Deliberate Practice - Excerpt From Handbook (2006)

P1: JzG052184097Xc38 CB1040B/Ericsson 0 521 84087 X February 28, 2006 6:17

C H A P T E R 38

The Influence of Experience andDeliberate Practice on the Development

of Superior Expert Performance

K. Anders Ericsson

There are several factors that influence thelevel of professional achievement. First andforemost, extensive experience of activi-ties in a domain is necessary to reach veryhigh levels of performance. Extensive expe-rience in a domain does not, however, invari-ably lead to expert levels of achievement.When individuals are first introduced to aprofessional domain after completing theirbasic training and formal education, theyoften work as apprentices and are supervisedby more-experienced professionals as theyaccomplish their work-related responsibili-ties. After months of experience they typi-cally attain an acceptable level of proficiency,and with longer experience, often years,they are able to work as independent pro-fessionals. At that time most professionalsreach a stable, average level of performance,and then they maintain this pedestrian levelfor the rest of their careers. In contrast, somecontinue to improve and eventually reachthe highest levels of professional mastery.

Traditionally, individual differences in theperformance of professionals have beenexplained by an account given by Galton(1869/1979, see Ericsson, 2003a, for a

description). According to this view, everyhealthy person will improve initially throughexperience, but these improvements areeventually limited by innate factors thatcannot be changed through training; henceattainable performance is constrained byone’s basic endowments, such as abili-ties, mental capacities, and innate talents.This general view also explains age-relateddeclines in professional achievement owingto the inevitable degradation of generalcapacities and processes with age (see alsoKrampe & Charness, Chapter 40). Morerecently, researchers of expert performancehave found that there are many types ofexperience and that these different typeshave qualitatively and quantitatively dif-ferent effects on the continued acquisitionand maintenance of an individual’s per-formance (Ericsson, 1996, 2002 ; Ericsson,Krampe, & Tesch-Romer, 1993). This frame-work proposes that some types of experi-ence, such as merely executing proficientlyduring routine work, may not lead to furtherimprovement, and that further improve-ments depend on deliberate efforts to changeparticular aspects of performance.

685


686 the cambridge handbook of expertise and expert performance

In this chapter I will review evidence onthe effects of experience and deliberate prac-tice on individual differences in the acqui-sition of skilled and expert performance. Iwill first describe the traditional account ofindividual differences in performance basedon experience and innate talent. Then I willreview evidence on the effects of varioustypes of experience on performance, espe-cially the effects of deliberate practice. Inthe last half of the chapter I will discusshow deliberate practice can account for thechanges in the structure of the mechanismsthat mediate the superior performance ofexperts.

The Traditional View of SkillAcquisition and ProfessionalDevelopment: History and SomeRecent Criticisms

Ideas about how experience and trainingcan explain individual differences in attainedlevel of performance have a long history. Thecontemporary view of lifespan development(Denney, 1982) is based on the assumptionthat children develop their abilities duringchildhood and can reach their innate poten-tial under favorable experiential conditions.Further, the general view is that the indi-vidual’s potential is limited by innate bio-logical capacities that will ultimately con-strain the highest level of achievement. SirFrancis Galton is often recognized for artic-ulating this view in the 19th century. Hispioneering book, Hereditary Genius (Galton,1869/1979), presented evidence that heightand body size were determined genetically.Most importantly, he argued that innatemechanisms also regulated size and char-acteristics of internal organs, such as thenervous system and the brain, and thusmust similarly determine mental capacities.Galton (1869/1979) clearly acknowledgedthe need for training and practice to reachhigh levels of performance in any domain.However, he argued that improvements ofperformance for mature adults are rapid onlyin the beginning of training and that sub-sequent increases diminish, until “Maximal

performance becomes a rigidly determinatequantity” (p. 15). According to Galton, therelevant heritable capacities determine theupper bound for the performance that anindividual can attain through practice, andreflect the immutable limit that “Nature hasrendered him capable of performing” (p. 16).According to Galton, the characteristics thatlimit maximal performance after all benefitsof training have been gained must, there-fore, be innately endowed. Galton’s argu-ments for the importance of innate factorsfor attaining the highest levels of perfor-mance were compelling and, thus, have hada lasting impact on our culture’s view of abil-ity and expertise.

Contemporary theories of skill acquisi-tion (Anderson, 1982 ; Fitts & Posner, 1967)are consistent with Galton’s general assump-tions about basic unmodifiable capacitiesand with observations on the general courseof professional development. When individ-uals are first introduced to a skilled activitysuch as driving a car, typing on a computer,or playing golf, their primary goal is to reacha level of proficiency that will allow themto perform these everyday tasks at a func-tional level. During the first phase of learn-ing (Fitts & Posner, 1967), beginners try tounderstand the requirements of the activ-ity and focus on generating actions whileavoiding gross mistakes. This phase is illus-trated in the lower arrow in Figure 38.1. Inthe second phase of learning, when peoplehave had more experience, noticeable mis-takes become increasingly rare, performanceappears smoother, and learners no longerneed to focus as intensely on their perfor-mance to maintain an acceptable level. Aftera limited period of training and experience –frequently less than 50 hours for most every-day activities such as typing, playing tennis,and driving a car – an acceptable level ofperformance is typically attained. As indi-viduals adapt to a domain during the thirdphase of learning, their performance skillsbecome automated, and they are able to exe-cute these skills smoothly and with minimaleffort (as is illustrated in the lower arrow inFigure 38.1). As a consequence of automati-zation, performers lose the ability to control


experience and deliberate practice 687

Figure 38.1. An illustration of the qualitative difference between the course ofimprovement of expert performance and of everyday activities. The goal foreveryday activities is to reach as rapidly as possible a satisfactory level that isstable and “autonomous.” After individuals pass through the “cognitive” and“associative” phases, they can generate their performance virtually automaticallywith a minimal amount of effort (see the gray/white plateau at the bottom ofthe graph). In contrast, expert performers counteract automaticity bydeveloping increasingly complex mental representations to attain higher levelsof control of their performance and will therefore remain within the “cognitive”and “associative” phases. Some experts will at some point in their career give uptheir commitment to seeking excellence and thus terminate regular engagementin deliberate practice to further improve performance, which results inpremature automation of their performance. (Adapted from “The scientificstudy of expert levels of performance: General implications for optimal learningand creativity” by K. A. Ericsson in High Ability Studies, 9, p. 90. Copyright 1998

by European Council for High Ability.)

the execution of those skills, making inten-tional modifications and adjustments diffi-cult (see Hill & Schneider, Chapter 37).In the automated phase of learning, perfor-mance reaches a stable plateau, and no fur-ther improvements are observed – in agree-ment with Galton’s (1869/1979) assumptionof a performance limit.

Similar phases of acquisition and autom-atization have been shown to account fordevelopment in professional domains, suchas telegraphy (Bryan & Harter, 1897, 1899)and typing (Book, 1925a, 1925b). Whereasinitial proficiency in everyday and profes-sional skills may be attained within weeksand months, development to very high lev-els of achievement appear to require manyyears or even decades of experience. In fact,Bryan and Harter claimed already in 1899

that over ten years are necessary for becom-ing an expert. In their seminal theory of

expertise, Simon and Chase (1973) proposedthat future experts gradually acquired pat-terns and knowledge about how to reactin situations by storing memories of theirpast actions in similar situations. Hence, per-formance is assumed to improve as a con-sequence of continued experience. Chaseand Simon’s (1973) research on chess mas-ters extended the pioneering work by deGroot (1946/1978) and demonstrated thatthe masters’ recall for briefly presented reg-ular game positions was vastly superior toless-skilled players. Simon and Chase (1973)argued that the masters must have acquiredsome 50,000 chunks or patterns to enablethem to retrieve the appropriate moves forthe current position in a chess game. Theyhighlighted the parallels between reachingthis highly skilled performance in chessand acquiring other cognitive skills, such asspeaking a foreign language with its large



vocabulary of many thousands of words.They found that players must have playedchess for at least ten years before they areable to win international chess tournaments.In a similar vein, every healthy child requiresmany years of experience of listening andspeaking before they are able to master theirfirst language with its extensive vocabulary.

