chapter 4 connecting cognitive science ... - …ddl/articlescopy/fischerdaley_execprocesses... ·...

CHAPTER 4

Connecting Cognitive Scienceand Neuroscience to EducationPotentials and Pitfalls

in Inferring Executive Processes

KURT W. FISCHERSAMANTHA G. DALEY

Scientific understandingof mind and brain is advancingquickly andenergetically, and society’s need to improve the quality of educationmakesheadlinesevery day. Naturally, these two trendscreatea broadinterestin usingresearchaboutthe brain andmind to guide educationalpractice.Knowing how our minds/brainsfunction,how weusethe brainand body to processandstore new information, how our minds/brainschangeanddevelop,andhow damageto our brains contributesto dis-abilities andotherproblems—alltheseresearchefforts havegreatpoten-tial for moving forward the scienceand practiceof learning.

This move to makeeducationalpracticemorescientific hasproper-ties that are similar to the history of medical practice. Medicine oncerelied on the collectedwisdom of culture but had no systematicproce-dure for testing which medical practiceswere actually effective. In thelast 200 years, especiallysince the innovations of Louis Pasteur inFrance,medicinehasestablisheda powerful basein researchin the bio-logical sciences.Medical practiceis now guided by what the biologicalsciencesknow about the body, and the biological sciencesconduct

55

fischeku

Text Box

Fischer, K. W., & Daley, S. (2006). Connecting cognitive science and neuroscience to education: Potentials and pitfalls in inferring executive processes. In L. Meltzer (Ed.), Understanding executive function: Implications and opportunities for the classroom (pp. 55-72). New York: Guilford.

56 THEORETICAL AND CONCEPTUAL FRAMEWORKS

researchthat is informedby what is importantto medicine.This processis interactiveand reciprocal.Scientistsdo not dictatemedical practice,but theycontributeto investigatingandunderstandinginnovativetreat-mentsandtechniquesthat often derive from the clinical skills and expe-riences of practitioners.In parallelfashion,educationmustclarify andstrengthenits relationshipwith the researchdisciplinesthat study devel-opment,learning,andthe brain.This relationshipshouldbe a reciprocalone in which educationalpracticeand scientific researchinform andlearnfrom eachother, as medicineand biology act symbiotically.

Although this relationshipis still emerging, the growth in knowl-edge of development,learning, and the brain alreadyprovides poten-tially productiveconnectionsbetweeneducationalpracticeandscientificresearch.Oneparticularlypromisingarenais analysisof possiblegeneralabilities that are proposedas an important focus for educationalprac-tice, such as the teachingof executivefunction, the topic of this book.Based on both researchand practice—scientificknowledgeabout howchildren learn and developand the history of efforts in educationtoteach a broad, general ability—we will argue that there is no tightlyorganizedexecutivefunction butonly looselycoupled,diverseexecutiveskills. A closer relationshipbetweeneducationalpracticeand researchcan provide a more accurateand practically useful view of executivefunction andothercandidatesfor generalabilities, suchas metamemory,metacognition,andtheory of mind (Fischer& Immordino-Yang,2002;Fischer,Immordino-Yang,& Waber, 2006).

Commonly in human developmentand in learning in schools,researchersandteachersobservethe regularoccurrenceof similarbehav-iors andchangesthat suggesta unified entity, suchasexecutivefunction.~—.-~Careful researchand practical observation~~f~j~allyfind that thesebehaviorsare more diversethanunified. This widely occurring patternpervadesall aspectsof human learning and development, includingmotor functioning, cognitive development,and brain development,aswell asexecutivefunction.We arguethat, like theothercases,executivefunction hasimportantgeneralcharacteristicsthatmakeit seemto be aunified entity, but at the level of detail importantfor educationalprac-tice it is diverseand variable,notunified. The practical implications ofsuch an interpretationare many.

A RECIPROCAL PARTNERSHIP: DEVELOPMENT,NEUROSCIENCE, AND EDUCATION

Overthe last two centuries,the biological scienceshavecometo form anatural partner for the improvementof medical practice.What fieldsmight play a similar role in relationto education?Thecomplexnatureof

Cognitive Science and Neuroscience 57

educationalpracticemeansthat severaltypes of researchcan produceeducationally usableknowledge (Fischer & Katzir, in press). In thischapter,we focusparticularlyon knowledgeabouthumandevelopment,the learningprocess,and brain functioning.

The studyof how human beingsdevelopand learn falls under theumbrella of the disciplineof humandevelopmentand its sibling, devel-opmentalpsychology.This is the field of researchthat investigateshowlearningtakesplaceand how peoplechangeas they growfrom infancythrough adulthood. Its researchmethodshave traditionally involvedcontrolled experimentsin laboratories as well as studiesof naturallyoccurringchangesin behaviorwith ageand setting. Researchquestionshaveoften focusedon narrowinquiries relatedto (1) normativepatternsfor particularagesor social groupsand (2) species-generalhumanchar-acteristicsof thought, memory,attention,emotion, and learning. Thefield is now experiencingmajor efforts to move it toward a broaderframeworkthat examinesdevelopmentas a function of the many com-ponentsthat affect humanbehavior,including biology, context,culture,and individual variation. Indeed, the first volume of the influentialHandbookofChild Psychologyhighlights this importantshift in empha-sis (Damon& Lerner,2006).

