Deliberate Practice and Acquisition of ExpertPerformance: A General OverviewK. Anders Ericsson, PhD

AbstractTraditionally, professional expertise has been judged by length of experience, reputation, and perceivedmastery of knowledge and skill. Unfortunately, recent research demonstrates only a weak relationshipbetween these indicators of expertise and actual, observed performance. In fact, observed performancedoes not necessarily correlate with greater professional experience. Expert performance can, however,be traced to active engagement in deliberate practice (DP), where training (often designed and arrangedby their teachers and coaches) is focused on improving particular tasks. DP also involves the provisionof immediate feedback, time for problem-solving and evaluation, and opportunities for repeated perfor-mance to refine behavior. In this article, we draw upon the principles of DP established in otherdomains, such as chess, music, typing, and sports to provide insight into developing expert performancein medicine.

ACADEMIC EMERGENCY MEDICINE 2008; 15:988–994 ª 2008 by the Society for Academic EmergencyMedicine

E ducation and training have a very long history.Ever since Greek civilization, there has been animportant distinction made between general

education and vocational training, which requires theacquisition of a high level of skill. Education, as proposedby Plato and Aristotle, discouraged specialization andfocused on developing thinking skills and general knowl-edge. With respect to skilled manual work, primarily per-formed by slaves in Athens,1 Plato proposed an earlystart of training and restricted engagement to a singlecraft.2,3 When discussing medical expertise, Plato con-trasted the routine application of medical procedures bya slave doctor to the thoughtful diagnosis and reasoningabout treatment and explanation to the patient by anexpert doctor.4 Competency among doctors was thusthought to be best assessed by examining their superiorknowledge5 and ability to teach others.6 The primaryadditional factor to achieve competence was believedto be due to accumulated experience, and thus agewas expected to be correlated with wisdom. Platoargued that the ideal doctor was one with significant

experience—ideally, personal experiences of recoveringfrom many diseases.7

Since Aristotle, the issues of how to evaluate and cer-tify expertise and the tension between theoreticalunderstanding and practical skill have remained. Thelevel of expertise of practitioners has been monitoredand credentialed by guilds and more recently by profes-sional organizations.8 In the United States, students typ-ically require a general education, such as a collegeeducation, before they can be admitted to study at pro-fessional schools. Following primarily theoretical train-ing at professional schools, graduates are trained asapprentices to experienced practitioners until they earnthe credentials to practice independently.

Consistent with the distinction between general theo-retical knowledge and professional skill, early tradi-tional models of skill and expertise9–13 distinguishdifferent stages of development. The first stage (novice)involves following the teachers’ instruction and apply-ing rules and procedures step by step. With increasingexperience, the student becomes more able to generatethe same outcomes faster and more efficiently. Afterextensive experience, individuals become experts andare able to respond rapidly and intuitively. Somedomains, such as driving a car, are simple and ‘‘almostall novices [beginners] can eventually reach the level wecall expert.’’11 In other more complex domains, such astelegraphy and chess, it may take decades to reach thehighest levels.13,14 Some researchers even explicitlyreject the idea that expert behavior and performanceneeds to be uniformly superior to less experiencedindividuals.11 The pioneering research on expertise13

emphasized improvements in performance due to

ISSN 1069-6563 ª 2008 by the Society for Academic Emergency Medicine988 PII ISSN 1069-6563583 doi: 10.1111/j.1553-2712.2008.00227.x

From the Department of Psychology Florida State University,Florida, Tallahassee, FL.Received July 7, 2008; revisions received July 9 and July 10,2008; accepted July 10, 2008.Presented at the 2008 Academic Emergency MedicineConsensus Conference, ‘‘The Science of Simulation in Health-care: Defining and Developing Clinical Expertise,’’ Washington,DC, May 28, 2008.Address for correspondence and reprints: K. Anders Ericsson,PhD; e-mail: ericsson@psy.fsu.edu.

experience in the domain. In studies of medical doctorsand nurses, it was typical to search for experts by usingpeer-nomination procedures among highly experiencedprofessionals.9,15,16