Some scientists started to consider thepossibility that expertise was an automaticconsequence of lengthy experience, and theyconsidered individuals with over ten yearsof full-time engagement in a domain to beexperts. These scientists typically viewedexpertise as an orderly progression fromnovice to intermediate and to expert, wherethe primary factors mediating the progres-sion through these stages were instruction,training, and experience. Thus, the primarycriteria for identifying experts were socialreputation, completed education, accumu-lated accessible knowledge, and length ofexperience in a domain (over ten years)(Chi, Glaser, & Farr, 1988; Hoffman, 1992).

Several reviews over the past decade(Ericsson et al., 1993 ; Ericsson & Kintsch,1995 ; Ericsson & Lehmann, 1996; Ericsson& Smith, 1991; Vicente & Wang, 1998)have raised issues about this characteriza-tion of expertise. Most importantly, whenindividuals, based on their extensive experi-ence and reputation, are nominated by theirpeers as experts, their actual performance isoccasionally found to be unexceptional. Forexample, highly experienced computer pro-grammers’ performance on programmingtasks is not always superior to that of com-puter science students (Doane, Pellegrino, &Klatzky, 1990), and physics professors fromUC Berkeley were not always consistentlysuperior to students on introductory physicsproblems (Reif & Allen, 1992). More gen-erally, level of training and experience fre-quently has only a weak link to objectivemeasures of performance. For example, thelength of training and professional experi-ence of clinical psychologists is not relatedto their efficiency and success in treatingpatients (Dawes, 1994), and extensive expe-rience with software design is not associatedwith consistently superior proficiency on

presented tasks (Rosson, 1985 ; Sonnentag,1998). Similarly, when wine experts arerequired to detect, describe, and discrimi-nate characteristics of a wine without knowl-edge of its identity (i.e., seeing the label onthe bottle), their performance is only slightlybetter than those generated by regular winedrinkers (Gawel, 1997; Valentin, Pichon, deBoishebert, & Abdi, 2000). More gener-ally, reviews of decision making (Camerer& Johnson, 1991; Shanteau & Stewart, 1992)show that experts’ decisions and forecasts,such as financial advice on investing instocks, do not show a reliable superiorityover novices and thus must not improve sim-ply with added experience. Similar absenceof improvement by experienced individu-als considered experts has been documentedin several other areas (Ericsson & Lehmann,1996; Ericsson, 2004). There are even exam-ples, such as diagnosis of heart sounds and x-rays by general physicians (Ericsson, 2004)and auditor evaluations (Bedard & Chi,1993), in which performance decreases sys-tematically in accuracy and consistency withthe length of professional experience afterthe end of formal training.

Once it is clear that social and sim-ple experience-based indicators of expertisedo not guarantee superior performance, analternative approach is required. Ericssonand Smith (1991) proposed that the focusshould not be on socially recognized experts,but rather on individuals who exhibit repro-ducibly superior performance on representa-tive, authentic tasks in their field. For exam-ple, the focus should be on physicians whocan diagnose and treat patients in a supe-rior manner, on chess players who can con-sistently select the best moves for chess posi-tions, and on athletes and musicians whoexhibit superior performance in competi-tions. The first step in a science of expert per-formance requires that scientists be able tocapture, with standardized tests, the repro-ducibly superior performance of some indi-viduals, and then be able to examine this per-formance with laboratory methods, as willbe described in the next sections (see also,Ericsson, Chapter 13 , for a more detailedtreatment).



Reproducibly Superior Performanceand Experience

In many domains of expertise, individualshave been interested in assessing and com-paring levels of performance under fair andcontrolled circumstances. For thousands ofyears athletes have competed under highlystandardized conditions in track and fieldevents, such as running, jumping, and throw-ing. These competitive conditions approachthe controlled conditions generated in mod-ern studies of performance in the labora-tory. In a similar manner, musicians, dancers,and chess players have a long history of dis-playing their performance under controlledconditions during competitions and tour-naments. Such competitions, together withsimilar tests, such as auditions, serve severalpurposes beyond identifying the best per-formers and presenting awards. For youngerand developing performers, successful per-formance at competitions and auditions isnecessary to gain access to the best teachersand training environments.

Ericsson and Smith (1991) discussed howone could use similar techniques to mea-sure various types of professional exper-tise. We argued that a complete under-standing of the structure and acquisition ofexcellence will be possible only in domainsin which experts exhibit objectively supe-rior performance, in a reproducible manner,for the representative activities that definethe essence of accomplishment in a givendomain (Ericsson, 1996, 2002). Expert per-formers are accustomed to performing inresponse to external demands, such as dur-ing emergencies in their professional prac-tice, or at competitions and exhibitions. Ifthey are able to reproduce their performancerepeatedly on these types of occasions aswell during training, they should be ableto reproduce them even under laboratoryconditions, a finding confirmed by recentresearch (Ericsson & Lehmann, 1996).

Unfortunately, expert performanceoccurs naturally in complex and uniquecontexts, where the conditions of perfor-mance differ between performers, makingcomparison difficult. For example, musi-

cians select their own pieces of music fortheir performance. Similarly, the sequenceof moves in a chess game is never thesame and, thus, players never encounterthe exact same positions during the mid-dle game. Fortunately, most domains ofexpertise require that experts be able toexhibit superior performance for presentedrepresentative situations. Ericsson and Smith(1991; Ericsson, 1996) proposed a way to findrepresentative situations that capture theessence of expert performance in a domainand call for immediate action. They alsodescribed general methods for recreatingthese situations in the laboratory and theninstructing experts and less skilled individ-uals to reproduce their performance undercontrolled laboratory conditions, so thatinvestigators can identify the responsiblemediating mechanisms.

Representative tasks that have beenfound to capture the essence of expertise inthree domains are illustrated in Figure 38.2(see treatments of these and other fields inSections 5). In each example, the measuredperformance is closely related to the natu-rally occurring performance. To study chessexpertise, players at different skill levels areasked to generate the best move for the samechess positions that have been taken fromactual games between chess masters, whichare not publicly available. Different typistsare presented the same material and askedto type as much as possible during a fixedtime period. Musicians are asked to playfamiliar or unfamiliar pieces of music whilebeing recorded, and are then asked to repeattheir performance exactly. When musiciansare instructed to repeat their original per-formance, experts’ consecutive renditionsshow much less variation than renditions byless-skilled musicians and, by implication,experts exhibit greater control over theirperformance.

When a review of evidence is restricted toonly the reproducible superior performanceof experts, obtained under directly compa-rable conditions, it is possible to examineseveral claims about the relation betweenexpert performance and experience thatgeneralize across domains. First, extensive



Figure 38.2 . Three examples of laboratory tasks that capture the consistently superior performanceof domain experts in chess, typing, and music. (From “Expertise,” by K. A. Ericsson and Andreas C.Lehmann, 1999, Encyclopedia of Creativity. Copyright by Academic Press.)

experience is shown to be necessary to attainsuperior expert performance. Second, onlysome types of domain-related experience areshown to lead to improvement of perfor-mance. In addition, many thousands of hoursof specific types of practice and training havebeen found to be necessary for reaching thehighest levels of performance.

The Necessity of Domain-SpecificExperience for Attaining ReproduciblySuperior Performance

Recent reviews (Ericsson, 1996; Ericsson& Lehmann, 1996) show that extendedengagement in domain-related activities isnecessary to attain expert performance inthat domain. The availability of standardizedtests allows us to measure the level of per-formance during development and to com-pare these longitudinal data to uniform adultstandards. Hence, we can describe the devel-opment of expert performance as a func-tion of age and years of experience, as fol-

lows. First, longitudinal assessments of per-formance reveal that all individuals improvegradually, as illustrated in Figure 38.3 . Thereis no objective evidence that a child oradult is able to exhibit a high level of per-formance without any relevant prior expe-rience and practice. Similarly, there is noevidence for abrupt improvements of repro-ducible performance when it is tested ona monthly or yearly basis. When the per-formance of child prodigies in music andchess are measured against adult standards,they show gradual, steady improvementover time. Second, elite performance keepsimproving beyond the age of physical matu-ration – the late teens in industrialized coun-tries (Ulijaszek, Johnston, & Preece, 1998) –and is, thus, not directly limited by thefunctional capacity of the body and brain.Peak performance of experts is nearly alwaysattained in adulthood – many years, andeven decades, after initial exposure to thedomain, as illustrated in Figure 38.3 . Theage at which performers typically reach their



Figure 38.3 . An illustration of the gradual increases in expert performance as afunction of age, in domains such as chess. The international level, which isattained after more than around ten years of involvement in the domain, isindicated by the horizontal dashed line. (From “Expertise,” by K. A. Ericsson andAndreas C. Lehmann, 1999, Encyclopedia of Creativity. Copyright by AcademicPress.)

highest level of performance in many vigor-ous sports is the mid- to late 20s. For thearts and science, it is a decade later, in the30s and 40s (see Schulz & Curnow, 1988,and Simonton, 1997, for reviews). The con-tinued and often extended development ofexpertise past physical maturity shows thatadditional experience is necessary to attainone’s highest level of performance.

