The ‘optimal stress’ curve began with mice learning to dodge shocks

The slide turns up in leadership workshops, sales trainings and test-prep seminars. A smooth hill rises and falls, with stress along the bottom axis and performance up the side, and the lesson seems obvious: too little pressure and you coast, too much and you crack, so aim for the top of the hump.

The hill has a name, the Yerkes-Dodson law, and it has been cited for well over a century. Its origin gets less attention. In 1908, Yerkes and Dodson measured how quickly mice learned to pick the right passage in a box when choosing the wrong one brought an electric shock. They were not studying stress, people or job performance.

Research on human stress tells a plainer story than the hill. A stressful moment takes a small but real bite out of working memory and mental flexibility, makes recall harder and nudges you toward habit. Stress also costs sleep, and lost sleep makes the next stressful day land harder. Here is what the evidence supports, and where the famous curve misleads.

Where the stress curve really came from

Yerkes and Dodson varied the strength of the shock and how hard the two choices were to tell apart. A 2007 history in the journal Neural Plasticity, by David M. Diamond and colleagues at the VA hospital in Tampa, Florida, describes how the results split by difficulty. When the choice was easy, black versus white, the mice learned faster as the shock grew stronger, with no downturn. When the choice was hard, black versus gray, a middling shock produced the fastest learning.

So the original finding had two shapes: a rising line for the easy task and a hump only for the hard one. And it measured how fast animals formed a habit, not how well anyone performed. Karl Halvor Teigen, a University of Oslo psychologist, noted in a 2017 retrospective that the groups of mice were very small and that the pattern seems to have emerged as the work went along rather than from a plan.

The single hump arrived decades later. Diamond’s team traced it to the 1950s, when Schlosberg, Hebb and Duffy, three prominent psychologists, each described a curved link between arousal and performance without citing Yerkes and Dodson. Hebb’s version later picked up the Yerkes-Dodson name, and from the 1960s on, textbooks taught a single upside-down U. Teigen adds that writers kept swapping the ingredients: shock strength became motivation, arousal, stress or anxiety, and learning speed became performance, coping or problem-solving. Diamond’s group called the law “a well-cited, but misunderstood, century-old principle…”

Correcting the record without overcorrecting

Not everyone who corrects the history throws the idea out. Diamond’s team argued that the original two-part pattern, in which strong pressure speeds simple learning and hampers difficult learning, held up in later animal work, starting with rat experiments by Broadhurst about five decades after 1908. There is even an upside-down U in brain chemistry: a 2009 review by Amy F. T. Arnsten of Yale describes how too little or too much of two signaling chemicals impairs working memory, drawing mainly on studies in monkeys and rats. That describes brain chemistry, not a dial anyone can set at work.

Others are blunter about the workplace version. In a 2015 review in the Journal of Managerial Psychology, Martin Corbett traced how a century-old animal experiment came to be treated as a law of human behavior. He concluded that the evidence doesn’t support it and argued that it still props up management habits of keeping pressure high to squeeze out performance. His advice to practitioners was not to chase performance by manipulating employees’ stress.

By the library’s evidence grades, the idea that a moderate dose of stress brings out your best, so you should dial it to the middle, rates as insufficient evidence, and not for lack of studies. As usually stated, it names no task, no amount of stress and no way to measure either, so no result could prove it wrong.

What a stressful moment does to working memory and flexibility

The most useful summary of the human evidence is a 2016 meta-analysis, a study that pools the results of many others, by Grant S. Shields, Matthew A. Sazma and Andrew P. Yonelinas at the University of California, Davis. They combined 51 studies covering 2,486 volunteers with no known psychiatric disorder. Each went through a brief lab stressor, ranging from upsetting videos and ice water to a public-speaking test, and was then tested on executive functions, the planning-and-control skills that the executive function piece takes apart.

The pattern was consistent but modest:

• Working memory, the mental scratchpad for whatever you’re actively juggling, was impaired across 34 studies and 1,353 people. The average effect was about a fifth of a standard deviation, which statisticians call small. Stressed and calm groups overlapped a great deal. The hit was larger when tasks demanded juggling a lot at once and when the stressor was more severe.
• Cognitive flexibility, switching rules or approaches, was impaired by a somewhat larger amount, about three-tenths of a standard deviation, but that estimate rests on just six studies and 280 people.
• Inhibition showed no overall effect, because it split in two. Stress improved people’s ability to hold back an action and worsened their ability to tune out distraction. The authors summed up the stressed state as more reactive to whatever grabs attention, yet with tighter control over actions.

The team also compared these results with studies that simply gave people doses of cortisol, the best-known stress hormone. Stress impaired executive functions overall; cortisol alone did not, and it had no effect at all on flexibility. Whatever stress does to thinking, it works through more than cortisol. The body’s stress response itself is covered elsewhere in the library.

