The historical debate on the relative influences of genes (i.e. nature) andAbstractenvironment (i.e. nurture) on human behaviour has been characterised by extremepositions leading to reductionist and polemic conclusions. Our analysis ofresearch on sport and exercise behaviours shows that currently there is littlesupport for either biologically or environmentally deterministic perspectives onelite athletic performance. In sports medicine, recent molecular biologicaladvances in genomic studies have been over-interpreted, leading to a questionable‘single-gene-as-magic-bullet’ philosophy adopted by some practitioners. Similar-ly, although extensive involvement in training and practice is needed at elitelevels, it has become apparent that the acquisition of expertise is not merely aboutamassing a requisite number of practice hours. Although an interactionist perspec-tive has been mooted over the years, a powerful explanatory framework has beenlacking. In this article, we propose how the complementary nature of degenerate
962 Davids & Baker
neurobiological systems might provide the theoretical basis for explaining theinteractive influence of genetic and environmental constraints on elite athleticperformance. We argue that, due to inherent human degeneracy, there are manydifferent trajectories to achieving elite athletic performance. While the greatesttraining responses may be theoretically associated with the most favourablegenotypes being exposed to highly specialised training environments, this is a rareand complex outcome. The concept of degeneracy provides us with a basis forunderstanding why each of the major interacting constraints might act in acompensatory manner on the acquisition of elite athletic performance.
We used to think our fate was in our stars. Now chology.[12-14] February 2001 marked the publicationwe know, in large measure, our fate is in our genes of the draft human genome, signalling the arrival of(James Watson, 1989 cited in Vitzthum[1]).1 biology in the genomic era. Whilst questions of the
relative roles of nature and nurture in the past mayWhat are sports scientists to make of this ge-have been addressed from the standpoint of academ-nocentric view of human behaviour? This articleic interest, in recent times there have been increasingconsiders one of the most complex and enduringreports of medical interventions and attempts to usecontroversies in science and medicine: the ‘nature-genetic knowledge for performance enhancement innurture debate’.[1,4-7] Science has a number of suchsport.[15,16]dualisms where strong theoretical arguments are
presented for two opposing viewpoints and the na- In particular, there have been some efforts inture-nurture issue is, arguably, the most pervasive in molecular biology to identify single gene variantsits history. For example, in psychology it is known with the potential to profoundly impact on individu-as the ‘nativism-empiricism issue’, developmental al performance or the propensity to lead to a specificscience terms it the ‘maturation versus learning de- disease (e.g. a gene responsible for physical power,bate’ and physiology and medicine considers how the propensity to exercise, or for breast cancer).heredity and environment account for variations in In sport, this ‘single gene as magic bullet’ philos-health and human performance. ophy, favoured by some molecular geneticists, has
Put simply, the nature-nurture debate concerns led to claims that elite performers are born to suc-the extent to which an individual is a product of her/ ceed. This approach is exemplified by attempts tohis genes or environment, generating questions over identify sprinters and endurance runners on the basisthe role of genes and environmental influences and of differing alleles (i.e. forms) of a single geneexperience in a range of human behavioural con- known as α-actinin-3.[17,18] There have been similartexts, including athletic, educational and musical claims on the roles of different variants of the ACEperformance and achievement. For this reason, in gene in endurance and power events such aspsychological medicine, behavioural geneticists are mountaineering and running.[19] In those responsibleinterested in human variations in key factors, includ- for managing elite sports performance, the belief ining health, development, cognitive and movement the potential impact of single genes is perhaps moreabilities, perceptual skills, social attitudes, psycho- prevalent. For example, in September 2000, Scien-pathology and personality, related to functioning in tific American published a vision of the Olympica range of domains.[8-11] Games in the year 2012, predicting the widespread
How environmental and genetic constraints cor- existence of gene-transfer technology (popularlyrelate or interact to shape performance variations in known as gene or cell doping) whereby specifiedsport and exercise is a question of increasing interest artificial DNA material is coupled to well under-in sports medicine, sports pedagogy and sports psy- stood viral delivery mechanisms and inserted direct-
1 Although research reviewed in this section is restricted to sport performance, similar noteworthy developments havealso been made in exercise psychology (e.g. Podewils et al.[2] and Schuit et al.[3]).
ly into the nuclei of muscle cells[20,21] (also see environments. Here we note how this is an aptMiah[15] and McCrory[22]). More recently, it was description for sports medicine and sport science towidely reported that the Sea Eagles, a rugby football understand the relationship between genes and envi-league team from Manly, Australia, tested 18 play- ronment. The ‘complementary nature’ of phenome-ers for variations in 11 different exercise-related na in the natural world suggests that it is highlygenes in order to enhance the specificity of training important for geneticists to identify many singleprogramme design.[23] The same article highlighted gene variants, although the role of these genes inthe possibility of sports clubs genetically profiling regulating behaviour needs to be framed by theirathletes to understand susceptibility for certain inju- overarching tendencies to network and to cooperateries, map genetic suitability for specific team posi- or compete with environmental constraints. In thistions and roles, and to gain insights into player respect, our article highlights the need for sportsdevelopment into various sports or physical activi- scientists to adopt a multidisciplinary perspective inties. order to gain a detailed understanding of the molec-
ular biological mechanisms of DNA transmissionThese alarming trends by some elite sport scien-and gene expression, and we discuss how sometists to use the ‘single-gene as magic bullet’ philoso-molecular biological advances may have beenphy as a platform for optimising performance places‘over-interpreted’ by scientists using the ACE genethe spotlight again on the nature-nurture debate.research literature as an example. Finally, we evalu-Besides obvious ethical issues, reports of gene-pro-ate the limitations of current understanding raisingfiling and gene-transfer technology raise more gen-questions over the current relevance of the debate.eral theoretical and practical questions about the
nature of genetic and environmental constraints on1. The Nature-Nurture Debate inskill acquisition and performance, and propensity toPsychological Scienceparticipate in physical activities.[24] Although most
knowledge on the effects of genes on athletic per- There is no escape from the conclusion that na-formance has been gained from research in molecu- ture prevails enormously over nurture when the dif-lar biology, genetics, genomics and medicine, this ferences of nurture do not exceed what is commonlyresearch also has relevance for many areas of sports found among persons of the same rank of societyscience and medicine, such as sport and exercise and in the same country.[27]
psychology.[25] Although it was Francis Galton[27] who first usedthe phrase ‘nature and nurture’ to describe theClearly, sports scientists need to understand thesources of individual differences in 1874 (and madecurrent state of the nature-nurture debate in sciencethe subsequent ‘conclusion’ outlined above inin order to make sense of the pending practical1883),[27] arguments over the role of environmentalapplications of gene-transfer technology. In this arti-influences versus innate characteristics can becle, we begin by historically reviewing the nature-traced back to the Platonic dialogues and Aristoteli-nurture debate in science, focusing specifically onan philosophy of ancient Greece. It is only in thepsychological medicine in which it has flourished.past 130 years, however, that these issues have comeSecondly, we evaluate evidence for both nurture andto dominate scientific (as well as non-scientific)nature constraints on sport performance anddiscussions.achievement. There is growing awareness in science
of ‘the complementary nature’, as Kelso and Eng- Increased interest in this area was undoubtedlystrom[26] so aptly entitled their critique of the histori- driven to a large extent by the publication ofcal tendency to attempt to explain phenomena on the Darwin’s Origin of Species in 1859, which informedbasis of dualist positions and theoretical stances. Galton’s[27,28] contributions most notably, but alsoTheir thesis is that a systems biological approach those of William James[29] and Lewis Terman.[30]
can help categorise natural phenomena as exhibiting Much of this work was based on the conclusion thatstrong cooperative tendencies to interact as well as biology was the chief constraint to expertise andtendencies to function separately, the emergence of achievement, known as biological determinism. In-which can characterise the relationship of genes and deed, Terman’s appropriately titled Genetic Studies
of Genius,[30-32] one of the longest and most ambi- concept of heritability defined as the transmission oftious longitudinal studies in history, was based pri- a trait from parent to progeny.[1] Regardless, fewmarily around this notion. In 1918, Fisher intro- researchers today would argue that neither environ-duced the statistic of ‘heritability’, which can be mental nor biological factors are important and theredefined as “the proportion of the total phenotypic is very little to suggest the nature-nurture debatevariance that is associated with genetic variance in a could be resolved in favour of one constraint to thespecific sample with a specific genetic composition complete exclusion of the other. More recent nature-and environmental context”.[1] The heritability sta- nurture studies seem focused on questions like:tistic to describe individual variation is captured in a • What factor is most important?biometric model, which includes the influence of • How much of the variability in condition ‘x’ canindividual alleles, dominant alleles and environmen- be accounted for by genetic (or environmental)tal variance in a studied sample. variables?