Finally, the most compelling evidencefor the role of vast experience in expertisecomes from investigators who have shownthat, even for the most talented individ-uals, ten years of experience in a domain(ten-year rule) is necessary to become anexpert (Bryan & Harter, 1899) and to winat an international level (Simon & Chase,1973). Subsequent reviews have shown thatthese findings extend to international-levelsuccess in music composition (Hayes, 1981),as well as to sports, science, and the arts(Ericsson, Krampe, & Tesch-Romer, 1993).A closer examination of the evidence forthe ten-year rule shows that the numberten is not magical. In fact, the number ofyears of intense training required to becomean internationally acclaimed performer dif-

fers across domains. For example, elite musi-cians (disregarding the biased standards forchild prodigies) need closer to 20 to 30 yearsof training and often peak when they arearound 30 to 40 years old. Further, outstand-ing scientists and authors normally pub-lished their first work at around age 25 , andtheir best work follows around ten years later(Raskin, 1936).

Other investigators have pointed topotential exceptions to the ten-year rule.Some of the exceptions are so close to theten-year rule that they support the neces-sity for around ten years to win at the inter-national level. For example, famous chessplayer Bobby Fischer required nine yearsof intense chess study before being recog-nized as a grand master in chess at age 16

(Ericsson et al, 1993). Other examples sug-gest clearer violations. Very tall basketballplayers (around seven feet) have been ableto reach the highest professional ranks inless than ten years of training – in aroundsix years. Research on training of memoryexperts has shown that individuals can reachthe highest level in the world after less thana couple of years training (Ericsson, 2003b;



Ericsson, Delaney, Weaver, & Mahadevan,2004). More generally, people are able toreach world-class levels in fewer than tenyears in activities that lack a history of orga-nized international competition. In addition,there is solid evidence that the highest levelsfor performance in a given domain are notstable but sometimes continue to increaseover historical time as a function of pro-gressively higher and more effective levelsof training and practice.

Increases in Performance over HistoricalTime: The Relation between Performanceand Improved Methods of Practice

In virtually every aspect of human activ-ity there have been increases in the effi-ciency and level of performance. Over cen-turies and millennia, across domains ofexpertise, people have developed methodsfor accumulating and preserving discoveredknowledge and skills and produced toolsand refined their technique of application.Hence, they have assembled a body of orga-nized knowledge that can be transferredfrom the current to the next generationthrough instruction and education (Ericsson,1996; Feldman, 1994). It is no longer nec-essary for individuals to discover the rel-evant knowledge and methods by them-selves. Today’s experts can rapidly acquirethe knowledge originally discovered by thepioneers. For example, in the 13 th centuryRoger Bacon argued that, using the then-known methods of learning (self-study), itwould be impossible to master mathemat-ics in less than 30 to 40 years (Singer,1958). Today, the roughly equivalent mate-rial (calculus) is taught in highly orga-nized and accessible form in every highschool. Today the development of expertlevels of achievement requires instruction byteachers that helps performers gain accessto the body of domain-specific knowledge,which is expressed and accumulated interms of predefined concepts, notation sys-tems, equipment, and measurement devices.The increases in the level of expert per-formance over historical time are oftentaken for granted in science and sports, but

the improvements in instrumentation andequipment make it difficult to find compara-ble tasks over large time-spans in which per-formance can be directly compared. How-ever, in domains with fewer changes in toolsand instruments, such as music performancewith the piano and violin, today’s perform-ers readily master music that was consid-ered unplayable by the best musicians inthe 19th century. They can match or ofteneven surpass the technical virtuosity of leg-endary musicians and music prodigies of thepast, such as Wolfgang Amadeus Mozart(Lehmann & Ericsson, 1998).

In sports, the increases in performanceover time are well known and even todayworld records are broken on a regular basis.In some events, such as the marathon, swim-ming, and diving, many dedicated amateursand college athletes perform at a muchhigher level in the 21st century than the goldmedal winners of the early Olympic Games.For example, after the IVth Olympic Gamesin 1908, organizers almost prohibited thedouble somersault in dives because theybelieved that these dives were dangerous,and no human would ever be able to controlthem. More generally, record-breakinglevels of performance are nearly alwaysoriginally attained by only a single eminentperformer. However, after some time, otherathletes are able to design training methodsthat allow them to attain that same levelof performance. Eventually, these trainingmethods become part of regular instruction,and all elite performers in the domain areexpected to attain the new higher standard.Perhaps the most well-known example isRoger Bannister’s first ever sub-four-minutemile. The earlier record for the mile hadbeen viewed as the ultimate limit forperformance, but after Bannister broke thefour-minute barrier, several other runnerswere able to do so within a couple of years(Denison, 2003). Over time, differencesin practice methods have become so greatthat Olympic swimmers from early in thelast century would not even qualify forswim teams at today’s competitive highschools (Schulz & Curnow, 1988). In somecompetitive domains, such as baseball, it is



sometimes difficult to demonstrate theincreased level of today’s performersbecause both the level of the pitcher andbatter has improved concurrently (Gould,1996). In spite of the increases in the averagelevel of elite performance over historicaltime, the variability in individual differencesin athletes’ performance remains large –a topic that will be addressed in the nextsection.

From Experience to Designed Practice

Many individuals seem satisfied in reachinga merely acceptable level of performance,such as amateur tennis players and golfers,and they attempt to reach such a levelwhile minimizing the period of effortful skillacquisition. Once an acceptable level hasbeen reached, they need only to maintaina stable performance, and often do so withminimal effort for years and even decades.For reasons such as these, the length of expe-rience has been frequently found to be aweak correlate of job performance beyondthe first two years (McDaniel, Schmidt, &Hunter, 1988). In addition, extensive watch-ing is not the same as extensive playing.Williams and Davids (1995), for example,found large differences in the ability to antic-ipate events in soccer between players andavid spectators. The select group of individ-uals who eventually reach very high levels donot simply accumulate more routine experi-ence of domain-related activities but extendtheir active skill-building period for years oreven decades, both forward and backwardin time. In particular, from retrospectiveinterviews of international-level performersin many domains, Bloom (1985a; see chap-ter by Sosniak, Chapter 16) showed thatelite performers are typically introduced totheir future realm of excellence in a play-ful manner at a young age. As soon as theyenjoy the activity and show promise com-pared to peers in the neighborhood, theirparents help them seek out a teacher andinitiate regular practice. Bloom and his col-leagues (Bloom 1985b) demonstrated thatperformers that reach an international level

have received remarkable support by theirparents and teachers (see also Mieg, Chap-ter 41). The parents of the future elite per-formers were even found to spend large sumsof money for teachers and equipment, and todevote considerable time to escorting theirchild to training and weekend competitions.In some cases, the performers and their fam-ilies even relocate to be closer to the chosenteacher and the training facilities. Based ontheir interviews, Bloom (1985a) argued thataccess to the best training resources was nec-essary to reach the highest levels.

At the same time the best training envi-ronments are not sufficient to produce thevery best performers, and there are sub-stantial individual differences even amongindividuals in these environments. Can dif-ferences in the amount and type of domain-related activities that individuals engage inexplain individual differences in music per-formance even among the best performers?Expert violinists at the music academy inBerlin kept a weekly diary on how muchtime they spent during a week on differentactivities (Ericsson et al., 1993). All groups ofexpert violinists were found to spend aboutthe same amount of time (over 50 hours)per week on music-related activities. How-ever, the best violinists were found to spendmore time per week on activities that hadbeen specifically designed to improve per-formance, which we call “deliberate prac-tice.” A prime example of deliberate prac-tice is the expert violinists’ solitary practicein which they work to master specific goalsdetermined by their music teacher at weeklylessons. The same groups of expert violin-ists, along with a group of professional vio-linists from world-class symphony orches-tras, were also interviewed to estimate theamount of deliberate practice in which theyhad engaged during their musical develop-ment. Even among these elite groups wewere able to find that the most accomplishedmusicians had spent more time in activi-ties classified as deliberate practice duringtheir development. Figure 38.4 shows thatthese differences were reliably observablebefore their admittance to the academy ataround age 18. By the age of 20, the best



Figure 38.4. Estimated amount of time for solitary practice as a function of agefor the middle-aged professional violinists (triangles), the best expert violinists(squares), the good expert violinists (empty circles), the least accomplished expertviolinists (filled circles), and amateur pianists (diamonds). (From “The role ofdeliberate practice in the acquisition of expert performance,” by K. A. Ericsson,R. Th. Krampe, and C. Tesch-Romer, 1993 , Psychological Review, 100(3), p. 379

and p. 384 . Copyright 1993 by American Psychological Association. Adaptedwith permission.)

musicians had spent over 10,000 hours prac-ticing, which averages 2 ,500 and 5 ,000 hoursmore than two less-accomplished groups ofmusicians at the same academy, respectively(Ericsson et al., 1993). In comparison to ama-teur pianists of the same age (Krampe &Ericsson, 1996), the best musicians from theacademy and the professionals had practiced8,000 more hours.

