The Yerkes-Dodson law: what pressure really does to your work

Most people know the feeling from both sides. With nothing at stake, a task drags: you check messages, reread the same line and drift. With far too much at stake, the same task falls apart in a different way: your thoughts race, you make careless slips and the simple steps you know well suddenly feel unfamiliar. Somewhere in between, the work goes well.

That middle ground is usually explained with the Yerkes-Dodson law, the idea that performance rises with arousal up to a point and then falls. It is one of the most reproduced graphs in psychology and management training. It is also one of the most misremembered. The original 1908 experiment was about mice, electric shocks and how quickly they learned to choose the right box. The tidy "inverted U" curve was added decades later by other researchers.

Short answer: the Yerkes-Dodson law says that performance improves with arousal or pressure up to a moderate level and then declines, and that the best level is lower for difficult tasks than for easy ones. The broad pattern has support, especially in brain research on attention and working memory. But "arousal" is not one dial, the classic curve is a simplification, and at least one review argues that it has no solid basis as a law of human work performance. The useful lesson is narrower: hard thinking tolerates less pressure than routine work, the kind of pressure matters, and how you interpret your own arousal can change how it affects you.

What is the Yerkes-Dodson law?

In the form taught today, the Yerkes-Dodson law describes the relationship between arousal and performance. Arousal here means general activation: how alert, energised, keyed up or stressed you are. The law makes two claims:

1. Too little and too much both hurt. Performance is weakest when arousal is very low or very high and best somewhere in the middle. Drawn as a graph, this is the familiar inverted U.
2. Task difficulty moves the peak. For simple or well-practised tasks, performance keeps improving at fairly high arousal. For complex or unfamiliar tasks, the best level is lower, and performance starts to fall sooner.

The second claim is the one people tend to forget, and it is the more useful one. It is also the claim closest to what Robert Yerkes and John Dodson actually reported.

Arousal level	How it often feels at work	Typical effect on hard thinking	Typical effect on routine work
Low	Bored, flat, easily pulled toward your phone	Slow starts, mind wandering	Slow, but errors are rare
Moderate	Alert, interested, a sense that the task matters	Usually the best range	Good
High	Racing, tense, watched or rushed	More slips, narrower attention, harder to hold several things in mind	Often still good, sometimes faster

Treat the table as a rough guide, not a measurement scale. The research below explains why the real picture is messier.

What the 1908 experiment actually tested

Yerkes and Dodson worked with mice in a discrimination task. Each mouse had to learn to choose one of two compartments, and a wrong choice was met with an electric shock of varying strength. The researchers also varied how hard the discrimination was. A 2007 review in Neural Plasticity summarises the results this way: when the choice was easy (black versus white), mice learned faster as the shock got stronger. When the choice was harder (black versus grey), learning was most efficient at an intermediate shock strength rather than the strongest one (Diamond and colleagues, 2007).

Three details from the original matter for anyone applying it to their own work:

• The study measured learning speed, not everyday performance. It counted how quickly animals formed a habit, not how well a person writes a report.
• The variable was shock strength, not "arousal". The word arousal, in the modern sense, was not part of the original paper.
• For easy tasks the relationship was close to a straight line. Stronger stimulation simply helped. The curve only appeared on the harder task.

The psychologist Karl Halvor Teigen traced how the law changed in a 1994 history. He showed that later researchers and textbook authors kept recasting it, as the effect of anything from reward and punishment to arousal, tension or stress, on outcomes ranging from learning to memory and coping. These changes were rarely discussed openly, and the new versions were often credited to Yerkes and Dodson themselves.

How a mouse study became a law about people

For several decades the 1908 result drew little attention. In the 1950s, interest in general arousal grew, and researchers such as Donald Hebb and Elizabeth Duffy argued that the relationship between arousal and performance takes the form of an inverted U. According to the 2007 review, they made this argument without citing Yerkes and Dodson, and Hebb's single curved graph later came to be known as the Yerkes-Dodson law. In the process, the linear part of the original finding, for easy tasks, largely disappeared from textbooks (Diamond and colleagues, 2007).

