Get Excited: Why Saying “I Am Excited” Beats “Calm Down” Before a Speech, and What the Replication Found

By Noam Shemla · The Studies

TL;DR Almost everyone’s advice for nerves before a speech is to calm down. Alison Wood Brooks tested the opposite: calling the nervous feeling excitement. Before singing karaoke, giving a speech on camera or doing maths under pressure, people who said or read “I am excited” or “get excited” felt more excited and did better than those told to stay calm. They did not feel less anxious. A direct replication of the speech study in 2025 found the same feelings but not the better speeches.

Paper: Brooks, A. W. (2014). Get excited: Reappraising pre-performance anxiety as excitement. Journal of Experimental Psychology: General, 143(3), 1144–1158.
Studies: A survey of 300 adults, then four experiments with university students: karaoke singing (113), a speech on camera (140), and a timed maths test (188, then 218).
Manipulation: A few words: saying “I am excited” or “I am calm” out loud, or reading “Try to get excited” or “Try to remain calm” on screen.
Measures: Performance scored objectively or by blind judges, self-reported emotions, heart rate, and whether people saw the task as a threat or an opportunity.

Before a talk, the standard advice is to relax. Alison Wood Brooks, at Harvard Business School, asked whether that is the best advice, and tested an alternative with a physiological logic: anxiety and excitement both come with a racing heart. Turning a racing heart into calm means changing the body; calling it excitement only means changing the label.1

What people believe

Brooks first asked 300 American adults, most of them in full-time work, to imagine they, or a coworker, had to give a keynote speech to the whole company tomorrow and felt extremely anxious. What was the best advice?

Speaking on camera

The public speaking experiment put 140 students in a situation designed to make them nervous: two minutes to prepare a persuasive speech on why they would be a good work partner, delivered to a camera for “a committee of peers” to judge. Just before speaking, each said one sentence out loud, assigned at random: “I am excited” or “I am calm”. Three judges who did not know the conditions rated the videos.

The most striking difference was in how long people spoke. They had to speak for between two and three minutes, and the excited group used much more of that time.

Singing and maths

The other experiments tested the same idea on different kinds of pressure. In the first, students sang a karaoke song in front of an experimenter and were paid by the singing-accuracy score the game computed. Before singing, they said “I am excited”, “I am anxious”, or nothing.

In the maths experiments, a message on screen before a difficult timed test was enough. Students who read “Try to get excited” answered more questions correctly than those who read “Try to remain calm” or “Please wait a few moments” (3.60 against 2.94 out of 8), and a later experiment replicated it (3.42 against 2.80). Heart rate rose when students learned about the test and stayed up throughout, whatever message they had read. Telling people to calm down did not calm them down.

The final experiment looked for the mechanism. Students told to get excited were more likely to describe the test as an opportunity rather than a threat, and that difference in outlook accounted for a significant part of their better scores.

Why excitement, not calm

Brooks’s explanation is what she calls arousal congruency. Anxiety and excitement are both high-arousal states: the heart races in both. Calm is low-arousal. Moving from anxiety to calm means bringing the body down, which is hard and, as the heart-rate data showed, usually does not happen. Moving from anxiety to excitement only means reinterpreting the same arousal, from threat to opportunity. The feeling stays; its meaning changes.

Has it held up?

Not fully, and the part that has not held is the one speakers care about most. In 2025 Poynter and Pasqualini repeated the public speaking experiment directly. They found nearly identical results for how people felt: saying “I am excited” again made speakers feel more excited without making them feel less anxious. But judges rated the speeches no differently in either condition, and an indirect measure of anxiety, the emotional Stroop task, showed no difference either.2

So the evidence now supports the feeling more firmly than the performance. Saying “I am excited” changes how the nerves feel; whether it reliably changes how the speech goes is, after the replication, an open question.

What this study cannot tell you

Whether the speech benefit is real. It rests on one experiment of 140 students, and a direct replication did not find it.
How long any effect lasts. Every test followed the self-statement within minutes.
Whether it helps people with severe anxiety. Participants were students in a laboratory, not people with social anxiety disorder.
Some of the statistics are loose. In the maths study the reported indirect effect (.34) lies outside its own reported confidence interval [.13, .28], and the final study calls a mediation “full” although the direct effect stayed significant.
Excitement or not-calm? The speech study compared excitement with calm, not with saying nothing, so it cannot separate helping from not harming.

What a speaker can take from it

The one result every study agreed on, including the replication, is about feeling. Before a talk your heart will race, and telling yourself to calm down will probably not slow it. Saying “I am excited” does not slow it either, but it changes what the racing heart means to you, from a threat to an opportunity, and people who did it felt more excited about what they were about to do.

That is worth having even if it does not, by itself, make the speech better. The speech gets better with preparation and practice; the reframe makes the minutes before it easier to live in.

This article summarises a published paper, read from the author’s posted copy of the journal article. Every percentage, mean, effect size and sample size is the author’s own, from the survey and Studies 1 to 4 and Figures 3 and 4. The replication is summarised from its published abstract. The worked d formula reproduces the paper’s own value from the means it reports.

Measured

Advice people chose for pre-speech nerves (survey)
Singing accuracy, judges’ ratings of speeches, speech length, maths scores and heart rate
Self-reported excitement, anxiety and confidence, and threat or opportunity mind-set

Inferred, not measured

That the word “excited” caused better performance: shown in the original experiments; not reproduced for speeches in the 2025 replication
That an opportunity mind-set explains the effect: a mediation analysis in one experiment

References

Brooks, A. W. (2014). Get excited: reappraising pre-performance anxiety as excitement. Journal of Experimental Psychology: General, 143(3), 1144–1158. https://doi.org/10.1037/a0035325
Poynter, M., & Pasqualini, M. S. (2025). Getting excited about public speaking: a replication. Psychology of Language and Communication, 254–265. https://doi.org/10.58734/plc-2025-0011