In the 1980s the definition of expertise based on accu-mulated knowledge, extensive professional experience,and peer nominations was becoming increasingly criti-cized. Numerous empirical examples were reportedwhere ‘‘experts’’ with extensive experience andextended education were unable to make better deci-sions than their less skilled peers or even sometimesthan their secretaries.17 Early studies were unable toestablish superior accuracy of the peer-nominated bestgeneral physicians, when compared to a group ofundistinguished physicians.15,16 Similar findings weresubsequently attained for clinical psychotherapists,where more advanced training and longer professionalexperience were unrelated to the quality and efficiencyof treatment outcomes.18 In addition, examinations ofthe cognitive mechanisms that mediated the actions ofindividuals exhibiting consistently superior perfor-mance revealed a complex structure that could not beaccounted for by a mere accumulation of experienceand knowledge.19 In response to these criticisms, Erics-son and Smith19 proposed the redirection of researchfrom studying the behavior of socially recognizedexperts toward the study of reproducibly superior per-formance in a given domain.

THE SCIENTIFIC STUDY OF EXPERTPERFORMANCE AND ITS ACQUISITION

Establishing a science of superior performance startswith the accumulation of a body of reproducible empir-ical phenomena.20 Methods for reproducing the supe-rior performance under standardized and experimentalconditions are thus needed. When superior perfor-mance can be replicated in the laboratory, its structurecan be examined and analyzed with process analysisand experimental methods to reveal the mechanismsunderlying superior performance.

Every day, professionals make superior decisions;soccer players make the deciding goal, and scientistsmake amazing discoveries. However, it is impossible toknow if a singular action was due to unique circum-stances (if placed in that same context many peoplewould have produced a similar outcome), luck, or astable superior ability to handle such situations. TheGreeks were particularly interested in determiningaspects of soldiers’ performance, such as who is thestrongest or the fastest and can throw a javelin or dis-cus the longest. The Greeks developed athletic competi-tions with standardized events. For example, ratherthan having athletes run from one point to another innatural terrain, they built straight, flat tracks that wereindistinguishable for all runners and devised methodsto force runners to start at the exact same time andcross the same finishing line. More recently, there haveemerged competitions in music, dance, and chess thathave objective performance measures to identify thewinners. In all of these traditional domains, eliteindividuals reliably outperform less accomplished indi-viduals. There are many groups of professionals who,

on a daily basis, perform similar tasks. For example,professional investors have equal access to investmentopportunities in the stock market, medical professionalstreat patients with similar symptoms, and psychothera-pists treat patients with similar reported problems.

An expert athlete, musician, chess master, nurse, ormedical doctor is expected to be able respond toemerging task demands. It is part of the definition of anexpert performer that they are able to perform at virtu-ally any time with relatively limited preparation. Eliteathletes must be ready to compete, even if the competi-tion is delayed for a few hours, or even a day, becauseof bad weather. Similarly, an emergency physician (EP)is expected to be able and ready to help when theyencounter a distressed patient in the clinic or even inpublic.

Ericsson and Smith19 proposed how naturally occur-ring events can be used to capture the essence ofexpertise in a given domain. For example, it would bepossible to re-create an actual clinical situation by film-ing a scenario from the responsible doctor’s perspec-tive and decision-making process. This film could thenbe presented to individual experts and less experienceddoctors, the film could be stopped at critical decisionpoints, and the observing doctor may be asked to pro-vide appropriate direction to personnel under normaltime constraints. By selecting scenarios where there isonly one or a couple of appropriate actions that are notdetrimental to the patient, it is possible to evaluate thedoctor’s actions and decisions and their ability to assessthe essential factors in the time-constrained situation.The pioneering work of selecting critical events wasintroduced by de Groot21 in the domain of chess. DeGroot extracted critical situations in games betweenchess masters and then set up a controlled laboratorysituation where chess players were sequentially pre-sented with the associated positions (see Figure 1). Inanother example, since expertise in typing should gen-eralize to any material, all typists may be provided withthe same text to type as fast and accurately as possible.The final example given in Figure 1 illustrates the skillof sight reading, where an accompanist is presentedwith a sheet of music and asked to accompany a singerwithout having a chance to prepare in advance. Theability of accurately playing as many notes as possiblefrom the music score is what differentiates skilledaccompanists from other pianists. Over the past fewdecades, standardized test situations, where perfor-mance can be assessed within an hour, have beendeveloped that are highly correlated to real-world per-formance, such as tournament performance in chess,golf, and Scrabble; performance in music competitions;and medical diagnosis.23–25 These findings are consis-tent with the hypothesis that there is an underlyingfactor of attained expertise in a domain, where themajority of the tasks can be ordered on a continuum ofdifficulty.26 In many domains of expertise, there is arank ordering of difficulty for mastery of different tasks.For example, dives are given a difficulty score, andmusic pieces are rated in the number of years of studyrequired before mastery. Similarly, gymnastics, martialarts, mathematics, and many of the sciences have aclear progression of levels defined by mastery of

