Weaknesses in executive functioning predict the initiating of adolescents’ alcohol use - 502154

(1)

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Weaknesses in executive functioning predict the initiating of adolescents’

alcohol use

Peeters, M.; Janssen, T.; Monshouwer, K.; Boendermaker, W.J.; Pronk, T.; Wiers, R.;

Vollebergh, W.

DOI

10.1016/j.dcn.2015.04.003

Publication date

2015

Document Version

Final published version

Published in

Developmental Cognitive Neuroscience

License

CC BY-NC-ND

Link to publication

Citation for published version (APA):

Peeters, M., Janssen, T., Monshouwer, K., Boendermaker, W. J., Pronk, T., Wiers, R., &

Vollebergh, W. (2015). Weaknesses in executive functioning predict the initiating of

adolescents’ alcohol use. Developmental Cognitive Neuroscience, 16, 139-146.

https://doi.org/10.1016/j.dcn.2015.04.003

General rights

It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)

and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open

content license (like Creative Commons).

Disclaimer/Complaints regulations

If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please

let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material

inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter

to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You

will be contacted as soon as possible.

(2)

ContentslistsavailableatScienceDirect

Developmental

Cognitive

Neuroscience

jou rn a l h om ep ag e : h t t p : / / w w w . e l s e v i e r . c o m / l o c a t e / d c n

Weaknesses

in

executive

functioning

predict

the

initiating

of

adolescents’

alcohol

use

Margot

Peeters

a,∗

_,

_Tim

_Janssen

b

_,

_Karin

_Monshouwer

a,c

_,

_Wouter

_Boendermaker

b

_,

Thomas

Pronk

b

_,

_Reinout

_Wiers

b

_,

_Wilma

_Vollebergh

a

a_Utrecht_University,_Utrecht,_The_Netherlands b_University_of_Amsterdam,_The_Netherlands c_Trimbos_Institute,_The_Netherlands

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received25September2014 Receivedinrevisedform8April2015 Accepted9April2015

Availableonline17April2015

Keywords: Executivefunctioning Alcoholuse Bingedrinking Adolescents Survivalanalyses Cognition

a

b

s

t

r

a

c

t

Recently,ithasbeensuggestedthatimpairmentsinexecutivefunctioningmightberiskfactorsfor theonsetofalcoholuseratherthanaresultofheavyalcoholuse.Inthepresentstudy,weexamined whethertwoaspectsofexecutivefunctioning,workingmemoryandresponseinhibition,predictedthe firstalcoholicdrinkandfirstbingedrinkingepisodeinyoungadolescentsusingdiscretesurvival anal-yses.AdolescentswereselectedfromseveralDutchsecondaryschoolsincludingbothmainstreamand specialeducation(externalizingbehavioralproblems).Participantswere534adolescentsbetween12 and14yearsatbaseline.Executivefunctioningandalcoholusewereassessedfourtimesoveraperiod oftwoyears.Workingmemoryuniquelypredictedtheonsetoffirstdrink(p=.01)andfirstbinge drink-ingepisode(p=.04)whileresponseinhibitiononlyuniquelypredictedtheinitiatingofthefirstdrink (p=.01).Theseresultssuggestthattheassociationofexecutivefunctioningandalcoholconsumption foundinformerstudiescannotsimplybeinterpretedasaneffectofalcoholconsumption,asweaknesses inexecutivefunctioning,foundinalcoholnaïveadolescents,predicttheinitiatingof(binge)drinking. Though,prolongedandheavyalcoholusemightfurtherweakenalreadyexistingdeficiencies.

1. Introduction

Recentstudieshavedemonstratedassociationsbetween cog-nitivedeficitsand(heavy)drinkinginadolescents(Fernieetal., 2013;Hansonetal.,2011;Khuranaetal.,2012;Niggetal.,2006; Squegliaetal.,2009;Tapertetal.,2002).Especiallybinge drink-ingseemstobeassociatedwithcognitivedeficitsinadolescents andyoungadults(Hermensetal.,2013;Peetersetal.,2014).An importantquestionthatarisesfromthesefindingsiswhetherthese deficitsarepresentpriortoandpredicttheonsetofdrinkingor whetherthe(heavy)useofalcoholinducesthesedeficits.Although thisrelationshiphasbeeninvestigatedinseveralstudies(Fernie etal.,2013;Hansonetal.,2011;Khuranaetal.,2012;Niggetal., 2006;Squegliaetal.,2009;Tapertetal.,2002),noneofthesestudies haveincludedalcohol-naïveadolescentsatbaseline,whichclearly limitstheinterpretationoftheassociationsfound.Inthepresent studywethereforeexaminedwhethertherelativeweaknessesin

∗ Correspondingauthorat:UtrechtUniversity,Heidelberglaan1,3584CSUtrecht, TheNetherlands.Tel.:+31302533056.