Conversely, work from the environmentalist • When, during the lifespan, is a genetic or envi-camp during the same period also reinforced this ronmental constraint more influential in shapingdualist ‘all or nothing’ approach – the idea that behaviour?individuals start as a ‘tabula rasa’ (blank slate) with Despite continued attention to these issues andno innate traits or characteristics, and that all forms considerable advancement in the methods of exami-of learning and behaviour result from interactions nation, it is unlikely they will be resolved soon. Thiswith our environment. This social deterministic is not to say that understanding of genetic and envi-viewpoint is succinctly demonstrated by Wat- ronmental influence has not advanced. It certainlyson’s[33] famous boast: has; moreover, these advances have obvious impor-
tance for understanding the development of exper-“Give me a dozen healthy infants and my owntise and talent identification in sports and physicalspecified world to bring them up in, and I’ll guaran-activities. Individual differences between perform-tee to take anyone at random and train him toers remains a central issue in many areas of sportsbecome any kind of specialist I might select – doc-medicine and science, leading to crucial questionstor, lawyer, artist … regardless of his talents,such as: why do some athletes benefit more frompenchants, tendencies, abilities, vocations and racetraining and practice than others? How can specificof his ancestors.”athletic potential be identified and developed? ToThe history of this debate in psychologicalwhat extent can genetic testing provide informationmedicine is marked by radical shifts in opinion,for specifically designing training and practiceusually driven by social/cultural factors.[34] For ex-programmes to suit individual athletes? Is there anample, the Nazi atrocities of the Second World Warinnate drive to exercise? Our aim in this article is towere strongly rooted in biological determinism andoverview current understanding of the role of theafter 1945, political and intellectual thoughtenvironment and genes in constraining physical per-changed to endorse positions based on the conceptformance and the acquisition of expertise. Next, wethat differences between individuals were the resultreview the current level of knowledge regardingof opportunities and experience, coinciding with thethese issues.dominance of the Marxist ideology of Dialectical
Materialism in the Eastern European block, whileopposing ideology that sought to separate individu- 2. Environmental Constraintsals on the basis of biology were met with derision on Performanceand scorn. A similar social upheaval occurred withthe publication of Herrnstein and Murray’s[35] The While ‘nature’ refers to the innate characteristicsBell Curve in 1994. an individual contributes to their performance, ‘nur-
Despite the prominence of these extreme views ture’ makes up those qualities that result from one’sin the history of psychological medicine, it has experiences. That experiential or environmental fac-become clear that many researchers misunderstand tors have a tremendous influence on learning andthe statistic of heritability and confuse it with the performance is undeniable. In this section we pro-
vide a synopsis of the range of environmental fac- concluded that the deliberate practice frameworktors that contribute to sport performance. also applies to the development of expert athletes.
To date, psychologists examining the applicationof the theory of deliberate practice sport perform-2.1 Quantity and Quality of Trainingance have investigated a range of sports. In general,these studies have encountered problems with ap-In behavioural science, there are few relation-plying the deliberate practice framework to the at-ships as robust as the one between time spent prac-tainment of sport skill. For example, athletes tend totising and improvements in performance. Examina-rate relevant practice activities as being very enjoya-tions of skills ranging from cigar rolling[36] to read-ble and intrinsically motivating,[45,46] which con-ing inverted text[37] have supported the strongtrasts with a key component of Ericsson et al.’s[41]
positive relationship between these variables. De-definition of deliberate practice activities. Further-spite the strength of this relationship, the specificmore, there is some concern about which forms ofprofile of this association is a topic of debate, withathletic training constitute deliberate practice. In thesome arguing that it best fits a power curve[38] or anoriginal work of Ericsson et al.,[41] only practiceexponential curve,[39] while others[40] argue thatalone was seen as meeting the requirements forlearning curves are varied and diverse due to thedeliberate practice. In studies of deliberate practicedifferent rates at which individuals satisfy uniquein sport, there are few, if any, training activities thatconstraints on them during practice. Although prac-meet the original criteria in the definition set out bytice eventually becomes asymptotic, longitudinalEricsson and others.[41] Helsen et al.[46] suggestedexaminations of practice over time indicate thatthat the specifications regarding what constitutesimprovements can continue even after years of in-deliberate practice in sport should be extended tovolvement.[36]
include all relevant forms of training.Much of our understanding about environmentalDespite these problems, the relationship betweeninfluences on performance comes from studies of
hours of training and level of attainment is typicallysports expertise. In 1993, Ericsson et al.[41] producedconsistent with the tenets of the deliberate practicea seminal paper on the role of practice and expertframework. Expert athletes accumulate more hoursdevelopment that shaped a great deal of the researchof training than non-experts.[45-47] Not only do ex-that followed. Their position (based on the work ofperts spend more time overall in practice, they alsoEricsson et al.,[42] Simon and Chase[43] and others)devote more time to participating in activitieswas that, with proper attention to what they calleddeemed most relevant to developing the essential‘deliberate practice’ (i.e. relevant, effortful activitiescomponent skills for the highest level of perform-done with the specific goal of improving perform-ance.[48,49] For example, Baker et al.[48] found thatance), healthy individuals could prevent perform-expert athletes from basketball, netball and fieldance improvements from tapering off. Ericssonhockey accumulated significantly more hours inet al.[41] advocated that the relationship between timevideo training, competition, organised team prac-spent in deliberate practice was monotonic (i.e. line-tices and one-on-one coach instruction than non-ar) rather than a power or exponential function.expert athletes.Their research with elite and non-elite musicians
supported the conclusion that the primary factor The approach advocated by Ericsson and hisdistinguishing performers at different skill levels colleagues,[41] where expertise is simply the endwas the number of hours spent in deliberate practice. result of amassing the requisite number of hours ofFor example, in their examination of violinists, ex- training, has come under some criticism. For exam-perts had accumulated >7400 hours of deliberate ple, this approach reinforces the idea that specialisa-practice by 18 years of age, compared with 5300 tion during early stages of development is necessaryhours for intermediate-level performers and 3400 in order to attain elite performance as an adult. Thehours for lower-level performers. Despite originat- early specialisation approach is associated with sev-ing with musicians, Ericsson and his colleagues (for eral negative consequences,[50,51] and research-a recent review see Ericsson[44]) have repeatedly ers[50,52] have recently challenged the necessity of
early specialisation in the development of elite per- and proficiency is more complex than originallythought.formance as an adult. They proposed that a more
Considerable debate remains over the extent todiversified involvement during early development iswhich training alone determines an athlete’s ulti-equally beneficial for acquiring requisite abilities.mate level of proficiency;[59,60] however, thereBaker et al.[52] studied experts from field hockey,seems to be little doubt that attention to high-qualitybasketball and netball and found that a wide range oftraining is critical for promoting the development ofsports was typically performed during early stagesexpert performance. The only issue seems to con-of development. This broad involvement was gradu-cern the nature of the practice constraints that inter-ally decreased as the athletes developed until theyact with and shape the quality of training processes‘specialised’ into their main sport. Moreover, Bakerin high-level sport. These constraints can be cap-et al.[52] found a negative correlation between thetured as environmental or organismic in nature,[12]number of other sports played and the number ofand we discuss several examples in section 2.2.sport-specific training hours required before making
their respective national teams, suggesting that par-2.2 Access to Resourcesticipation in other indirectly relevant activities may
augment the physical and cognitive skills necessary In a recent review, Baker and Horton[61] identi-in the athletes’ primary sport. There is evidence to fied a number of critical secondary factors that mod-suggest that some general capacities are transferable erate the influence of primary factors such as train-across similar activities.[53-56]
ing. The availability of essential resources can sig-Furthermore, Cote et al.[57] also suggest that play- nificantly influence the ability to engage in the
like activities during the early stages of training are required amounts of high-quality training. We con-beneficial for skill development in many sports. In sider several of these resources in sectionsearly development, activities that are inherently en- 2.2.1–2.2.5.joyable and motivating may be necessary to provide
2.2.1 Coachesan impetus to continue training during times whenCoaches play a critical role in optimising anmore diligent, effortful practice is required. Without
athlete’s training time, in some cases having com-this pleasurable involvement, athletes run the risk ofplete control over the practice environment. Indeed,dropping out of sport.[58]
a key characteristic of expert coaches is meticulousThe notion that play may be equally beneficial to planning.[62] Researchers have dissected practices in
skill development during initial stages of involve- sports ranging from individual sports such as wres-ment is not unsupported. During early development, tling and figure skating,[49,63] to team sports such asprogress comes rapidly and easily because there is volleyball[49] and basketball.[64,65] What seems clearso much room for performance improvements. Dur- is that the structure and content of expert athletes’ing this period, it is likely that any form of relevant practice is superior to that of non-experts.participation would provide improvement, regard- In addition to the ability to maximise trainingless of whether this participation is in the form of time, proper coaching is critical for promoting adap-direct involvement through sport-specific training or tations to more advanced levels of learning andindirect involvement through sports that share com- development. As learners advance in skill develop-mon basic characteristics. However, as performance ment, cognitive, physical and emotional needsprogresses, enhancements become increasingly dif- change and the role of the coach changes according-ficult to achieve until a point where focused training ly. For example, Cote et al.[57] suggested that, duringon specific areas of weakness becomes the only early phases of development, coaches should em-means of advancement. At this point, highly struc- phasise opportunities to develop fundamental motortured practice becomes the most effective form of skills and general abilities that will form the basistraining. Although work in this area continues and for more advanced development in the future. Onlythese results should not be seen as conclusive, the later in development should the focus change todata suggest that the relationship between practice more technical aspects of training and a greater
2.2.4 Relative Ageattention to ‘deliberate’ types of practice. The abilityAnother factor that appears to influence the ac-of the coach to use domain-specific knowledge to
quisition of expertise is the ‘relative age effect’ (seeoptimally structure practice is essential to athleteMusch and Grondin[71] for a detailed review of thisprogression and development.concept). Many sporting codes group children byage to moderate evaluation and competition; howev-2.2.2 Parentser, research indicates that older individuals in a
Researchers have highlighted the critical role that given age cohort (i.e. that have a greater ‘relativeparents play in promoting the athletic development age’), report higher levels of proficiency. To date,of their children. In an examination of exceptional this effect has been reported in various educationalindividuals across a range of sports and other activi- settings[72,73] and in sports ranging from professionalties, Bloom[66] reported that parental resources, such ice hockey and baseball[74,75] to junior football.[76]