ACAD EMERG MED • November 2008, Vol. 15, No. 11 • www.aemj.org 989

increasingly difficult tasks. In sum, a collection of repre-sentative tasks that capture the essence of expertise ina given domain may be identified and administered toall participants under controlled and standardized con-ditions to objectively measure performance.

THE ACQUISITION OF SUPERIOR REPRODUCIBLE(EXPERT) PERFORMANCE

Once we are able to measure individual performance, itis possible to measure the time course of improvementand identify several characteristics that generalizeacross different domains of expertise.23,24,27 In somedomains there is no demonstrable improvement in per-formance as a function of years of professional experi-ence after completed training. For example, theaccuracy of diagnosis of heart sounds and many typesof measurable activities of nurses and general physi-cians do not improve as a function of professionalexperience, and sometimes the performance even grad-ually decreases after graduation.28–30 In contrast, manytraditional domains of expertise, such as arts and sci-ences, games, and sports, demonstrate improvementsthat continue for decades.

Based on an analysis of many different domains,consistent patterns of performance level over time (Fig-ure 2) have been observed.27 When the same uniformadult performance standards are used, abrupt improve-ments in performance do not occur, and changes overtime are gradual. In addition, the age at which expertstypically reach their peak performance is in the thirdand fourth decades for the arts and sciences and some-what earlier for vigorous sports. Finally, all performers,even the most ‘‘talented,’’ need around 10 years ofintense involvement before they reach an internationallevel in established sports, sciences, and arts.13,31 Most

elite individuals take considerably longer to reach thatlevel. The necessity for years and even decades ofrequired engagement in domain-related activities is themost compelling evidence for the crucial role of experi-ence required to attain high levels of performance.Some of the best evidence for the necessity of improvedtraining methods and expanded practice durationscomes from historical comparisons.24,27 The most dra-matic increases in the level of attained performanceover historical time are found in sports. In competitionssuch as the marathon and swimming events, a largenumber of today’s serious amateurs could easily beatthe gold medal winners of the past.

Figure 1. Three examples of laboratory tasks that capture the consistently superior performance of domain experts in chess,typing, and music (adapted from Ericsson and Lehmann22).

Figure 2. Illustration of the gradual increases in expert perfor-mance as a function of age, in domains such as chess. Theinternational level, which is attained after more than around10 years of involvement in the domain, is indicated by the hori-zontal dashed line (from Ericsson and Lehmann22).

990 Ericsson • DELIBERATE PRACTICE AND ACQUISITION OF EXPERT PERFORMANCE

MERE EXPERIENCE VERSUS DELIBERATEPRACTICE

To reconcile the virtual absence of a relation betweenamount of experience and objective performance inmany professional domains on the one hand, and thenecessity of many years or even decades of full-timeengagement in training for reaching high levels of per-formance, my colleagues and I31 tried to identify thosedomain-related activities necessary for improving per-formance and classified them as deliberate practice(DP).

Based on a review of research on skill acquisition,we31 identified a set of conditions where practice hadbeen uniformly associated with improved performance.Significant improvements in performance were realizedwhen individuals were 1) given a task with a well-defined goal, 2) motivated to improve, 3) provided withfeedback, and 4) provided with ample opportunities forrepetition and gradual refinements of their perfor-mance. Deliberate efforts to improve one’s performancebeyond its current level demands full concentration andoften requires problem-solving and better methods ofperforming the tasks.31