E-mailaddress:m.peeters1@uu.nl(M.Peeters).

executivefunctions(i.e.,workingmemoryandresponseinhibition) predicttheinitiationoftheﬁrstalcoholicdrinkandtheﬁrstbinge drinkingepisodeinyoungadolescents.

1.1. Cognitivedevelopmentinadolescence

During adolescence several important cognitive functions matureanddevelop.Executivefunctionssuchasworking mem-ory,attentionandresponseinhibition,continuetomatureuntillate adolescence(BlakemoreandChoudhury,2006;Caseyetal.,2008). Thesefunctionsareimportantfortheplanning,organizationand coordinationofothercognitiveprocesses(Baddeley,1983;Miyake etal.,2000),andinvolveneuralnetworksincludingdifferentareas intheprefrontalcortex.Incontrasttootherbrainregions(e.g.,the visualcortex),theprefrontalcortexundergoeschangesuntillate adolescence(BlakemoreandChoudhury,2006).Workingmemory, inparticular,continuestomatureuntillate adolescence(18–21 years;Lunaetal.,2004).Inaddition,basiclevelsofresponse inhibi-tionarealreadypresentinearlychildhood;youngchildrenalready abletoinhibitresponses,however,improvementofthisfunction continuesintoadolescence(Lunaetal.,2004).Bothworking mem-oryandresponseinhibitionareimportantforcognitivecontrol;the

http://dx.doi.org/10.1016/j.dcn.2015.04.003

(3)

140 M.Peetersetal./DevelopmentalCognitiveNeuroscience16(2015)139–146

abilitytosuppressthoughtsandimpulsesinordertoachieve long-termgoals(BungeandWright,2007;Lunaetal.,2004;Steinberg, 2007).Cognitivecontrolhasoftenbeenrelated torisk behavior (Caseyetal.,2008;Steinberg,2007).Ithasbeenarguedthatthe slowermaturationoftheprefrontalcortex,andtheaccompanying gradualdevelopmentofcognitivecontrol,mightunderlietheoften observedincreasesinriskybehaviorsduringadolescence,including bingedrinking(Caseyetal.,2008;Steinberg,2007).Adifferentline ofresearchhassuggestedthattheincreaseinrisk-takingbehavior isamotivationalissue(CroneandDahl,2012).Itismore reward-ing,andthusmoremotivating,foradolescentsthanitisforadults toengageinriskybehaviorsbecauseoffactorssuchaspeer accep-tanceorpopularity,whichmightexplaintheincreasedrisk-taking behaviorduringadolescence(CroneandDahl,2012).Thedelayed developmentoftheprefrontalcortexandassociatedcognitive con-trolskillsinadolescentsmightthereforenotrepresenta“deﬁcit” orimmaturefunctioning,butratherahighlyﬂexibleandadaptive development,tunedtothesocialdemandsadolescentsencounter (foradetaileddescriptionseeCroneandDahl,2012).Nevertheless, inbothlinesofresearch,thedevelopmentofcognitivecontrolin lateadolescenceplaysanimportantroleinexplainingrisk-taking behaviorsinadolescents.

1.2. Executivefunctioningandalcoholuse

Working memory (i.e., keeping information active) and responseinhibitionaretwoimportantexecutivefunctionsoften examinedin relationtoalcoholuse(Khuranaetal.,2012; Nigg etal.,2006;Verdejo-Garciaetal.,2008;Wiersetal.,2007).Deﬁcits inresponse inhibitionand workingmemoryappear toincrease automaticandimpulsiveresponseinsuchawaythatbehavioral responsesinrelationtoalcoholstimuliaremorelikelydirectedby immediatesatisfactionofneedsthanpursuanceoflong-termgoals (Grenardetal.,2008;Thushetal.,2008;Peetersetal.,2012,2013; Wiersetal.,2007).Inotherwords,deﬁcitsinexecutivefunctioning limitindividualstorespondinacontrolledandplannedmannerto alcoholstimuli,leadingtomoreimpulsiveandautomaticresponses whichhavebeenassociatedwithincreasedalcoholuse(Grenard etal.,2008;Thushetal.,2008;Peetersetal.,2012).

