Article
Reference
Expertise and generalization: lessons from action video games
BAVELIER, Daphné, BEDIOU, Benoît, GREEN, C. Shawn
Abstract
Training is typically characterized by a trade-off between developing efficient performance, or expertise, and maintaining the ability to generalize one's knowledge beyond the trained domain. Here we ask whether it may be possible to train individuals to enhance their generalization abilities despite this natural trade-off. We first review the proposal that enhanced attentional control and cognitive flexibility may be potential mechanisms that will produce broad generalization. We then consider the case of action video game play which has been associated with enhancements in both attentional control and cognitive flexibility as well as generalization beyond the trained intervention.
BAVELIER, Daphné, BEDIOU, Benoît, GREEN, C. Shawn. Expertise and generalization:
lessons from action video games. Current Opinion in Behavioral Sciences , 2018, vol. 20, p.
169-173
DOI : 10.1016/j.cobeha.2018.01.012
Available at:
http://archive-ouverte.unige.ch/unige:103295
Disclaimer: layout of this document may differ from the published version.
Expertise and generalization: lessons from action video games
Daphne´ Bavelier
1, Benoit Bediou
1and C Shawn Green
2Trainingistypicallycharacterizedbyatrade-offbetween developingefficientperformance,orexpertise,andmaintaining theabilitytogeneralizeone’sknowledgebeyondthetrained domain.Hereweaskwhetheritmaybepossibletotrain individualstoenhancetheirgeneralizationabilitiesdespitethis naturaltrade-off.Wefirstreviewtheproposalthatenhanced attentionalcontrolandcognitiveflexibilitymaybepotential mechanismsthatwillproducebroadgeneralization.Wethen considerthecaseofactionvideogameplaywhichhasbeen associatedwithenhancementsinbothattentionalcontroland cognitiveflexibilityaswellasgeneralizationbeyondthetrained intervention.
Addresses
1Faculte´ dePsychologieetSciencesdeL’Education(FPSE),Universite´
deGene`ve,CampusBiotech,baˆt.H8-2,ChemindesMines,9, 1202Geneva,Switzerland
2DepartmentofPsychology,UniversityofWisconsin-Madison,Madison, WI53706,USA
Correspondingauthor:Bavelier,Daphne´ ([email protected])
CurrentOpinioninBehavioralSciences2018,20:169–173 ThisreviewcomesfromathemedissueonHabitsandskills EditedbyBarbaraKnowltonandJo¨rnDiedrichsen
https://doi.org/10.1016/j.cobeha.2018.01.012 2352-1546/ã2018ElsevierLtd.Allrightsreserved.
Expertise: a double-edged sword
Expertise,orreachinganoutstandinglevelofskillinone domain, is typically the result of thousands of hours of practice in the given domain. And while the path to expertise is commonly multi-faceted, it isnearly always the case thatsome degreeof automatizationof function canbefoundattheheartofexpertlevelperformance.Such automatization allows what were initially complex and cognitively demanding sequences of actions or thoughts to beexecuted automatically withminimal effort and/or cognitiveload.Thisnotonlyservestoproduceactionsor thoughtsthatarefastandaccurate,butcritically,releases limitedcognitiveresourcesforalternativetasks.Forexam- ple, a novice who is first learning to drive a manual transmissioncarmayneedtoexpendconsiderablecogni- tive effort determining what pedal to press, which
directionto movethe shiftknob,amongothers.Thisin turnleavesnoresourcesfreeforconversingwiththefront seatpassengerorfollowingamap.Conversely,inanexpert driver,basicdrivingfunctionshavebeenlargelyautoma- tized and cognitive resources are thusavailable to com- pletetheseadditionaltasks.
However, while there is clear value to automatizing function, ithas long beenaccepted thatautomatization alsocomesatacost[1,2].Inparticular,byrepeatingthe sametaskoverandover,expertsmaydevelopskillsthat aresotask-specificthattheylacktheflexibilitytoadaptto alterationsintheautomatizedtasksandsub-tasks[3–7].