Neuroscienceinvolvesthe studyof thebrain, especiallyits organiza-tion, functioning, andunderlyingphysiology,includingthe neurons,syn-apses,and neuralnetworksthat it comprises.Neuroscienceemphasizesbrain functioning but not necessarilyoutwardly noticeable behavior.Most questionsin neurosciencefocuson specific,experimentallytracta-ble hypothesesaboutthe brain’s responseto simplestimuli (Turk et al.,2002).Studyof the brainoften requirescombinationsof fields in permu-tationssuchas cognitiveneuroscience,behaviorgenetics,andbehavioralneurochemistry.Our focusin thischapteris on researchthat makescon-nectionsbetweenthe brain’s activity andpeople’sactionsandthoughts.

Educationis different from developmentandneuroscience,as it isnot only an academicareaof study, but also a practical field. We con-sidereducationbroadly to include traditional classroomlearning,adultlearning, and informal learningactivities.

Education, human development, and neurosciencehave yet toestablish a truly reciprocal partnershipdespite continually increasinginteractionstending in that direction.A handfulof casesshow the enor-mouspotentialof reciprocalinteractionsfor benefitingeducationalprac-tice. For example, researchin dyslexia has led to major advancesnotonly in understandingthe basesof specific readingdisabilities,but alsoin the designof interventionsto help studentswith dyslexialearnto readand write effectively. Maryanne Wolf and her colleagues (Wolf &Bowers, 1999;Wolf & Katzir-Cohen,2001; Wolf, Miller, & Donnelly,2000) havedevelopeda curriculum to support studentswith dyslexia


that integratesknowledge from neuroscience,development,cognitivescience,andeducationin innovative andmeaningfulways. David Roseand his colleaguesuse principles from thesedisciplines to inform thedevelopmentof softwareand othereducationaltools that supportread-ing, writing, andinstructionthat is flexible enoughfor a variety of learn-ers, following theprinciples of what they call “universal design” (Rose,Meyer, Strangman,& Rappolt, 2002;Rose, Chapter13, this volume).Theseare but two examplesthatshowwhat is possiblewhenexpertsci-entistsand educatorsfrom different disciplineswork togetherto studyand inform educationalactivities.

Such efforts move forwardthe reciprocalrelationshipsof educationwith cognitivedevelopmentalscienceandneuroscience,producingmajoradvancesand innovationsin educationalpractice. Yet caution remainsimperativein basingeducation-relateddecisionson basicresearch,espe-cially whenthereare one-sidedrelationshipsratherthanreciprocalpart-nerships.For example,conceptsabout executivefunction in cognitivesciencehaveled to educationalpracticesthat areoverly simple and donot engagethe variability that teachersencounterevery day with stu-dentsin their classrooms.

SIMILARITIES ACROSS DOMAINS: SIMILAR PATTERNSDO NOT SIGNIFY A UNITARY ABILITY

In a commontype of unwarrantedleap, researchersuncoversimilar pat-terns of behavioror brain functioning and assumethat the similaritiesreflect a singleunderlying processor structure.The discoveryof thesesimilarities, such as parallel patternsof learningor development,pro-vides the basis for many importantscientific discoveries,so researchersshouldseekthem,but the interpretationof theparallelsrequirescaution:Even whenthe similarities point to some commonprocessor function,theytypically do not imply a unified or singularcapacity,so implicationsfor educationalpracticearenot simple(Fischer& Bidell, 2006). Oneofthe most generalcharacteristicsof humanfunctioning (both body andbehavior) is that many componentsoperatemostly independentlywhileat the sametime havingsome important links andsimilarities.

In medicine,whena personexperiencesa suddenhigh fever, doctorsdare not assumethat this symptom indicates a singularcause.A fevercan come from bacterial infection, viral infection, overheatingof thebody, insufficient cooling of the body, malfunction of the immunesys-tem, and many other diversecauses.In every case,the body’s tempera-ture regulation systemis involved, but there is no single causeacrosscases.A veterinariancan build her practice on the understandingthatboth poodlesand schnauzersaredogs,but shedarenot assumethat the


various dogsare identical, or she will makecritical mistakes in treat-ment.We will now discussseveralexamplesof similarities in patternsofbiological andcognitivedevelopmentandthenhow to interpretthesimi-larities andwhat the insights from these examplesimply for analyzingexecutivefunction.

Motor System(s)The motor system—thefunctions of the body that allow and controlmovement—canin many ways be viewed as a single entity. Like thedigestivesystemor the respiratorysystem, studentstend to study themotorsystemas a unit. Understandingthe functioning of muscles,ten-dons,ligaments,andthe restwill providea solid foundationfor under-standinghow people can tap their toes,nod their heads,or throw abaseball.Given the common functions and the strong connectionsofsomecomponents,it makessenseto view the motor functionsas a sys-tem.

At the sametime, the sharedfunctionsof the motor systemdo notmakeit unitary or uniform, and assumingsucha unity leadsto criticalmisunderstanding.For example,an impairment or injury in one aspectof motoractivity often haslittle or no influenceon anotheraspect.Themotorsystemis highly differentiatedinto grossandfine motoractivities,voluntary and involuntary muscles,distinct organs(arms, legs, heart,stomach,motor cortex, cerebellum,etc.) and evenfurther specializedwithin eachof thesecategories.