as support, were imperative in nurturing talent. In a Two main explanations have been offered tomore recent examination, Cote[67] corroborated account for the relative age effect. The first is thatBloom’s findings. older players are bigger, stronger, faster and better
Both researchers noted that athletes go through coordinated than the younger players and thus expe-stages of development and that the resources provid- rience more success and rewards and are more likelyed by parents shift as athletes mature. During early to remain involved. Younger peers, on the othersport development, parents typically provide the hand, experience failure and frustration and with-initial opportunity to get involved in sports. In es- draw. A second hypothesis is that older players aresence, parents play a leadership role by initiating more likely to be selected to higher competitivesport involvement. As the athlete matures, parental representational teams where they receive improvedinvolvement decreases and the performer takes coaching, facilities and playing time compared withgreater control of the decision-making process re- their peers. Each of these explanations have cleargarding their future career. Parents continue to pro- implications for the development of elite athletes.vide support in a background role, as providers of
2.2.5 Size of Birthplacefinancial support and, more importantly, emotional
Preliminary research has indicated that the size ofsupport.[67]
the city where an athlete spends their developmentalyears may also affect their likelihood of attaining
2.2.3 Culture elite-level performance.[77,78] In an examination ofA significant and often overlooked factor in the birthplace size in professional baseball, basketball,
development of athletes is culture,[61] which some ice-hockey players and golfers, Cote et al.[79] foundhave argued evolves at a faster rate than genetic trait that the optimal city size for athlete developmentdispositions.[68] The importance that a nation or appears to be between 1000 and 500 000 people.community places on a particular sport can have a What sets this study apart is that the effect wasdramatic influence on any success achieved. For consistent across team and individual sports andexample, in Canada, the sport of ice hockey has across countries (Canada and US).become a fundamental component of the national While the mechanisms for this ‘birthplace effect’identity.[69] Recent estimates[70] suggest that Canada are not known, it is likely that a range of factors playhas three times more children playing ice hockey a role. For example, adolescents in smaller commu-than Russia, Sweden, Finland, the Czech Republic nities receive more social support, have higherand Slovakia combined. Given these factors, levels of self-efficacy, and experience fewer con-Canada’s international success is not surprising. flicts with others than those from larger cities.[80]
Other countries have similar sport-related associa- Furthermore, smaller cities may provide an environ-tions including Austria with skiing and Kenya with ment that is more conducive to the development ofdistance running. More research is needed to devel- sports skills by virtue of having more and saferop our understanding of how culture constrains the recreational space.[81] Further research is required todevelopment of expertise in sport. confirm the specific source(s) of this effect, but
these findings reinforce the role that contextual fac- One of the most productive research programmestors play in promoting sports achievement. examining relationships between genetic factors and
Although research on sports expertise has provid- psychological outcomes has been MISTRA (Minne-ed a wealth of information about the environmental sota Study of Twins Reared Apart).[83] In addition toconstraints on the development of expert level per- examinations of general intelligence[84] and person-formers, less is known about how these variables ality,[85,86] MISTRA has also considered the influ-facilitate or impede performance at lower levels of ence of genes on psychological measures such asability. Further research is needed with intermedi- attitudes towards work[87] and job satisfaction,[88] asate-level performers to complete our understanding well as score on the Stroop Color-Word Test.[89]
of the process of skill acquisition and the role of Without exception, genes account for a significantvarious environmental constraints. portion of the inter-individual variation in these
measures. While the proportion of variance account-3. Genetic Constraints on ed for by genes varies depending on the measure andSports Performance the study sample, no psychological measure has
been found that has zero heritability.This section examines evidence for genetic con-A primary limitation of genetic studies of psy-straints on performance variability.[1] We focus on
chological variables is that their role in performancethe literature in psychological medicine and physiol-is relatively unknown. Physiological traits such asogy/molecular biology, which take different ap-muscle fibre type and aerobic capacity have beenproaches to this issue.clearly linked to performance by specific underlying
3.1 Psychological Research mechanisms. However, the links between psycho-logical traits and performance are not as clear (e.g.
The field of behavioural genetics in psychologi- how does one’s attitude towards work specificallycal medicine focuses on individual differences. De- affect performance?). This state of affairs may existspite the, at times, acrimonious debate over the role because (i) research in the field of sports medicineof genes and environment in psychological has not evolved to the point where there is a clearmedicine and related fields, it has been largely ig- understanding of the mechanisms of psychologicalnored in theoretical sport and exercise psychology effects; or (ii) the mechanisms of effect are toowhere systematic research programmes on genetic complex for current levels of understanding. What isconstraints have been conspicuous by their absence. clear, however, is that without unambiguous under-As with many other contentious issues in psycholog- standing of how psychological factors affect per-ical research, the dualist approach has failed to formance, researchers will be unable to replicate thecapture the ‘complementary nature’ of the relation-
study designs used in population genetics and mo-ship between genetic and environmental constraints,
lecular biology.that is, subtle nuances of the complex interrelationsFurthermore, while certain general traits havebetween genes and environment.
been linked to heritability, it is now widely acceptedAlthough research in this area is still limited,in behavioural genetics research that refinement ofrequiring replication and validation, there is consis-these traits into domain-specific abilities (e.g. pat-tent evidence supporting the role of genes in deter-tern recognition, strategic thinking) occurs throughmining intelligence (as measured by IQ tests) andexposure to optimal preparation in specific environ-personality. Behavioural genetics studies havements. Plomin and Colledge[90] have argued thatplaced the heritability of IQ at around 50%. Perhapsgenetic research in psychological science is movingmore interesting is the heritability of personality.beyond heritability, especially beyond the dualistPopulation examinations of a range of personality‘how much’ issue (i.e. how much of the phenotype ismeasures have reported that heritability ranges frompredicted by either genetic or environmental con-0.30 to 0.50, regardless of the instrument used or thestraints) towards a greater understanding of thepersonality trait being measured (for a review see
Plomin et al.[82]). ‘how’ question (i.e. focusing more clearly on the
mechanisms by which genes can be expressed in Other work on motor learning and performancespecific environments). has been more ambitious. A study of pursuit rotor
tracking performance from MISTRA[97] examined3.2 Studies of Genetic Constraints on Skill performance and learning of MZ (n = 64 pairs) andAcquisition and Performance DZ (n = 32 pairs) twins reared apart. The perform-
ance outcome score was time on target over 75 trialsThere have been some attempts to understand(expressed as a proportion of the perfect score of 20whether expertise, for example captured in the abili-seconds). Fox et al.[97] observed that performance ofty to recognise and perform a musical note or collec-the groups was highly similar with both showingtion of notes when sounded, known as perfect orsubstantial improvements over the five trial blocksabsolute pitch,[91] has a genetic or environmentalof the first day. Patterns of variability for bothbasis. Absolute pitch is over-represented in profes-groups were also similar. Over practice, some par-sional musicians, compared with the general popula-ticipants improved more than others leading to in-tion where its incidence has been estimated at 1 increases in the within-group variability by day 3 of10 000.[92] Using surveying techniques, investiga-the practice regime. However, it is important to notetors have shown strong familial[93] and racial dispo-that statistical analysis did not reveal significantsitions for the ability[94] in expert musicians. Inter-differences between the variances of the MZ and DZestingly, Slonimsky[95] pointed out that lacking per-twins over trials. The authors noted greater variabili-fect pitch did not hinder the careers of geniusty in correlations with task performance in the DZcomposers such as Wagner or Tchaikovsky.group over trials, although this effect may have beenSome early attempts to examine the genetic con-due, in part, to the smaller number of DZ pairstributions to performance and acquisition of motorstudied. The slope of the regression line for the DZskills also did not support a genocentric view. Forintraclass correlations for the last 2 days was close toexample, Williams and Gross[96] studied the per-zero, implying a decreasing contribution of environ-formance of 22 monozygotes (MZ; same genomemental factors as practice continued. Despite thetwins) and 41 dizygotes (DZ; non-identical twins)large inter-group differences in participants, it wason a stabilometer balance task over 6 days to ex-concluded that the consistently larger intraclass cor-amine the genetic contribution to performance andrelations for performance in the MZ group com-learning. It was expected that between-individualpared with DZ twins pointed to a significant geneticvariation in performance and learning would be lesscomponent of performance.in the MZ group compared with the DZ group. This
Fox et al.[97] proposed that a model in whichprediction was supported by data indicating agenetic and environmental effects were combinedgreater intra-pair resemblance for the MZ groupbest fit the data. The influence of heritability (re-only when learning profiles of the twins were com-flecting both genetic and environmental factors) waspared over time. Intraclass correlations were used tohigh from the first (proportion of performance vari-provide an estimate of the proportion of total pheno-ance explained = 0.66) to last trial block (0.69). Thetypic variance in performance and learning account-fact that the influence of heritability was high for theed for by heritability. Heritability effects were re-first of the initial five trial blocks (0.66, 0.53, 0.52,ported as low during the earliest stages of learning,0.55, 0.52, respectively) might be taken as evidencebut became increasingly influential later in practice.to imply that individuals rely on innate capacitiesMoreover, the proportion of variance in perform-when they attempt to perform a novel task for theance accounted for by systematic variation of thefirst few practice trials. Conclusions of a clear MZenvironment, due to manipulation of constraintsversus DZ distinction for dependent variables suchduring learning by coaches and teachers, peakedas percentage time on target, rate of improvement ofearly in practice. Although, heritability is made upperformance over trial block and improvement afterof genetic and environmental components, thesea period of rest, were based on genetic influence.findings imply that there is some potential for influ-
encing performance and learning by manipulating This line of research could be most relevant fortask constraints during practice. clinicians interested in motor learning and the acqui-
sition of expertise. However, more work is needed times separation was not at birth and intermittentto clarify the influence of heritability, since in the contact was maintained during upbringing.[1,99]
study by Fox et al.[97] it seems that motor perform- These design weaknesses have led to questionsance was confused with skill acquisition. They only whether there have been strong and clear demonstra-studied participants over 75 trials and it could be tions of behavioural similarities between MZAsargued that the skill acquisition processes were not reared in uncorrelated environments. According toexamined, but rather motor performance. Finally, Joseph[99] “… significant MZA personality andmore recent work has examined heritability in neu- behavioural correlations can be explained plausiblyromuscular coordination.[98] A twin model was used on the basis of the various environmental similari-to investigate MZ-DZ variations in accuracy and ties shared by separated identical twins and by in-economy of effort, assessed by analysing kinematic flated figures resulting from bias and error in thedata and electromyographical records of muscle ac- various studies”.tivity. Data showed that heredity accounted for the