The core assumption of deliberate prac-tice (Ericsson, 1996, 2002 , 2004 ; Ericssonet al., 1993) is that expert performanceis acquired gradually and that effectiveimprovement of performance requires theopportunity to find suitable training tasksthat the performer can master sequentially –typically the design of training tasks andmonitoring of the attained performance isdone by a teacher or a coach. Deliberatepractice presents performers with tasks thatare initially outside their current realm ofreliable performance, yet can be masteredwithin hours of practice by concentratingon critical aspects and by gradually refiningperformance through repetitions after feed-

back. Hence, the requirement for concentra-tion sets deliberate practice apart from bothmindless, routine performance and playfulengagement, as the latter two types of activ-ities would, if anything, merely strengthenthe current mediating cognitive mechanismsrather than modify them to allow increasesin the level of performance. Research iscurrently reevaluating claims that someindividuals can improve their level of per-formance without concentration and delib-erate practice. Even the well-known fact thatmore “talented” children improve faster inthe beginning of their music developmentappears to be in large part due to the fact thatthey spend more time in deliberate prac-tice each week (Sloboda, Davidson, Howe& Moore, 1996). In a recent study of singersGrape, Sandgren, Hansson, Ericsson, andTheorell (2003) revealed reliable differencesof skill in the level of physiological and psy-chological indicators of concentration andeffort during a singing lesson. Whereas theamateur singers experienced the lesson asself-actualization and an enjoyable release



of tension, the professional singers increasedtheir concentration and focused on improv-ing their performance during the lesson.

In other domains it has been more diffi-cult to isolate practice activities that meetall the criteria for deliberate practice inmusic. In sports, several studies have founda consistent relation between attained per-formance and amount of practice (Helsen,Starkes, & Hodges, 1998; Hodges & Starkes,1996; Starkes et al., 1996). In recent reviewsof deliberate practice in sports (Cote,Ericsson, & Law, 2005 ; Ericsson, 2003c;Ward, Hodges, Williams, & Starkes, 2004),several issues have been discussed concern-ing the relation between different domain-related practice activities and improvementsin performance. In a study of insuranceagents Sonnentag and Kleinc (2000) foundthat engagement in deliberate practice pre-dicted higher performance ratings.

For example, whereas solitary trainingwas found to distinguish elite and less-skilledperformers in some sports, the amountof time spent in team-related deliberatepractice activities correlates reliably withskill level in team sports (Helsen et al.,1998; Ward et al., 2004). Contrary to someevidence suggesting that playful activities,sporting diversity, and late specialization areassociated with elite level sport, a quasi-longitudinal study by Ward et al. (2004)demonstrated that elite-level youth soccerplayers did not spend more time in play-ful activities or in other sports or activitiesthan their less-skilled counterparts, nor didthey specialize any later. Instead, while less-skilled players spent the majority of theirtime in “play,” elite players spent signifi-cantly longer per week and accrued moretotal time in deliberate practice. They per-ceived themselves to be more competentthan the less-skilled players, and rated oneof their parents as the most influential per-son in their career.

Rare longitudinal studies of elite perform-ers (some of them world class, Schneider,1993) have found that the most potentvariables linked to performance and futureimprovements of performance involved

parental support, acquired task-specific (inthis case, tennis) skills, and motivationalfactors including concentration. Similarlyin chess, Charness and his colleagues(Charness, Krampe, & Mayr, 1996; Charness,Tuffiash, Krampe, Reingold, & Vasyukova,2005) found that the amount of solitarychess study was the best predictor of chessskill, and when this factor was statisticallycontrolled, there was only a very small ben-efit from the number of games played inchess tournaments. Similar findings havebeen obtained by Duffy, Baluch, and Eric-sson (2004) for dart throwing. In a par-ticularly interesting study McKinney andDavis (2004) examined successful handlingof emergency situations during flying byexpert pilots. They found that if prior to theemergency event the expert pilots had prac-ticed the same emergency situation in thesimulator, they were reliably more successfulin dealing with the actual event. More gener-ally, Deakin, Cote, and Harvey (Chapter 17)review evidence on methods for recordingthe amount and structure of deliberate prac-tice, using diary methods and other kinds ofobservations.

In this handbook several chapters dis-cuss the role of deliberate practice in rela-tion to self-regulated learning (Zimmerman,Chapter 39), to successful training in simu-lators (Ward, Williams, & Hancock, Chap-ter 14), to maintained performance in olderexperts (Krampe & Charness, Chapter 40),and in creative activities (Weisberg, Chap-ter 42). Other chapters review evidenceon the relation between deliberate prac-tice and the development of expertise inmany domains, such as professional writing(Kellogg, Chapter 22), music performance(Lehmann & Gruber, Chapter 26), sports(Hodges, Starkes, & MacMahon, Chap-ter 27), chess (Gobet & Charness, Chap-ter 30), exceptional memory (Wilding &Valentine, Chapter 31), and mathematicalcalculation (Butterworth, Chapter 32). Thenext section will focus on the microstruc-ture of deliberate practice and how it leadsto changes in the mechanisms that mediateexpert performance.



Deliberate Practice and theAcquisition of Complex MechanismsMediating Expert Performance

The fundamental challenge for theoreti-cal accounts of expert performance is topropose how expert performers can avoidreaching a performance asymptote within alimited time period, as predicted by contem-porary theories of skill acquisition and exper-tise (Anderson, 1982 ; Fitts & Posner, 1967),and keep improving their performance foryears and decades.

In the introduction of this chapter thestages of everyday skill acquisition weredescribed. At the first encounter with a taskpeople focus on understanding it and care-fully generating appropriate actions, as illus-trated in the lower arm of the previously dis-cussed Figure 38.1. With more experience,individuals’ behaviors adapt to the demandsof performance and become increasinglyautomatized, people lose conscious controlover the production of their actions and areno longer able to make specific intentionaladjustments to them. For example, peoplehave difficulty describing how they tie theirshoelaces or how they get up from sitting in achair. When the behaviors are automatized,mere additional experience will not lead toincreased levels of performance.

In direct contrast to the acquisition ofeveryday skills, expert performers continueto improve their performance with moreexperience as long as it is coupled with delib-erate practice. The key challenge for aspiringexpert performers is to avoid the arresteddevelopment associated with automaticityand to acquire cognitive skills to supporttheir continued learning and improvement.By actively seeking out demanding tasks –often provided by their teachers and coaches– that force the performers to engage inproblem solving and to stretch their per-formance, the expert performers overcomethe detrimental effects of automaticity andactively acquire and refine cognitive mech-anisms to support continued learning andimprovement, as shown in the upper arm ofFigure 38.1. The expert performers and theirteachers identify specific goals for improv-

ing particular aspects of performance anddesign training activities that allow the per-former to gradually refine performance withfeedback and opportunities for repetition(deliberate practice). The performers willgradually acquire mechanisms that increasetheir ability to control, self-monitor, andevaluate their performance in representa-tive situations from the domain and thusgain independence from the feedback oftheir teachers (Ericsson, 1996, 2002 ; Glaser,1996). Although the overall structure ofthese mechanisms reflects general principles,the detailed structure and practice activitiesthat mediate their acquisition will reflect thedemands of that particular activity and thusdiffer from one domain of expertise toanother.