Any basic anatomyandphysiologytextbook makesclear the com-plexity of the processesthat enablehumanmovement.In anyonepartofthe motor system,such as muscles,crucial distinctions must be made.Onetextbook highlights the needto considerbothdifferencesandsimi-larities amongtypesof muscles(Marieb,2006).Thereare threetypesofmuscletissue,which differ in cell structure,body location, andtype ofstimulation to causecontraction,but all typeshavethesamekinds of fil-amentsthat participatein contraction.In anotherpieceof the motorsys-tem,different joints are capableof different typesof rotation,eachwithdifferent implicationsfor injury and treatment(Mader, 2005). Regard-ing how componentswork together, thereare distinct categoriesthatspecify characteristicpatternsof coordination,such as voluntary andinvoluntary movementsand gross and fine motor skills.

Viewing the motorsystemasa unified entity, then,is useful for ana-lyzing how bodily movementhappens,but the systemis composedofmany different parts and processes.It cannot be treated as a unitarystructure.The parts function independentlyin mostways,althoughtheyare partly connectedand coordinated.The same is true of behavioralsystems.


Cognitive DevelopmentThe traditional view of developmentassumesthat componentsthatshow similargrowth functionsinvolve the sameunitary underlyingpro-cessor capacity—asinglestageof logic for Piaget’s (1983) theory or asinglebuffer of short-termmemoryfor classicalinformation-processingviews (Case,1974;Klahr & Wallace, 1976).This traditional view treatsdevelopmentas a ladderon which peoplemove upwardstep by steptosuccessivelyhighergeneralcognitivestagesin a linear fashion.An indi-vidual functionsat a single generalstageacrossdomainsand no longerusesearlierones,accordingto this view. The stageson this metaphoricalladder assumea common stateof developmentacrossall domainsoflearningand behavior,from the ability to solve arithmeticproblems tothe maturity to respondto social challenges.A S-year-oldwill show thesame(pre-operational)stageof cognitive developmentin arithmetic andsocial understanding,and an adult will show the same(formal opera-tional) stagein bothdomains.

Peopledo not showthis kind of consistency.Researchstronglydoc-umentsthat the laddermetaphoris wrongwhenappliedacrossdomains(Fischer & Bidell, 2006), and any experiencedteacheror observerofchildren knows that studentsshow different capacities in differentdomains.Only within a domaindo children develop along a relativelyunified, consistentpathway.

A dynamicview of developmentmoves from the traditional ladderview to a differentmetaphorthat includesbothconsistencyandvariabil-ity, consistentpathwayswithin a domainanddifferentpathwaysamongdifferent domains,Developmentproceedsalongthe strandsof a web asshownin Figure 4.1. Eachstrand in the web representsa different spe-cific domainof development.Dependingon the breadthof the contentonechooses,the strandsmight representbroaddomainssuchas motorskills, arithmetic knowledge,and literacy, or they might representspe-cific subdomainswithin a narrowerdomain such as simple arithmeticproblems,with additionon onestrand,subtractionon another,andmul-tiplication on a third. The strandsin the figure specify domainsin thedevelopmentof executiveprocesses.Developmentproceedsfrom thetopof the diagramto the bottom, but a personcandevelopalong differentstrandsat different paces.While the laddermetaphorand the theorybehindit emphasizethe normativecommonalitiesin developmentacrossdomains,the web andthe dynamicview capturevariations in develop-ment acrossdomains,as well as connectionsand separations(repre-sentedby intersectionsand branches).

Besidesthis variability acrossdomainsof functioning,the webmet-aphoralso allowsfor variability within domainsfor individual learners.A personworkingon a specifictask doesnot stayfixed at onepoint on a


Domains

Regulating Monitoring

learning comprehension

CCDCCD0-aBCD

Reflectivejudgment

FIGURE 4.1. A developmentalweb for domainsof executivefunction.

strand but varies his or her activity dependingon context and state(Fischer& Bidell, 2006;Fischer,Bullock, Rotenberg,& Raya, 1993).A1-year-oldlearnsto walk on a level carpetinside thehomebut is unableto makea fewstepsacrossthegrassin the backyard.Everyonehasexpe-riencedsituationssuchas being ablein practiceto rememberlines for aplay or shoota free throw andthenfailing with the samebehaviorin theperformancethat matters.Whatdoesit meanto “know” materialor tohave“mastered”a skill? On any particulartask,a personactsat a widevarietyof levels,rangingfrom the functional,or typical, level to the opti-mal level (whatcanbe donewith contextualsupport).The roleof asup-portive environmentin causingvariation alonga strandin the web hasbeenwidely documentedfor tasks as diverse as telling a story aboutsocial interactionsor predictingwhetherobjectswill sink or float.

The variability in behaviorcapturedby the web helpsilluminate theuniformity seenin somedevelopmentalchanges.Children demonstraterapid changesin performancein specific ageregionsfor optimal condi-tions in familiar domains-changesthat havesomeof the propertiesofstages.Such spurtshave beendocumentedin studiesof, for example,


reflectivejudgmentin adolescentsandadults (Kitchener,Lynch, Fischer,& Wood,1993) and useof pronounsin the earlyspeechof Dutch chil-dren(Ruhland& van Geert,1998),asshownin Figure4.2.Thesespurtsand other kinds of discontinuities tend to cluster at particular ageregions for optimal performance,such as approximately2 years forspurts in vocabulary,useof sentences,and pretendplay. In Figure 4.1,look carefullyat whenthe strandschangedirection,branch,or join, andyou will see that thesediscontinuitiescluster in specific regions.