3.4 Physiological/Moleculargreatest proportion of MZ-DZ twin differences inBiological Researchmovement accuracy and economy for fast elbow
flexion movements, not slow velocity movements. Research in the disciplines and sub-disciplines ofbiological psychology, medicine, physiology and3.3 Issues with Twin Studiescell biology has favoured a molecular approach inattempting to identify the genetic influence on indi-In the literature on genetic variations and per-vidual differences in human behaviour, rather thanformance, there are a number of substantial method-relying on the twin study methodologies favouredological concerns with twin and adoption stud-by psychologists. As highlighted in the introduction,ies.[1,12,99] Classical twin research methodologya number of candidate genes that contribute to per-compares the correlation or concordance rates withformance variability continue to be identified in thethe same measures from same sex DZ twins rearedmolecular biology and exercise and sports physiolo-together and MZ twins reared together. Identicalgy literature (for a review see Davids et al.[100]). The(MZ) twins share 100% of the same genes andmolecular search for the genetic bases of humanfraternal or non-identical (DZ) pairs only 50% oncapacities, such as physical performance, has engen-average. Greater MZ similarity is taken as evidencedered strong rhetoric in some quarters, with someof the powerful influence of genetic constraints. Amolecular biologists calling it the ‘biological coun-popular method has been to study identical twinsterpart’ to the holy grail,[101] and some sport scien-separated at birth (MZA: monozygotic, reared apart;tists asserting that genes are responsible for up toe.g. MISTRA), and raised under different socioeco-half the variation in physical performance betweennomic and cultural constraints. Such a comparison isindividuals within a population.[102] But does currentbelieved to provide an ideal analysis of nature andscientific understanding support the philosophy ofnurture effects. Genetic inferences from separatedbiological determinism? Is the ‘single gene as magictwin studies are based on the assumption that theirbullet’ rationale a valid basis for investigating theshared environments were not systematically morecomplexities of genetic and environmental con-similar than a group of unrelated and randomlystraints on performance in sport? In this section weselected paired individuals, the so-called ‘unequalexamine some evidence for the genetic bases ofenvironment assumption’. The problem, accordingphysical performance in humans showing that theto Joseph,[99] is that these comparisons are almostbio-deterministic viewpoint is not supported.impossible to achieve in reality. It is a daunting task
to obtain ‘pure’ separated twin samples that fit the There has been an ongoing, concerted endeavourstringent criteria needed for a good test of genetic to produce a human gene map for physical perform-and environmental constraints. This is because sepa- ance and health-related fitness.[103] Most effort torated twins are often reared by different parts of the identify single gene variants influencing the genet-same extended family, or are brought up in highly ics of fitness and physical performance has beensimilar cultures and socioeconomic contexts. Some- conducted on ACE insertion/deletion (I/D) poly-
morphisms in different groups of exercisers includ- exercise training down- and upregulated expressioning elite athletes and non-exercisers. However, the of type I myosin heavy-chain messenger RNAACE gene is one of a number of candidate genes (mRNA), respectively, whereas both environmentalresearchers have associated with inter-individual va- factors upregulated expression of type IIx myosinriability in physical performance.[103,104] Other stud- heavy-chain mRNA.[110]
ies have examined the relationship between the insu- Twin-based research has revealed for some timelin-like growth factor-1 (-I) genotype and responses that at least 50% of variance in baseline strength andto muscle training.[105] Some research has attempted lean body mass is accounted for by genetic fac-to determine mitochondrial DNA and α3-actinin tors,[111,112] with muscle adaptation to strength train-(ACTN3) genotypes in elite national-level athletes ing also long considered to be high in heritabili-in endurance events (n = 52) and sprints (n = 89). A ty.[113,114] Genes that constrain production of pro-trend towards higher ACTN3 X/X genotype fre- teins are believed to be influential in regulatingquency was observed in the endurance athletes.[106] muscle strength and muscle mass.[105] For this rea-A detailed overview of all the work on single gene son, IGF-I is believed to be a prime candidate genet-variants for fitness and performance is beyond the ic marker for these purposes, although the environ-scope of this article. Instead, we shall focus on mental influence of training is clearly implicated inresearch in key areas of skeletal muscle and muscle its effect on muscle mass and strength gains.[115] Inenergy systems in humans as well as the ACE gene research on muscle fibre gene expression,to exemplify the benefits of a model of nature and calcineurin has been proposed to stimulate the slownurture interactions to explain fitness and perform- and fast muscle fibre-type phenotype. This is impor-ance variations in humans. tant given the role of intracellular calcium levels in
regulation of slow twitch fibres.[116,117] In more3.5 Skeletal Muscle and Skeletal Muscle health-related research, the influence of environ-Energy Systems mental factors on gene expression has also been
demonstrated on different muscle fibre types. ForThe effects of various environmental factors onexample, dietary regulation of fat oxidative genegene expression in fibre typology of striated muscleexpression has been examined for type I (soleus)have been studied in humans and animals, withand type II (extensor digitorum longus) muscle fib-many studies reporting interactive effects betweenres in humans.[118] Evidence showed that the expres-expression of single gene variants and a number ofsion of an array of genes was increased regardless ofenvironmental factors including microgravity condi-muscle fibre type.tions, training, illness, obesity and aging.[107,108] In
Finally, the role of the enzyme adenosineparticular, onset of sarcopenia with aging has stimu-monophosphate deaminase (AMPD) has been iden-lated increased effort to study the inter-individualtified in regulating muscle energy metabolism dur-variability in adaptation to muscle strength training,ing intense endurance exercise, with the skeletalcausing some researchers to seek a genetic explana-muscle-specific isoform of AMPD encoded by thetion for the individual variations.[109] Other work hasgene AMPD1, located on chromosome 1p13-demonstrated how environmental constraints suchp21.[119] Although a deficiency of AMPD was con-as training or enforced immobilisation (simulatingsidered to be responsible for exercise limitation andinjury or illness) interact with gene expression indecrements in functional capacity, reports suggestmuscle fibre type. Adaptation to environmental con-that this was not the case, as long as training andstraints appears to mediate gene expression in mus-experience to an elite level had occurred.[119]
cle fibre and myofibril, although the interactioneffects can differ extensively. Myosin heavy chain is 3.6 The ACE Genea highly significant protein, with three functionallydistinct forms, and research has attempted to under- Over the last decade, the role of ACE gene vari-stand how its expression in muscle is affected by ants has received considerable attention in the exer-environmental constraints such as diet, training and cise physiology, molecular biology and sportsrest.[107] For example, enforced immobilisation and medicine literature bases.[120-123] Significant inter-
individual variations in response to training of the levels of ACE in muscle than the DD allele, and ancardiovascular system have led investigators to increased response to physical training.[128] Recruitsquestion the extent to which genetic diversity may with ACE genotype II differed by as much as 1100%be responsible for the data.[124,125] In muscle, ACE in response to repetitive upper-arm exercises com-has the role of degrading vasodilators (i.e. pared with DD genotype peers. Individuals with abradykinin and tachykinin) and stimulating produc- heterogeneous genotype (DI) were associated withtion of the vasoconstrictor angiotensin II during levels of performance between both homozygousphysical performance.[126] Three variants of the pol- genotypes. In sport, a higher prevalence of the IIymorphism of the human ACE gene have been genotype has been found in elite endurance athletesfound. The presence or absence of a 287 bp frag- including mountaineers able to resist effects of hyp-ment, characterises the I (insertion) and D (deletion) oxia and climb to 7000m without the aid of oxygen,allele respectively, leading to three variants (II, ID as well as Olympic-standard endurance runners andand DD). elite rowers.[128-130] In their compendium of all genes
and markers associated with health-related fitnessIncreasing ACE activity has been linked with theand performance, Rankinen et al.[103] revealed sixD allele, affecting the degradation of bradykinin andstudies reporting a positive relationship betweensynthesis of angiotensin II. DD participants showpower performance and muscle strength-relatedincreased conversion of angiotensin I to angiotensinphenotypes.II, the latter having a vasoconstriction effect. How-
ever, since the conviction of early research, there Do these data imply that athletes are born and nothas been less certainty in the outcomes of studies on made? Despite the rhetoric in the popular scienceACE gene polymorphisms on endurance perform- literature, it seems that there is currently no clearance. For example, work on Kenyan elite middle- evidence to support that implication. The appropri-and long-distance runners, compared with non-ath- ateness of this conclusion in the face of claims thatlete controls, found no differences in genotype fre- human physical performance is ‘strongly’ influ-quencies between the groups.[121] Clearly more work enced by genetic factors[130] was underscored in ais needed in this area since Rankinen et al.[103] noted study by Bouchard et al.[131] who attempted to estab-that three cross-sectional association studies pub- lish the proportion of influence attributable to genet-lished in 2005 reported a positive performance ef- ic and environmental constraints on familial resem-fect for endurance-related phenotypes in the ACE blance for maximal oxygen uptake during exercisegene. on a cycle ergometer in sedentary individuals.[131]
Exercise performance of fathers, mothers, sons andAngiotensin II seems to have a stimulatory influ-daughters was measured in 86 nuclear families.ence on endogenous factors for muscle cell growth,Maximum heritability including genetic and non-contributing to a hypertrophic training response use-genetic causes for physical performance accountedful for power development.[127] Degradation offor 51% of the total adjusted phenotype variance.bradykinin results in lower substrate metabolismSeveral models of interacting constraints were test-and less efficient vasodilation. Therefore, lowered and results showed that there was 2.6–2.9 timeslevels of ACE activity may be associated with in-more variance between families than within fami-creased half-life bradykinin, which alters substratelies.metabolism. Increased angiotensin II is associated
with the DD genotype and may facilitate muscle Yet again, difficulty in disambiguating the spe-bulk for power sport performance. It is estimated cific constraints of genes and environments wasthat 25% of the population have the II genotype, apparent in that study. Genetic and familial environ-50% the ID genotype and 25% the DD genotype.[104] mental influences could not be ‘fully quantified
Do particular variants of the ACE gene genotype separately’ although ‘inferences about their respec-occur more frequently in specific populations (e.g. tive contributions to the phenotype variance couldendurance or power athletes) compared with con- be made by inspection of the pattern of familialtrols? Early work with army recruits found that the II correlations’. Of course correlations do not implyform of the gene was associated with lower activity causation. The emphasis on the constraints imposed
by the shared familial environment is also important of genetic transmission, including transcription andto note. While this explanation may hold, it does not translation, revealing that expression levels of singlepreclude influence of ‘wider’ environmental con- genes or networks of genes are far more open tostraints such as sociocultural changes in society, stochastic fluctuations in constant environmentsincluding the impact of media images, ‘stereotype than previously thought.[134] It is now proposed thatthreat’,[132] government education programmes and molecular interactions are inherently random lead-peer group pressure. ing to stochasticity in the biochemical processes of