One result ran against the authors’ expectations: working memory impairment was larger in studies with more men, though that link faded once study design was taken into account. Individuals vary widely around these averages.

Arnsten’s review adds a point from older research. Some of the first stress experiments followed the Second World War, after skilled pilots crashed in combat because of mental errors. Many of those studies found that stress impaired complex, flexible thinking while sparing or even improving simple, well-rehearsed tasks, and that people who felt in control, even falsely, were often not impaired.

Stress puts you on autopilot

Stress also changes how you choose. In a 2009 experiment at Ruhr University Bochum in Germany, Lars Schwabe and Oliver T. Wolf had some volunteers hold a hand in ice water while being watched and evaluated, then taught everyone two actions that each earned a different snack. Next, participants ate one snack until they were full of it. Unstressed people shifted away from the action tied to the food they’d had their fill of. Stressed people kept going, and they knew less about which action produced which snack. Schwabe and Wolf concluded that “stress promotes habits at the expense of goal-directed performance in humans.”

That is one laboratory study, but the everyday version is familiar: under pressure, you reach for the usual route, the usual reply, the usual fix, even when the situation has changed.

Why your mind goes blank at the worst moment

If you’ve ever known an answer the night before and lost it in the exam room, timing matters. A 2017 meta-analysis by the same UC Davis group, in Psychological Bulletin, pooled 113 studies with 6,216 participants and sorted them by when the stress hit:

• Before or during learning, stress usually impaired later memory. The exception was a stressor that came just before learning material directly related to it.
• After learning, stress usually improved memory, unless the stress happened in a different setting. How memories settle in after learning is a separate subject.
• Just before or during recall, stress impaired memory, more so for emotionally charged material.

Two details matter for real life. How much cortisol rose did not track how much memory changed. And memory effects were generally smaller in women taking hormonal contraceptives.

One of the first experiments to show stress disrupting recall itself used rats. In a 1998 study in Nature, D. J. de Quervain, B. Roozendaal and J. L. McGaugh at the University of California, Irvine, gave rats a foot shock before a memory test in a water maze. Shock 30 minutes before the test impaired the rats; shock 2 minutes or 4 hours before did not. The impairment tracked stress-hormone levels at test time. That timing comes from rats and doesn’t map neatly onto your 9 a.m. interview.

Practice that holds up under pressure

One experiment suggests how you learn may matter as much as how stressed you feel. In 2016, Amy M. Smith, Victoria A. Floerke and Ayanna K. Thomas at Tufts University had volunteers learn material either by rereading it or by testing themselves on it. A day later, half were put through a stressor. The rereaders showed the usual stress-related memory slump; the self-testers did not.

That is a single study. A small 2026 follow-up in Brazil, with 56 men, found stress didn’t impair recall of word pairs learned by repeated self-testing, but it had no rereading group to compare. Self-testing as a study method gets fuller treatment alongside other study habits.

Two famous exam-day fixes, and how they held up

In 2011, Gerardo Ramirez and Sian L. Beilock at the University of Chicago reported in Science that a short writing exercise, done right before a high-stakes test, raised scores. Students briefly wrote down their thoughts and worries about the exam ahead. The paper described two laboratory and two randomized classroom experiments, with the clearest gains among students who were habitually anxious about tests.

Then came a larger check. In 2018, a large research team, with Colin F. Camerer of Caltech as first author, published an effort to repeat 21 experimental social science studies that had appeared in Nature and Science between 2010 and 2015. Each repeat was preregistered, meaning its methods and analysis plan were filed publicly in advance, and the original authors reviewed the plans. Samples averaged about five times the original size. Thirteen of the 21 replicated, and across all 21 the repeat effects averaged about half the size of the originals.

The test-anxiety writing study was among the eight that did not. Under the project’s rules, only one experiment from each paper was repeated, the first that reported a significant effect. A failed repeat doesn’t erase the other experiments. It does mean the writing exercise belongs in the contested column, not the proven one.

A second idea rests on a single study. In 2010, in research led by Jeremy P. Jamieson at Harvard, people preparing for the GRE, the graduate school admissions test, were told before a practice exam that the keyed-up feeling of physical arousal improves performance. They showed a significant rise in salivary alpha-amylase, a marker of the body’s fight-or-flight activation, and outperformed a comparison group on the math section. One to three months later, their official GRE math scores were also higher. The gains were reported only on math, and one study can’t settle whether the idea works.

Stress and sleep keep each other going

Stress doesn’t hit everyone’s sleep equally. A 2018 review in the Journal of Sleep Research by David A. Kalmbach, Jason R. Anderson and Christopher L. Drake, centered at Henry Ford Hospital’s sleep center in Detroit, describes “sleep reactivity“: a lasting, trait-like tendency for stress to wreck a person’s sleep. Highly reactive sleepers sleep far worse when stressed, while low-reactivity sleepers get through stressful stretches largely undisturbed. Genetics, a family history of insomnia, being female and stressful surroundings all shape it.

