Deliberate practice

Also known as purposeful practice, the 10,000-hour rule (incorrectly)

Practice only improves you when it targets something you cannot yet do and tells you immediately whether you did it.

The claim Expert performance is built by effortful, goal-directed practice with immediate feedback at the edge of current ability — not by accumulated experience.

Mixed evidence The distinction between deliberate practice and mere repetition is well supported; the strong claim that it accounts for most of expert performance did not survive meta-analysis.

Anders Ericsson's contribution was to show that time spent is the wrong unit. Studying violinists at a Berlin academy in 1993, he and his colleagues found that what separated the best students was not hours with the instrument but hours of a particular kind: solitary, effortful, aimed at a specific weakness, with immediate feedback, at a level just beyond comfortable. Doctors, typists and amateur musicians can spend decades on an activity and plateau early, because experience without that structure is repetition rather than practice.

The framework became famous in a mangled form. The '10,000-hour rule' is Malcolm Gladwell's, not Ericsson's, and Ericsson spent the rest of his career objecting to it: 10,000 was an average for one group at one age in one domain, there was no threshold, and the number was never the point. The point was the structure of the hours.

Then came the correction. Macnamara, Hambrick and Oswald's 2014 meta-analysis asked how much of the variation in performance deliberate practice actually explains. The answers were 26% in games, 21% in music, 18% in sports, 4% in education and under 1% in professions. Practice matters, and it matters much less than the strong version claimed — most of what separates performers is something else, or several something elses, including starting age, opportunity and whatever it is that IQ and personality measures pick up.

Ericsson replied that the meta-analysis counted studies whose 'practice' measures were nothing like deliberate practice in his sense — self-reported hours, group lessons, ordinary rehearsal — and so measured the dilution rather than the thing. That objection has force, and it also makes the theory harder to test, because the cleanest way to identify deliberate practice after the fact is that it worked.

What survives the argument intact is the practical core, and it is worth separating from the dispute about ceilings. Repetition of what you can already do produces almost nothing; targeted work on a specific weakness with fast feedback produces almost all the improvement most people ever get. Whether that gets you to world class is a question most learners will never need answered.

What to do differently

  1. Name the specific thing you cannot yet do, at a resolution small enough to attempt today. 'Get better at speaking' is not one; 'stop pausing before every past-tense verb' is.
  2. Build in feedback that arrives in seconds or minutes, not weeks. Without it you are rehearsing your errors.
  3. Work just past the edge of comfortable. If the session is enjoyable and error-free, it is performance, not practice.
  4. Keep sessions short and undivided — the literature consistently finds a few hours a day is the ceiling for this kind of effort.
  5. Separate practice from performing. Both are necessary, and doing only the second is the standard way to plateau.

Limits and trade-offs

  • The strong version — that practice explains most of expert performance — is contradicted by meta-analysis, and repeating it as settled science is not honest.
  • It is unfalsifiable at the edges: any failure to improve can be reclassified as practice that was not really deliberate.
  • It requires a coach or an objective standard to supply the feedback, which most self-directed learners do not have and cannot easily manufacture.
  • It is exhausting and unpleasant by construction, and a regime you abandon in three weeks is worse than a mediocre one you keep for a year.
  • The research is concentrated in domains with clear, measurable performance — chess, music, sport. Its transfer to open-ended skills is assumed more than demonstrated.

What this means for a language

Language learning has an unusually severe version of the plateau this framework describes: after a couple of years many learners can be understood, stop being corrected, and simply stop improving. That is repetition without feedback, exactly as predicted. The deliberate-practice reading is to isolate one weakness at a time — a sound you produce wrongly, a tense you avoid, a register you cannot hear — and get correction fast enough to act on. It is also worth naming the limit: comprehension seems to grow from sheer volume of input in a way this framework does not describe well, so a language learner needs both the targeted hour and the untargeted ten.

Methods that lean on this Shadowing Speak From Day One Bidirectional Translation

The research

Related

Builds on Desirable difficulties Limits Transfer and "learning to learn" Contrasts with Growth mindset

Read next

  • Peak: Secrets from the New Science of Expertise Book

    Anders Ericsson, Robert Pool · 2016 · Houghton Mifflin Harcourt

    Deliberate practice from the researcher who defined it, and his own correction of the '10,000 hours' story that was built on his work. Read alongside the Macnamara meta-analysis, which puts a much smaller number on how much practice explains.

  • Deliberate Practice and Performance in Music, Games, Sports, Education, and Professions: A Meta-Analysis Paper

    Brooke N. Macnamara, David Z. Hambrick, Frederick L. Oswald · 2014 · Psychological Science 25(8), 1608–1618

    The correction to the practice-is-everything story. Deliberate practice explained 26% of performance variance in games, 21% in music, 18% in sports, 4% in education and under 1% in professions.

  • The Role of Deliberate Practice in the Acquisition of Expert Performance Paper

    K. Anders Ericsson, Ralf Th. Krampe, Clemens Tesch-Römer · 1993 · Psychological Review 100(3), 363–406

    The violinist study. The origin of both the deliberate-practice framework and — via Gladwell, not via this paper — the '10,000 hours' claim the authors have spent thirty years disowning.

  • Ultralearning: Master Hard Skills, Outsmart the Competition, and Accelerate Your Career Book

    Scott H. Young · 2019 · Harper Business

    Aggressive self-directed projects — an MIT curriculum in a year, a language in three months — turned into nine principles. Strongest on directness and project design; weakest as evidence, since the case studies are selected on success.

  • Range: Why Generalists Triumph in a Specialized World Book

    David Epstein · 2019 · Riverhead Books

    Popular-press, and it argues a thesis rather than surveying a field — but its chapters on interleaving, desirable difficulties and the limits of early specialisation are the most readable version of that literature anywhere.

  • Deep Work: Rules for Focused Success in a Distracted World Book

    Cal Newport · 2016 · Grand Central Publishing

    About the conditions for learning rather than the mechanics of it. The argument that concentration is a trainable and increasingly rare capacity is well made; the supporting research is thinner than the confident tone suggests.

The full reading list →

← All learning principles