The Yerkes–Dodson curve: what a 1908 mouse study really showed, and a better idea for pre-performance nerves

Almost every book on stress, exams or stage fright shows the same picture: an upside-down U. Too little arousal and you are flat, too much and you fall apart; somewhere in the middle is your peak. It is usually credited to the "Yerkes–Dodson law" of 1908. The original study, though, was about mice, electric shocks and learning to tell black from white — and the tidy curve came later. What research has found since is messier, more modest, and more useful for anyone about to walk on stage.

The original study

Robert Yerkes and John Dodson worked with dancing mice. Each mouse had to choose between two boxes, one white and one black, and always pick the white one; entering the black one gave it an electric shock. The question was simple: does the strength of the shock change how quickly a mouse learns, and which strength works best?

They ran three series that differed in how hard the discrimination was — how different the two boxes looked in brightness. When the task was easy, the mice learned faster the stronger the shock (they stopped before reaching strengths that would be harmful). When the task was hard, learning first got faster as the shock increased and then got slower again: both weak and strong shocks led to slow learning. Their conclusion was that the harder the task, the weaker the most helpful stimulus.

The paper is open about its limits. The authors note that in one series they trained four mice per shock strength and that the curve would probably have been smoother with ten; in another, only three strengths were tested, so the best one may have been missed. There is no mention of anxiety, stress, arousal or human performance anywhere in the paper.

What happened next

From mice to a "law" of human stress. A 1994 review traced how the finding was retold over the century (Teigen, 1994). Later researchers and textbook authors turned "strength of stimulus" into punishment, reward, motivation, drive, arousal, anxiety, tension or stress, and "habit-formation" into performance, problem-solving or memory. Task difficulty — the heart of the original result — was sometimes dropped altogether. These changes, the review notes, are rarely discussed and are often attributed to Yerkes and Dodson themselves.

From law to folklore. A 2015 analysis went further, arguing that the evidence for the inverted-U as a law of human performance is "found wanting", that its simple graph helped it spread, and that it still informs management practices that aim to keep workplace stress up in the name of performance (Corbett, 2015). That is a critical essay, not a new experiment, but it fits the review: the picture travelled much further than the data.

A different question: not "how much", but "what does it mean?" From around 2010, researchers led by Jeremy Jamieson tested something else: instead of trying to lower arousal, change how people interpret it. In a study of students preparing for a graduate admissions exam, those told that arousal can improve performance scored higher on the maths section than controls — in the lab and, one to three months later, on their real exam (Jamieson et al., 2010). Interestingly, their sympathetic arousal went up, not down. A 2012 experiment found that people who reappraised their stress as useful had healthier cardiovascular responses and paid less attention to threatening information (Jamieson, Nock & Mendes, 2012). In 2014, Alison Wood Brooks reported that most people believe calming down is the best way to handle pre-performance anxiety, but that people who told themselves "I am excited" felt more excited and performed better at karaoke, public speaking and maths than those who tried to calm down (Brooks, 2014).

What the meta-analyses say. Pooled results are more modest than the headlines. A 2019 meta-analysis found that reappraisal reduced how stressed people felt (a standardised difference of about 0.43), but not their physiological stress responses (Liu et al., 2019). A 2024 meta-analysis of randomised trials on task performance found a small overall benefit (d = 0.23), a little larger for public-performance tasks (d = 0.34), and concluded that these interventions are "not silver bullets" (Bosshard & Gomez, 2024). And a carefully pre-registered study in competitive gaming (300 players, with Jamieson among the authors) found no benefit at all — partly because the players were not very stressed in the first place (Behnke et al., 2024).

A cautionary neighbour: "power poses". Performance advice also absorbed a more dramatic claim: that holding an expansive "power pose" for a minute raises testosterone, lowers cortisol and makes you bolder (Carney, Cuddy & Yap, 2010). A large replication found no effect on hormones or risk-taking (Ranehill et al., 2015). A 2022 meta-analysis of 73 studies found that posture does affect mood and behaviour, but mainly because slumped, contracted postures make things worse; expansive versus neutral postures made almost no difference (g = 0.06), while contracted versus neutral did (g = 0.45) (Elkjær et al., 2022). The useful part survived; the hormonal miracle did not.