transcription, protein synthesis and the basis of geneOther research on human behaviour has revealedtransmission.[135] Intracellular processes involvethat genes work in combined networks to influencethousands of chemical interactions and a high levelbiological function, refuting the idea of successfulof random fluctuations has been identified in RNAathletes being differentiated on the presence of alevels and expressed proteins with the potential forsingle gene variant (for a similar argument in devel-considerable unpredictability in molecularopmental theory see Johnston and Edwards[133]).events.[136] As Kauffman[134] noted, random chemi-Rankinen at al.’s[103] record on the progress in mo-co-electrical fluctuations are a serious issue in ge-lecular biology towards a human gene map, notednetic networks since cellular regulatory processes inthat although 165 autosomal entries, 5 X chromo-biological systems are inherently noisy. The use ofsome assignments and 17 mitochondrial markersnegative feedback loops helps to stabilise systemwere obtained in 2005 alone, there is still littledynamics against effects of perturbing fluctua-understanding of the role that such genes might playtions.[136] Simulation modelling of the gene trans-in individual differences in health-related fitnessmission process has shown that deterministic viewsand physical performance. Therefore, whilst re-of protein concentration and synthesis, the basis ofsearch may be capable of identifying single genegene expression, are untenable due to the potentialvariants believed to be involved in variations infor fluctuations in these critical variables, bothfitness and physical performance, many scientistslinked with population heterogeneity.[135] Tradition-have moved away from an extreme biologicallyal views hold that stochasticity is an index of cellulardeterministic stance. Despite this, there has beennoisiness, disruptive to cell regulation processes.inadequate progress in understanding how singleSome more recent interpretations of stochasticitygene variants may interact with different environ-suggest that it is a mechanism for cellular and phe-mental constraints, such as enforced bed rest, ill-notypic diversification, providing an evolutionaryness, aging, space travel and sports training.[121]
advantage for organisms that benefit from it.[135]These more complex research investigations areprecisely what are needed with respect to under- Although some geneticists have pointed to thestanding how practical interventions may interact propensity of some physiological processes forwith distinctive genotypical profiles. destabilising the phenotype, the main argument pro-
posed is that there is a ‘redundancy and reserve3.7 Stochasticity in Cellular capacity’[122] in the design of neurobiological sys-Regulation Processes tems to cope with these perturbations. Later in this
article we argue that a more accurate characterisa-The arguments against a biologically determinis-tion in neurobiology may be that genetic regulatorytic view of processes of gene transmission are im-systems are ‘degenerate’ and not replete with redun-portant to note because of the prevalence of thedancy.[137,138] This is because the propensity for re-implicit assumptions underlying the ‘single-gene asentrant trajectories between network states has pro-magic bullet’ philosophy held by some sport practi-vided valuable insights into the functional role oftioners. Somehow over the years, the view of DNAstochasticity in stabilising diversified cell types asas information bearer has been replaced with theattractors within cellular networks.[134]
mechanistic fallacy of DNA as blueprint, plan, ormaster molecule.[7] Fortunately, the ‘gene as What are the implications of this new knowledgeblueprint’ traditional rationale is giving way to bet- from molecular biology for sport scientists interest-ter understanding of the mechanisms and processes ed in individual differences between performers? It
is critical that scientists and clinicians understand The shift in understanding of the cellularmechanics of gene expression should lead to a cor-the biological significance of this previously unob-rection in the relentless advance to find the ‘singleserved stochasticity in gene expression. Further-gene as magic bullet’. Some molecular biologistsmore, other data mitigating against the blueprintare leading the way in advocating a less determinis-conceptualisation of genetic transmission processestic view of gene expression based on their researchcan be observed in evidence that identical twins arefindings. For example, Jones et al.[104] have notednot actually identical. Evidence from the study ofthat “The ACE I/D polymorphism should not bephenotypically identical twins showed that their fin-considered a ‘gene for human performance’, but agerprints differ and the shape of their brains canmarker for modulation such that one would expectdiffer by as much as 40%.[139] This is important foran excess of the I allele in the truly elite endurancesport clinicians to note because typically they areathlete, with a concordant excess of the D alleleexposed to research findings from behavioural ge-represented in the more power-oriented events”.netics, which are concerned with explanations of
With respect to understanding the nature of ge-hereditary influences at the level of populations, notnetic constraints on variability in sport performance,individuals. Heritability estimates are specific to ait is not clear how useful this information could besample studied, and changes to the environment canfor sport practitioners designing specific trainingchange the heritability estimate from the same sam-and selection programmes. Jones et al.[104] also stat-ple of individuals with the same genotypes.[1]
ed that “The ACE genotype has never been associat-Clearly, therefore, as sequencing of genomes ined with endurance performance in the untrainedbiological organisms gathers pace, it is becomingstate. Any effect appears to require a period of gene-apparent that genes should not be viewed as aenvironment interaction. A high level of aerobicblueprint for success in sport. As Johnston and Ed-fitness is an essential, but not sole, requirement forwards[133] have pointed out, it is “a very long stepelite endurance”. A final point to note according tofrom polypeptide sequences to behaviour – a stepJones et al.[104] is that “There will always be elitethat covers much incompletely understood territo-endurance athletes who are of the ACE DD geno-ry”. Attempts to see genes as building plans are onetype, and many champions in anaerobic sports of theof the great artificialisms of human conceptualisa-II genotype. Whatever the data may conclude, elitetions of nature.[140] It has become a ‘central dogma’athletes are still made and not born, though perhapsof how people think about the process of evolu-some may be made elite in one discipline moretion.[141] Genes simply contain the information toeasily than others”.synthesise proteins with properties leading them to
cluster together. Lewontin[7] criticised the stance of 4. The Fallacy of a Dualist ‘Debate’:‘biological determinism’, the medical model’s re- Alternative Approaches to Naturejection of polymorphism and the implicit notion of and Nurturevariability as deviation from a ‘perfect ideal’. Ge-netic diversity is the norm and biological systems Our review of the strengths and weaknesses ofare not DNA determined. There is no single, stan- theoretical ideas and empirical research for theoriesdard, normal DNA sequence that human beings of learning and performance has revealed majorshare and it is estimated that DNA sequencing in implications for understanding how environmentalindividuals varies by 0.1% (≈3 million nucleotides) and genetic constraints function. It can be concludedincluding inherited sequences from parents. It takes that neither specific approach provides enough ex-more than DNA to produce a living organism, which planatory power to account for all the data on per-cannot be computed from DNA sequences. Accord- formance variability. There is now clear evidenceing to Lewontin,[7] “A living organism at any mo- rejecting the idea that single gene variants can pre-ment of its life is the unique consequence of a dispose an athlete to superior performance manifest-developmental history that results from the interac- ed in a specific domain, without clear and detailedtion of and determination by internal and external consideration of the performance context (e.g. aforces”. gene for soccer performance). Edelman and Gal-
ly[137] have highlighted the fallacy of this view by formance remains an important focus for futureresearch on genetic and environmental constraints.arguing that “All observable properties of an organ-
ism are determined by the workings of a degenerate These ideas on the dynamic relationships be-network of many genes”. Degeneracy exists at all tween genes and environment and the functionallevels of biological movement systems and is tech- role of degeneracy in the human nervous system arenically defined as non-isomorphic components pro- harmonious with a dynamical systems theoreticalducing isofunctional outcomes, effects or solu- perspective on the influence of interacting con-tions.[138] Degeneracy in gene networks promotes straints. In this regard, science needs to move be-evolutionary fitness of a species by ensuring that yond a characterisation of natural phenomena in
‘dualisms’ to understand the ‘complementary na-genetic diversity supports functional adaptation toture’.[26] The application of dynamical systems theo-variable environments. Neurobiological degeneracyry to the study of neurobiological performance canprovides the basis for gene expression as an inher-provide a substantial theoretical framework for in-ently stochastic process.[135] We also noted that theterpreting the complementary nature of genetic andimplicit basis of the deliberate practice perspectiveenvironmental constraints on each individual per-is the adage ‘all healthy individuals are createdformer.[12,100] From this theoretical standpoint, ge-equal’. Analysis of the literature on genetic con-netic diversity may be responsible for a small part ofstraints on variability of performance does not sup-training or performance response differences be-port this conclusion, but this interpretation of thetween individuals, and only when there is a favour-literature should not be taken to imply that physicalable interaction with important environmental con-performance is biologically determined. Rather, ef-straints are performance benefits observed. For ex-fects of interacting constraints on acquisition ofample, there is growing consensus in the study ofskill, expertise, health and physical performancehuman obesity that the contribution of genetic fac-have been noted, since despite variations in genetictors is exacerbated in environments that differ instructure, maximal heritability of particular traitscaloric availability.[143] Genetic propensity towardsincludes strong environmental components. Theadiposity has less of a constraining influence onpresence of dynamically varying environments in-individuals in environments where caloric availabil-teracts with an inherent property of human move-ity is lower, whereas these same individuals would
ment systems, degeneracy, to signal a new view onbe at greater risk in caloric-rich environments. Such
variability. Degeneracy is an omnipresent property environments can be categorised as high or low risk,of many complex, neurobiological systems that fa- depending on prevalence of other significant cultur-cilitates achievement of functional movement out- al constraints including lack of training facilities,comes in variable environments through diversity of work patterns imposed on traditional meal times,processes.[138] This new perspective on variability and the fall in popularity of physically active pas-reveals that compensatory adaptation in perform- times, leading to a greater emphasis on more staticance achievement occurs as the result of system activities such as playing computer games and TVtrade-offs between specificity and diversity of watching.behaviours.[137]
Strong interaction effects on endurance perform-Research considered in this article has shown ance of amphibians have been clearly demonstrated
that, in the development of individuals, biological in evolutionary physiology studies manipulating di-dispositions and environments should not be con- et. A high initial heritability component (0.40) forceptualised as separately evolving, static entities, locomotory endurance capacity (running time tobut rather as dynamically yoked. Biological disposi- exhaustion just after birth) was altered with dietarytions and environmental influences are responsive to changes, showing how expression of genetic tenden-each other as exemplified in research on the chang- cies for physical performance is dependent on envi-ing structure of the CNS with learning exper- ronmental contexts.[144] Although this work needs toiences.[142] Determining the impact of neurobiologi- be replicated with humans, it appears that phenotyp-cal degeneracy on achievement of elite sport per- ic expression of exercise behaviours might be best