When people are introduced to an unfamiliar domainof activity, such as a new job, sport, or game, they fre-quently encounter situations where they cannot reactfast enough or where they are unable to produce func-tional actions, resulting in obvious failures. Over time,they are able to figure out adequate responses by prac-tice, problem-solving, and trial-and-error or with helpfrom supervisors, teachers, or colleagues. With furtherexperience they become increasingly able to generaterapid adequate actions with less and less effort—consis-

tent with the traditional theories of expertise and skillacquisition11,12 as is illustrated at the lower arm inFigure 3. After some limited training andexperience—frequently less than 50 hours for mostrecreational activities, such as skiing, tennis, anddriving a car—an individual’s performance is adaptedto the typical situational demands and is increasinglyautomated, and they lose conscious control overaspects of their behavior and are no longer able tomake specific intentional adjustments. For example,people have automated how they tie their shoelaces orhow they stand up from sitting in a chair. When perfor-mance has reached this level of automaticity andeffortless execution, additional experience will notimprove the accuracy of behavior nor refine the struc-ture of the mediating mechanisms, and consequently,the amount of accumulated experience will not berelated to higher levels of performance.

In direct contrast, aspiring experts continues toimprove their performance as a function of more expe-rience because it is coupled with DP. The key challengefor aspiring expert performers is to avoid the arresteddevelopment associated with automaticity. These indi-viduals purposefully counteract tendencies towardautomaticity by actively setting new goals and higherperformance standards, which require them to increasespeed, accuracy, and control over their actions, as isshown in the upper arm of Figure 3. The experts delib-erately construct and seek out training situations toattain desired goals that exceed their current level ofreliable performance.

The clearest cases of DP are found when children oradolescents have recently gotten involved in active par-ticipation in sports, singing, or playing a music instru-ment. Many of the more ‘‘talented’’ will be encouragedby their parents or coaches to start seeing a profes-sional teacher, who can assess their current perfor-mance level and design training activities to improvecertain aspects of performance. In these cases, it will beobvious that performance has improved from one weekto the next if the students are able to reach a highertarget performance, which had previously been outsidethe range of their performance ability.

After years of daily practice, the aspiring expert per-formers become able to monitor their performance sothey can start taking over the evaluative activity of theteacher and coach. They acquire and refine mechanismsthat permit increased control, which allow them tomonitor performance in representative situations toidentify errors as well as improvable aspects.23,24 Thereis compelling evidence for these complex cognitivemechanisms from studies in expert performance. Forexample, chess masters can select the best move for achess position. When the chess position is removed,they are able to report their thoughts and also recallthe locations of all the pieces on the chess boardvirtually perfectly. The superior incidental memory ofexperts for relevant information for representative taskshave been demonstrated in a large number of domains,such as sports, music, ballet, and medicine.33,34 Whenexpert performers are working on appropriately chal-lenging tasks, there is compelling evidence that theiractions are cognitively mediated.23

Figure 3. Illustration of the qualitative difference between thecourse of improvement of expert performance and of everydayactivities. The goal for everyday activities is to reach as rapidlyas possible a satisfactory level that is stable and ‘‘autonomous.’’After individuals pass through the ‘‘cognitive’’ and ‘‘associa-tive’’ phases, they can generate their performance virtuallyautomatically with a minimal amount of effort (see the gray ⁄ -white plateau at the bottom of the graph). In contrast, expertperformers counteract automaticity by developing increasinglycomplex mental representations to attain higher levels of con-trol of their performance and will therefore remain within thecognitive and associative phases. Some experts will at somepoint in their career give up their commitment to seeking excel-lence and thus terminate regular engagement in deliberatepractice to further improve performance, which results inpremature automation of their performance (adapted fromEricsson32).

ACAD EMERG MED • November 2008, Vol. 15, No. 11 • www.aemj.org 991

Once a professional reaches an acceptable skill level,more experience does not, by itself, lead to improve-ments. For example, tennis players will not improvetheir backhand volley in tennis by playing more games.However, a tennis coach can provide opportunities forDP. Initially, the coach can place relatively easy volleys,followed by increasingly unpredictable and difficultones, and then later, integrate rallies with backhandvolleys into regular game contexts. High-fidelity simula-tors can provide beginning and advanced EPs withopportunities to improve their performance of medicalprocedures.