Peopledevelopin spurtsunderoptimalconditions,buttypically notunderordinaryconditions,which lack contextualsupportand/or exten-sive familiarity andpractice(Fischeret al., 1993).Figure 4.3 illustratesatypical patternfor developmentof optimal andfunctional (ordinary) lev-els in a domain such as reflective judgmentor representationof socialinteractions.Skills developin spurtsfor optimal level (high support,topline) but more slowly and smoothly for functional level (low support,bottomline).The samepersonshowsbothoptimal andfunctionallevels,which comeandgo with variationsin contextualsupportand state. Inthis way, eachpersonacts at multiple levels from momentto moment,evenfor a singledomain(strandin theweb), moving up anddown with-in a rangeof skill levels as a function of supportand state.

When observersnote only the spurts and other discontinuities,which clusterat a specific ageregionin the developmentalweb, they seewhat appearsto be a singleability emergingat that age.Examinedmore

C00)0)CD0)a)a0)CD0C00.

0

0)nEDz

120

100

80

60

40

20

080

Age in weeks

100 120 140 160

FIGURE 4.2. Developmentalspurt in pronounuseby Tomas,a Dutchboy.


P

AbS

Ab2

a)Abl

Co

RpS

Rp2

Rpl

Age in years

FIGURE 43. Optimal and functional levels in cognitive development.Skillsdevelop along a common complexity scalemarked by a seriesof skill levels.With high support, they grow in spurtsfor optimal level. Withoutsupport,theygrowmore continuouslyfor functional level.

broadly (in the whole web), this patternbecomesone regularity withinthe broaderpictureof variability. Thereis clearly no single,unitary abil-ity emergingacrossall skills anddomainsat one age.Instead,a personbuilds skills along each strand,following its domain-specificdevelop-mentalprogression,andat certainpoints alongthe strand,spurtsahead.Thisspurt is a local processin the domain,nota shift in a single,unitarynew ability.

Developmentis a complexphenomenonthat encompassesboth (1)elementsof uniformity, such as regions of common change acrossstrandsor domains,and (2) elementsof individuality andvariation. Aswith motor functioning, assumingunity neglectsthe variation that ispresentin cognitive developmentand, thus, oversimplifies anddistortsthe natureof development,making it seemlike a ladder. An accurateview of developmentmustaccountfor bothuniformity andvariability.

The uniform aspectsof cognitive developmentprovide a valuableheuristicbasedon large-scalepatternsof developmentalchange.Indeed,

Abstractsystems

Abstractmappings\e’J~”

Singleabstractions

Representation~systems

— —S— —

\&~‘

3 5 7 g 11 13 15 17 ig 21


they have led to the identification of a general developmentalscaleunderlying both cognitive developmentand learning (Dawson-Tunik,Commons,Wilson, & Fischer,2005;Fischer & Bidell, 2006; Fischer&Immordino-Yang,2002).At the sametime, focuson only the uniformityleads to distortions, especiallyin education.The unitary view producesan emphasison norms and a neglectof the variationthat is inevitablypresentin educationalsettings.Children within a classroomwill not allreachthe samereadinglevel at the sametime, despiteladder-likeviewsof readingthat mark a text as at one specific gradelevel. This perspec-tive makesindividual variationappearabnormalandproblematicratherthana phenomenonto be exploredandexplained.If a studentis abletocompletea pageof algebraproblemsone day but seemsto haveforgot-teneverythingthe nextday, the behaviorappearsabnormaland inexpli-cable(althoughsensitiveteachersknow to expectsuch variation). Withthe dynamic view of development,suchvariation is understandableandpotentially predictablefrom context and emotional state (e.g., Did heskipbreakfastsohecannotconcentratetoday?Is a testnextperiod caus-ing anxiety?Did he havea supportivealgebralessonright beforehe didthe problemsyesterday?).

A dynamicview of developmentrecognizesthe similarities in devel-opment acrossdifferent domainsand simultaneouslyinterpretsthem intermsof the patternsof variation. This dynamicview is moreuseful andaccuratethana traditional view that assumesa unitary processbecauseit dealsdirectly with the complexities of human learning and action.Educatorsanddevelopmentalscientistsworking togethercan (1) illumi-nate the understandingof developmentby connectingit to variations instudents’behaviorsin schoolsandfamiliesand (2) simultaneouslycreateresearchthat feeds back to practitionersto help them usecognitive anddevelopmentalanalysisto facilitate learningand teachingin schools.

Brain DevelopmentThe scienceof brain developmentis muchless maturethan thatof cogni-tive development.Yet earlyevidencesuggeststhat the model for cogni-tive developmentappliesstraightforwardlyto importantaspectsof braindevelopmentas well. Brain growth andcognitive growth seemto showthe samekind of web patternand the sametype of recurringgrowthcycle,with multiple developingstrandsandspurtsand otherdiscontinu-ides in growthalong eachstrand.For instance,the partof the prefrontalcortexthat supportsworkingmemory(holding information on-line for atime) developsseparatelyfrom the part of the occipital cortexthat ana-lyzesvisual information,althoughbothdevelopwith similar discontinu-ities (Fischer & Rose, 1996). (Scholarsfrequently nominate the pre-frontal cortex as the key brain region for executivefunction.)


