Yerkes-Dodson Law of Arousal and Performance

How the Law Works

The Yerkes–Dodson law describes the relationship between arousal and performance, showing that a moderate level of mental or physical alertness produces the best results.

In psychology, arousal refers to a general state of readiness and alertness – driven by factors like excitement, focus, or mild stress – not just feeling nervous or anxious.

Performance improves as arousal rises, but only up to an optimal point—after that, too much arousal causes it to drop.

This is known as the inverted-U model.

The curve’s shape depends on the task: complex or unfamiliar tasks are best tackled with lower arousal, while simple or well-practised tasks often benefit from higher arousal.

For example, a little pre-exam tension can sharpen focus, but overwhelming anxiety can make you forget what you know.

Practical Examples

How can students use it to improve exam performance?

• Find your sweet spot. If you feel flat (sleepy, unfocused), briefly raise arousal: stand and stretch, do 20–60 seconds of brisk movement, review a high-yield summary card, or set a 10–15 minute Pomodoro with a clear micro-goal.

• Dial down when jittery. If you feel wired (racing heart, mind blanking), use 60–120 seconds of slow breathing (4–6 breaths/min, longer exhale), progressive muscle relaxation, or the reframe “I’m excited” to channel arousal into focus.

• Match arousal to task. Use higher arousal for quick recall and practice tests; use lower arousal (quiet, steady pace) for complex problem-solving and multi-step calculations.

• Exam-day routine. 5–10 minutes pre-test: light movement → two calming breaths → skim a confidence list (3 things you know cold) → pick a “first easy win” question to lock in momentum.

How does it apply in sports, public speaking, or workplace productivity?

• Sports. Explosive, well-learned skills (sprinting, powerlifting) usually benefit from higher arousal (hype music, dynamic warm-up, energizing cue words). Fine-motor or novel/complex skills (golf putt, gymnastics routine, new play) often need lower arousal (slower breathing, quiet focus, consistent pre-shot routine).

• Public speaking. Aim for medium arousal: convert nerves into energy with brisk walking and power posture, then settle with two slow breaths. Use a simple opening script, anchor points on your outline, and a tempo cue (“slow and clear”) to avoid rushing.

• Workplace. Routine tasks (inbox triage, filing) can tolerate higher arousal and short sprints. Deep work (analysis, writing, coding) needs lower arousal: silence notifications, time-box 25–50 minutes, set a single outcome, and start with a 60-second calming reset.

What are everyday signs you’re over- or under-aroused for a task?

• Under-arousal (too low): Boredom, drifting attention, procrastination, slow starts, repeated rereads with nothing sticking.

• Quick fix: Add stakes (timer, micro-deadline), stand or walk 2 minutes, switch to an easier starter task, play neutral background noise if silence feels sleepy.

• Optimal arousal (just right): Clear goal, steady pace, present-moment focus, small errors caught quickly, time passing normally.

• Maintain: Brief breaks before fatigue, keep the environment stable, stick to the current goal.

• Over-arousal (too high): Racing thoughts, muscle tension, twitchy pacing, rushing, frequent mistakes, mind blanks on steps you know.

• Quick fix: Two minutes of slow breathing with long exhales, relax shoulders/jaw, lower stimulation (quieter room, dimmer screen), break the task into smaller chunks and re-enter with the easiest step.

Links to Psychology Theories

How does it connect to stress theories like the fight-or-flight response?

• The fight-or-flight response is a surge of sympathetic arousal; at moderate levels (often called eustress), attention sharpens and performance improves, matching the rising side of the inverted-U.

• When arousal becomes excessive (distress), prefrontal functions like working memory and flexible thinking degrade, producing tunnel vision and more errors—the falling side of the curve.

• Practical read: a little time pressure can help you lock in, but panic (racing heart, breathlessness, blanking) pushes you past the peak.

How does it relate to motivation theories such as Self-Determination Theory or Drive Theory?

• Drive Theory (classic “more arousal → better performance”) fits well-learned, simple tasks; Yerkes–Dodson refines it by adding the decline at high arousal, especially for complex or novel tasks.