For example, the performance of skilled typists plum- metswhencertainseeminglyminorchangesaremadein thekeyboardcharacteristics[8].Together,theevidence stronglysupportstheideathatexpertiseisoftenaccom- panied bythedevelopment of skillsthat are incredibly specifictoonlythosepreciseactions/thoughtsassociated withskilledperformance.Whatislesscleariswhetherit ispossibletotrainindividualstobecomeexperts,notata specifictaskordomain,butratheratflexiblyadaptingand transferring their skills and knowledge as new circum- stancesarise?
Training that fosters generalization — lessons from action video game play
Overthepast15years,evidencehasaccumulatedshow- ingthatplayingoneparticularformofvideogame,known asactionvideogames(primarilywhatareknownasfirst- person or third-person shooter video games), leads to rather broad generalization,with performanceenhance- mentsnotedindomainsasvariedasvisualperception(e.
g., enhanced contrast sensitivity & better peripheral detection—[9–11]), top-down attention (e.g., better change detection, reductionin attentional capture [12–
15]),visuo-spatialcognition(e.g.,bettermentalrotation;
enhanced visualshort-term memory [16,17]) and finally multi-taskingandtaskswitching(e.g.,lesserswitchcost, greaterease atmulti-tasking[18–20]).Inarecentmeta- analysiscombiningtheresultsof73studiesand3773total participants,self-declaredactionvideogameplayerswere foundtooutperformnon-gamersbyabouthalfastandard deviationacrossallaspectsofcognitioncombined.Impor- tantly, the same trendwas seen whenconsidering only true experiments,wherein non-gamerparticipants were specifically trained on an action video game and any cognitive gains were contrasted against those seen in anactivecontrolgroupthatwastrainedonacommercial non-action video game. Indeed, asecond meta-analysis
on 21 intervention studies involving 609 participants showed that action game training produced a benefit on cognition of around 1/3 of a standard deviation (as compared to control-trained individuals) [21]. Such results indicate that action video game play is not just correlatedwithenhancedcognitiveabilities,butdoesin fact,causeimprovementsin theseabilities.
Thebroadgeneralizationseentoresultfromactionvideo gametrainingisconsistentwiththe factthatsuchgames naturallymeshcomplexity,novelty,andvariability,ensur- ingthegameplaycanneverbefullyautomatized[22,23].
Actiongamerscertainlydevelopsomegame-specificbeha- viorsakintoexpertiseafterhoursspentonaparticulartitle, aswhengamersanecdotallyreportthattheycannotplaya game where the y-axis mapping is different from their practiced routine (i.e., whether pushing the mouse up makestheplayerlookuporlookdown).Yet,theincredible diversity of situationsencountered across various action videogametitlesensuresconstantengagementoftwokey cognitiveprocesses:attentionalcontrolandcognitiveflex- ibility.Forinstance,becauseenemiescanappearatvirtu- allyanylocation,atanytime,inanynumber/combinations, itisnotpossibletolearnanautomaticsequenceofactions that will produce game success. Instead, action video games continuously challenge attention allocation and the flexible evaluation of goals and sub-goals [24,25].
Akeypredictionofthisworkisthatplayingsuchgames resultsinenhancedattentionalcontrolandcognitiveflex- ibilitywhichinturnfostersgeneralization.
Attentionalcontrolandcognitiveflexibilityaretwocen- tral and complementary executive functions [26,27], which,aspredicted,areenhancedafteractionvideogame play. Manybehavioral studiesnow document enhance- mentsassociatedwithactionvideogameplayonarange oftaskstappingattentionalcontrol—fromanenhanced abilityto redirect eye-gaze andattention when initially wronglyallocated[15,28],to superior visualsearchper- formance [14,29], to better distractor suppression [30].
