Anders Ericsson

Anders Ericsson

Anders Ericsson, 1947 to 2020, was a Swedish-born psychologist, Conradi Eminent Scholar and Professor of Psychology at Florida State University, and the man who built the research programme on expert performance. His central idea is deliberate practice: structured, effortful, feedback-driven work at the edge of current ability, which is a different thing from accumulated time. He is also this site's clearest single case of the famous-name attribution error, because the number the public attaches to his work is not his, and two books in this collection use his findings without naming him at all.

Who they are

The founding paper is Ericsson, Krampe and Tesch-Römer (1993), and he set the position out for a general reader in Peak, written with Robert Pool in 2016. Deliberate practice as he defined it is solitary, effortful, designed by a teacher, aimed at a specific weakness, and fed by immediate feedback. Almost none of the popular uses of his work preserve all five of those conditions.

The ten-thousand-hour rule is the clearest case. Malcolm Gladwell drew it from Ericsson's violinist data in Outliers in 2008 and made it famous. Ericsson objected publicly and repeatedly, in Peak and elsewhere, and his objections were specific rather than general: the round number is arbitrary, it was never a threshold, and nothing in the work licenses transplanting a figure from violin study to every other domain. What his research says carries the effect is the quality and structure of the practice rather than the hours logged. Gladwell's side of it is that he was popularising a real finding, and that a memorable figure is what makes a finding travel, which is a fair account of what popularisers are for. This site agrees with Ericsson. The number is his popularisers' invention rather than his result, and substituting hours for deliberate practice reverses the actual claim.

One correction made here against the 1993 paper's own text, because this site got it wrong first. The ten-thousand-hour figure does not appear anywhere in that paper. The best group's accumulated solitary practice by age eighteen was about 7,410 hours, and the paper invokes Simon and Chase's ten-year rule rather than any hour count. The framing was built by extrapolating one group in one study forward to age twenty.

The scientific dispute about effect size is separate from the attribution dispute, and correcting the second does not settle the first. Brooke Macnamara and colleagues have argued meta-analytically that practice explains far less of the variation in performance than the strongest readings of Ericsson imply, and that in sport it separates sub-elite from average without reliably separating elite from elite, which is where the popular claim lives. A 2019 direct replication of the 1993 violinist study, using the same design, the same population and the original definition, found that the group rated best had practised fewer hours than the group rated merely good.

His reply is definitional and it is real science. Macnamara's inclusion criteria count structured practice rather than deliberate practice as defined in 1993, sweeping in group activities, competitions and play, and many of the studies included never use the term. A reanalysis restricted to studies meeting the original definition puts the share of variance much higher. Ericsson and Harwell (2019) makes that case at length, and Miller and colleagues (2021) argue that the meta-analyses underestimate the effect by neglecting individualisation. Macnamara, Hambrick and Moreau replied that his definition has shifted over time in ways that make the theory unfalsifiable, and that objection is live and is not answered anywhere in the record this site has read.

Where this site lands, in this order. With Macnamara on the numbers, and with Ericsson on the definition. He is right that measuring the wrong thing gives the wrong number. She is right that the retrospective self-reported hour counts underneath the whole literature are weak, and that the effect shrinks towards nothing at the elite end. Practice is necessary and it is not sufficient.

One point of fairness belongs to him. The 2019 replication authors, who are his critics, report that the confidence interval around the original mean of about 7,410 hours runs from roughly 2,894 to 11,926 hours and contains the good group's mean. The original study was never powered to separate the top two groups. That is a criticism of the 1993 design, and it is also a reason not to read the replication as a refutation of him personally.

Key papers

Ericsson, K. A., Krampe, R. T., & Tesch-Römer, C. (1993), "The role of deliberate practice in the acquisition of expert performance," Psychological Review, 100(3), 363-406. The founding paper, and the source of the violinist data everything downstream rests on.

Ericsson, K. A., & Pool, R. (2016), Peak: Secrets from the New Science of Expertise. The general-reader statement of the position, and where he sets out his objection to the ten-thousand-hour framing.

Duckworth, A. L., Kirby, T. A., Tsukayama, E., Berstein, H., & Ericsson, K. A. (2011), "Deliberate Practice Spells Success," Social Psychological and Personality Science, 2(2), 174-181. One hundred and ninety finalists at the 2006 National Spelling Bee. Deliberate practice predicted advancement strongly, being quizzed did not survive the full model, and leisure reading was a flat null. Participants rated deliberate practice as the least enjoyable and most effortful of the three activities, which is the finding most consistent with his definition and the one most often left out.

Ericsson, K. A. (2016), "Summing Up Hours of Any Type of Practice Versus Identifying Optimal Practice Activities," Perspectives on Psychological Science, 11, 351-354. His fullest reply to the meta-analysts, answering Macnamara, Moreau and Hambrick's sports meta-analysis in the same issue. Macnamara, Hambrick and Moreau replied at 355 to 358.

Ericsson, K. A., & Harwell, K. W. (2019), Frontiers in Psychology. The reanalysis restricted to studies meeting the 1993 definition.

Key ideas

Deliberate practice is not practice. It is solitary, effortful, teacher-designed, aimed at a named weakness and fed by immediate feedback. Drop any of those and the construct is a different one.

The ten-thousand-hour rule is publishing history rather than a finding. The figure is not in the 1993 paper.

The definitional objection and the effect-size objection are different arguments, and both are live. Practice is necessary and it is not sufficient.

The measurement problem runs under the whole literature. Retrospective self-reported hour counts are weak evidence on either side of the dispute.

The 1993 study was underpowered to separate its top two groups, which its own critics state. That cuts against reading the 2019 replication as a personal refutation.

Where a book uses the mechanism without the name, a reader gets the advice and no route to the dispute. That is the specific cost of the omission.

Sources

  1. Ericsson, K. A., Krampe, R. T., & Tesch-Römer, C. (1993), "The role of deliberate practice in the acquisition of expert performance," Psychological Review, 100(3), 363-406.
  2. Ericsson, K. A., & Pool, R. (2016), Peak: Secrets from the New Science of Expertise.
  3. Duckworth, A. L., Kirby, T. A., Tsukayama, E., Berstein, H., & Ericsson, K. A. (2011), "Deliberate Practice Spells Success," Social Psychological and Personality Science, 2(2), 174-181.
  4. Ericsson, K. A. (2016), "Summing Up Hours of Any Type of Practice Versus Identifying Optimal Practice Activities," Perspectives on Psychological Science, 11, 351-354. Reply by Macnamara, Hambrick and Moreau in the same issue, 355-358.
  5. Ericsson, K. A., & Harwell, K. W. (2019), Frontiers in Psychology.
  6. Gladwell, M. (2008), Outliers: The Story of Success. Where the ten-thousand-hour framing originates.
  7. Vault note: [[Anders Ericsson]], which records the attribution audit and the correction made against the 1993 paper's own text.