Where it stands today

Evidence: Moderate evidence — for the modern idea that re-reading arousal as a sign of readiness gives a small boost to how people feel and perform under pressure. The effect is small, varies a lot between studies, shows mainly in self-report rather than physiology, and may only matter when people are genuinely stressed. The Yerkes–Dodson curve itself is best read as a rough metaphor, not a law: the original data were few mice, and the human "inverted-U" was added later.

What this means for you

You do not need to hunt for a magic "optimal level" of nerves. Two simpler moves have better support:

• Change the label, not only the level. A racing heart before a concert or exam is your body getting ready. Read What anxiety actually is and practise answering the thought "I'm too nervous" with "my body is preparing": Thoughts are not facts.
• Turn it down only if it is too much to work with. If you are shaking so hard you cannot play or speak, a couple of minutes of paced breathing can take the edge off; then go back to "ready", not "calm".
• Have a short routine for the last minutes, so you are not improvising under pressure: STOP: a pause between the trigger and the reaction and your three-skill Pressure Kit.
• Skip the power pose if it feels silly — but avoid slumping and shrinking just before you go on.

If nerves before performing are so strong that you avoid exams, auditions or competitions, or they affect your sleep and mood for weeks, a licensed mental health professional or a performance psychologist can help.

What this does NOT mean

• It does not mean stress is good for you and should be kept high: the "law" has been misused exactly that way at work.
• It does not mean saying "I'm excited" will transform your result; average effects are small.
• It does not mean calming techniques are useless — they help when arousal is overwhelming; they are just not the only option.
• It does not mean every famous psychology finding is wrong — but a single striking study, especially one with a dramatic body-chemistry story, is a reason to wait for replications.

Sources

• Yerkes RM, Dodson JD. The relation of strength of stimulus to rapidity of habit-formation. J Comp Neurol Psychol, 1908;18:459–482. DOI: 10.1002/cne.920180503 · Open text: Classics in the History of Psychology
• Teigen KH. Yerkes–Dodson: a law for all seasons. Theory & Psychology, 1994. DOI: 10.1177/0959354394044004
• Corbett M. From law to folklore: work stress and the Yerkes–Dodson law. J Managerial Psychol, 2015. DOI: 10.1108/JMP-03-2013-0085
• Jamieson JP, Mendes WB, Blackstock E, Schmader T. Turning the knots in your stomach into bows: reappraising arousal improves performance on the GRE. J Exp Soc Psychol, 2010. DOI: 10.1016/j.jesp.2009.08.015
• Jamieson JP, Nock MK, Mendes WB. Mind over matter: reappraising arousal improves cardiovascular and cognitive responses to stress. J Exp Psychol Gen, 2012. DOI: 10.1037/a0025719
• Brooks AW. Get excited: reappraising pre-performance anxiety as excitement. J Exp Psychol Gen, 2014. DOI: 10.1037/a0035325
• Liu JJW, Ein N, Gervasio J, Vickers K. The efficacy of stress reappraisal interventions on stress responsivity: a meta-analysis and systematic review. PLoS One, 2019. DOI: 10.1371/journal.pone.0212854
• Bosshard M, Gomez P. Effectiveness of stress arousal reappraisal and stress-is-enhancing mindset interventions on task performance outcomes: a meta-analysis of randomized controlled trials. Sci Rep, 2024. DOI: 10.1038/s41598-024-58408-w
• Behnke M, Lakens D, Petrova K, et al. Applying a synergistic mindsets intervention to an esports context. R Soc Open Sci, 2024. DOI: 10.1098/rsos.240691
• Carney DR, Cuddy AJC, Yap AJ. Power posing: brief nonverbal displays affect neuroendocrine levels and risk tolerance. Psychol Sci, 2010. DOI: 10.1177/0956797610383437
• Ranehill E, Dreber A, Johannesson M, et al. Assessing the robustness of power posing: no effect on hormones and risk tolerance in a large sample of men and women. Psychol Sci, 2015. DOI: 10.1177/0956797614553946
• Elkjær E, Mikkelsen MB, Michalak J, et al. Expansive and contractive postures and movement: a systematic review and meta-analysis. Perspect Psychol Sci, 2022. DOI: 10.1177/1745691620919358