understood at the level of individual interactions accuracy. Given these constraints on performancewith key environmental constraints and associated and behaviour, prediction of athletic performancerisk, rather than as defective behaviour in a ‘medical will always be limited by uncertainty and com-model’.[100] This is particularly relevant when con- pounded by degeneracy. However, understandingsidering effects of time spent in practice in sport. the range of influences that affect sports perform-Given differences in genetic contributions, perform- ance is within our grasp, provided we ask the rightance variations are more likely to assert themselves questions.under intensive practice regimes. In contrast to the bullish search by some molecu-
An implication of these findings and of other data lar biologists for the single gene variant thatreviewed in this article for understanding individual characterises elite performance in specific sportsvariations in physical performance is that there may and physical activities, psychologists generallybe diverse ways for neurobiological systems to seem to have accepted the idea of co-action of genesachieve similar performance outcome levels. For and environment, preferring to highlight the viewexample, athletes of a putatively less favourable that no psychological measure shows zero heritabili-genotypic disposition might succeed as long as they ty.[90] Furthermore, it is possible to interpret existingare exposed to appropriate training environments. data on inter-individual variability in the health andHowever, it can be concluded that performers with a physical activity domain based on the interaction ofmore favourable genotype, who interact with appro- genetic and environmental constraints. For example,priate training environments, are more likely to re- an interesting issue concerns the relationship be-ceive a greater training response. Current data on tween propensity to engage in daily physical activitygenetic constraints on motor skill acquisition are and inter-individual differences in resting metabolicunclear due to a number of methodological weak- rate. The heritability estimate of the tendency tonesses and conflicting findings, and there is a need exercise and participate in sports and physical activ-for more work to identify genetic mechanisms un- ity has been provided[24] and this issue was consid-derlying performance variations. Moreover, the ered with groups of five different types of twinsgreater emphasis on constraints in the dynamical differing on MZ-DZ and male-female dimensions insystems perspective implies that the nature-nurture adolescents aged 15 years.[146] Although, the origi-argument can be superseded, as we note in section 5. nal biometric model traced to Fisher in 1918 for
estimating the heritability statistic did not include a5. Where to from Here? The Future of thecomponent for interaction of genotype and environ-Nature-Nurture Debatement, recent evidence suggests that an interactionistmodel exhibiting the co-influence of genetic factors,Based on evidence reviewed, one of the mostas well as unique and shared environments provideslikely reasons the nature-nurture debate has not beena best fit with the data on time spent in physicalresolved is that it cannot be resolved using currentactivity. In females, the genetic component contrib-dualist paradigms that separate influences into dis-uted 44% of the variance explained, with sharedtinct categories (i.e. either nature or nurture). At itsenvironmental factors mainly contributing the rest.root, behavioural research will never predict eventsFor males, the genetic component was more domi-with the same level of certainty as fields like biolo-nant, contributing 83% of the variance. The authorsgy, physics and chemistry. Dynamic neurobiologi-explained this major difference in the data as ema-cal systems, such as athletes, are dramatically affect-nating from the importance of the social environ-ed by any uncertainty (or chaos) in the behaviour ofment for female members of twins and physicalperformance-related variables. According to Gollubmorphological factors for males at the age studied.and Solomon,[145] “any uncertainty [in a complexOther evidence has found that genes may influencesystem]… no matter how small, will lead to rapidlythe disposition towards higher intensity exercise ingrowing errors in any effort to predict the futurephysical activity choices, although attempts to im-behaviour”. The behaviour of complex systems isplicate marker genes with individual differences inonly completely predictable when the componentsresting metabolic rate have found only weak andof performance are known to an infinite degree of
inconclusive support.[146] The implication is that the teractive approach for analysing relative contribu-genes-environment interaction may have different tions of genes and environment in degeneratedimensions for males and females, which could neurobiological systems may shed light on a numberhave important implications for planning physical of significant issues in sports science and medicineactivity participation programmes. related to the acquisition of expertise in sport.
6. Conclusions AcknowledgementsFrans de Waal, in his address to the 2001 confer- No sources of funding were used to assist in the prepara-
ence of the American Psychological Association, tion of this review. The authors have no conflicts of interestthat are directly relevant to the content of this review. Theadvocated for an approach to human behaviour thatauthors wish to acknowledge the help of Pam Smith of theintegrated the effects of learning and the environ-School of Human Movement Studies, Queensland Universityment with Darwinian perspectives of biology andof Technology, in preparing this manuscript for publication.
genetics.[147]
Allied to this recommendation, the literature re-Referencesviewed in this article suggests that many clinicians
1. Vitzthum VJ. A number no greater than the sum of its parts: theand sports scientists may be expecting more from use and abuse of heritability. Hum Biol 2003; 75: 539-882. Podewils LJ, Guallar E, Kuller LH, et al. Physical activity,quantitative molecular studies than can be delivered.
APOE genotype, and dementia risk: findings from the Cardio-A requisite step forward would be to discard bothvascular Health Cognition Study. Am J Epidemiol 2005; 161:
the dualist approach to nature versus nurture and the 639-513. Schuit AJ, Feskens EJM, Launer LJ, et al. Physical activity and‘gene as magic bullet’ philosophy. It is becoming
cognitive decline: the role of the apolipoprotein e4 allele. Medclear that sports performance is the result of interac- Sci Sports Exerc 2001; 33: 772-7tions among a host of genes and environmental 4. Wilson EO. Sociobiology: the new synthesis. Cambridge (MA):
Harvard University Press, 1975constraints. There are important implications for5. Hellman H. Great feuds in science: ten of the liveliest disputes
advising clinicians, teachers, coaches, administra- ever. New York: Wiley, 19986. Ceci S, Williams W. The nature-nurture debate: the essentialtors and athletes on the seductive, but highly sim-
readings. New York: Blackwell Publishing Co., 2000plistic, myths surrounding ‘single-gene-as-magic-7. Lewontin RC. It ain’t necessarily so: the dream of the human
bullet’ approaches, which have become more preva- genome and other illusions. London: Granta Books, 20008. De Geus EJC, Boomsma DI. A genetic neuroscience approachlent in sports medicine. These implications relate to
to human cognition. Eur Psychologist 2001; 6: 241-53the uncritical acceptance of general models for tal- 9. Bouchard Jr TJ, McGue M. Genetic and environmental influ-ent identification and development programmes ences on human psychological differences. J Neurobiol 2003;
54: 4-45across the world. For example, there is no single10. Kimble GA. Evolution of the nature-nurture issue in the history
gene for ‘hand-eye’ or ‘hand-foot’ coordination in of psychology. In: Plomin R, McClearn GE, editors. Nature,nurture and psychology. Washington, DC: American Psycho-children. But there are nervous system thresholdslogical Association, 1993: 3-25and differences in children’s opportunities to exploit 11. Petrill SA, Lipton PA, Hewitt JK, et al. Genetic and environ-
critical maturational phases in developing timing mental contributions to general cognitive ability through thefirst 16 years of life. Dev Psychol 2004; 40: 805-12behaviour in interceptive actions.[148] Sport re-
12. Baker J, Davids K. Genetic and environmental constraints onsearchers are encouraged to consider interactionist variability in sport performance. In: Davids K, Bennett SJ,
Newell KM, editors. Variability in the movement system: adesigns when investigating the range of influencesmultidisciplinary perspective. Champaign (IL): Human Kinet-on the development of skilled performance. Never- ics, 2006: 129
theless, establishing the validity of moderating and 13. Brutsaert TD, Parra EJ. What makes a champion? Explainingvariation in human athletic performance. Resp Physiolmediating relationships will be a long-term processNeurobiol 2006; 151: 109-23
due to the complexity of the relationships suggested 14. Pitsiladis Y, Bale J, Sharp C, et al., editors. East Africanrunning: towards a cross-disciplinary perspective. London:and the difficulties associated with identifying inter-Routledge, Taylor & Francis, 2006active relationships of this nature (for a review see 15. Miah A. Genetically modified athletes: biomedical ethics, gene
Wahlsten[149]). Cronbach[150] stated, “serious inter- doping and sport. London: Routledge, Taylor & Francis, 200416. Tamburrini C, Tansjo T, editors. Genetic technology and sport:action research ought to be a program of several
ethical questions. London: Routledge, Taylor & Francis, 2005years’ duration, although there are few examples in 17. Coghlan A. Elite athletes born to run. New Scientist 2003; 30:4-5behavioural science”. Although challenging, an in-
18. Yang N, MacArthur DG, Gulbin JP, et al. ACTN3 genotype is 46. Helsen WF, Starkes JL, Hodges NJ. Team sports and the theoryassociated with human elite athletic performance. Am J Hum of deliberate practice. J Sport Exerc Psychol 1998; 20: 12-34Genetics 2003; 73: 627-31 47. Hodge T, Deakin J. Deliberate practice and expertise in the
19. Coghlan A. How the right genes may get you off to a flying start. martial arts: the role of context in motor recall. J Sport ExercNew Scientist 1998 May; 23: 4 Psychol 1998; 20: 260-79
20. Anderson JL, Schjerling P, Saltin B, et al. Muscle, genes and 48. Baker J, Cote J, Abernethy B. Learning from the experts:athletic performance. Scientific Am 2000; 283: 31-7 practice activities of expert decision-makers in sport. Res Q
Exerc Sport 2003; 74: 342-721. Friedmann T, Koss JO. Gene transfer and athletics: an impend-ing problem. Mol Ther 2001; 3: 819-20 49. Deakin JM, Cobley S. An examination of the practice environ-
22. McCrory P. Super athletes or gene cheats. Br J Sports Med ments in figure skating and volleyball: a search for deliberate2003; 37 (3): 192-3 practice. In: Starkes JL, Ericsson KA, editors. Recent advances
in the study of sport expertise. Champaign (IL): Human Kinet-23. Dennis C. Rugby team converts to give genes test a try. Natureics, 2003: 115-352005; 434: 260
50. Baker J. Early specialization in youth sport: a requirement for24. Stubbe JH, Boomsma DL, De Geus EJC. Sport participationadult expertise? High Ability Stud 2003; 14: 85-94during adolescence: a shift from environmental to genetic
factors. Med Sci Sports Exerc 2005; 37: 563-70 51. Wiersma LD. Risks and benefits of youth sport specialization:perspectives and recommendations. Pediatr Exerc Sci 2000;25. Unal M, Unal DO. Gene doping in sports. Sports Med 2004; 34:12: 13-22357-62
26. Kelso JAS, Engstrom DA. The complementary nature. Cam- 52. Baker J, Cote J, Abernethy B. Sport specific training, deliberatebridge (MA): MIT Press, 2006 practice and the development of expertise in team ball sports. J
Appl Sport Psychol 2003; 15: 12-2527. Galton F. Inquiries into human faculty and its development.London: Macmillan, 1883 53. Abernethy B, Baker J, Cote J. Transfer of pattern recall skills as
a contributor to the development of sport expertise. Appl28. Galton F. On men of science: their nature and their nurture.Cognitive Psychol 2005; 19: 705-18Nature 1874; 9: 344-5
54. Flynn MG, Carroll KK, Hall HL, et al. Cross training: indices of29. James W. Principles of psychology. New York: Holt, 1890training stress and performance. Med Sci Sports Exerc 1998;30. Terman L. Genetic studies of genius. Stanford (CA): Stanford30: 294-300University Press, 1925
55. Millet GP, Candau RB, Barbier B, et al. Modelling the transfers31. Terman LM, Oden MH. The gifted child grows up. Stanfordof training effects on performance in elite triathletes. Int J(CA): Stanford University Press, 1947Sports Med 2002; 23: 55-6332. Terman LM, Oden MH. The gifted group at mid-life. Stanford
56. Mutton DL, Loy SF, Perry DM, et al. Effect of run vs combined(CA): Stanford University Press, 1959cycle-run training on aerobic capacity and running perform-33. Watson JB. Behaviorism. New York: Norton, 1924ance. Med Sci Sports Exerc 1993; 25: 1393-734. Pinker S. The blank slate: the modern denial of human nature.