• Self-Determination Theory (SDT) explains quality of motivation: autonomy, competence, and relatedness tend to keep arousal in a productive, self-regulated range; controlling pressure elevates anxiety and risks overshooting the peak.

• In practice: autonomy-supportive study or coaching fosters steady, optimal arousal; harsh evaluation or micromanagement spikes arousal and impairs complex performance.

How does it fit with the concept of “flow” in positive psychology?

• Flow occurs when skill and challenge are high and balanced, with clear goals and immediate feedback—subjectively “calm but energized” focus.

• Flow typically sits near the optimal zone of the inverted-U: aroused enough for intensity, low enough in anxiety to keep precision and flexibility.

• Key distinction: Yerkes–Dodson models a performance–arousal curve; flow describes the quality of experience. They align at the peak, but you can feel “amped” without flow, or be in flow at different absolute arousal levels depending on the task.

Why Arousal Affects Performance

The sections above describe what happens as arousal rises. Two theories explain why.

Drive Theory

Hull’s (1943) drive theory argues that arousal energises whichever response is dominant, meaning the one a person already knows best. Extended by Spence (1956) and applied to social facilitation by Zajonc (1965), the theory predicts that for a well-practised skill, more arousal simply means better performance.

Novices are different. Their dominant response is often the wrong one, so arousal amplifies mistakes instead of skill.

A footballer taking a penalty he has drilled thousands of times has the right action as his dominant response, so the arousal of the moment can sharpen it. An inexperienced player, without that dominant response yet, is more likely to mishit under the same pressure.

The difference is decisive.

Drive theory cannot explain the downturn at very high arousal, since it predicts performance keeps rising indefinitely. This gap is exactly what the inverted-U was introduced to fill.

Cue-Utilisation and Attentional Narrowing

Easterbrook’s (1959) cue-utilisation hypothesis, rooted in the psychology of selective attention, is the most influential explanation for the whole curve. As arousal rises, attention narrows like a beam. At first it screens out irrelevant distractions, which helps performance. Pushed further, it excludes relevant cues too, hurting performance.

This single idea explains both limbs of the inverted-U. It also explains why harder tasks peak at lower arousal: a complex task needs a wide set of cues, so it suffers as soon as narrowing begins.

Aim: Bahrick, Fitts and Rankin (1952) tested whether raising incentive, and so arousal, narrows the range of cues a person attends to.

Method: Participants tracked a demanding central task while also monitoring peripheral signal lights, with monetary incentive varied to raise arousal.

Results: The central task improved; peripheral responses dropped.

Conclusion: Rising arousal narrows the range of cues a person uses, exactly as Easterbrook’s hypothesis predicted.

Incentive is only a proxy for arousal, and the artificial dual-task setting may exaggerate narrowing compared with real skills. Even so, the direction of the effect is clear, and the “tunnel vision” it describes shows up again in applied findings on weapon focus under stress.

The Physiological View: An Optimal Level

A third account looks at the brain itself.

Hebb (1955) reframed arousal in neurophysiological terms, as the general activation level of the brain. He argued for an inverted-U between arousal and behavioural efficiency, with an optimal level that the brain is motivated to seek. Too little activation is boring; too much is stressful.

A modern candidate mechanism sits in the brainstem’s locus coeruleus-noradrenaline system. Aston-Jones and Cohen’s (2005) adaptive-gain theory proposes that moderate, steady firing in this system optimises engagement and performance, while the high, sustained firing linked to stress produces distractible, disorganised behaviour.

The takeaway is the same.

This view also connects to the wider idea that people are motivated to seek their own optimal level of stimulation. That helps explain why some people thrive on excitement while others avoid it.

Research Evidence

The Yerkes–Dodson law has been interpreted in different ways since it was first proposed in 1908.

In their original paper, Robert Yerkes and John Dodson studied discrimination learning in “dancing mice.”

The mice had to choose between a white and a black box.

Entering the white box delivered a mild (non-injurious) electric shock; the black box did not.

First series (baseline task).

With very weak shocks, mice took many tries to meet the learning goal (picking correctly 10/10 times for three days in a row).