Similarly, although less well established, action video gameplayhasalsobeen positivelyassociatedwithtasks requiringcognitiveflexibilitysuchasmulti-tasking,task switching,or formsof workingmemory [16,20].Aneffi- cientdiagnostictasktoassessthetwokeycomponentsof attentionalcontrolandcognitiveflexibilityappearstobe theMultipleObjectTracking task, whereinindividuals must trackmovingobjects fromamongstadisplay con- taining many visually identical distractor objects. This task requires both attentional tracking over time and flexibleworkingmemoryindexingandupdating.Accord- ingly, thistaskhasrecentlybeenshown to loadontwo orthogonalfactors: onerelatedto ageneralizedcapacity forefficient perceptionandawareness—akeyfunction of attentional control—the other related to cognitive flexibility—orthecapacitytoholdandflexiblymanipu- lateinformationin workingmemory[31].
A prediction: an inverted U-shape curve for generalization as training proceeds
Enhanced attentional control and cognitive flexibility havebeenproposedtoprovideamechanismforgenerali- zation, while expertise is achievedat least partially via automatization,aprocessthatinherentlyentailsreleasing demandsonattentionalcontrolandcognitiveflexibility.
As expertise sets in, these two functions become less challenged and their enhancement is expected to fade away. Indeed, much as physical fitness decays in the absenceof continued physicaldemand, sodo enhance- mentsincognitive abilitiesintheabsenceofcontinued cognitive demand [32,33]. Generalization is therefore expectedtodecreaseaslearningprogressestowardexper- tise,apredictionin linewith thehighlyspecific skilled performancenoted inexperts.
Generalization is also expected to be rather limited during the earlieststages of training. This earlyperiod is characterized by a quickly saturating learning phase that mostly corresponds to the learners’ mastering the basicrulesorstrategiesrequiredbythenewtask.Learn- ingthesebasicrulesiscertainlydemanding,tappinglong- termmemorysystemsinparticular.Yet,untiltheserules are somewhat consolidated, it will be unclear to the learner to what, or how to direct their processing resources, thus minimizing the overall load on both attentional control and cognitive flexibility. To clarify thispoint,consideranathleteplayinganewsport.Until thefundamentalrules, goals, andstrategiesof thesport areunderstood,thephysicalchallengewillnotbemaxi- mal.Thesameideaappliesherewithrespecttocognitive challenges.
The phase of learning predicted to induce thegreatest generalization is therefore after this earliest phase, but before expertise sets in. It corresponds to a phase of learning which is rather slow and often associated with theviewthat‘practicemakesperfect.’Earlyintheslow learningphase,thetaskissufficientlywellunderstoodto result in load being placed on attentional control and cognitive flexibility—load which will, of course, be slowlyreleased asthe learning movestowardexpertise.
Thus, generalization as a function of training time is expectedtobeanasymmetricU-shapedcurve(Figure1).
Although this remains a prediction, it is instructive to consider such a view in the context of the impact of playingthevideogameTetrisonmentalrotation.Given theaboveframework,wewouldexpectnaı¨veparticipants whoareaskedtoplayanintermediate amountofTetris (e.g.,10–30hours),toshowsomedegreeoftransferfrom theirTetristrainingtonewmentalrotationtasks.Sucha finding has indeed been documented in the existing literature[34,35].Atthesametime,wewouldalsoexpect that expert Tetris players—i.e., individuals who play competitivelyandwhohavehundreds,ifnotthousands,
170 Habitsandskills
of hours of Tetris experience—should notshow stun- ning enhancements in mental rotation ability. Instead, they would be expected to mainly excel only at the rotation of Tetris-like shapes, as would be facilitated by automatizing the action sequences that linkeach of thevariousboardconfigurationstothe7possible Tetris shapes.Althoughsuchautomatizationwouldreleasecog- nitive load and greatly facilitate performance, it would obviouslybe oflittle valuefor any mental rotationtask that does notemploy Tetris shapes as documented by previouswork[36,37].Althoughtherehasnotyetbeena systematicinvestigationofmentalrotationgeneralization as afunctionofthenumberofhoursofTetris play,the availabledataisinlinewiththisviewofthetimecourseof generalization. Moretitratedstudiesthough wouldpro- vide avaluabletestofthetrade-offbetweengeneraliza- tionand automatizationastrainingproceeds.