According to the expert-performanceapproach (Ericsson, 1996, 2002 , 2004), skillacquisition is viewed as an extended series ofgradual changes of the physiological and cog-nitive mechanisms that allow the observableperformance to show associated improve-ments. The acquisition of expert perfor-mance can thus be described as a series ofrelatively stable states, where each state hasa set of mechanisms that mediate the exe-cution of the associated performance (seeFigure 38.5). The primary differencesbetween two adjacent states can be physio-logical, where the subsequent states differ inthe level of strength, endurance, or speed ofcritical muscular systems. Alternatively, thedifference between the mechanisms of thetwo adjacent states might be primarily cog-nitive. For example, the performer might beable to better represent and monitor inter-nal and external states during performance,which in turn allows the performer to gener-ate and select better actions, or initiate andcomplete actions faster, or execute motoractions more consistently and accurately.

Another fundamental challenge to a the-oretical account of the acquisition of expertperformance involves describing plausibleexplanations of how a certain type of prac-tice activity (deliberate practice) can changeany complex State[I] into the directly fol-lowing complex State [I + 1]. First, thepractice activities that mediate improved



Figure 38.5 . A schematic illustration of the acquisition of expert performance asa series of states with mechanisms for monitoring and guiding futureimprovements of specific aspects of performance.

physiological function will be discussed, fol-lowed by a description of how practice canimprove performance by changes in cogni-tive mechanisms that mediate performanceand further learning.

Improving Adaptations by StrainingPhysiological Systems

Measurable increases in physical fitness donot simply result from wishful thinking.Instead people have to engage in intense aer-obic exercise that pushes them well beyondthe level of comfortable physical activityif they are to improve their aerobic fit-ness (Ericsson, 2003a; Ericsson et al., 1993 ;Robergs & Roberts, 1997). Specifically, inorder to increase their aerobic fitness youngadults have to exercise at least a couple oftimes each week, for at least 30 minutesper session, with a sustained heart rate thatis 70% of their maximal level (around 140

beats per minute for a maximal heart rate of200). When the human body is put underexceptional strain, a range of dormant genesin the DNA are expressed and extraordi-nary physiological processes are activated.Over time the cells of the body, includ-ing the brain (see Hill & Schneider, Chap-ter 37) will reorganize in response to the

induced metabolic demands of the activityby, for example, increases in the number ofcapillaries supplying blood to muscles andchanges in metabolism of the muscle fibersthemselves. These adaptations will eventu-ally allow the individual to execute the givenlevel of activity without greatly strainingthe physiological systems. To gain furtherbeneficial increases in adaptation, the ath-letes need to increase or change their weeklytraining activities to induce new and perhapsdifferent types of strain on the key phys-iological systems. For example, improve-ments of strength are attained when indi-viduals lift weights to induce brief maximalefforts of the targeted muscle groups. Moregenerally, athletic training involves push-ing the associated physiological systems out-side the comfort zone to stimulate phys-iological growth and adaptation (Ericsson,2001, 2002 , 2003a, 2003c, 2003d). Further-more, recent reviews (Gaser & Schlaug,2003 ; Hill & Schneider, Chapter 37; Kolb& Whishaw, 1998) show that the func-tion and structure of the brain is far moreadaptable than previously thought possi-ble. Especially, early and extended traininghas shown to change the cortical mappingof musicians (Elbert, Pantev, Wienbruch,Rockstoh, & Taub, 1995), the development



of white matter in the brain (Bengtsson et al.,2005), the development of “turn out” ofballet dancers, the development of perfectpitch, and flexibility of fingers (Ericsson &Lehmann, 1996).

In sum, elite performers search continu-ously for optimal training activities, with themost effective duration and intensity, thatwill appropriately strain the targeted physi-ological system to induce further adaptationwithout causing overuse and injury.

The Acquisition of MentalRepresentations for Performanceand Continued Learning

One of the principal challenges to contin-ued improvement of expert performance isthat the acquired representations and mech-anisms mediating expert performance mustbe modifiable to allow gradual changes thatincrementally improve performance. Theyneed to allow for improvements of spe-cific aspects of the performances as wellas for the coordination of necessary adjust-ments required by the associated changes.The experts’ mental representations thusserve a dual purpose of mediating the supe-rior expert performance while also providingthe same mechanisms that can be incremen-tally altered to further enhance performanceafter practice and training. Finally, individu-als must engage in deliberate-practice activi-ties to continue to stretch their performance.The dual role of representations has beenmost extensively documented in chess andtyping (Ericsson, 1996, 2002 , 2004). Chessplayers rely on planning out consequencesof potential moves in order to select the bestmove during matches in a tournament. Dur-ing deliberate practice, the same chess play-ers will rely on the same planning mecha-nisms to improve their ability to select thebest moves. Similarly, typists rely on thesame representations during performanceand when they attempt to increase theirspeed of typing during deliberate practice.After these two examples of deliberate prac-tice, broader issues of deliberate practice ina wide range of domains will be discussed.

chess

Expertise in chess was proposed by Simonand Chase (1973) to be a prototype for manydomains of expertise. In his pioneering workon chess expertise, de Groot (1946/1978)uncovered the detailed processes that allowworld-class chess players to analyze chesspositions and to find their best move for eachposition. He instructed expert and world-class players to “think aloud” while theyselected the best move in a set of unfa-miliar chess positions taken from gamesby chess masters (see Figure 38.2). Sub-sequent reviews related to this researchshowed that the quality of the selectedmoves was closely associated with the per-formers’ play in tournaments and, there-fore by inference, captured the essence ofchess skill (Ericsson, Patel, & Kintsch, 2000).De Groot’s (1946/1978) analysis of experts’“think aloud” protocols revealed that theyfirst formed a rapid impression of the chessposition in order to retrieve potential movesfrom memory. These promising moves werethen evaluated by mentally planning theconsequences of potential options. Duringthe course of this exercise, even the world-class players would discover better moves,indicating that they continued to improve.

A major challenge for successful planningand chess skill is that the chess players beable to represent the chess positions in work-ing memory in a manner that allows evalu-ation and flexible exploration of sequencesof moves. The skills required to representand manipulate chess positions in long-termmemory appear to develop slowly as a func-tion of increased chess skill (Ericsson &Kintsch, 1995 ; Ericsson, et al., 2000). Con-sequently, more-skilled chess players havebeen shown to be able to plan more thor-oughly and to represent chess positionsmore effectively. In addition, their mem-ory for briefly presented chess positions isvastly superior to those of less-skilled players(Gobet & Charness, Chapter 30). However,this superior recall performance is limitedto representative chess positions and disap-pears almost completely when chess posi-tions are randomly rearranged.



The central challenge to an account of thecontinued improvement of a chess experts’ability is to understand how they plan andselect the best action in a given game sit-uation. Chess players typically practice thistask by studying chess openings and analyz-ing published games between the very bestchess players in the world. They typicallyanalyze the games by playing through thegames, one move at a time, to determineif their selected move matches to the cor-responding move originally selected by themasters. If the master’s move in the studiedchess game differed from their own selec-tion, this would imply that their planningand evaluation must have overlooked someaspect of the position. It is important to notethat this learning process allows the playerto diagnose the source of suboptimal movesand thus make a local change that improvesthe selection of related moves without caus-ing interference with other aspects of theexisting skill.

By more careful and extended analysis,the chess expert is generally able to discoverthe reasons for the chess master’s superiormove. Serious chess players spend as muchas four hours every day engaged in this typeof solitary study (Charness et al., 1996, 2005 ;Ericsson et al., 1993). By spending additionaltime analyzing the consequences of movesfor a chess position, players can increase thequality of their selections of moves. Withmore study, individuals refine their repre-sentations and can access or generate thesame information faster. As a result, chessmasters can typically recognize a superiormove virtually immediately, whereas a com-petent club player requires much longer tofind the same move by successive planningand evaluation rather than recognition. Thesame type of improvement based on deliber-ate practice and increased depth of planningcan explain gradually increased performancein a wide range of domains, such as billiards,golf, music, and surgery (Ericsson, 2004).

typing

If chess is viewed as one of the most intel-lectually challenging tasks, then typing is

typically viewed as a diametrically differ-ent, mundane, habitual activity. Many adultsare able to type, yet there are often largeindividual differences in the speed attained.The standardized measure of typing speedinvolves having skilled typists and unskilledparticipants type passages from a collectionof unfamiliar texts as fast as they can with-out making errors. High-speed films of fingermovements show that the faster typists startmoving their fingers toward their desiredlocations on the keyboard well before thekeys are struck. The superior typists’ speedadvantage is linked to their perceptual pro-cessing of the text beyond the word that theyare currently typing (Salthouse, 1984). Bylooking ahead in the text to identify letters tobe typed, they can prepare future keystrokesin advance. This evidence for anticipationhas been confirmed by experimental stud-ies where expert typists have been restrictedfrom looking ahead. Under such conditionstheir typing speed is dramatically reducedand approaches the speed of less-skilledtypists.