The strongestempirical evidenceof these brain growth patternscomesfrom researchon the developmentof electricalactivity in the cor-tex, measuredthroughthe electroencephalogram(EEG).The moststud-ied propertyof the EEG is its energy (called “power”), which developsthrough fits and startsat specific agesthat correspondto the agesofemergenceof optimal levels in cognitivecapacity from infancy throughearly adulthood(Fischer& Bidell, 2006; Somsen,van ‘t Klooster, vander Molen, van Leeuwen,& Licht, 1997; Thatcher, 1994). Figure 4.4showsthe resultsof onenormativestudy (Matousek& Petersen,1973)for the relativeenergy in the alphabandof EEG in the backof thecor-tex, with spurtsand plateausclearly evidentat approximately4, 8, 12,IS, and 20 yearsof age,apparentlymarking the cognitive levels that aremost relevant for the school years. Note the similarity to the growthcurve for optimal level in Figure 4.3.

The similarity of growth curvesfor EEG energyand cognitiveper-formance suggest a connectionbetween developmentof brain andbehavior, but few studieshave looked at brain and behavior concur-rently to test the correspondencedirectly. For the current argument,assumethat the correspondenceis real—thatgrowth spurtsin the EEG

80 - Abstractsystems

70 -

60 -

~40

~ 30

20

10

010

Age in years

FIGURE 4.4. Developmentof relativepowerin alphaEEG in occipito-parietalarea in Swedishchildrenand adolescents.The energy (relative power) in theEEGgrowsin spurtsthatparallelthepatternfor optimal level in Figure4.3. Thisgraph showsrelative powerin the occipital-parietalareaof the cortexfor thealphabandof the EEG in Swedishsubjects(Matousek& Petersen,1973).

AbstractmappingsSingle

abstractions

Representationalsystems

Representational- mappings

I I I I I I I I I I

5I I I I I 1

15 20


indeedreflectbrain reorganizationsthat relateto the new capacitiesthatemergeat specific ages. Even if this scientific hypothesisproves true,thereare major issuesabout the implications for educationalpractice.Cautionis requiredin drawingconclusionsaboutthe natureof learningand development.

In the I970s and 1.980s,severalAmericanbiologistsandeducatorsusedevidenceaboutage-relatedspurtsin headcircumferenceand EEGenergy,which they called “phrenoblysis” (Epstein,1974, 1978) to drawconclusionsabouthow schoolchildrenlearn(Fischer& Lazerson,1984).They treatedtheseconclusionsas factsandused them to makeextensiverecommendationsto schoolboards,teachers,andparents.For example,the scholarswent directly from theii findings about spurts in headgrowth andEEG to conclusionsthatwhen the headis in a growth pla-teau(a periodof little change,not a spurt), no learningcan occur.Theytold educatorsthat instructionin newconceptsshouldfocuson periodsof growth becausethat was when new learningcould occur. Yet therewas absolutely no researchtestinghow learning related to periods ofbrain andheadgrowth,andtherewassubstantialevidencethat childrenlearnnew materialat all agesduring the schoolyears,with no flat peri-ods where learningdoesnot occur.

One reasonfor the popularity of theserecommendationsin educa-tion was that few educatorsknew muchabout the biology of the brain,somany of them simply acceptedthe claims of phrenoblysisas scientificfact. With strongerreciprocalconnectionsbetweenneuroscience,cogni-tive developmentalscience,and education,the scientists’ hypothesisabout the relation of spurts andplateausto learningwould havebeensubjectto empiricaltestbeforebeing usedto makerecommendationsforeducationalpolicy andpractice.

The model of phrenoblysisassumedthat the brain and cognitionworkedtogetheras a unitarysysteminsteadof beingcomposedof manyparts,most of which areonly looselycoupled. Contraryto thatassump-tion, developmentdoesnot happenin a singleprocessacrossall regionsof the brain, althoughthereare importantsimilarities in someaspectsofbrain developmentacrossmany brain regions.An overly simple look atthe EEG evidencecan lead to the conclusion that the entire brain isdeveloping during a growth spurt. In reality, developmenttakes placealong separatestrands(in a developmentalweb), andoneof the goalsofneuroscientificresearchis to characterizerelationsamongthe growingstrands.Early evidence indicates that the growth processoccurs incycles,movingsystematicallyaroundlocationsin the brain, not as a sin-gle spurtat the sametime acrossall brain regions(Thatcher,1994).Theleft andright hemispheresseemto developin different sequences,whichappearto repeat for each cycle of cortical reorganization(Fischer &


Rose,1996).Again, what at first appearslike a unified processis in facta diverseset of individual processesacting in concert with each other.Scientistsand educatorscan understandhow the “system” works onlyby examiningthe partsand how they vary, which will eventuallyleadtoan explanationof the neuroscientificprinciples of brain developmentand learning.

Executive Function Is Not UnitaryExecutivefunction is typically conceivedas a broadcognitive capacityand is subjectto thesamekindsof pitfalls in interpretationasothercon-cepts aboutcognitionand brain. The idea of a single,unified executivefunction falls into the sametrap as conceptsof unitary motorabilities,cognitive development,and brain development.The continuing lack ofconsensusregardinga definitionof executivefunctionarisesin largepartfrom the problemsthat result from treatingit as a single,unitary cogni-tive ability. Teuber (1972) was oneof the first to addressthe questiondirectly, in his article entitled “Unity and Diversity of Frontal LobeFunctions,” and a numberof researchershavetakenup the issuemorerecently (Duncan,Johnson,Swales,& Freer, 1997; Miyake, Friedman,Emerson,Witzki, & I-Iowerter, 2000). As Baddeley(1996) putsit, thequestionremainswhetherit will “prove moreappropriateto regardtheexecutiveas a unified systemwith multiple functions,or simply as anagglomerationof independentthough interacting control processes”(p. 5).