We note that a major challenge in understanding the trade-offbetweengeneralizationandexpertiseisthatthe timecourseoflearning,andthusthetimeneededtoreach expertise, varies widely across domains. Expertise in playingsimple brain gamesdevelops quitefast; accord- ingly learning to play simple brain games has been associated with limitedgeneralization [38].By contrast, expertise when learning to playmusical instrumentsor other long-termactivitiessuchas chessdevelops slowly
andaccordingly thelearningof thesecomplexactivities hasbeenassociatedwithgreatergeneralization[39,40].
Inline withthekeyprediction fromtheproposedview training paradigms that are likely to produce the most generalization are the ones that keep a high load on attentional controlandcognitiveflexibility.
The mechanisms by which attentional control and cognitive flexibility may favor
generalization
To understand how attentional control and cognitive flexibilitymayfostergeneralization,ithelpstodifferen- tiatebetweentwodifferentwaysthatgeneralizationcan be assessed. The first, and most common, approach to assessinggeneralizationinvolvestraininganindividualon agiventask,andthenexaminingtheextenttowhichthe trainingproducesimmediatebenefitsonnew,untrained tasks. Thisapproach goesbackto Thorndike’s‘common elements’hypothesis,where theprediction isthatimme- diatelybetterperformanceonthegeneralizationtaskwill only be observed to the extent that the training and generalization tasks share key processing components [41–43].Aless commonlyemployedapproach to exam- ining generalization involves examining the extent to which trainingfacilitatesthelearning ofnew tasks (i.e., bycontrasttoonlyexamininginitialperformanceonthe new tasks—[24]).Thisapproach recognizesthatit is possibleforprevioustrainingtofacilitatetheacquisition of thenewtask(whether ornotimmediate benefitsare also observed on the new task). This latter form of generalizationhasbeenreferredtoas‘learningtolearn’.
Of course, these two forms of generalization are not mutually exclusiveandcouldco-occur.
Critically,enhancedattentionalcontrol/cognitiveflexibil- ity offersanatural mechanism for the learningto learn form of generalization. Enhanced attentional control allows for better extraction of task-relevant information once theindividualunderstandsthebasicrules/goalsof thetask.Thisinturnnotonlyallowsformoreinformed decisionstobemadeoneachtrialofthetask,itwillalso facilitate learning asmore information aboutthetaskis extractedoneachtrial[44,45].Andbecausetheratioof signal-to-noiseaffectsperformanceinmanydistincttasks, thiswouldservetoaccountfortheratherbroadgenerali- zationproducedbytrainingknowntoenhanceattentional control and cognitive flexibility—such as action video game play [46].Greater cognitive flexibility also allows one to more gracefully adapt to new tasks, to update memory information and to re-evaluate goals and sub- goalsas theychange,allof whichare similarlylikelyto facilitate learning or asymptotic performance in many tasks anddomains.
Enhanced attentional control and cognitive flexibility therefore offers a complementary mechanism to the prevalentviewofgeneralization,whichfocusesprimarily
Figure1
early practice effortful
late expertise automatized
Time
Amount of generalization
Expert Novice
Current Opinion in Behavioral Sciences
Proposedconceptualizationoftherelationshipbetweenlearning phasesandamountofgeneralization.Intheveryearliestphaseof training,theexpectedamountofgeneralizationislowforthesimple reasonthatnotmuchhasyetbeenlearnedthatcouldgeneralize.As participantsmovethroughtheearly-to-intermediatephaseoftraining, thetaskisexpectedtobedemandingintermsofprocessing resourcesresultinginenhancementsinattentionalcontroland cognitiveflexibility,andasaresult,greatergeneralizationisexpected.