Reactivity also predicts trouble ahead. In one study the review describes, 1,449 good sleepers were followed for two years. Those with high reactivity were nearly 60% more likely to develop insomnia symptoms and twice as likely to develop chronic insomnia, even after accounting for their sleep history, depressive symptoms and stress exposure. In another group of 2,892 people with no history of insomnia disorder or major depression, 15.7% of those above a risk cutoff developed insomnia disorder within a year, nearly double the 9.1% rate for the group as a whole. The review also describes rumination, going over worries again and again, as amplifying the risk in people with sensitive sleep.

The loop runs the other way too. A 2024 Psychological Bulletin meta-analysis, with Cara A. Palmer of Montana State University as first author, pooled 154 experimental studies with 5,717 healthy participants aged 7 to 79 whose sleep was cut short, interrupted or skipped entirely. Every form of sleep loss reduced positive mood. Sleep loss also raised anxiety symptoms by a moderate amount. Effects on negative mood and depressive symptoms were mixed.

Put the two together and the cycle is plain. A stressful week costs sleep, especially for reactive sleepers. Short sleep leaves you more anxious and less upbeat. The next stressful day starts from behind. What weeks of short nights do to alertness is a fatigue question, one the library’s fatigue coverage answers.

If sleep trouble hardens into chronic insomnia, two professional bodies agree on the starting point. The American College of Physicians in 2016 and the American Academy of Sleep Medicine in 2021 both gave a strong recommendation to cognitive behavioral therapy for insomnia, a structured program known as CBT-I, and the physicians’ group named it the initial treatment, rating the evidence behind that advice as moderate in quality. How it stacks up against sleeping pills belongs to the sleep and brain health article.

Getting ready for a high-stakes day: what holds up

Everything on this list comes from the research described here, with a note on how much weight it can carry.

1. Protect the nights before, not just the morning. Experiments consistently show that sleep loss raises anxiety and lowers positive mood. They measured mood, not test or interview scores, but walking in calmer is worth something on its own.
2. Rehearse by recalling, not rereading. One experiment found that self-tested material resisted stress-related forgetting. Preliminary, but it costs nothing extra.
3. Take load off your working memory. Stress hit working memory hardest when tasks required juggling a lot. Writing down figures, steps or names you’d otherwise hold in your head follows from that finding, though no study here tested it as a strategy.
4. Reframe the racing heart, if it helps you. One study found benefits on math scores. Low cost, preliminary.
5. Write out your worries just before, if you like, but don’t count on it. The original finding failed its largest repeat. Contested.
6. Don’t try to manufacture the right amount of stress. No evidence supports treating stress as a dial, for yourself or for a team.

Signs that stress needs a professional’s eye

The National Institute of Mental Health separates stress, a response to an outside cause, from anxiety, which can persist even when no threat is present. It says you may be at risk for an anxiety disorder if stress feels unmanageable and its symptoms interfere with everyday life, lead you to avoid things, or never seem to let up. NIMH lists sleep loss among the shared symptoms of stress and anxiety and suggests talking to a professional when symptoms won’t go away.

Concrete situations that call for that conversation:

• Worry that keeps you from work, school or seeing people, or that has you steering around situations you used to handle.
• Sleep trouble that outlasts the stressful stretch that started it. Whether it has become insomnia disorder, and which treatment fits, is for a clinician to judge.
• Memory or concentration problems that persist after the pressure is gone. The brain fog article walks through what a clinician checks.

Whether persistent stress has become an anxiety disorder or depression is for a clinician or mental health professional to sort out, not an online quiz. If you are in immediate distress or thinking about hurting yourself, call or text 988 to reach the 988 Suicide & Crisis Lifeline. In a life-threatening emergency, call 911.

Stress is not a performance dial

The curve on the slide was built from shocked mice learning a habit, then redrawn into a rule about people that the original experiment never tested. The better evidence says a stressful moment trims working memory, flexibility and recall a little, on average, and pushes you toward habit. Sleep is the part you can most protect in advance, so if a big day is coming, the nights before it deserve as much planning as the morning of.