In sum, the superior speed of reactionsby expert performers, such as typists andathletes, appears to depend primarily oncognitive representations mediating skilledanticipation (see also, Endsley, Chapter 36),rather than faster basic speed of their ner-vous system (Abernethy, 1991). For instance,expert tennis players are able to antici-pate where a tennis player’s shots will landeven before the player’s racquet has con-tacted the ball (Williams, Ward, Knowles, &Smeeton, 2002). Eye movements of experttennis players show that they are able topick up predictive information from sub-tle, yet informative, motion cues, such aship and shoulder rotation, compared to theirnovice counterparts. They can also use later-occurring and more-deterministic cues, suchas racket swing, to confirm or reject their ear-lier anticipations.

Research on instruction in typing(Dvorak, Merrick, Dealey, & Ford, 1936) hasso far provided the best initial insights intohow speed of performance can be increasedthrough deliberate practice that alters andimproves the representations mediating



anticipation and coordination of fingermovements. The key empirical observationis that people can increase their typingspeed by exerting full concentration towardimprovement. Regular typists can typicallymaintain this level of concentration foronly 15 to 30 minutes per day. Whentypists concentrate and strain themselvesto type at a faster rate (typically around10 to 20% faster than their normal speed),they strive to anticipate better, possiblyby extending their gaze further ahead.

The increased tempo also brings outkeystroke combinations for which the typ-ists are comparatively slow, thus restricting afluent higher speed. These challenging com-binations can then be trained in special exer-cises and incorporated into the typing of reg-ular text. This is in order to assure that anymodifications can be integrated with the rep-resentations mediating typical typing tasks.By increasing anticipation and successivelyeliminating weaknesses, typists can increasetheir average speed in practice at a rate thatis still 10 to 20% faster than their new aver-age speed attained after such practice. Thegeneral approach of finding methods to pushperformance beyond its normal level – evenif that performance can be maintained onlyfor a short time – offers the potential foridentifying and correcting weaker compo-nents that will improve performance as wellas for enhancing anticipation.

A Broader View of Expert Performanceand Deliberate Practice

The theoretical framework of deliberatepractice asserts that improvement in perfor-mance of aspiring experts does not happenautomatically or casually as a function of fur-ther experience. Improvements are causedby changes in cognitive mechanisms mediat-ing how the brain and nervous system con-trol performance and in the degree of adap-tation of physiological systems of the body.The principal challenge to attaining expertlevel performance is to induce stable spe-cific changes that allow the performance tobe incrementally improved.

Once we conceive of expert performanceas mediated by complex integrated systemsof representations for the planning, analysis,execution, and monitoring of performance(see Figure 38.5), it becomes clear that itsacquisition requires a systematic and delib-erate approach. Deliberate practice is there-fore designed to improve specific aspectsof performance in a manner that assuresthat attained changes can be successfullymeasured and integrated into representativeperformance. Research on deliberate prac-tice in music and sports shows that contin-ued attempts for mastery require that theperformer always try, by stretching perfor-mance beyond its current capabilities, tocorrect some specific weakness while pre-serving other successful aspects of function.This type of deliberate practice requires fullattention and concentration, but even withthat extreme effort, some kind of failure islikely to arise, and gradual improvementswith corrections and repetitions are nec-essary. With increased skill in monitoring,skilled performers in music focus on master-ing new challenges by goal-directed deliber-ate practice involving problem solving andspecialized training techniques (Chaffin &Imre, 1997; Ericsson, 2002 ; Gruson, 1988;Nielsen, 1999).

In their research on sports, Deakin andCobley (2003) found that ice skaters spenda considerable portion of their limited prac-tice time on jump-combinations they havealready mastered, rather than working onthe yet-to-be-mastered combinations, wherethere is the largest room for improve-ment. More generally, they found that withincreasing levels of attained skill the skatersspent more time on jumps and other chal-lenging activities that had the potential toimprove performance.

Practice aimed at improving integratedperformance cannot be performed mind-lessly nor independent of the representativecontext for the target performance. In addi-tion, more-accomplished individuals in thedomain, such as professional coaches andteachers, will always play an essential rolein guiding the sequencing of practice activi-ties for future experts in a safe and effective



manner. Research on self-regulated learning(Zimmerman, Chapter 39) has documentedeffective study methods that are related tosuperior academic performance, especiallyin high schools. More recent work has shownthat engagement in study methods consis-tent with deliberate practice has been foundto predict achievement in both undergradu-ate college students (Plant, Ericsson, Hill, &Asberg, 2005) as well as in students in med-ical school (Moulaert, Verwijnen, Rikers, &Scherpbier, 2004).

The deliberate-practice framework canalso explain the necessity for further delib-erate practice in order for individuals sim-ply to maintain their current level of skill. Itis well known that athletes and musicianswho reduce or stop their regular practicewill exhibit a reduced level of performance– a maintained level of challenge and strainappear necessary to preserve the attainedphysiological and cognitive adaptations. Thesame type of account has been developed toexplain age-related reductions in music per-formance and how they can be counteractedby maintained levels of deliberate practice(Krampe & Ericsson, 1996; see Krampe andCharness, Chapter 40).

Concluding Remarks: GeneralCharacteristics of Deliberate Practice

The perspective of deliberate practiceattributes the rarity of excellence to thescarcity of optimal training environmentsand to the years required to develop thecomplex mediating mechanisms that sup-port expertise. Even children considered tohave innate gifts need to attain their supe-rior performance gradually, by engaging inextended amounts of designed deliberatepractice over many years. Until most indi-viduals recognize that sustained training andeffort is a prerequisite for reaching expertlevels of performance, they will continue tomisattribute lesser achievement to the lackof natural gifts, and will thus fail to reachtheir own potential.

The effects of mere experience differgreatly from those of deliberate practice,

where individuals concentrate on activelytrying to go beyond their current abilities.Consistent with the mental demands ofproblem solving and other types of com-plex learning, deliberate practice requiresconcentration that can be maintained onlyfor limited periods of time. Although thedetailed nature of deliberate practice willdiffer across domains and as a function ofattained skill, there appear to be limits onthe daily duration of deliberate practice, andthis limit seems to generalize across domainsof expertise. Expert performers from manydomains engage in practice without rest foronly around an hour, and they prefer to prac-tice early in the morning when their mindsare fresh (Ericsson et al., 1993). Elite musi-cians (Ericsson, 2002) and athletes (Ericsson,2001, 2003c) report that the factor that lim-its their deliberate practice is primarily aninability to sustain the level of concentrationthat is necessary. Even more interestingly,elite performers in many diverse domainshave been found to practice, on the average,roughly the same amount every day, includ-ing weekends, and the amount of practicenever consistently exceeds five hours per day(Ericsson, 1996; Ericsson et al., 1993). Thelimit of four to five hours of daily deliber-ate practice or similarly demanding activ-ities holds true for a wide range of eliteperformers in different domains, such aswriting by famous authors (Cowley, 1959;Plimpton, 1977), as does their increasedtendency to take recuperative naps. Fur-thermore, unless the daily levels of prac-tice are restricted, such that subsequentrest and nighttime sleep allow the individ-uals to restore their equilibrium, individ-uals often encounter overtraining injuriesand, eventually, incapacitating “burnout.” Insome domains of sports, such as gymnas-tics, sprinting, and weight lifting, the max-imal effort necessary for representative per-formance is so great that the amount of dailydeliberate practice is even further limited byfactors constraining the duration of produc-tion of maximal power and strength.

The scientific study of deliberate prac-tice will enhance our knowledge about howexperts optimize the improvements of their



performance (and motivation) through ahigh level of daily practice they can sustainfor days, months, and years. The emerginginsights should be relevant to any motivatedindividual aspiring to excel in any challeng-ing domain (Ericsson, 2004). Although weare already gaining understanding about howperformers improve with deliberate practiceand reach expert levels, it is unlikely thatwe will ever be able to fully understand andpredict future innovations. We may be ableto reproduce the path of development thatelite performers have taken to reach theirhighest levels of performance in the past. Wemay also be able to help performers in onedomain of expertise, such as surgery, learnabout the best training methods that havebeen developed in domains with a longertradition, such as violin performance. Wemay even be able to work in collaborationwith world-class performers who are work-ing on improving their performance to newand undiscovered heights. At the highestlevels of expert performance, the drive forimprovement will always involve search andexperimentation at the threshold of under-standing, even for the masters dedicated toredefining the meaning of excellence in theirfields.