How is it possible to improve one’s ability to planand select the best action in a given situation? Chessplayers typically solve this problem by studying pub-lished games from chess tournaments between thevery best players in the world. They play through thegames, one move at the time, to select the best movefor a given position before they look at the move madeby the chess master. If their move matches the corre-sponding move selected by the masters, then there isnothing obvious to change. If, on the other hand, thechess master’s move differed from their own selection,it would imply that their planning and evaluation over-looked some aspect of the position. By more carefuland extended analysis, the aspiring player is generallyable to discover the reasons for the chess master’smove. Most importantly, they can then reflect on theirthought process during their faulty move selection andexamine how they need to change their planningmethods when encountering other related situations infuture games. Similarly, it should be possible to collectauthentic information about medical cases encounteredby surgeons, general physicians and EPs. With thebenefit of hindsight, it should be possible to recreatethe original patient encounter in a time-constrainedcontext and have the aspiring expert doctor maketreatment decisions and get immediate feedback bycomparing their decisions with those of expert physi-cians. Extensive research on how speed of perfor-mance can be increased through DP has beenconducted on typing.35 The amount of daily typing bycollege students is essentially unrelated to current typ-ing speed during a test. The key findings are that spe-cial training and efforts to increase typing speed arepredictive of the typing speed attained. More gener-ally, individuals can improve typing speed by pushingthemselves, as long as they can maintain full concen-tration. Straining to type at a faster speed—typicallyaround 10%–20% faster than their normal speed—facil-itates typists to better anticipate the copy, possibly byextending their gaze further ahead. Similarly, athleteswho push themselves overload their muscles, whichenact gene expressions that lead to physiologic andanatomic changes, which increase their strength andpower.36,37

Several studies and reviews23,24 report a consistentrelationship between the amount and quality of solitaryactivities meeting the criteria of DP and performance ina wide range of domains of expertise. To reach a levelwhere one can win international competitions, it is esti-mated that over 10,000 hours of DP have been gener-ated for several domains.

PROFESSIONAL EXPERTS’ RELIANCE ONINTUITION VERSUS DP WITH CHALLENGINGCASES

Expert performers are able to report their thoughtprocesses and critical aspects of the encountered situa-tions.23,24 This seems to be inconsistent with the claimsfor experts’ use of intuition: ‘‘When things proceednormally, experts don’t solve problems and don’t makedecisions, they do what normally works.’’11 Commonsituations, such as treating an otherwise healthy adultfor a cold or flu, can be handled successfully by virtu-ally anyone, and expert performers would not beexpected to demonstrate greater treatment success thanother medical service providers. According to theexpert performance approach, we need to study chal-lenging task situations with reliable individual differ-ences in performance. Challenging task situations haverarely or even never been experienced first hand. In thedomain of medicine, this would involve treatingpatients with poor prognosis, rare diseases, and condi-tions involving the interaction of several medical prob-lems and medications. For these types of challengingpatients, there are professionals who provide treat-ments with superior outcomes than their colleagues.Some of these professionals’ performance can be linkedto special training and practice motivated by increasedstandards, such as certification in specialties.30

It is not immediately obvious how other professionalsgain their performance advantage. When people areengaged in professional activities, public performances,and competitions, it is difficult to engage in learningand training because the priority is on performing con-sistently at a high level. Even if an individual is awareof a mistake or a failed step, it is not possible to stopthe ongoing activity during a public music perfor-mance. Instead, the professionals need to proceed andmake any necessary adjustments to minimize the per-ceptible effects of the disruption and maximize thechances for a successful overall outcome. There aresome medical domains, such as surgery, where mis-takes lead to observable consequences that need to beimmediately addressed. Interestingly, in surgery thereis evidence for improvements in performance as a func-tion of surgeries of a given type.29,38

In a professional environment with real-timedemands, it is generally necessary to wait until the endof ongoing activity before one is able to reflect on howthe mistake happened and what could be changed toavoid a similar, future problem. In other cases, the pro-fessionals may notice the problem, such as internalbleeding in surgery, during the activity. If the bleedingproblem is discovered hours later, it may be difficult forthe professionals to remember what they were thinkingat the time the actual problem occurred.