As the shock got stronger, they learned faster—until the strongest level, where learning slowed again.

If you plot shock strength vs. number of tries (where fewer tries = better), you get a U-shape.

If you flip that to overall performance (where higher = better), you get the familiar inverted-U.

This went against the authors’ simple “more stimulation = better learning” expectation.

Second and Third Series: How Task Difficulty Changed the Curve

Yerkes and Dodson then made the choice easier by increasing the light contrast between boxes and tested five shock levels.

Now, the strongest shocks produced the fastest learning, and the weakest produced the slowest, more like a straight increase than a U-shape (Yerkes & Dodson, 1908; Teigen, 1994).

Because the task made few demands on attention, extra stimulation did not disrupt performance the way it would on a harder task.

Then they raised the stakes.

Next, they made the choice harder by reducing the light contrast and tested four shock levels.

Here, a moderate shock (the second-weakest) led to the best learning (fewest tries), again suggesting a U-shaped link between stimulation and learning rate (Teigen, 1994).

This is the clearest evidence for the “law”: task difficulty determined whether stronger stimulation helped performance or hurt it.

What they took from this.

Both too little and too much stimulation can hurt learning, and the best level depends on how difficult the task is.

As tasks get harder, the optimal stimulation shifts downward toward a gentler level (Yerkes & Dodson, 1908; Teigen, 1994).

Later theorists forgot this nuance.

Later authors, not Yerkes and Dodson themselves, generalised these two findings into the broader “law” linking arousal to performance across species and tasks.

The distinction matters in practice.

This is why a single prescription for the right amount of nerves is wrong.

A sprinter chasing a personal best benefits from a jolt of arousal that would ruin a golfer’s short putt.

The optimum is a moving target, set by how much information the task demands.

The pattern itself is only descriptive.

It says performance falls at high arousal; it does not say why, which is a separate question this article turns to next.

---

Replication Studies

Early follow-ups generally supported the idea that task difficulty changes the effect of stimulation on learning.

• Chicks (Cole, 1911): On easy tasks, stronger shock helped; on medium difficulty, very strong shock slowed learning; on difficult tasks, strong shock led to mixed results. Cole saw this as broadly consistent with Yerkes–Dodson.

• Kittens (Dodson, 1915): A medium shock beat a strong one on the standard task; when the task was made easier, medium and strong shocks worked about equally well, and learning overall improved as shock increased.

• Reward strength (Dodson, 1917):  A similar U-shape appeared with rewards: moderate deprivation (e.g., mild hunger) boosted learning, but excessive deprivation made it worse.

• Humans (Vaughn & Diserens, 1930; summarized in Young, 1936). People learned a maze best with light or medium punishment, not with none or heavy punishment—again pointing to an optimum rather than a straight line. Young concluded: for any activity, there is an optimal degree of punishment.

---

From Punishment–Learning to Motivation–Performance

By the 1930s–1940s, researchers began to reframe the idea.

Punishment was no longer seen as a core driver of learning (e.g., Thorndike, 1932; Skinner, 1938; Estes, 1944), and scientists distinguished learning from performance more carefully.

Mid-century accounts treated the pattern as a link between drive/arousal (motivation) and performance, noting that too much arousal can sometimes get in the way (e.g., Hilgard & Marquis, 1961).

Textbooks and reviews (e.g., Bourne & Ekstrand, 1973) popularized the inverted-U: performance is highest at moderate arousal and lower at both extremes.

The idea kept evolving.

Note: Early animal work often measured tries needed to learn (lower is better), which makes the graph look U-shaped.

When we talk about performance level (higher is better), the same pattern appears as an inverted-U.

Rats were tested next.

A stronger test came from Broadhurst (1957), who used several motivation levels and three difficulty levels with larger groups of rats.

He found that the best motivation level moved with task difficulty, higher for easy tasks and lower for hard ones, matching Yerkes and Dodson’s core claim (Broadhurst, 1957; Teigen, 1994).

He also suggested looking at individual differences (e.g., “emotionality”) that might shift where the peak sits on the curve.

Critical Evaluation

1. Situations where performance doesn’t follow the inverted-U pattern

The inverted-U is not universal; some tasks show different shapes (linear increases, linear decreases, thresholds, or plateaus).