Analysesof executivefunction havetakenpositions of both unityand divergence and various stancesin between.The Norman andShallice (1986) model of the control of action positsa relatively unifiedsystemduringcompletionof non-routineactivities. It posits two modesof control,oneresponsiblefor routine activitiesandonefor non-routineones.Routinetasksare triggeredwheneverappropriatestimuli are pres-ent, andthe systemproceedsautomaticallywithout further monitoring.Tasksthat are morecomplexor novel requirehigher-ordercontrol by anexecutivesystemthat regulatesthe executionof the activities.While theearly version of this model was stronglyunitary, it has moved towarddifferentiation in revisedversions(Miyake et al., 2000).

Toward the otherendof the spectrumfrom unified to independentis PenningtonandOzonoff’s (1996) model,which treatsexecutivefunc-tion as a useful functional constructbut moves away from the broadfrontal cortexmetaphor,in which all typesof executivetasksare seenasreflectinga singlebrain function.The authorsinterpretthe metaphorasa logical outgrowth of findings about deficits in patients with frontallobe damage.Many patients have difficulty with planning or prob-


lem solving, but their intelligenceis often preserved.PenningtonandOzonoffproposea clusterof weaklycoupledfunctionsconvergingupon“planning or programmingfuture actions,holding thoseplans or pro-gramson-line until executed,andinhibiting irrelevantactions”(p. 55).

A clear indication of the shift even further toward separatepro-cessesis to refer to executivefunctions in the plural (Burmeisteret al.,2005; Fischer,Barkley, Smallish,& Fletcher,2005;Manchester,Priestley,& Jackson,2004).Recall from the discussionsof the motorsystem,cog-nitive development,and brain developmentthat this kind of model ofseparatecomponentsthat work togetheris pervasivein cognitivescienceand biology. A unified executivefunction may be useful as a construct(Zelazo, Mueller, Frye, & Marcovitch,2003), but it is misleadingas arepresentationof the true natureof the system.Severalpiecesof evidencesupportthe stancethat executivefunctioningconsistsof diversecompo-nentsthat function independentlyin many ways.

First, peopleperform differentially on measuresof separateaspectsof executive function. Just as a theory of unitary stagesof cognitivedevelopmentpredictssimilar capabilitiesacrossdomains,aunitary viewof executivefunction predictssimilar performancein different compo-nents,such as planningandinhibition, Severalstudiesshow distinct dif-ferencesin thesecomponents.

For example,Carlson,Moses, and Claxton (2004) found that 3-and4-year-oldsperformeddifferently on testsof planningandinhibitorycontrol, showing largely independentprocesses.In a landmark study,Miyake and colleagues(2000) focusedon differential performanceonelementsof executivefunction in collegestudents.Most researchusessimple correlations between tasks, which could reflect differences inaspectsof the taskthatdo notinvolve executivefunctioning,suchaslan-guageuse.Miyake andcolleaguesuseda morepowerful statisticalanal-ysis, a latent variable approach,to investigatethreeseparateelementsofexecutivefunction: shifting, updating,andinhibition. With confirmatoryfactor analysis,they found that the threeconstructswereclearly distin-guishableand that they demonstratedsomeunderlying commonality.They concludedthat the results indicated“both unity and diversity ofexecutivefunctions” (Miyakeet al., 2000,p. 87). Justlike thestrandsinthe developmentalweb, they are mostly independentbut loosely cou-pled.

The hypothesizedcomponentsof executivefunction neitherappearat the samelevel of masterywithin individuals nor do they necessarilydeveloptogether.Anderson(2002) suggeststhat individual elementsofexecutivefunction show different developmentaltrajectories,includingattentionalcontrol, cognitive flexibility, goal setting, and informationprocessing.Eachdomain involvesa distinct developmentalstrandin the


web for executivefunction, as shown in Figure 4.1, and evidencesug-geststhat separateprocessesdevelopat different rates,reachingskilledlevelsat differentages.The precisenatureof thesetrajectoriesneedstobe investigatedempirically, but clearlythe evidencepoints to diversity inexecutivefunctions throughoutdevelopment.

EDUCATIONAL IMPLICATIONS

Doesthe debateandconfusionaboutthe definitionof executivefunctionmakeany differencefor educationalpractice?Yes. As in the examplesof motor functioning, cognitive development,and brain development,assumptionof a unity that is not presentleads quickly to dangersinpracticalimplications.Oneof the mostpotentpitfalls involvesdecisionsabouthow to supportstudentswith deficitsin executivefunction,whoseprofiles vary dramatically.

Many developmentaldisabilities,such as attention-deficit/hyperac-tivity disorder(ADHD) andautism,involve deficits in executivefunction(Pennington& Ozonoff, 1996).Although thesedeficits make the disor-ders appearsimilar, the executivedysfunction manifestsdifferently indistinct disorders,and diverseinterventionsare required.For example,the processof inhibition showslessimpairmentin many individualswithautismthan in thosewith ADHD, eventhoughthe impairmentin otherexecutivefunctions,such as planning, is more severein autism (Hill,2004;Pennington& Ozonoff, 1996). Following the web model in Fig-ure4.1, different aspectsof executivefunction developalong separatestrands(which sometimesintersector branch).