In an ideal learning environment based on simula-tions and re-created old cases, performers can waituntil they are fully rested before confronting challeng-ing situations. Based on our discussions of measure-ment of performance, students and advancedperformers should be presented with cases just abovetheir current level of ability. Ideally, the cases shouldalso have known correct actions—where retrospective

992 Ericsson • DELIBERATE PRACTICE AND ACQUISITION OF EXPERT PERFORMANCE

analysis has reveals the best courses of appropriateaction. The best training situations focus on activities ofshort duration with opportunities for immediate feed-back, reflection, and corrections. Each completed trialshould be followed by another similar brief task withfeedback, until this type of task is completed with con-sistent success. At this point of mastery, the activityshould be embedded in more complex contexts andalternated with other types of cases until the skill hasbeen integrated in the performer’s repertoire. (For amore extended and detailed discussion on training inmedical simulators and DP see the artc3e by McGag-hie39.)

CONCLUSION

Based on recent advances in the scientific analysis ofreproducibly superior (expert) performance, we knowthat superior performance does not automaticallydevelop from extensive experience, general education,and domain-related knowledge. Superior performancerequires the acquisition of complex integrated systemsof representations for the execution, monitoring, plan-ning, and analyses of performance. Educators shouldtherefore create training opportunities for DP, appro-priate for a given individual at given level of skill devel-opment. Performers may then make the necessaryadjustments to improve specific aspects of performanceto assure that attained changes will be successfully inte-grated into representative performance.

References

1. Robb FC. Aristotle and education. Peabody J Educ.1943; 20:202–213.

2. Plato. Laws. The Dialogues of PLATO (846–847)[Translated by Benjamin Jowett]. EncyclopaediaBritanica, Chicago, 1952.

3. Anderson LF. Some facts regarding vocationaltraining among the ancient Greeks and Romans.School Rev. 1912; 20:191–201.

4. Plato. Plato Laws. The Dialogues of PLATO (720)[Translated by Benjamin Jowett]. EncyclopaediaBritanica, Chicago, 1952.

5. Plato. Charmides. The Dialogues of PLATO (172)[Translated by Benjamin Jowett]. EncyclopaediaBritanica, Chicago, 1952.

6. Aristotle. Metaphysics. The works of ARISTOTLE(982a29–30) [Translated under the editorship of WDRoss]. Encyclopaedia Britanica, Chicago, 1952.

7. Plato. Republic. The Dialogues of PLATO (408)[Translated by Benjamin Jowett]. EncyclopaediaBritanica, Chicago, 1952.

8. Ericsson KA. An introduction to Cambridge Hand-book of Expertise and Expert Performance: its devel-opment, organization, and content. In: EricssonKA, Charness N, Feltovich P, Hoffman RR (eds).Cambridge Handbook of Expertise and ExpertPerformance. Cambridge, UK: Cambridge UniversityPress, 2006: pp. 3–19.

9. Benner P. From Novice to Expert: Excellence andPower in Clinical Nursing Practice. Upper SaddleRiver, NJ: Prentice Hall Health, 1984.

10. Dreyfus SE, Dreyfus HL. A five-state model of themental activities involved in directed skill acquisi-tion. University of California, Berkely, CA: Opera-tion Research Center Report 80-2, 1980.

11. Dreyfus HL, Dreyfus SE. Mind Over Machine: ThePower of Human Intuition and Expertise in the Eraof the Computer. New York: The Free Press, 1986.

12. Fitts P, Posner MI. Human Performance. Belmont,CA: Brooks ⁄ Cole, 1967.

13. Simon HA, Chase WG. Skill in chess. Am Sci. 1973;61:394–403.

14. Bryan WL, Harter N. Studies on the telegraphiclanguage: the acquisition of a hierarchy of habits.Psychol Rev. 1899; 6:345–375.

15. Elstein AS, Shulman LS, Sprafka SA. Medical Prob-lem Solving: An Analysis of Clinical Reasoning.Cambridge, MA: Harvard University Press, 1978.

16. Ericsson KA. An expert-performance perspective ofresearch on medical expertise: the study of clinicalperformance. Med Educ. 2007; 41:1124–30.

17. Camerer CF, Johnson EJ. The process-performanceparadox in expert judgment: how can the expertsknow so much and predict so badly? In: EricssonKA, Smith J (eds). Towards a General Theory ofExpertise: Prospects and Limits. Cambridge: Cam-bridge University Press, 1991, pp 195–217.

18. Dawes RM. House of Cards: Psychology and Psy-chotherapy Built on Myth. New York: Free Press,1994.

19. Ericsson KA, Smith J. Prospects and limits in theempirical study of expertise: an introduction. In:Ericsson KA, Smith J (eds). Toward a General The-ory of Expertise: Prospects and Limits. Cambridge:Cambridge University Press, 1991, pp 1–38.