Finally,atsomepoint,taskfunctionsbegintobeautomatized.
Althoughthetaskitselfisperformedwithincreasingefficiency,the learningthatsubservessuchchangesinperformanceistightlytiedto thespecificofthetaskreleasingthepressureonattentionalcontrol andcognitiveflexibilityandthuslesserdegreesofgeneralizationare expected.
on the extent to which the training and generalization tasks share processing components. Importantly, while theformermechanism(learningtolearn)allowsforforms of relatively ‘far’ generalization, the latter exclusively predicts ‘near’ generalization (as, by definition, only
‘near’taskswillsharethemostprocessingcomponents).
Notethatafundamentalissueremainsin characterizing generalizationasfarversusnearinthatwecurrentlylacka theoreticalframeworktosystematicallyidentifytherele- vantlevelofoverlapbetweenanytwotasksforpredicting generalization [47,48]. Clearly, this is a computational challengethatawaitstobeaddressedinfutureresearch.
Conclusion
Wehave consideredhow the delicatebalance between thedemandsoflearningandthedevelopmentofautom- atization as expertise sets in may affect generalization.
Wepropose thatgeneralization will be broadest during early-to-intermediatephasesoflearning,wherethebasic taskrulesandstructuresareknown,buthighdemandson attentionalcontroland cognitiveflexibilityremain.The case of action video games is interesting because it representsacaseofpersistentextremeloadonattentional control and cognitive flexibility, forcing individuals to make decisions in an ever-changing environment at a pace that is difficult to match in other activities. We propose here that the incredible diversity of situations encountered across various video game titles ensures constantengagementofattentionalcontrolandcognitive flexibility,fosteringinturngreateradaptabilityandgen- eralizationwhenfacingnewtasks ordomains.
Conflict of interest statement
BavelierisonthescientificadvisoryboardofAkiliInter- active,Boston
Acknowledgements
ThisworkwasfundedbygrantsfromtheSwissNationalScience FoundationtoDB(SNF100014_159506)andfromtheOfficeofNaval ResearchtoCSG(GrantN00014-17-1-2049).
References and recommended reading
Papersofparticularinterest,publishedwithintheperiodofreview, havebeenhighlightedas:
ofspecialinterest ofoutstandinginterest
1. ChiMTH:Twoapproachestothestudyofexperts’
characteristics.In TheCambridgeHandbookofExpertiseand ExpertPerformance.EditedbyEricssonNCKA,FeltovichPJ, HoffmanRR.CambridgeUniversityPress;2006:21-30.
2. SternbergRJ:Costsofexpertise.In TheRoadtoExcellence.
EditedbyEricssonKA. LawrenceErlbaumAssociates;1996:347- 354.
3. FulvioJM,GreenCS,SchraterPR:Task-specificresponse strategyselectiononthebasisofrecenttrainingexperience.
PLoSComputBiol2014,10:e1003425.
4. LoganGD:Towardaninstancetheoryofautomatization.
PsycholRev1988,95:492-527.
5. SchneiderW,FiskAD:Attentiontheoryandmechanismsfor skilledperformance.AdvanPsychol1983,12:119-143.
6. SchneiderW,ShiffrinRM:Controlledandautomatichuman information-processing.1.Detection,search,andattention.
PsycholRev1977,84:1-66.
7. ShiffrinRM,SchneiderW:Controlledandautomatichuman information-processing.2.Perceptuallearning,automatic attending,andageneraltheory.PsycholRev1977,84:127-190.
8. CrumpMJ,LoganGD:Warning:thiskeyboardwill
deconstruct—theroleofthekeyboardinskilledtypewriting.
PsychonBullRev2010,17:394-399.