Profiles of executivefunction deficits canevenbe uniqueandunde-tectableby traditional measures.Multiple studieshavereportedcasesofindividualswith frontal lobe damagewho performedwell on executivetasksin a laboratorysettingbuthadcleardifficulty with executivetasksin reallife (Burgess,Alderman,Evans,Emslie, & Wilson, 1998). Profilesof executivefunction performanceare highly variable and do not war-rant a unitary concept.

Recognizingthat executivefunction hasmultiple aspectsand is notunitary thus has practical implications in the classroom.Researchandpracticethat separatedistinct aspectsof a phenomenon,moving beyondthe vaguefrontal metaphorfor executive function,will help educatorsdevise more useful,differentiateddiagnosesand interventions.Blanketstatementsof deficits in executivefunction are certainlyless useful thanfocused ones highlighting particular componentskills like attention,inhibition, andplanning. The remarkablesuccessof researchand prac-tice that specify the particular functions underlying dyslexia and other


readingdifficulties, describedearly in this chapter,providesan excellentmodel for this endeavor(Fischeret al., 2006).

An interactive relationship between education and the learning,brain, anddevelopmentalscienceswill fosterthis type of nuancedunder-standingandimprove the work of botheducatorsandresearchers.Justas doctors facing a particular challengeor novel discoveryin patientscan inform the work of biological scientists,educatorscan help fosterusefulandproductiveresearchin complementarydisciplines.The bene-fits are reciprocalbecauseinnovativeresearchfindings canbe appropri-ately translatedinto educationalpractice.

REFERENCES

Anderson,J. R. (2002). Spanningsevenordersof magnitude:A challengeforcognitivemodeling. Cognitive Science,26, 85—112.

Baddeley, A. (1996). Exploring the central executive. Quarterly Journal ofExperimentalPsychology:SectionA, 49, 5—28.

Burgess,P. W., Alderman,N., Evans,J., Emslie,H., & Wilson, B. A. (1998). Theecological validity of testsof executive function.Journal of the Interna-tional NeuropsychologicalSociety,4, 547—558.

Burmeister R., Hannay,H. J, Copeland,K., Fletcher,J. M., Boudousquie,A., &Dennis,M. (2005). Attention problemsand executivefunctionsin childrenwith spina bifida and hydrocephalus.Child Neuropsychology,11, 265—283.

Carlson,S. M., Moses,L. J., & Claxton, L. J. (2004). Individual differencesinexecutive functioning and theory of mind: An investigationof inhibitorycontrol and planning ability. Journal of ExperimentalChild Psychology,87, 299—319,

Case,R. (1974). Structuresand strictures: Somefunctional limitations on thecourseof cognitive growth. Cognitive Psychology,6, 544—574.

Damon,W., & Lerner, R. M. (Eds.). (2006). Handbookof child psychology:Theoreticalmodelsof human development(6th ed., Vol. 1). New York:Wiley.

Dawson-Tunik, T. L., Commons,M., Wilson, M., & FischeçK. W (2005). Theshapeof development.EuropeanJournalof DevelopmentalPsychology,2,163—195.

Duncan,J., Johnson,R., Swales,M., & Freer, C. (1997). Frontal lobe deficitsafter headinjury: Unity anddiversity of function. CognitiveNeuropsychol-ogy, 14, 713—741.

Epstein,H. T. (1974). Phrenoblysis:Special brain and mind growth periods.DevelopmentalPsychobiology,7, 207—224.

Epstein,H. T. (1978). Growth spurtsduringbraindevelopment:Implications foreducationalpolicy andpractice.In J. S. Chall& A. E Mirsky (Eds.),Educa-tion andthe brain (Yearbookof theNationalSociety for the Studyof Edu-cation). Chicago: University of ChicagoPress.


Fischer,K. W, & Bidell, T. R. (2006).Dynamicdevelopmentof action, thought,and emotion. In W. Damon& K. M. Lerner(Ed.),Handbookof child psy-chology; Theoretical models of human development(6th ed., vol. 1,pp. 313—399). New York; Wiley.

Fischer,K. W., Bullock, D. H., Rotenberg,E.J., & Raya, P. (1993).The dynam-ics of competence:How context contributes directly to skill, In R. H.Wozniak & K. W. Fischer (Eds.), Developmentin context: Acting andthinking in specificenvironments(pp. 93—117). Hillsdale, NJ: Erlbaum.

Fischer,K. W., & Immordino-Yang,M. H. (2002). Cognitive developmentandeducation:From dynamicgeneralstructureto specific learningand teach-ing. In E. Lagemann(Ed.), Traditions of scholarshipin education.Chicago;SpencerFoundation.

Fischer, K. W., Immordino-Yang,M. H., & Waber, D. P. (2006). Toward agroundedsynthesisof mind, brain, andeducationfor readingdisorders:Anintroductionto the field and this book.In K. W. Fischer,J. H. Bernstein,&M. H. Immordino-Yang(Eds.),Mind, brain, and educationin readingdis-orders. Cambridge,UK; CambridgeUniversity Press.