Sources

1. Neural Plasticity. “Diamond DM, Campbell AM, Park CR, Halonen J, Zoladz PR. "The temporal dynamics model of emotional memory processing: a synthesis on the neurobiological basis of stress-induced amnesia, flashbulb and traumatic memories, and the Yerkes-Dodson law." Neural Plast, 2007. PubMed. DOI: 10.1155/2007/60803.” pubmed.ncbi.nlm.nih.gov
2. Tidsskrift for Norsk psykologforening. “Teigen KH. "En artikkel for alle årstider" (An article for all seasons), 2 May 2017.” psykologtidsskriftet.no
3. Journal of Managerial Psychology (Emerald). “Corbett M. "From law to folklore: work stress and the Yerkes-Dodson Law." Journal of Managerial Psychology, 2015;30(6):741-752. DOI: 10.1108/JMP-03-2013-0085.” doi.org
4. Nature Reviews Neuroscience. “Arnsten AFT. "Stress signalling pathways that impair prefrontal cortex structure and function." Nat Rev Neurosci, 2009. PubMed. DOI: 10.1038/nrn2648.” pubmed.ncbi.nlm.nih.gov
5. Neuroscience & Biobehavioral Reviews. “Shields GS, Sazma MA, Yonelinas AP. "The effects of acute stress on core executive functions: A meta-analysis and comparison with cortisol." Neurosci Biobehav Rev, 2016. PubMed. DOI: 10.1016/j.neubiorev.2016.06.038.” pubmed.ncbi.nlm.nih.gov
6. Psychological Bulletin. “Shields GS, Sazma MA, McCullough AM, Yonelinas AP. "The effects of acute stress on episodic memory: A meta-analysis and integrative review." Psychol Bull, 2017. PubMed. DOI: 10.1037/bul0000100.” pubmed.ncbi.nlm.nih.gov
7. Journal of Neuroscience. “Schwabe L, Wolf OT. "Stress prompts habit behavior in humans." J Neurosci, 2009. PubMed. DOI: 10.1523/JNEUROSCI.0979-09.2009.” pubmed.ncbi.nlm.nih.gov
8. Nature. “de Quervain DJ, Roozendaal B, McGaugh JL. "Stress and glucocorticoids impair retrieval of long-term spatial memory." Nature, 1998. PubMed. DOI: 10.1038/29542.” pubmed.ncbi.nlm.nih.gov
9. Science. “Smith AM, Floerke VA, Thomas AK. "Retrieval practice protects memory against acute stress." Science, 2016. PubMed. DOI: 10.1126/science.aah5067.” pubmed.ncbi.nlm.nih.gov
10. Frontiers in Cognition. “Klier C, Buratto LG. "Does retrieval practice protect memories against the amnesic effects of stress? An evaluation of the role of stimulus difficulty." Front Cognit, 2026. PubMed. DOI: 10.3389/fcogn.2026.1838176.” pubmed.ncbi.nlm.nih.gov
11. Science. “Ramirez G, Beilock SL. "Writing about testing worries boosts exam performance in the classroom." Science, 2011. PubMed. DOI: 10.1126/science.1199427.” pubmed.ncbi.nlm.nih.gov
12. Nature Human Behaviour. “Camerer CF, Dreber A, Holzmeister F, et al. "Evaluating the replicability of social science experiments in Nature and Science between 2010 and 2015." Nat Hum Behav, 2018. PubMed. DOI: 10.1038/s41562-018-0399-z.” nature.com
13. Journal of Experimental Social Psychology. “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. PubMed. DOI: 10.1016/j.jesp.2009.08.015.” pubmed.ncbi.nlm.nih.gov
14. Journal of Sleep Research. “Kalmbach DA, Anderson JR, Drake CL. "The impact of stress on sleep: Pathogenic sleep reactivity as a vulnerability to insomnia and circadian disorders." J Sleep Res, 2018. PubMed. DOI: 10.1111/jsr.12710.” pubmed.ncbi.nlm.nih.gov
15. Psychological Bulletin. “Palmer CA, Bower JL, Cho KW, et al. "Sleep loss and emotion: A systematic review and meta-analysis of over 50 years of experimental research." Psychol Bull, 2024. PubMed. DOI: 10.1037/bul0000410.” pubmed.ncbi.nlm.nih.gov
16. American College of Physicians (Annals of Internal Medicine). “Qaseem A, Kansagara D, Forciea MA, et al. "Management of Chronic Insomnia Disorder in Adults: A Clinical Practice Guideline From the American College of Physicians." Ann Intern Med, 2016. PubMed. DOI: 10.7326/M15-2175.” pubmed.ncbi.nlm.nih.gov
17. American Academy of Sleep Medicine (Journal of Clinical Sleep Medicine). “Edinger JD, Arnedt JT, Bertisch SM, et al. "Behavioral and psychological treatments for chronic insomnia disorder in adults: an American Academy of Sleep Medicine clinical practice guideline." J Clin Sleep Med, 2021. PubMed. DOI: 10.5664/jcsm.8986.” pubmed.ncbi.nlm.nih.gov
18. National Institute of Mental Health. “I'm So Stressed Out! Fact Sheet (NIH Publication No. 20-MH-8125).” nimh.nih.gov