power posing: what happened when science checked the superhero stance

The study

In 2010 Dana Carney, Amy Cuddy and A. J. Yap published "Power posing: Brief nonverbal displays affect neuroendocrine levels and risk tolerance" in Psychological Science. Carney was then at Columbia Business School.

Participants held either two expansive "high-power" poses (open, taking up space) or two contracted "low-power" poses (closed, making themselves small), for one minute each. The researchers measured testosterone and cortisol in saliva, asked how powerful participants felt, and offered them a gamble to measure risk tolerance. Carney later noted that the final sample was 42 people.

What it found

The abstract reported that high-power posers showed higher testosterone, lower cortisol, more feelings of power and more tolerance for risk, while low-power posers showed the opposite. It concluded that two simple one-minute poses could make a person more powerful, with real-world implications.

It was an appealing message: quick, cheap and easy to teach.

Did it hold up?

For hormones and behaviour, no. For self-reported feelings of power, the picture is mixed.

2015: a large replication. Eva Ranehill and colleagues at the University of Zurich, the Stockholm School of Economics and Dartmouth College ran a similar study with 200 people, 102 men and 98 women, published in Psychological Science. Power poses had no effect on testosterone or cortisol, and no effect on willingness to take financial risks. Participants who held the expansive poses did report feeling more powerful. The authors' reading was that power posing mainly changes how confident people say they feel, without evidence that this carries into behaviour or physiology.

2015 to 2017: checking the published record. Joe Simmons and Uri Simonsohn of the University of Pennsylvania used a method called p-curve, which looks at the pattern of results across many studies to judge whether they contain real evidence or mostly reflect selective reporting. In a 2015 analysis on their blog, Data Colada, they applied it to 33 published studies cited in support of power posing. The curve was flat, which is what you would expect if the true effect were zero or too small for those studies to detect. They developed the analysis into a 2017 paper in Psychological Science.

2016: Dana Carney's statement. In a document on her faculty website, Carney, the study's first author, explained that the evidence since 2015 had changed her view. She wrote: "I do not believe that 'power pose' effects are real." She described the original sample as tiny and the effects as flimsy, and listed choices made during the analysis that she now sees as stretching the evidence, including adding participants in batches while checking results, picking the more favourable of two statistical tests, and selecting the self-report questions that worked. She also pointed to confounds, such as experimenters knowing the hypothesis.

It is rare and admirable for a scientist to write something like this about her own best-known paper. It is a large part of why this story is now taught as an example of how science corrects itself.

2017 to 2019: the debate about feelings. A special set of six preregistered studies, analysed together by Q. F. Gronau and colleagues in Comprehensive Results in Social Psychology in 2017, found very strong evidence that expansive poses increase how powerful people feel. When the analysis was limited to people who had not heard of power posing, the evidence was only moderate.

In 2018 Amy Cuddy, with S. J. Schultz and Nathan Fosse, published a reply in Psychological Science. Applying p-curve to a wider set of 55 studies of postural feedback, they reported clear evidential value, chiefly for the effect of expansive postures on feelings of power. In a 2019 commentary in Meta-Psychology, Marcus Credé argued that most of those studies compared expansive poses with contracted ones, not with a normal stance, and that the few which did use a neutral comparison suggest the difference comes mainly from slumped poses making people feel worse, rather than power poses making them feel better.

2022: the largest summary. R. Körner and colleagues pooled about 130 experiments with nearly 10,000 participants in Psychological Bulletin. Upright and dominant postures were linked to a more positive view of oneself. Effects on behaviour were weaker and possibly due to study design. There was no evidence that poses raise testosterone or other hormones. They also noted that most studies lacked a neutral control group and came from Western, industrialised countries.

So the verdict, in plain words: the hormone and risk-taking claims did not replicate. A small effect on how confident people say they feel is supported by some analyses and questioned by others.

How it gets misused

• "Two minutes in a superhero pose raises your testosterone." The best evidence says it does not.
• "Power posing will make you perform better." There is no reliable evidence for that.
• "The researchers were frauds." Nothing in the record suggests the data were invented. Carney states the data are real. The problems were a small sample and flexible analysis choices that, as Carney put it, did not seem like p-hacking back then. Mocking the people involved misses the more useful lesson about how evidence works.