Fischer,K. W., & Katzir, T. (Eds.).(in press).Building usableknowledgein mind,brain, and education,Cambridge,UK: CambridgeUniversity Press.

Fischer,K. W., & Lazerson,A. (1984). Research:BrainspurtsandPiagetianperi-ods. EducationalLeadership,41(5), 70.

Fischer,K. W., & Rose,S. P. (1996). Dynamicgrowth cycles of brain andcogni-tive development.In K. Thatcher,G. K. Lyon, J. Rumsey,& N. Krasnegor(Eds.), Developmentalneuroimaging:Mapping the developmentof brainand behavior (pp. 263—279).New York: AcademicPress.

Fischer,M., Barkley, K. A., Smallish,L., & Fletcher,K. (2005). Executivefunc-tioning in hyperactivechildren as young adults: Attention, inhibition,responseperseveration,and the impact of comorbidity. DevelopmentalNeuropsychology,27, 107—133.

Hill, E. L. (2004). Executivedysfunctionin autism.Trendsin CognitiveSciences,8, 26—32.

Kitchener,K. S., Lynch, C. L., Fischer,K. W., & Wood, P. K. (1993). Develop-mental rangeof reflective judgment:The effect of contextual supportandpracticeon developmentalstage.DevelopmentalPsychology,29, 893—906.

Klahr, D., & Wallace, J. G. (1976). Cognitive development:An information-processingview. Hillsdale, NJ: Erlbaum.

Mader,S. S. (2005). Understandinghumananatomyand physiology(5th ed.).NewYork: McGraw-Hill.

Manchester,D., Priestley,N., & Jackson,H. (2004).The assessmentof executivefunctions; Coming out of the office. Brain Injury, 18, 1067—1081.

Marieb, E. N. (2006). Essentialsof humananatomyand physiology(8th ed).NewYork: PearsonBenjaminCummings.

Matousek,M., & Petersen,I. (1973). Frequencyanalysisof theEEGin normalchildrenandadolescents.In P. Kellaway & I. Petersen(Eds.), Automationof clinical electroencephalography(pp. 75—102).New York: RavenPress.

Miyake, A., Friedman,N. P., Emerson,M. J., Witzki, A. H., & Howerter,A.(2000). Theunity and diversity of executivefunctionsand their contribu-


tions to complex“frontal lobe” tasks:A latentvariableanalysis.CognitivePsychology,41, 49—100.

Norman,D. A., & Shallice,T. (1986).Attention to action: Willed andautomaticcontrolof behavior.In K. J. Davidson,G. E. Schwartz,& D. Shapiro(Eds.),Consciousnessandself-regulation:Advancesin researchandtheory (Vol. 4,pp. 1—18). New York: Plenum,

Pennington,B. E, & Ozonoff, S. (1996). Executivefunctionsanddevelopmentalpsychopathology.Journalof Child PsychologyandPsychiatry,37, S1—87.

Piaget,J. (1983). Piaget’s theory. In P. H. Mussen& W. Kessen(Ed.), Handbookof child psychology:History, theory,andmethods(4th ed.,vol.l, pp. 103—126). New York: Wiley.

Rose,D. H., Meyer, A., Strangman,N., & Rappolt,G. (2002). Teachingeverystudent in the digital age: Universal designfor learning. Alexandria,VA:ASCD.

Kuhland, K., & van Geert,P. (1998). Jumpinginto syntax: Transitionsin thedevelopmentof closedclass words.British Journal of DevelopmentalPsy-chology, 16(Pt. 1), 65—95.

Somsen,K. J. M., van ‘t Klooster, B. J., van derMolen,M. W., van Leeuwen,H.M. P., & Licht, K. (1997). Growthspurts in brain maturationduring mid-dle childhood asindexedby EEGpower spectra.Biological Psychology,44,18 7—209.

Teuber, H.-L. (1972). Unity and diversity of frontal lobe functions.ActaNeuro-biologiae Experimentalis,32, 615—656,

Thatcher,K. W. (1994). Cyclic cortical reorganization:Origins of humancogni-tive development.In G. Dawson& K. W. Fischer (Eds.),Humanbehaviorand the developingbrain (pp. 232—266). New York: Guilford Press.

Turk, D. J., Heatherton,T. E, Kelley, W. M., Funnell,M. G., Gazzaniga,M. S.,& Macrae, C. N. (2002). Mike or me? Self-recognition in a split-brainpatient.Nature Neuroscience,5, 841—842.

Wolf, M., & Bowers,P. (1999).The “double-deficithypothesis”for thedevelop-

mentaldyslexias.Journalof EducationalPsychology,91, 1—24.Wolf, M., & Katzir-Cohen,T. (2001). Readingfluency andits intervention.Sci-

entific Studiesof Reading,5, 211—239.Wolf, M , Miller, L & Donnelly K (2000) etrieval Automaticity, Vocab~6 — C) ~ IV ~ (RAVE-O): A comprehensivefluency-

—zr basedreadinginterventionprogram.Journalof LearningDisabilities, 33,375—386.

Zelazo, P. D., Mueller, U., Frye, D., & Marcovitch,S. (2003). The developmentof executive function in early childhood.Monographsof the SocietyforResearchin Child Development,68(3).

chapter 4 connecting cognitive science ... - …ddl/articlescopy/fischerdaley_execprocesses... ·...

Documents