The first modern test came in 1957. Peter Broadhurst trained rats to escape from water through a discrimination task at different levels of difficulty, with different levels of motivation to escape (Broadhurst, 1957). As summarised in the 2007 review, rats on an easy discrimination learned fastest at the highest level of motivation, while an intermediate level worked best on a harder one. That is the same two-part pattern Yerkes and Dodson described, again in animals.

How strong is the evidence for humans at work?

This is where the confident version of the law runs into trouble.

In a 2015 paper in the Journal of Managerial Psychology, Martin Corbett examined how a "speculative animal behaviour modification experiment" turned into a law of human behaviour cited in management texts. He concluded that the empirical evidence for the law is wanting, that it has no basis in empirical fact as a general law, and that it continues to justify management practices that try to raise or maintain stress in order to boost performance. His practical recommendation was direct: managers should not try to increase performance by manipulating employee stress levels (Corbett, 2015). This is one author's critical review rather than a new data set, but it is a useful corrective to the poster version of the curve.

Part of the problem is that "arousal" and "stress" are not single quantities. A 2017 meta-analysis of 113 studies with 6,216 participants looked at how acute stress affects episodic memory. It did not find one curve. Stress before or during learning usually impaired memory, unless the stress and the material were closely linked in time and content. Stress just after learning usually improved memory. Stress just before recall impaired it. The size of the cortisol response did not predict the effect on memory (Shields and colleagues, 2017). In other words, timing and context changed the direction of the effect, which a single inverted U cannot capture.

So a fair summary is this: the general idea that hard tasks suffer under high pressure is reasonable, but there is no precise, universal curve that tells you how much pressure to apply to a given person or job.

What brain research adds

The strongest modern support for an inverted-U pattern comes from neuroscience on the prefrontal cortex, the part of the brain most involved in planning, holding information in mind and controlling attention.

In a 2009 review in Nature Reviews Neuroscience, Amy Arnsten described how noradrenaline and dopamine each have an inverted-U influence on working memory: too little or too much of either impairs prefrontal function. She also reported that even quite mild acute stress that a person cannot control can cause a rapid loss of prefrontal abilities. Tasks that depend on this region suffered under stress, while well-learned habits were spared or even improved. Early studies also found that people who felt in control of a stressor, even if that sense of control was an illusion, were often not impaired by it (Arnsten, 2009).

That neatly matches the task-difficulty half of the original law. Demanding, flexible thinking relies on circuits that are sensitive to too much stress. Routine, practised behaviour relies on circuits that are not.

A 2025 study in PNAS with 28 adults tracked spontaneous arousal through pupil size during a decision-making task. Performance typically peaked at moderate arousal. When the researchers raised arousal with a drug that acts on noradrenaline and dopamine, the whole arousal-performance curve shifted, which they interpreted as performance adapting to arousal changes within each overall state (Beerendonk and colleagues, 2025). This is a laboratory perceptual task with a small sample, and the drug condition is a research tool, not a suggestion. Its value here is that a modern, measured version of the curve exists, and that where the peak sits is not fixed.

Not all pressure is the same

Workplace research suggests that the type of demand matters as much as the amount. Organisational psychologists distinguish challenge stressors, such as high workload, time pressure and responsibility that feel like a chance to grow, from hindrance stressors, such as red tape, unclear roles and obstacles that feel like they block progress.

In a study of 696 learners, Jeffery LePine and colleagues found that stress linked to challenges in the learning environment was positively related to learning performance, while stress linked to hindrances was negatively related to it. Both kinds of stress were related to more exhaustion, and exhaustion was related to worse performance. Challenge stress went with more motivation to learn; hindrance stress went with less (LePine, LePine and Jackson, 2004). The study is correlational, so it shows associations rather than cause and effect.

The practical reading is that two days with the same "stress level" can affect your work in opposite ways. A tight deadline on work you care about is not the same as an afternoon lost to unclear requests. And even useful pressure has a cost in fatigue.

Reframing arousal: the part you can change

If the Yerkes-Dodson curve were the whole story, the only advice would be to calm down before hard tasks. A line of research led by Jeremy Jamieson and Alison Wood Brooks suggests that how you interpret arousal can matter too.