9. BuckleyDetal.:Actionvideogameplayersanddeafobservers havelargerGoldmannvisualfields.VisionRes2010,50:548- 556.
10. LiRJetal.:Enhancingthecontrastsensitivityfunctionthrough actionvideogametraining.NatNeurosci2009,12:549-551.
11. AchtmanRL,GreenCS,BavelierD:Videogamesasatoolto trainvisualskills.RestorativeNeurolNeurosci2008,26:435-446.
12. ClarkK,FleckMS,MitroffSR:Enhancedchangedetection performancerevealsimprovedstrategyuseinavidaction videogameplayers.ActaPsychol(Amst)2011,136:67-72.
13. GreenCS,BavelierD:Learning,attentionalcontrol,andaction videogames.CurrBiol2012,22:R197-R206.
14. WestGLetal.:Visuospatialexperiencemodulatesattentional capture:evidencefromactionvideogameplayers.JVision 2008,8.
15. ChisholmJD,KingstoneA:Improvedtop-downcontrolreduces oculomotorcapture:thecaseofactionvideogameplayers.
AttenPerceptPsychophys2012,74:257-262.
16. BlackerKJ,CurbyKM:Enhancedvisualshort-termmemoryin actionvideogameplayers.AttenPerceptPsychophys2013, 75:1128-1136.
17. SpenceI,FengJ:Videogamesandspatialcognition.RevGen Psychol2010,14:92-104.
Areviewofhowactionvideogameplayaffectsperformanceinabroad arrayoftasksthatarekeyforspatialcognition.Theoreticalandpractical implicationsofthesefindingsforlearningandeducationarediscussed.
18. CainMS,LandauAN,ShimamuraAP:Actionvideogame experiencereducesthecostofswitchingtasks.Attention PerceptionPsychophys2012,74:641-647.
19. ColzatoLSetal.:DOOM’dtoswitch:superiorcognitive flexibilityinplayersoffirstpersonshootergames.Front Psychol2010,1:8.
20. StrobachT,FrenschPA,SchubertT:Videogamepractice optimizesexecutivecontrolskillsindual-taskandtask switchingsituations.ActaPsychol(Amst)2012,140:13-24.
21. BediouB,AdamsDM,MayerRE,TiptonE,GreenCS,BavelierD:
Meta-analysisofactionvideogameimpactonperceptual, attentional,andcognitiveskills.PsychologicalBulletin2018, 144:77-110.
Ameta-analysistakingstockofmorethan15yearsofstudiesonthe impact ofactionvideo game playoncognition. Cross-sectionaland interventionstudiesareconsideredseparately,withtherolesofseveral keymoderators,suchascognitivedomainsorage,beingexamined.
22. MoreauD,ConwayAR:Thecaseforanecologicalapproachto cognitivetraining.TrendsCognSci2014,18:334-336.
Thispaperattemptstoresolvethedebatearoundtheeffectsofworking memorytrainingonintelligence,byhighlightingnotonlydifferencesin methodsbut alsointheecologicalvalidityoftraining regimens.It is arguedthatnovelty,complexity,anddiversityarekeyelementsoftraining regimensthatfostergeneralization.
23. SchmidtRA,BjorkRA:Newconceptualizationsofpractice— commonprinciplesin3paradigmssuggestnewconceptsfor training.PsycholSci1992,3:207-217.
24. BavelierDetal.:Brainplasticitythroughthelifespan:learning tolearnandactionvideogames.AnnuRevNeurosci2012, 35:391-416.
172 Habitsandskills
Areview ofhowcomplextrainingenvironmentssuchasactionvideo game play may fostergeneralization by producingenhancements in attentioanl controlandcognitive flexibilitywhichinturnincreasesthe abilitytolearn.
25. DaleG,GreenCS:Thechangingfaceofvideogamesandvideo gamers:futuredirectionsinthescientificstudyofvideogame playandcognitiveperformance.JCognEnhance2017,1:280- 294.
26. DiamondA:Executivefunctions.AnnuRevPsychol2013, 64:135-168.
27. MiyakeA,FriedmanNP:Thenatureandorganizationof individualdifferencesinexecutivefunctions:fourgeneral conclusions.CurrDirPsycholSci2012,21:8-14.