What it means for you

Posture is still worth attention, just for gentler reasons.

How you stand affects how you breathe, and breath affects how your voice sounds. If you are hunched over a phone before a pitch, a client call or a stage talk, your chest is tight and your voice may come out thin. Standing or sitting tall, shoulders loose, feet steady, gives your lungs room and your voice support. Some people also simply feel more settled that way, which fits the modest self-report findings.

Think of a nurse in Doha before a difficult conversation with a family, or a team lead in Pune walking into a town hall. A quiet routine helps more than a dramatic stance:

"Feet flat. Shoulders down. One slow breath in, a longer breath out. First line ready."

"I don't need to look powerful. I need to be clear and present."

If you are speaking in a second language, this matters even more. A steady breath slows your pace, and a slower pace makes you easier to understand, whatever your accent.

Sources

1. D. R. Carney, A. J. Cuddy and A. J. Yap, Power Posing: Brief Nonverbal Displays Affect Neuroendocrine Levels and Risk Tolerance, Psychological Science, 21(10), 1363 to 1368, 2010. doi.org/10.1177/0956797610383437
2. E. Ranehill, A. Dreber, M. Johannesson, S. Leiberg, S. Sul and R. A. Weber, Assessing the Robustness of Power Posing: No Effect on Hormones and Risk Tolerance in a Large Sample of Men and Women, Psychological Science, 26(5), 653 to 656, 2015. doi.org/10.1177/0956797614553946
3. University of Zurich news release on the Ranehill study, 1 April 2015. www.news.uzh.ch/en/articles/2015/posen-der-macht-weniger-maechtig-als-gedacht.html
4. Dana R. Carney, My position on "Power Poses", statement on her faculty website, University of California, Berkeley, 2016. faculty.haas.berkeley.edu/dana_carney/pdf_My%20position%20on%20power%20poses.pdf
5. J. P. Simmons and U. Simonsohn, Power Posing: P-Curving the Evidence, Psychological Science, 28(5), 687 to 693, 2017. doi.org/10.1177/0956797616658563
6. Joe Simmons and Uri Simonsohn, Data Colada post 37 on power posing, 8 May 2015. datacolada.org/37
7. Q. F. Gronau, S. J. van Erp, D. W. Heck, J. Cesario, K. J. Jonas and E.-J. Wagenmakers, A Bayesian Model-Averaged Meta-Analysis of the Power Pose Effect with Informed and Default Priors: The Case of Felt Power, Comprehensive Results in Social Psychology, 2(1), 123 to 138, 2017. research.tilburguniversity.edu/en/publications/a-bayesian-model-averaged-meta-analysis-of-the-power-pose-effect-/
8. A. J. C. Cuddy, S. J. Schultz and N. E. Fosse, P-Curving a More Comprehensive Body of Research on Postural Feedback Reveals Clear Evidential Value for Power-Posing Effects: Reply to Simmons and Simonsohn (2017), Psychological Science, 29(4), 656 to 666, 2018. doi.org/10.1177/0956797617746749
9. Amy Cuddy, official website, list of 2018 journal articles (summary of the reply). www.amycuddy.com/articles/journal-articles/category/2018
10. Marcus Credé, A Negative Effect of a Contractive Pose is not Evidence for the Positive Effect of an Expansive Pose: Comment on Cuddy, Schultz, and Fosse (2018), Meta-Psychology, 3, 2019. open.lnu.se/index.php/metapsychology/article/view/1723
11. Martin Luther University Halle-Wittenberg and University of Bamberg press release on R. Körner, L. Röseler, A. Schütz and B. J. Bushman, Dominance and Prestige: Meta-Analytic Review of Experimentally Induced Body Position Effects, Psychological Bulletin, 2022. www.uni-bamberg.de/presse/pm/artikel/pm-powerpose-psychologie/

Published by The Full Stop, the journal of Brightside English, in the The Experiments series. Every factual claim links to a source below; where accounts or findings are disputed, we say so. Sources checked on 26 Oct 2026. Reviewed by a Brightside senior coach. Corrections: see our editorial standard.