20. Ericsson KA. The acquisition of expert perfor-mance: an introduction to some of the issues. In:Ericsson KA (ed). The road to excellence: the acqui-sition of expert performance in the arts andsciences, sports, and games. Mahwah, NJ:Lawrence Erlbaum, 1996, pp 1–50.

21. de Groot A. Thought and Choice in Chess. TheHague: Mouton (Original work published 1946),1978.

22. Ericsson KA, Lehmann AC. Expertise. In: RuncoMA, Pritzer SR (eds). Encyclopedia of Creativity.San Diego, CA: Academic Press, 1990, pp 695–707.

23. Ericsson KA. Protocol analysis and expert thought:concurrent verbalizations of thinking duringexperts’ performance on representative task. In:Ericsson KA, Charness N, Feltovich P, Hoffman RR(eds). Cambridge Handbook of Expertise andExpert Performance. Cambridge, UK: CambridgeUniversity Press, 2006, pp 223–242.

24. Ericsson KA. The influence of experience and delib-erate practice on the development of superiorexpert performance. in: Ericsson KA, Charness N,Feltovich P, Hoffman RR (eds). Cambridge Hand-book of Expertise and Expert Performance. Cam-bridge, UK: Cambridge University Press, 2006, pp685–706.

25. Ericsson KA, Williams AM. Capturing naturally-occurring superior performance in the laboratory:

ACAD EMERG MED • November 2008, Vol. 15, No. 11 • www.aemj.org 993

translational research on expert performance. J ExpPsychol Appl. 2007; 13:115–123.

26. van der Maas HLJ, Wagenmakers EJ. A psychomet-ric analysis of chess expertise. Am J Psychol. 2005;118:29–60.

27. Ericsson KA, Lehmann AC. Expert and exceptionalperformance: evidence on maximal adaptations ontask constraints. Annu Rev Psychol. 1996; 47:273–305.

28. Choudhry NK, Fletcher RH, Soumerai SB. System-atic review: the relationship between clinical experi-ence and quality of health care. Ann Intern Med.2005; 142:260–73.

29. Ericsson KA. Deliberate practice and the acquisitionand maintenance of expert performance in medi-cine and related domains. Acad Med. 2004; 79(10Suppl):S70–81.

30. Ericsson KA, Whyte J, Ward P. Expert performancein nursing: reviewing research on expertise in nurs-ing within the framework of the expert-performanceapproach. ANS Adv Nurs Sci. 2007; 30:E58–1.

31. Ericsson KA, Krampe RT, Tesch-Römer C. The roleof deliberate practice in the acquisition of expertperformance. Psychol Rev. 1993; 100:363–406.

32. Ericsson KA. The scientific study of expert levels ofperformance: general implications for optimal learn-ing and creativity. High Abil Stud, 1990; 9:75–100.

33. Ericsson KA, Kintsch W. Long-term working mem-ory. Psychol Rev. 1995; 102:211–45.

34. Ericsson KA, Patel V, Kintsch W. How experts’adaptations to representative task demands accountfor the expertise effect in memory recall: commenton Vicente and Wang (1998). Psychol Rev. 2000;107:578–92.

35. Keith N, Ericsson KA. Evidence for a deliberatepractice account of typing proficiency in everydaytypists. J Exp Psychol Appl. 2007; 13:135–145.

36. Ericsson KA. Deliberate practice and the modifiabil-ity of body and mind: toward a science of thestructure and acquisition of expert and elite perfor-mance. Int J Sport Psychol. 2007; 38:4–34.

37. Ericsson KA. Deliberate practice and the modifiabil-ity of body and mind: a reply to the commentaries.Int J Sport Psychol. 2007; 38:109–123.

38. Seymour NE, Gallagher AG, Roman SA, O’BrienMK, Bansal VK, Anderson DK, Satava RM. Virtualreality training improves operating room perfor-mance. Ann Surg. 2002; 236:458–64.

39. McGaghie WC. Research opportunities in simula-tion-based medical education using deliberatepractice. Ann Emerg Med. 2008; in press.

994 Ericsson • DELIBERATE PRACTICE AND ACQUISITION OF EXPERT PERFORMANCE