• Exam study. Students preparing for the GRE were either told that arousal can improve performance or given no such information. The reappraisal group showed a rise in a saliva marker of sympathetic nervous system activity and scored higher on the practice maths section. One to three months later, their actual GRE maths scores were also higher than those of the control group (Jamieson and colleagues, 2010).
• Speech study. Fifty adults gave a stressful speech in front of evaluators who frowned and crossed their arms. Those told to see their arousal as functional and adaptive showed more efficient cardiovascular responses and less attentional bias toward negative information than two control groups (Jamieson, Nock and Mendes, 2012).
• Excitement study. Across studies involving karaoke singing, public speaking and maths, people who reappraised nervous arousal as excitement, for example by saying "I am excited" out loud, felt more excited and performed better than those who tried to calm down (Brooks, 2014).

These are individual laboratory and field studies with modest samples, so treat them as promising rather than settled. They fit the brain research on perceived control, though: arousal that feels like readiness seems to cost less than arousal that feels like threat. For your working day, that suggests a two-part approach. Reduce pressure that adds nothing, and reinterpret the pressure that comes with work you care about.

When to raise the pressure and when to lower it

Situation	Likely position on the curve	What to try
Routine admin, clearing a backlog	Often too low	A short timer, a visible target, background company or a mild deadline. Our article on social facilitation explains why company helps familiar work.
Writing, analysis, learning something new	Easily pushed too high	Remove audiences and interruptions, split the work into steps, and protect a block of time, as in deep work.
A presentation, exam or difficult conversation	High, and some of it is unavoidable	Prepare thoroughly so more of the task becomes routine, then try reframing the arousal as readiness rather than forcing calm.
Late in a long day	Low energy, sometimes paired with high strain	Rather than adding pressure, take a real break. Our guides to the afternoon slump and brain breaks cover options.
Pressure from unclear goals or obstacles	Hindrance, not challenge	Fix the obstacle first: clarify the request, cut steps, ask for a decision. Effort spent pushing through a hindrance tends to drain energy without improving output.

Signs you may be under-aroused

• You keep starting other tasks or reaching for your phone, and mind wandering is frequent.
• The work feels pointless rather than hard.
• You find it easier to begin once a deadline is close, a pattern discussed in our guide to why you can't start tasks.

Signs you may be over-aroused

• You reread the same sentence or recheck the same number because nothing seems to stick.
• You make small slips in steps you normally do without thinking.
• You feel unable to keep several parts of a problem in mind at once, which is a sign of cognitive overload.

A two-week test

Because the curve differs between people and tasks, your own records will tell you more than a general rule.

1. Pick two task types. One routine (for example, email, data entry or editing) and one demanding (for example, drafting, problem-solving or learning a new tool).
2. Vary the pressure on purpose. On some days, add a little pressure to each type: a timer, a stated deadline or a colleague nearby. On other days, remove it: no timer, no audience, notifications off. Keep session length the same.
3. Rate arousal before you start. Score from 1 (flat) to 5 (racing). Add a note on whether the pressure feels like a challenge or a hindrance.
4. Measure output. Record what you finished, how long it took and the errors you found later. How productive you felt is a weaker guide than what you actually produced.
5. Compare by task type. Look for the range where each type of work went best. Many people find that routine work tolerates more pressure than they expected, and that their hardest work needs less.

Where Zone fits

Zone is a pair of EEG earbuds with an app that shows app brain states: In Flow, Locked In, Running Hot, Drifting and Depleted. The earbuds read signals through an in-ear EEG array, which our article on in-ear EEG explains. These app states are not a Yerkes-Dodson scale. Running Hot, for example, is the app's estimate that effort is outpacing mental energy, and Drifting is its estimate that engagement has dropped while energy holds. Neither is a measurement of arousal, and the app does not tell you where you sit on a performance curve.

What the app can add to the two-week test is a second record. You can place your session summaries from higher-pressure days beside those from lower-pressure days and see, for example, whether Running Hot appeared more often during demanding work under a deadline, or whether Drifting was more common during routine work with no pressure at all. Our guide to focus tracking explains how sessions work. Treat these as app estimates, and keep your output, timings and error counts as the main evidence. Zone does not diagnose, treat, or manage any medical condition.

