Effect on Hormones and Risk Tolerance in a Large Sample of Men and Women
- J. P. Simmons et al. (2017). Power Posing: P-Curving the Evidence
- M. Kosfeld et al. (2005). Oxytocin increases trust in humans
- G. Nave et al. (2015). Does Oxytocin Increase Trust in Humans? A Critical Review of Research
- C. H. Declerck et al. (2020). A registered replication study on oxytocin and trust
Cite this page
What Causes What. “Poses, sprays, and the replication wave.” https://whatcauseswhat.org/cases/pose-and-spray (atlas updated 2026-08-31).