References

Abernethy, B. (1991). Visual search strategies anddecision-making in sport. International Journalof Sport Psychology, 2 2 , 189–210.

Anderson, J. R. (1982). Acquisition of cognitiveskill. Psychological Review, 89, 369–406.

Bedard, J., & Chi, M. T. H. (1993). Expertise inauditing. Auditing, 12 , (Suppl.), 1–25 .

Bengtsson, S. L., Nagy, Z., Skare, S., Forsman,L., Forsberg, H., & Ullen, F. (2005). Extensivepiano practicing has regionally specific effectson white matter development. Nature Neuro-science, 8, 1148–1150.

Bloom, B. S. (1985a). Generalizations about talentdevelopment. In B. S. Bloom (Ed.), Developingtalent in young people (pp. 507–549). New York:Ballantine Books.

Bloom, B. S. (Ed.) (1985b). Developing talent inyoung people. New York: Ballantine Books.

Book, W. F. (1925a). Learning to typewrite. NewYork: The Gregg Publishing Co.

Book, W. F. (1925b) The psychology of skill. NewYork: The Gregg Publishing Co.

Bryan, W. L., & Harter, N. (1897). Studies in thephysiology and psychology of the telegraphiclanguage. Psychological Review, 4 , 27–53 .

Bryan, W. L., & Harter, N. (1899). Studies on thetelegraphic language: The acquisition of a hier-archy of habits. Psychological Review, 6, 345–375 .

Camerer, C. F., & Johnson, E. J. (1991). Theprocess-performance paradox in expertjudgment: How can the experts know so muchand predict so badly? In K. A. Ericsson andJ. Smith (Eds.), Towards a general theory ofexpertise: Prospects and limits (pp. 195–217).Cambridge: Cambridge University Press.

Chaffin, R., & Imreh, G. (1997). “Pulling teethand torture”: Musical memory and prob-lem solving. Thinking and Reasoning, 3 , 315–336.

Charness, N., Krampe, R. Th., & Mayr, U.(1996). The role of practice and coaching inentrepreneurial skill domains: An internationalcomparison of life-span chess skill acquisition.In K. A. Ericsson (Ed.), The road to excellence:The acquisition of expert performance in the artsand sciences, sports, and games (pp. 51–80).Mahwah, NJ: Erlbaum.

Charness, N., Tuffiash, M. I., Krampe, R., Rein-gold, E., & Vasyukova E. (2005). The role ofdeliberate practice in chess expertise. AppliedCognitive Psychology, 19, 15 1–165 .

Chi, M. T. H., Glaser, R., & Farr, M. J. (Eds.)(1988). The nature of expertise. Hillsdale, NJ:Erlbaum.

Cote, J., Ericsson, K. A., & Law, M. (2005). Trac-ing the development of athletes using retro-spective interview methods: A proposed inter-view and validation procedure for reportedinformation. Journal of Applied Sport Psychol-ogy, 17 , 1–19.

Cowley, M. (Ed.) (1959). Writers at work: TheParis review interviews. New York: Viking.

Dawes, R. M. (1994). House of cards: Psychologyand psychotherapy built on myth. New York:Free Press.

Deakin, J. M., & Cobley, S. (2003). A searchfor deliberate practice: An examination of thepractice environments in figure skating and vol-leyball. In J. Starkes & K. A. Ericsson (Eds.),Expert performance in sport: Recent advances in



research on sport expertise (pp. 115–135). Cham-paign, IL: Human Kinetics.

de Groot, A. (1978). Thought and choice in chess(Original work published in 1946). The Hague:Mouton.

Denison, J. (2003). Bannister and beyond: Themystique of the four minute mile. Breakawaybooks.

Denney, N. W. (1982). Aging and cognitivechanges. In B. B. Wolman (Ed.), Handbookof developmental psychology (pp. 807–827).Englewood Cliffs, NJ: Prentice Hall.

Doane, S. M., Pellegrino, J. W., & Klatzky, R. L.(1990). Expertise in a computer operatingsystem: Conceptualization and performance.Human-Computer Interaction, 5 , 267–304 .

Duffy, L. J., Baluch, B., & Ericsson, K. A. (2004).Dart performance as a function of facets ofpractice amongst professional and amateurmen and women players. International Journalof Sport Psychology, 35 , 232–245 .

Dvorak, A., Merrick, N. L., Dealey, W. L., & Ford,G. C. (1936). Typewriting behavior. New York:American Book Company.

Elbert, T., Pantev, C., Wienbruch, C., Rockstroch,B., & Taub, E. (1995). Increased cortical repre-sentation of the fingers of the left hand in stringplayers. Science, 2 70, 305–307.

Ericsson, K. A. (1996). The acquisition of expertperformance: An introduction to some of theissues. In K. A. Ericsson (Ed.), The road toexcellence: The acquisition of expert performancein the arts and sciences, sports, and games(pp. 1–50). Mahwah, NJ: Erlbaum.

Ericsson, K. A. (2001). The path to expertperformance: Insights from the masters on howto improve performance by deliberate practice.In P. Thomas (Ed.), Optimizing performance ingolf (pp. 1–57). Brisbane, Australia: AustralianAcademic Press.

Ericsson, K. A. (2002). Attaining excellencethrough deliberate practice: Insights from thestudy of expert performance. In M. Ferrari(Ed.), The pursuit of excellence in education(pp. 21–55). Hillsdale, NJ: Erlbaum.

Ericsson, K. A. (2003a). The search for generalabilities and basic capacities: Theoretical impli-cations from the modifiability and complexityof mechanisms mediating expert performance.In R. J. Sternberg & E. L. Grigorenko (Eds.),Perspectives on the psychology of abilities, compe-tencies, and expertise (pp. 93–125). Cambridge:Cambridge University Press.

Ericsson, K. A. (2003b). Exceptional memo-rizers: Made, not born. Trends in Cognitive Sci-ences, 7 , 233–235 .

Ericsson, K. A. (2003c). The development ofelite performance and deliberate practice: Anupdate from the perspective of the expert-performance approach. In J. Starkes & K. A.Ericsson (Eds.), Expert performance in sport:Recent advances in research on sport exper-tise (pp. 49–81). Champaign, IL: HumanKinetics.

Ericsson, K. A. (2003d). How the expert-per-formance approach differs from traditionalapproaches to expertise in sports: In searchof a shared theoretical framework for studyingexpert performance. In J. Starkes & K. A. Eric-sson (Eds.), Expert performance in sport: Recentadvances in research on sport expertise (pp. 371–401). Champaign, IL: Human Kinetics.

Ericsson, K. A. (2004). Deliberate practice andthe acquisition and maintenance of expert per-formance in medicine and related domains.Academic Medicine, 10, S1–S12 .

Ericsson, K. A., Delaney, P. F., Weaver, G., &Mahadevan, R. (2004). Uncovering the struc-ture of a memorist’s superior “basic” memorycapacity. Cognitive Psychology, 49, 191–237.

Ericsson, K. A., & Kintsch, W. (1995). Long-term working memory. Psychological Review,102 , 211–245 .

Ericsson, K. A., Krampe, R. Th., & Tesch-Romer,C. (1993). The role of deliberate practice in theacquisition of expert performance. Psychologi-cal Review, 100, 363–406.

Ericsson, K. A., & Lehmann, A. C. (1996). Ex-pert and exceptional performance: Evidenceon maximal adaptations on task constraints.Annual Review of Psychology, 47 , 273–305 .

Ericsson, K. A., Patel, V. L., & Kintsch, W. (2000).How experts’ adaptations to representativetask demands account for the expertise effectin memory recall: Comment on Vicente andWang (1998). Psychological Review, 107 , 578–592 .

Ericsson, K. A., & Smith, J. (1991). Prospectsand limits in the empirical study of expertise:An introduction. In K. A. Ericsson & J. Smith(Eds.), Toward a general theory of expertise:Prospects and limits (pp. 1–38). Cambridge:Cambridge University Press.