Common questions

What does the Yerkes-Dodson law say in simple terms?

Performance improves as arousal or pressure rises, up to a point, and then gets worse. The best level of pressure is lower for hard tasks than for easy ones.

Is the Yerkes-Dodson law true?

Partly. Brain research supports an inverted-U effect of noradrenaline and dopamine on working memory, and hard tasks do tend to suffer more under stress. But the original study was about mice learning a discrimination, "arousal" is not a single measurable quantity, and at least one critical review argues the law has no solid basis as a rule for managing human work.

What is an example of the Yerkes-Dodson law?

A looming deadline can help you finish routine paperwork quickly, while the same deadline can make a complex piece of writing or analysis harder, because demanding thinking is more sensitive to pressure.

What does the Yerkes-Dodson graph look like?

The common version is an upside-down U, with arousal on the horizontal axis and performance on the vertical axis. A more faithful version shows two curves: for an easy task, performance keeps rising with arousal or levels off; for a hard task, it peaks earlier and drops.

Is some stress good for performance?

Some pressure can help, especially for routine work and for challenges that feel meaningful. Pressure from obstacles and unclear demands is linked to worse performance, and both kinds are linked to exhaustion. Pressure is not a reliable tool for improving other people's work.

How is the Yerkes-Dodson law different from eustress?

Eustress is a popular term for stress experienced as positive. The Yerkes-Dodson law is about the amount of arousal and task difficulty. The research on challenge and hindrance stressors and on reappraisal suggests that how stress is experienced matters alongside how much of it there is.

Sources

1. Diamond, D. M., Campbell, A. M., Park, C. R., Halonen, J., and Zoladz, P. R. (2007). The temporal dynamics model of emotional memory processing: a synthesis ... and the Yerkes-Dodson law. Neural Plasticity, 2007, 60803. PubMed record
2. Broadhurst, P. L. (1957). Emotionality and the Yerkes-Dodson law. Journal of Experimental Psychology, 54(5), 345 to 352. PubMed record
3. Corbett, M. (2015). From law to folklore: work stress and the Yerkes-Dodson Law. Journal of Managerial Psychology, 30(6), 741 to 752. DOI
4. Shields, G. S., Sazma, M. A., McCullough, A. M., and Yonelinas, A. P. (2017). The effects of acute stress on episodic memory: a meta-analysis and integrative review. Psychological Bulletin, 143(6), 636 to 675. PubMed record
5. Arnsten, A. F. T. (2009). Stress signalling pathways that impair prefrontal cortex structure and function. Nature Reviews Neuroscience, 10(6), 410 to 422. PubMed record
6. Beerendonk, L., Mejías, J. F., Nuiten, S. A., de Gee, J. W., Zantvoord, J. B., Fahrenfort, J. J., and van Gaal, S. (2025). Adaptive arousal regulation: pharmacologically shifting the peak of the Yerkes-Dodson curve by catecholaminergic enhancement of arousal. Proceedings of the National Academy of Sciences, 122(28), e2419733122. PubMed record
7. LePine, J. A., LePine, M. A., and Jackson, C. L. (2004). Challenge and hindrance stress: relationships with exhaustion, motivation to learn, and learning performance. Journal of Applied Psychology, 89(5), 883 to 891. PubMed record
8. Jamieson, J. P., Mendes, W. B., Blackstock, E., and Schmader, T. (2010). Turning the knots in your stomach into bows: reappraising arousal improves performance on the GRE. Journal of Experimental Social Psychology, 46(1), 208 to 212. PubMed record
9. Jamieson, J. P., Nock, M. K., and Mendes, W. B. (2012). Mind over matter: reappraising arousal improves cardiovascular and cognitive responses to stress. Journal of Experimental Psychology: General, 141(3), 417 to 422. PubMed record
10. Brooks, A. W. (2014). Get excited: reappraising pre-performance anxiety as excitement. Journal of Experimental Psychology: General, 143(3), 1144 to 1158. PubMed record