Very simple, well-learned speeded tasks sometimes improve almost linearly as arousal rises (consistent with classic drive and social-facilitation findings).

In contrast, highly complex or novel tasks can worsen as arousal increases, producing a mostly downward slope with little or no “rising” phase (e.g., multi-step reasoning under intense time pressure).

Vigilance or monotonous monitoring tasks can also show threshold/plateau behavior, where raising arousal from very low levels helps, but further increases add little benefit.

These departures occur because arousal affects attention breadth, working memory, and motor control in task-specific ways.

Practitioners should avoid one-size-fits-all prescriptions.

Target arousal to the task: it may be optimal to increase stimulation for dull, overlearned activities but decrease it for precision or novel problem-solving. Treat the inverted-U as a heuristic starting point, not a binding rule.

2. Personality differences (introversion vs. extraversion)

Baseline arousal and reactivity vary with personality, shifting where the “optimal zone” sits on the curve.

Introverts typically have higher baseline cortical arousal and greater sensitivity to stimulation, so they may reach overload at lower levels of noise, caffeine, audience size, or time pressure.

Eysenck’s theory predicts this pattern.

Extraverts often have lower baseline arousal and seek stimulation, so additional intensity (livelier environment, tighter deadlines, energizing routines) can move them toward peak performance.

The optimum truly is personal.

Traits like trait anxiety/neuroticism and sensation-seeking further tilt the curve: height, width, and peak position can all change person-to-person.

Personalize conditions: quieter rooms, longer ramps, and calming routines for many introverts or high-anxiety individuals; brisk warm-ups, stronger cues, and lively environments for many extraverts or high sensation-seekers.

In education and workplaces, build flexibility (choice of noise levels, pacing, and break structure) rather than enforcing a single “optimal” arousal recipe.

3. Methodological flaws and oversimplifications in the original work

The 1908 Yerkes & Dodson evidence has limitations that weaken claims of a universal “law.”

Hebb (1955) broadened the idea further.

The original studies used mice learning a discrimination task with electric shock as the “arousal” manipulation, conflating punishment magnitude, motivation, stress, and learning rate.

Physiological arousal wasn’t directly measured, statistical methods were rudimentary by modern standards, and generalization from a narrow animal paradigm to diverse human performances was assumed rather than demonstrated.

Later popularizations often depict a smooth inverted-U without specifying moderators (task type, skill level, personality, context) or providing precise functional forms, encouraging over-simple applications.

The criticism goes deeper.

Brown (1965), in a paper titled “The Yerkes-Dodson law repealed,” argued the evidence was far weaker than confident textbook accounts suggest.

Winton (1987) found that introductory textbooks often misrepresent the law, leaving out the task-difficulty clause or presenting one tidy curve the original data do not support.

Teigen (1994) called it “a law for all seasons,” invoked so loosely that it becomes almost impossible to disprove.

Treat it as a heuristic, not a fixed rule.

Modern evaluation should rely on task- and person-specific data (e.g., error rates, response variability) and, where possible, physiological indicators of arousal (heart-rate variability, pupil dilation).

Interventions should be tested and iterated for the particular context instead of assumed from the generic curve.

Contemporary Research

Recent work asks a sharper question: is the arousal-performance link as strong, and as universal, as the classic curve suggests?

The strongest evidence comes from meta-analysis.

Aim: von der Embse, Jester, Roy and Post (2018) set out to synthesise three decades of research linking test anxiety to exam performance.

Method: Their meta-analysis pooled 238 studies published from 1988 onward, calculating combined effect sizes across standardized tests, university-entrance exams and grade point average.

The effect held at scale.

Results: Test anxiety was significantly linked to worse performance, but the effect was small to moderate, not the dramatic collapse the popular curve implies.

Conclusion: High evaluative arousal reliably hurts performance, though the effect is modest and depends on the student’s age and how high-stakes the test is.

As a meta-analysis of 238 studies, this sits near the top of the evidence hierarchy, though most of the underlying studies are correlational and rely on self-reported anxiety.