57. Cote J, Baker J, Abernethy B. From play to practice: a develop-New York: Viking, 2002mental framework for the acquisition of expertise in team35. Herrnstein RJ, Murray C. The bell curve: intelligence and classsports. In: Starkes J, Ericsson KA, editors. Expert performancestructure in American life. New York: Free Press, 1994in sports: advances in research on sport expertise. Champaign36. Crossman ERFW. A theory of the acquisition of speed-skill.(IL): Human Kinetics, 2003: 89-110Ergonomics 1959; 2: 153-66
58. Petlichkoff LM. Coaching children: understanding the motiva-37. Kolers PA. Memorial consequences of automatized encoding. Jtional process. Sport Sci Rev 1993; 2: 49-61Exper Psychol Hum Learn Memory 1975; 1: 689-701
59. Abernethy B, Farrow D, Berry J. Constraints and issues in the38. Newell A, Rosenbloom PS. Mechanisms of skill acquisition anddevelopment of a general theory of expert perceptual-motorthe law of practice. In: Anderson JR, editor. Cognitive skillsperformance. In: Starkes JL, Ericsson KA, editors. Expertand their acquisition. Hillsdale (NJ): Erlbaum 1981: 1-55performance in sports: advances in research on sport expertise.39. Heathcote A, Brown S, Mewhort DJK. The power law repealed:Champaign (IL): Human Kinetics, 2003: 349-69the case for an exponential law of practice. Psychonomic Bull
60. Sternberg RJ. Costs of expertise. In: Ericsson KA, editor. TheRev 2000; 7: 185-207road to excellence: the acquisition of expert performance in the40. Newell KM, Liu Y-T, Mayer-Kress G. Time scales in motorarts and sciences, sports and games. Mahwah (NJ): Lawrencelearning and development. Psychological Rev 2001; 108:Erlbaum Associates, 1996: 347-5357-82
61. Baker J, Horton S. A review of primary and secondary influ-41. Ericsson KA, Krampe RT, Tesch-Romer C. The role of deliber-ences on sport expertise. High Ability Stud 2004; 15: 211-28ate practice in the acquisition of expert performance. Psychol
62. Voss J, Green T, Penner B. Problem-solving in social sciences.Rev 1993; 100: 363-406In: Bower G, editor. The psychology of learning and motiva-42. Ericsson KA, Chase WG, Faloon S. Acquisition of a memorytion: advances in research theory. New York: Academic Press,skill. Science 1980; 208: 1181-21983: 165-21343. Simon HA, Chase WG. Skill in chess. Am Sci 1973; 61:
63. Starkes JL. The road to expertise: is practice the only determi-394-403nant? Int J Sport Psychol 2000; 31: 431-5144. Ericsson KA. Development of elite performance and deliberate
64. Bloom GA, Crumpton R, Anderson JE. A systematic observa-practice: an update from the perspective of the expert perform-tion study of the teaching behaviors of an expert basketballance approach. In: Starkes JL, Ericsson KA, editors. Expertcoach. Sport Psychologist 1999; 13: 157-70performance in sports: advances in research on sport expertise.
65. Tharp RG, Gallimore R. What a coach can teach a teacher.Champaign (IL): Human Kinetics, 2003: 49-83Psychol Today 1976; 9: 74-845. Starkes JL, Deakin J M, Allard F, et al. Deliberate practice in
66. Bloom BS. Developing talent in young people. New York:sports: what is it anyway? In: Ericsson KA, editor. The road toBallantine Publishing Group, 1985excellence: the acquisition of expert performance in the arts,
sciences, sports and games. Mahwah (NJ): Erlbaum, 1996: 67. Cote J. The influence of the family in the development of talent81-106 in sports. Sport Psychologist 1999; 13: 395-417
68. Ehrlich PR. Human natures: genes, cultures, and the human 91. Schlaug G, Jancke L, Huang Y, et al. In vivo evidence ofprospect. Washington, DC: Island Press, 2000 structural brain asymmetry in musicians. Science 1995; 267:
699-70169. Russell S. Ice time: a Canadian hockey journey. Toronto (ON):Viking, 2000 92. Sacks O. Musical ability [letter]. Science 1995; 268: 621
70. Robinson L. Crossing the line: violence and sexual assault in 93. Baharloo S, Service SK, Risch N, et al. Familial aggregation ofCanada’s national sport. Toronto: McCelland & Stewart Inc., absolute pitch. Am J Hum Genet; 2000; 67: 755-81998 94. Gregerson PK, Kowalsky E, Kohn N, et al. Absolute pitch:
71. Musch J, Grondin S. Unequal competition as an impediment to prevalence, ethnic variation, and estimation of the geneticpersonal development: a review of the relative age effect in component. Am J Hum Genet 1999; 65: 911-3sport. Dev Rev 2001; 21: 147-67 95. Slonimsky N. Perfect pitch. New York: Oxford University
72. Barnsley RH, Thompson AH. Gifted or learning disabled? The Press, 1998age of entering school may make the difference. Early Child- 96. Williams L, Gross JB. Heritability of motor skill. Acta Genethood Educ 1985; 18: 11-4 Med 1980; 29: 127-36
73. Hauck AL, Finch AJ. The effect of relative age on achievement 97. Fox PW, Hershberger SL, Bouchard TJ, et al. Genetic andin middle school. Psychol Schools 1993; 30: 74-9 environmental contributions to the acquisition of a motor skill.