Feldman, D. H. (1994). Beyond universals in cog-nitive development (2nd. Ed.). Norwood, NJ:Ablex.



Fitts, P., & Posner, M. I. (1967). Human perfor-mance. Belmont, CA: Brooks/Cole.

Galton, F., Sir (1869/1979). Hereditary genius:An inquiry into its laws and consequences(Originally published in 1869). London: JulianFriedman Publishers.

Gaser, C., & Schlaug, G. (2003). Brain struc-tures differ between musicians and non-musicians. Journal of Neuroscience, 2 3 , 9240–9245 .

Gawel, R. (1997). The use of language by trainedand untrained experienced wine tasters. Jour-nal of Sensory Studies, 12 , 267–284 .

Glaser, R. (1996). Changing the agency forlearning: Acquiring expert performance. InK. A. Ericsson (Ed.), The road to excellence: Theacquisition of expert performance in the arts andsciences, sports, and games (pp. 1–50). Mahwah,NJ: Erlbaum.

Gould, S. J. (1996). Full house: The spread of excel-lence from Plato to Darwin. New York: Harmonybooks.

Grape, C., Sandgren, M., Hansson, L.-O.,Ericson, M., & Theorell, T. (2003). Doessinging promote well-being?: An empiricalstudy of professional and amateur singers dur-ing a singing lesson. Integrative Physiological &Behavioral Science, 38, 65–71.

Gruson, L. M. (1988). Rehearsal skill and musi-cal competence: Does practice make perfect?In J. A. Sloboda (Ed.), Generative processesin music (pp. 91–112). Oxford, UK: ClarendenPress.

Hayes, J. R. (1981). The complete problem solver.Philadelphia, PA: Franklin Institute Press.

Helsen, W. F., Starkes, J. L., & Hodges, N. J.(1998). Team sports and the theory of delib-erate practice. Journal of Sport and Exercise Psy-chology, 2 0, 12–34 .

Hodges, N. J., & Starkes, J. L. (1996). Wrestlingwith the nature of expertise: A sport spe-cific test of Ericsson, Krampe and Tesch-Romer’s (1993) theory of “Deliberate Prac-tice.” International Journal of Sport Psychology,27, 400–424 .

Hoffman, R. R. (Ed.) (1992). The psychology ofexpertise: Cognitive research and empirical AI.New York: Springer-Verlag.

Krampe, R. Th., & Ericsson, K. A. (1996). Main-taining excellence: Deliberate practice andelite performance in young and older pianists.Journal of Experimental Psychology: General,12 5 , 331–359.

Kolb, B., & Whishaw, I. Q. (1998). Brain plasticityand behavior. Annual Review of Psychology, 49,43–64 .

Lehmann, A. C., & Ericsson K. A. (1998).The historical development of domains ofexpertise: Performance standards and innova-tions in music. In A. Steptoe (Ed.), Geniusand the mind (pp. 67–94). Oxford, UK: OxfordUniversity Press.

McDaniel, M. A., Schmidt, F. L., & Hunter, J. E.(1988). Job experience correlates of job perfor-mance. Journal of Applied Psychology, 73 , 327–330.

McKinney, E. H., & Davis, K. J. (2004). Effectsof deliberate practice on crisis decision perfor-mance. Human Factors, 45 , 436–444 .

Moulaert, V., Verwijnen, M. G. M., Rikers, R.,& Scherpbier, A. J. J. A. (2004). The effectsof deliberate practice in undergraduate medicaleducation. Medical Education, 38, 1044–1052 .

Nielsen, S. (1999). Regulation of learning strate-gies during practice: A case study of a singlechurch organ student preparing a particularwork for a concert performance. Psychology ofMusic, 2 7 , 218–229.

Plant, E. A., Ericsson, K. A., Hill, L., & Asberg, K.(2005). Why study time does not predict gradepoint average across college students: Implica-tions of deliberate practice for academic perfor-mance. Contemporary Educational Psychology,30, 96–116.

Plimpton, G. (Ed.) (1977). Writers at work: TheParis review. Interviews, Second Series. NewYork: Penguin.

Raskin, E. (1936). Comparison of scientific andliterary ability: A biographical study of eminentscientists and letters of the nineteenth century.Journal of Abnormal and Social Psychology, 31,20–35 .

Reif, F., & Allen, S. (1992). Cognition for inter-preting scientific concepts: A study of acceler-ation. Cognition and Instruction, 9, 1–44 .

Robergs, R. A., & Roberts, S. O. (1997). Exer-cise physiology: Exercise, performance, and clin-ical applications. St. Louis, MO: Mosby-YearBook.

Rosson, M. B. (1985). The role of experi-ence in editing. Proceedings of INTERACT ‘84IFIP Conference on Human-Computer Interac-tion (pp. 45–50). New York: Elsevier.

Salthouse, T. A. (1984). Effects of age and skillin typing. Journal of Experimental Psychology:General, 113 , 345–371.



Schneider, W. (1993). Acquiring expertise: Deter-minants of exceptional performance. In K. A.Heller, J. Monks, & H. Passow (Eds.), Interna-tional handbook of research and development ofgiftedness and talent (pp. 311–324). Oxford, UK:Pergamon Press.

Shanteau, J., & Stewart, T. R. (1992). Whystudy expert decision making? Some histori-cal perspectives and comments. OrganizationalBehaviour and Human Decision Processes, 53 ,95–106.

Schulz, R., & Curnow, C. (1988). Peak perfor-mance and age among superathletes: Track andfield, swimming, baseball, tennis, and golf. Jour-nal of Gerontology: Psychological Sciences, 43 ,113–120.

Simon, H. A., & Chase, W. G. (1973). Skill inchess. American Scientist, 61, 394–403 .

Simonton, D. K. (1997). Creative productivity: Apredictive and explanatory model of career tra-jectories and landmarks. Psychological Review,104 , 66–89.

Singer, C. (1958). From magic to science. NewYork: Dover.

Sloboda, J. A., Davidson, J. W., Howe, M. J. A., &Moore, D. G. (1996). The role of practice in thedevelopment of performing musicians. BritishJournal of Psychology, 87 , 287–309.

Sonnentag, S. (1998). Expertise in professionalsoftware design: A process study. Journal ofApplied Psychology, 83 , 703–715 .

Sonnentag, S., & Kleine, B. M. (2000). Deliber-ate practice at work: A study with insuranceagents. Journal of Occupational and Organiza-tional Psychology, 73 , 87–102 .

Starkes, J. L., Deakin, J., Allard, F., Hodges, N. J.,& Hayes, A. (1996). Deliberate practice insports: What is it anyway? In K. A. Ericsson(Ed.), The road to excellence: The acquisitionof expert performance in the arts and sciences,

sports, and games (pp. 81–106). Mahwah, NJ:Erlbaum.

Ulijaszek, S. J., Johston, F. E., & Preece, M. A.(Eds.) (1998). The Cambridge encyclopedia ofhuman growth and development. Cambridge,UK: Cambridge University Press.

Valentin, D., Pichon, M., de Boishebert, V., &Abdi, H. (2000). What’s in a wine name? Whenand why do wine experts perform better thannovices? Abstracts of the Psychonomic Society, 5 ,36.

Vicente, K. J., & Wang, J. H. (1998). An ecologicaltheory of expertise effects in memory recall.Psychological Review, 105 , 33–57.

Ward. P., Hodges, N. J., Williams, A. M., &Starkes, J. L. (2004). Deliberate practice andexpert performance: Defining the path toexcellence. In A. M. Williams & N. J. Hodges(Eds.), Skill acquisition in sport: Research, the-ory and practice (pp. 231–258). London, UK:Routhledge.

Williams, M., & Davids, K. (1995). Declarativeknowledge in sport: A by-product of experi-ence or a characteristic of expertise? Journal ofSport & Exercise Psychology, 17, 259–275 .

Williams, A. M., Ward, P., Knowles, J. M., &Smeeton, N. J. (2002). Anticipation skill ina real-world task: Measurement, training, andtransfer in tennis. Journal of ExperimentalPsychology: Applied, 8, 259–270.

Author Notes

This research was supported by theFSCW/Conradi Endowment Fund of FloridaState University Foundation. The author wantsto thank Andreas Lehmann, Len Hill, RobertHoffman, and Paul Ward for the helpfulcomments on earlier drafts of the chapter.


706

Deliberate Practice - Excerpt From Handbook (2006)

Documents

Transcript of Deliberate Practice - Excerpt From Handbook (2006)