Language:English|Svenska

Voice and body language: what the research actually says

Written by Ted Scheiman, banking and fintech executive, guest lecturer. Last reviewed 16 November 2026. How we research and fact-check

Advice on body language in presentations is full of confident numbers. "Only 7% of communication is words." "Stand like a superhero for two minutes and your hormones change." Some of this advice comes from real studies. Much of it stretches those studies far beyond what they tested.

This article checks the most common claims against the original research. You will see which findings hold up, which are contested, and what that means for how you use your voice and body in a guest lecture, panel or meetup talk.

Key takeaways

  • The "7-38-55 rule" does not say that words carry 7% of every message. Mehrabian himself says his equations apply only when people communicate feelings or attitudes [1].
  • The original power-posing study reported hormone changes in 42 people [2]. A larger replication with 200 people found no hormone or risk-taking effects, only a small effect on feeling powerful [3]. The debate continues [4].
  • Gestures give a moderate boost to understanding, especially when they show actions or add information that the words do not [6].
  • Hearing a person's voice made evaluators judge them as more intelligent than reading the same words [7].
  • Audiences form impressions fast, but nervous speakers usually overestimate how visible their nerves are [5][8].

Myth 1: "93% of communication is nonverbal"

What people claim

You have probably seen the figure: 7% words, 38% tone of voice, 55% body language. It is often presented as a law of all communication.

What the source says

The numbers come from the psychologist Albert Mehrabian. On his own website, he presents the split as a finding about inconsistent messages, expressed as an equation: "Total Liking = 7% Verbal Liking + 38% Vocal Liking + 55% Facial Liking" [1].

He adds a clear warning. The equations "were derived from experiments dealing with communications of feelings and attitudes (i.e., like-dislike)". And: "Unless a communicator is talking about their feelings or attitudes, these equations are not applicable" [1].

What this means for speakers

When you explain a payments architecture, a court ruling or a marketing result, your words carry the content. The 7-38-55 split says nothing about that. What it does suggest is narrower: when your words and your tone or face send different signals about how you feel, listeners may lean on the nonverbal signals when judging your feelings or attitude.

Practical version: if you say "I'm excited about this" in a flat voice while looking at your notes, do not expect the words to win.

Myth 2: "Power poses change your hormones"

The original study

In 2010, Carney, Cuddy and Yap reported that 42 participants who held two expansive "high-power" poses for one minute each showed higher testosterone, lower cortisol, more risk taking and stronger feelings of power than people who held contractive poses [2]. The authors concluded that embodiment extends "to physiology and subsequent behavioral choices" [2].

The replication

Ranehill and colleagues repeated the core design with 200 participants. They "found no significant effect of power posing on hormonal levels or in any of the three behavioral tasks". They did find a significant effect on self-reported feelings of power, with an effect size of d = 0.34 [3].

The ongoing debate

Cuddy and colleagues later argued, using a statistical method called p-curve, that research on postural feedback has evidential value. In a published commentary, Credé argued that most of those studies did not compare expansive poses with a neutral pose. In the few studies that did, effects on feelings of power and emotion seemed to come mainly from the harmful effect of contractive, hunched poses, not from a benefit of expansive ones [4].

What this means for speakers

  • Do not rely on posing to change your body chemistry before a talk. That claim did not hold up in the larger study [3].
  • Standing in an open, comfortable posture is harmless and may help you feel slightly more confident. Hunching over a laptop may make you feel worse [4].
  • For evidence-based ways to manage nerves, see how to calm nerves before a talk.

What gestures actually do

Hostetter's 2011 meta-analysis in Psychological Bulletin combined 63 samples that compared speech alone with speech plus gesture. Gestures gave a "significant, moderate benefit to communication" [6]. The benefit depended on what the gestures did:

  • Topic. Gestures that depict motor actions communicated better than gestures about abstract topics.
  • Redundancy. Gestures helped more when they added information not fully expressed in speech.
  • Age. Children benefited more than adults.

What this means for speakers

Use gestures to show things: sizes, sequences, directions, how a process flows. For example, a hypothetical logistics manager might trace the route of a shipment in the air while describing where delays happen. Decorative hand movements repeated on every sentence add little. Pointing at what you mean, or showing a shape, adds most.

Why your voice matters

Schroeder and Epley asked MBA students to record short job pitches. Volunteers and professional recruiters then heard, watched or read the pitches. Across five experiments, evaluators judged candidates as smarter and more favourable when they heard them than when they read identical words. Recruiters were more interested in hiring candidates they had heard. Adding video to audio made no difference to recruiters' evaluations [7].

The authors suggest that changes in tone, cadence and pitch reveal a speaker's thinking as it happens. APS's summary notes that natural variation in pitch and cadence conveys enthusiasm and intellect better than the monotone people often use when reading aloud [7].

What this means for speakers

  • Speak your ideas rather than reading a script. Reading tends to flatten your voice.
  • Let your pitch and pace follow your meaning. Slow down for the key point; vary your tone when you contrast two ideas.
  • Your voice carries much of your credibility in online talks and podcasts too, where your body is less visible.

First impressions: fast, but not the whole story

Ambady and Rosenthal showed students silent video clips of 13 college teachers, each 30 seconds in total. Ratings of the teachers' nonverbal behaviour correlated strongly with end-of-semester student evaluations (r = .76). Even 2- and 5-second clips predicted evaluations, and accuracy did not differ significantly between 10-, 5- and 2-second clips [5].

This "thin slices" study is famous, and often overstated. The samples were small, 13 teachers per study. The criterion was ratings, not how much students learned, and the authors themselves called for studies of learning and of other channels such as voice [5]. It shows that people form impressions fast and consistently. It does not show that those impressions are accurate about competence.