74. Barnsley RH, Thompson AH. Birthdate and success in minor Nature 1996; 384: 356-8hockey: the key to the NHL. Can J Behav Sci 1988; 20: 167-76 98. Missitzi J, Geladas N, Klissouras V. Heritability in neuromuscu-
75. Thompson AH, Barnsley RH, Steblelsky G. ‘Born to play ball’: lar coordination: implications for motor control strategies.the relative age effect and major league baseball. Sociol Sport J Med Sci Sports Exerc 2004; 36: 233-401991; 8: 146-51 99. Joseph J. Separated twins and the genetics of personality differ-
76. Barnsley RH, Thompson AH, Legault P. Family planning: ences: a critique. Am J Psychol 2001; 114: 1-30football style. The relative age effect in football. Int Rev Sociol 100. Davids K, Glazier P, Araujo D, et al. Movement systems asSport 1992; 27: 77-87 dynamical systems: the role of functional variability and its
77. Carlson RC. The socialization of elite tennis players in Sweden: implications for sports medicine. Sports Med 2003; 33: 245-60an analysis of the players’ backgrounds and development. 101. Kevles DJ, Hood L. The code of codes: scientific and socialSociol Sport J 1988; 5: 241-56 issues in the human genome project. Boston (MA): Harvard
78. Curtis JE, Birch JS. Size of community of origin and recruit- University Press, 1992ment to professional and olympic hockey in North America. 102. Hopkins WG. Genes and training for athletic performance [on-Sociol Sport J 1987; 4: 229-44 line]. Available from URL: http://www.sportsci.org/jour/
79. Cote J, MacDonald D, Baker J, et al. When ‘where’ is more 0101/wghgene.htm [Accessed 2005 Sep 17]important than ‘when’: birthplace effects on the achievement 103. Rankinen T, Bray MS, Hagberg JM, et al. The human gene mapof sporting expertise. J Sports Sci 2006; 24: 1065-73 for performance and health-related fitness phenotypes: the
80. Elgar FJ, Arlett C, Groves R. Stress, coping, and behavioural 2005 update. Med Sci Sports Exerc 2005; 38: 1863-88problems among rural and urban adolescents. J Adolesc 2003; 104. Jones A, Montgomery HE, Woods DR. Human performance: a26: 574-85 role for the ACE genotype? Exerc Sport Sci Rev 2002; 30:
81. Kytta M. Affordances of children’s environments in the context 184-90of cities, small towns, suburbs, and rural villages in Finland 105. Kostek MC, Delmonico MJ, Reichel JB, et al. Muscle strengthand Belarus. J Environ Psychol 2002; 22: 109-23 response to strength training is influenced by insulin-like
82. Plomin R, DeFries JC, McClearn GE, et al. Behavioural genet- growth factor 1 genotype in older adults. J Appl Physiol 2005;ics. 4th ed. New York: Freeman, 2001 98: 2147-54
83. Bouchard Jr TJ, Lykken DT, McGue M, et al. Sources of human 106. Niemi AK, Majaama K. Mitochondrial DNA and ACTN3 ge-psychological differences: the Minnesota Study of Twins notypes in Finnish elite endurance and sprint athletes. Eur JReared Apart. Science 1990; 250: 223-8 Hum Genet 2005; 13: 965-9
84. Bouchard Jr TJ. IQ similarity in twins reared apart: findings and 107. Allen DL, Harrison BC, Sartorius C, et al. Mutation of the IIBresponse to critics. In Sternberg RJ, Grigorenko E, editors. myosin heavy chain gene results in muscle fibre loss andIntelligence, heredity, and environment. Cambridge (MA): compensatory hypertrophy. Am J Physiol Cell Physiol 2001;Cambridge University Press, 1997: 126-60 280: C637-45
85. DiLalla DL, Carey G, Gottesman II, et al. Heritability of MMPI 108. Mitch WE, Price SR. Transcription factors and muscle cachexia:personality indicators of psychopathology in twins reared is there a therapeutic target? Lancet 2001; 357: 734-5apart. J Abnormal Psychol 1996; 105: 491-9 109. Ivey FM, Roth SM, Ferrell RE, et al. Effects of age, gender, and
86. Tellegen A, Lykken DT, Bouchard Jr TJ, et al. Personality mysostatin genotype on the hypertrophic response to heavysimilarities in twins reared apart and together. J Personality resistance strength training. J Gerontol A Biol Sci Med SciSocial Psychol 1988; 54: 1031-9 2000; 55: 641-8
87. Keller LM, Bouchard Jr TJ, Arvey RD, et al. Work values: 110. Hortobagyi T, Dempsey L, Fraser D, et al. Changes in musclegenetic and environmental influences. J Appl Psychol 1992; strength, muscle fibre size and myofibrillar gene expression77: 79-88 after immobilization and retraining in humans. J Physiol 2000;
524: 293-30488. Arvey RD, Bouchard Jr TJ, Segal NL, et al. Job satisfaction:environmental and genetic components. J Appl Psychol 1989; 111. Calvo M, Rodas G, Vallejo M, et al. Heritability of explosive74: 187-92 power and anaerobic capacity in humans. Eur J Appl Physiol
2002; 86: 218-2589. Johnson W, Bouchard Jr TJ, Segal NL, et al. The Stroop Color-Word Test: genetic and environmental influences: reading, 112. Forbes GB, Sauer EP, Wetkamp LR. Lean body mass in twins.mental ability, and personality correlates. J Educ Psychol Metabolism 1995; 44: 1442-62003; 95: 58-65 113. Thibault M, Simoneau J, Cote C, et al. Inheritance of human
90. Plomin R, Colledge E. Genetics and psychology: beyond herita- muscle enzyme adaptation to isokinetic strength training. Humbility. Eur Psychologist 2001; 6: 229-40 Heredity 1986; 36: 341-7
114. Thomis MA, Bennen GP, Maes HH, et al. Strength training: 133. Johnston TD, Edwards L. Genes, interactions and the develop-importance of genetic factors. Med Sci Sports Exerc 1988; 30: ment of behaviour. Psychol Rev 2002; 109: 26-34724-31
134. Kauffman S. Understanding genetic regulatory networks. Int J115. Chagnon YC, Rice T, Perusse L, et al. Genomic scan for genes Astrobiol 2003; 2: 131-9
affecting body composition before and after training in Cauca-135. Kaerns M, Elston TC, Blake W, et al. Stochasticity in genesians from HERITAGE. J Appl Physiol 2001; 90: 1777-87
expression: from theories to phenotypes. Nature Rev Genet116. Naya FJ, Mercer B, Shelton J, et al. Stimulation of slow skeletal 2005; 6: 451-64
muscle fibre gene expression in calcineurin in vivo. J Biol136. Gardner T, Collins JJ. Neutralizing noise in gene networks.Chem 2005; 275: 4545-8
Nature 2000; 405: 520-1117. Swoap SJ, Bridge Hunter R, Stevenson EJ, et al. The137. Edelman GM, Gally JA. Degeneracy and complexity in biologi-calcineurin-NFAT pathway and muscle fibre-type gene ex-
pression. Am J Physiol Cell Physiol 2000; 279: C915-24 cal systems. Proc Natl Acad Sci 2001; 98: 13763-8
138. Tononi G, Sporns O, Edelman GM. Measures of degeneracy118. McAinch AJ, Lee JS, Bruce CR, et al. Dietary regulation of fatoxidative gene expression in different skeletal muscle fiber and redundancy in biological networks. Proc Natl Acad Scitypes. Obesity Res 2003; 11: 1471-9 1999; 96: 3257-62
119. Rubio JC, Martin MA, Rabadan M, et al. Frequency of C34T 139. Yates FE. Self-organizing systems. In: Boyd C, Noble D, edi-mutation of the AMPD1 gene in world-class endurance ath- tors. Logic of life. Oxford: Oxford University Press, 1993:letes: does this mutation impair performance? J Appl Physiol
189-2182005; 98: 2108-12140. van Geert P. Dynamic systems of development: change between
120. Cam FS, Colakoglu M, Sekuri S, et al. Association between thecomplexity and chaos. New York: Harvester Wheatsheaf,ACE I/D gene polymorphism and physical performance in a1994homogenous non-elite cohort. Can J Appl Physiol 2005; 30:
74-86 141. Oyama S. The ontogeny of information: developmental systemsand evolution. 2nd ed. Durham (NC): Duke University Press,121. Scott RA, Moran C, Wilson RH, et al. No association between
angiotensin converting enzyme (ACE) gene variation and en- 2000durance athlete status in Kenyans. Comp Biochem Physiol
142. Dodge KA. The nature-nurture debate and public policy. Mer-2005; 141 Pt A: 169-75rill-Palmer Q 2004; 50: 418-27
122. Silva GJ, Moreira ED, Pereira AC, et al. ACE gene dosage143. Barsh GS, Farooqi IS, O’Rahilly S. Genetics of body-weightmodulates pressure-induced cardiac hypertrophy in mice and
regulation. Nature 2000; 404: 644-51men. Physiol Genomics 2006; 27: 237-44144. Le Gaillard J-F, Clobert J, Ferriere R. Physical performance and123. Sayed-Tabatabuei FA, Oostra BA, Isaacs A, et al. ACE
Darwinian fitness in lizards. Nature 2004; 432: 502-5polymorphisms. Circ Res 2006; 98: 1123-33145. Gollub J, Solomon T. Chaos theory. In: Ranson KA, editor.124. Feitosa MF, Gaskill SE, Rice T, et al. Major gene effects on
Academic American encyclopedia. Danbury (CT): Grolier In-exercise ventilatory threshold: the HERITAGE Family Study.J Appl Physiol 2002; 93: 1000-6 corporated, 1996
146. Beunen G, Thomis M. Genetic determinants of sports participa-125. Williams AG, Day SH, Folland JP, et al. Circulating angiotensinconverting enzyme activity is correlated with muscle strength. tion and daily physical activity. Int J Obes Relat Metab DisordMed Sci Sports Exerc 2005; 37: 944-8 1999; 23: Suppl. 3: S55-63
126. Sonna LA, Sharp MA, Knapik JJ, et al. Angiotensin-converting 147. Carpenter S. Psychology is bound to become more Darwinian,enzyme genotype and physical performance during US Army says eminent primatologist [letter]. Monitor Psychol 2001; 32:basic training. J Appl Physiol 2001; 91: 1355-63
4127. Pescatello LS, Kostek MA, Gordish-Dessman H, et al. ACE ID 148. Rosengren KS, Savelsbergh G, Van der Kamp J. Development
genotype and the muscle strength and size response to unilater-and learning: a TASC-based perspective of the acquisition ofal resistance training. Med Sci Sports Exerc 2006; 38: 1074-81perceptual-motor behaviours. Infant Behav Dev 2003; 26:
128. Montgomery HE, Marshall R, Hemingway H, et al. Human gene 473-94for physical performance. Nature 1998; 393: 221-2
149. Wahlsten D. Insensitivity of the analysis of variance to heredity-129. Gayagay G, Yu B, Hambly B, et al. Elite endurance athletes and environment interactions. Behav Brain Sci 1990; 13: 109-62
the ACE I allele: the role of genes in athletic performance.150. Cronbach LJ. Emerging views on methodology. In: Wachs TD,Hum Genet 1998; 103: 48-50
Plomin R, editors. Conceptualization and measurement of130. Myerson S, Hemingway H, Budget R, et al. Human angiotensin
131. Bouchard C, Daw EW, Rice T, et al. Familial resemblance forVO2max in the sendentary state: the HERITAGE Family Study.
Correspondence: Prof. Keith Davids, School of HumanMed Sci Sports Exerc 1998; 30: 252-8Movement Studies, Queensland University of Technology,132. Baker J, Horton S. East African running dominance revisited: aVictoria Park Road, Kelvin Grove, QLD 4059, Australia.role for stereotype threat? Br J Sports Med 2003; 37: 553-5