28. ChisholmJD,KingstoneA:Actionvideogamesandimproved attentionalcontrol:disentanglingselection-andresponse- basedprocesses.PsychonBullRev2015,22:1430-1436.
29. Hubert-WallanderBetal.:Changesinsearchratebutnotinthe dynamicsofexogenousattentioninactionvideogame players.AttenPerceptPsychophys2011,73:2399-2412.
30. MishraJetal.:Neuralbasisofsuperiorperformanceofaction videogameplayersinanattention-demandingtask.JNeurosci 2011,31:992-998.
31. EayrsJ,LavieN:Establishingindividualdifferencesin perceptualcapacity.JExpPsycholHumPerceptPerform,in press.
32. SabolMA,WisherRA:Retentionandreacquisitionofmilitary skills.MilitaryOperatRes2001,6:59-80.
33. HealyAF,EricssonKA,Bourne LEJr:OptimizingtheLong-Term RetentionofSkills:StructuralandAnalyticApproachestoSkill Maintenance.AnnualReport,1991–1992.C.U.A.BOULDER;1999.
34. OkagakiL,FrenschPA:Effectsofvideogameplayingon measuresofspatialperformance:gendereffectsinlate adolescence.JApplDevPsychol1994,15:33-58.
35. TerleckiMS,NewcombeNS,LittleM:Durableandgeneralized effectsofspatialexperienceonmentalrotation:gender differencesingrowthpatterns.ApplCognPsychol2008, 22:996-1013.
36. SimsVK,MayerRE:Domainspecificityofspatialexpertise:the caseofvideogameplayers.ApplCognPsychol2002,16:97-115.
37. LewandowskyS,LittleD,KalishML:Knowledgeandexpertise.
HandbookofAppliedCognition.2007:125-155.
38. OwenAMetal.:Puttingbraintrainingtothetest.Nature2010, 465:775-778.
39. SalaGetal.:Checkingthe“AcademicSelection”argument.
Chessplayersoutperformnon-chessplayersincognitive skillsrelatedtointelligence:ameta-analysis.Intelligence2017, 61:130-139.
40. SchellenbergEG,WeissMW:Musicandcognitiveabilities.In ThePsychologyofMusic.EditedbyD.D..Elsevier;2013:499-550.
Thischapterreviewsworkontheimpactofmusiconcognitiveabilities.It highlightsthefactthat,althoughmusictrainingtransferstocognition, studiesinexpertsshowthatexperiencedmusiciansonlyoutperformnon- musicians onmusical abilities(consistent witha U-shapedcurvefor generalization).
41. DeveauJetal.:Howtobuildbettermemorytraininggames.
FrontSystNeurosci2014,8:243.
42. SingleyMK,AndersonJR:TheTransferofCognitiveSkill.Cognitive ScienceSeries.HarvardUniversityPress.viii;1989:300.
43. ThorndikeEL,WoodworthRS:Theinfluenceofimprovementin onementalfunctionupontheefficiencyofotherfunctions.
PsycholRev1901,8:247-261.
44. GreenCS,PougetA,BavelierD:Improvedprobabilistic inferenceasagenerallearningmechanismwithactionvideo games.CurrBiol2010,20:1573-1579.
45. RoelfsemaPR,vanOoyenA,WatanabeT:Perceptuallearning rulesbasedonreinforcersandattention.TrendsCognSci2010, 14:64-71.
46. BejjankiVRetal.:Actionvideogameplayfacilitatesthe developmentofbetterperceptualtemplates.ProcNatlAcad SciUSA2014,111:16961-16966.
47. GershmanSJ,NivY:Learninglatentstructure:carvingnature atitsjoints.CurrOpinNeurobiol2010,20:251-256.
48. OsgoodCE:Thesimilarityparadoxinhumanlearning;a resolution.PsycholRev1949,56:132-143.