And your nerves are less visible than you think

In a study of students giving speeches to real peer audiences, speakers believed their nervousness was more visible than the audience rated it. In the authors' words, "Public speakers believe their nervousness is more apparent to others than is actually the case" [8]. The effect was stronger for more anxious speakers, and about a third of participants did not show it [8].

A summary of the evidence

Claim Status What to do
7% words, 38% voice, 55% body Misused. Applies only to communicating feelings or attitudes [1] Make sure tone and face match what you say about how you feel
Power poses change hormones Not replicated in a larger study [3] Stand comfortably; do not rely on hormones
Power poses make you feel more powerful Small effect; contested whether expansive poses help or hunched poses hurt [3][4] Avoid hunching; open posture is fine
Gestures aid understanding Moderate benefit in a meta-analysis [6] Gesture to show actions, shapes and sequences
Voice signals intelligence Supported across five experiments [7] Speak, do not read; vary pitch and pace
Impressions form in seconds Supported, small samples, ratings not learning [5] Start with energy and clarity

Frequently asked questions

Is the 7-38-55 rule true?

Only in a narrow sense. Mehrabian derived it from experiments on communicating feelings and attitudes, and he says the equations do not apply when a person is not talking about their feelings or attitudes. It does not mean that words carry only 7% of a presentation.

Does power posing work?

The original claim that power poses change testosterone and cortisol did not hold up in a larger replication. People did report feeling slightly more powerful, but researchers still debate whether that comes from expansive poses helping or hunched poses hurting. Use an open posture if it feels good, but do not expect a hormonal effect.

Should I use hand gestures when presenting?

Yes, when they show something. A meta-analysis found gestures help understanding moderately, most when they depict actions or add information the words do not. Constant gesturing that repeats your words adds little.

How can I improve my voice for presentations?

Speak your ideas instead of reading them, because reading tends to flatten intonation. Vary pitch and pace to match your meaning, and slow down for key points. Research found that hearing a voice made people seem more intelligent than the same words in writing.

Can the audience see how nervous I am?

Usually less than you think. Studies of student speakers found that people overestimate how visible their nervousness is to the audience. If nerves are persistent or severe, talking to a doctor or psychologist can help.

Practise where it counts

Body language and voice improve with real audiences, not mirrors. Our free search finds guest lectures, student societies, associations and meetups where you can practise in front of people who care about your topic. For more on reading and reaching a room, see emotions on stage. Start a free search

Sources

  1. Mehrabian, A. "'Silent Messages': A Wealth of Information About Nonverbal Communication (Body Language)". Author's website. https://kaaj.com/psych/smorder.html (accessed 16 November 2026)
  2. Carney, D. R., Cuddy, A. J. C., & Yap, A. J. "Power posing: Brief nonverbal displays affect neuroendocrine levels and risk tolerance". Psychological Science, 21(10), 1363–1368, 2010. DOI: 10.1177/0956797610383437. Journal abstract page (Association for Psychological Science): https://www.psychologicalscience.org/journals/psychological-science/0956797610383437/; full text hosted by Boston University: https://www.bu.edu/wgs/files/2014/12/Carney-et-al.-2010-PowerPosingAffectsNeuroendocrineHormones.pdf (accessed 16 November 2026)
  3. Ranehill, E., Dreber, A., Johannesson, M., Leiberg, S., Sul, S., & Weber, R. A. "Assessing the robustness of power posing: No effect on hormones and risk tolerance in a large sample of men and women". Psychological Science, 2015. DOI: 10.1177/0956797614553946. PDF hosted by University of Zurich: https://www.zne.uzh.ch/dam/jcr:e5fc4c3c-d50e-4aa9-96ab-4389e426d20d/Psychological%20Science-2015-Ranehill-0956797614553946.pdf (accessed 16 November 2026)
  4. Credé, M. "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. DOI: 10.15626/MP.2019.1723. https://open.lnu.se/index.php/metapsychology/article/view/1723 (accessed 16 November 2026)
  5. Ambady, N., & Rosenthal, R. "Half a minute: Predicting teacher evaluations from thin slices of nonverbal behavior and physical attractiveness". Journal of Personality and Social Psychology, 64(3), 431–441, 1993. PDF copy: https://www.math.stonybrook.edu/~bishop/classes/math598.F25/Half_a_Minute.pdf (accessed 16 November 2026)
  6. Hostetter, A. B. "When do gestures communicate? A meta-analysis". Psychological Bulletin, 137(2), 297–315, 2011. DOI: 10.1037/a0022128. Abstract record: https://bvbvf1.bib-bvb.de/vufind/PrimoRecord/cdi_proquest_miscellaneous_906330129 (accessed 16 November 2026)
  7. Schroeder, J., & Epley, N. "The sound of intellect: Speech reveals a thoughtful mind, increasing a job candidate's appeal". Psychological Science, 2015. DOI: 10.1177/0956797615572906. Journal abstract page: https://www.psychologicalscience.org/journals/psychological-science/0956797615572906/; Association for Psychological Science summary, "The Sound of Employability: Interviewers Judge Your Voice", 10 April 2015: https://www.psychologicalscience.org/?p=111335 (accessed 16 November 2026)
  8. MacInnis, C. C., Mackinnon, S. P., & MacIntyre, P. D. "The illusion of transparency and normative beliefs about anxiety during public speaking". Current Research in Social Psychology, 15(4), 2010. https://crisp.org.uiowa.edu/sites/crisp.org.uiowa.edu/files/2020-04/15.4.pdf (accessed 16 November 2026)

Find out where you could speak this term

Join the early-access list and be among the first to see your matches. Free, no card needed.

Get early access