A separate meta-analysis complicates the picture further. Shields, Sazma, McCullough and Yonelinas (2017) pooled 113 studies and over 6,000 participants on stress and memory. They found no single pattern.

Stress before learning usually hurt memory, stress right after learning helped it stick, and stress at recall made it harder to retrieve. Cortisol levels barely tracked the size of these effects at all, undermining the idea that one “arousal” quantity drives performance in a simple curve.

The takeaway is simple. Taken together, the field increasingly treats Yerkes-Dodson as a heuristic rather than a precise law. Corbett (2015) traces this shift directly, describing how the “law” moved from science into workplace folklore.

References

Anderson, K. J., & Revelle, W. (1983). The interactive effects of caffeine, impulsivity and task demands on a visual search task. Personality and Individual Differences, 4(2), 127-134.

Bourne, L. E., & Ekstrand, B. R. (1973). Psychology: Its principles and meanings (Dryden, Hinsdale, IL).

Broadhurst, P. L. (1957). Emotionality and the Yerkes-Dodson law. Journal of experimental psychology, 54(5), 345.

Brown, W. P. (1965). The Yerkes-Dodson law repealed. Psychological reports, 17(2), 663-666.

Cole, L. W. (1911). The relation of strength of stimulus to rate of learning in the chick. Journal of Animal Behavior, 1(2), 111.

Corbett, M. (2015). From law to folklore: work stress and the Yerkes-Dodson Law. Journal of Managerial Psychology.

Dodson, J. D. (1915). The relation of strength of stimulus to rapidity of habit-formation in the kitten. Journal of Animal Behavior, 5(4), 330.

Dodson, J. D. (1917). Relative values of reward and punishment in habit formation. Psychobiology, 1(3), 231.

Estes, W. K. (1944). An experimental study of punishment. Psychological Monographs, 57(3), i.

Geen, R. G. (1984). Preferred stimulation levels in introverts and extroverts: Effects on arousal and performance. Journal of Personality and Social Psychology, 46(6), 1303.

Gigerenzer, G., & Murray, D. J. (2015). Cognition as intuitive statistics. Psychology Press.

Hebb, D. O. (1955). Drives and the CNS (conceptual nervous system). Psychological review, 62(4), 243.

Hilgard, E. R., & Marquis, D. G. (1961). Hilgard and Marquis” conditioning and learning.

Levitt, E. E. (2015). The psychology of anxiety.

Loftus, E., & Ketcham, K. (1991). Witness for the defense: The accused, the eyewitness, and the expert who puts memory on trial. Macmillan.

Matthews, G. (1985). The effects of extraversion and arousal on intelligence test performance. British Journal of Psychology, 76(4), 479-493.

Revelle, W., Amaral, P., & Turriff, S. (1976). Introversion/extroversion, time stress, and caffeine: Effect on verbal performance. Science, 192(4235), 149-150.

Shields, G. S., Sazma, M. A., McCullough, A. M., & Yonelinas, A. P. (2017). The effects of acute stress on episodic memory: A meta-analysis and integrative review. Psychological Bulletin, 143(6), 636-675.

Skinner, B. F. (2019). The behavior of organisms: An experimental analysis. BF Skinner Foundation.

Teigen, K. H. (1994). Yerkes-Dodson: A law for all seasons. Theory & Psychology, 4(4), 525-547.

Thorndike, E. L. (1932). The fundamentals of learning.

Vaughn, J., & Diserens, C. M. (1930). The relative effects of various intensities of punishment on learning and efficiency. Journal of Comparative Psychology, 10(1), 55.

von der Embse, N., Jester, D., Roy, D., & Post, J. (2018). Test anxiety effects, predictors, and correlates: A 30-year meta-analytic review. Journal of Affective Disorders, 227, 483-493.

Winton, W. M. (1987). Do introductory textbooks present the Yerkes-Dodson Law correctly?. American Psychologist, 42(2), 202.

Yerkes, R. M., & Dodson, J. D. (1908). The relation of strength of stimulus to rapidity of habit-formation. Punishment: Issues and experiments, 27-41.

Young, P. T. (1936). Social motivation.