Severallongitudinalstudies haveexamined thedirect effect ofrelativelypoorexecutivefunctioningondrinkingbehavior in adolescence.Niggandcolleagues(2006),forinstance,foundthat poorinhibitionskillsinearlyadolescents(12–14years,N=498) predictedlater (15–17years)problematicalcohol use.Likewise, Khuranaandcolleagues(2012)foundthatrelativelypoor work-ing memorypredicted growth in the frequency of alcohol use ina community sample ofadolescents (N=358).Moreover,the authorsfoundthattheeffectofworkingmemoryonalcoholusewas mediatedbyimpulsivity suggestingthat poorworkingmemory functioningmightincreaseimpulsivebehaviorwhichsubsequently increasesalcoholuseamongadolescents.Inaddition,Fernieand colleagues(2013)foundthatseveralmeasuresofimpulsivity(i.e., delaydiscounting,risktakingandresponseinhibition)predicted increasein alcoholuseamong youngadolescents(12–13years, N=287).Thequestionremains,however,whetherthesedeﬁcits precedeandpredicttheonsetofadolescents’drinking,sincethese studies have examined executive functioning after adolescents alreadyconsumedtheirﬁrstdrink,thusleavingopenthe possibil-itythattheimpairedexecutivefunctioningwasalreadytheresult ofearlierdrinking.

1.3. Currentstudy

In the present study we examined the predictive effect of workingmemoryandresponseinhibition,twoexecutivefunctions undertheheadingofcognitivecontrol,andfrequentlyassociated

withriskbehavior,ontheinitiatingofthefirstalcoholicdrinkina sampleofyoungadolescents(12–15years,beforethelegalalcohol buyingageof16–intheNetherlandsatthetimeofthestudy). Althoughcognitioninvolvesmany moreaspects thanexecutive functions,wedecidedtofocusonthesetwocognitions,since partic-ularlythesefunctionsareassumedtoplayavitalroleinadolescent alcoholuseastheycontinuetomatureinadolescence( Verdejo-Garciaetal.,2008).Bingedrinkingisacommondrinkingpattern amongyoungadolescents,andithasbeenlinkedtoseveral(health) riskbehaviors(Milleretal.,2007).Wethereforealsoexaminedthe predictiveeffectofexecutivefunctioningontheinitiatingofthe firstbingedrinkingepisode(i.e.,fiveormoreglassesonaday). Bothadolescentsform mainstreamas wellasadolescentsfrom specialeducation(SE)wereincludedinthestudytoascertain vari-ationindrinkingbehavior.Adolescentsin SEschoolshavebeen foundtodrinkmoreheavily(Kepperetal.,2011)comparedto ado-lescentsfrommainstreameducation.AdolescentsattendtheseSE schoolswhentheyhavebehavioralproblems(e.g.,conduct prob-lems,hyperactivity,attentionproblems),althoughtheymaynot necessarilybediagnosedwithabehavioraldisorder.These exter-nalizingbehavioralproblemsareassociatedwithrelativelyweak WMfunctioning(Barkley,1997;Pihletal.,1990),andmightplace adolescentsatriskforproblematicalcoholuseandlateralcohol dependence.

We hypothesizedthat relativelyweakworkingmemoryand responseinhibitionwouldpredicttheinitiatingofthefirstalcoholic drinkandthefirstbingedrinkingepisodeinthissample. Discrete-timesurvivalanalyseswasusedtodeterminetheinitiatingofthe firstdrinkandbingedrinkepisodeduringthetwo-yearfollow-up ofthestudyandtoevaluatethepredictiveeffectofexecutive func-tioning.Moreover,bothfunctionshavepreviouslybeenassociated withadolescentalcoholuseandproblemdrinking(Khuranaetal., 2012;Niggetal.,2006),althoughtheiruniquecontributiontothe initiatingofdrinkinghasnotyetbeenexaminedsimultaneously.

2. Materialsandmethod

2.1. Participants

Thisstudywaspartofalargerstudyinwhichweassessed self-reportedandbehavioralcognitions,andpersonalitystylesrelated toalcoholuseandotherriskbehaviors.Participantswererecruited fromseveralmainstreamand 17SEschoolsintheNetherlands. Maleswereslightlyoverrepresented(69%boysversus31%girls) becausemoreboysthangirlsattendSEschools(Oswaldetal.,2003). Forbothsamples,informedconsentofthechildaswellas pas-siveparentalconsentwasrequested.Inthemainstreamsample, 37parentsdeclinedparticipationoftheirchild.FortheSEsample, 15parentsand7studentsdeclinedparticipationinthisstudy.The mainstreamsampleincluded250adolescents(69%girls),selected fromanationalsurveystudy,whichwaspartoftheInternational Health Behavior of School-aged Children-survey (HBSC, Zanotti etal.,2012).AdolescentswhoparticipatedintheDutchnational studywerecontactedforadditionalassessmentsafterthe comple-tionofthemainsurvey.TheSEsampleincluded374adolescents (12%girls).Forthepurposeofthisstudy,adolescentsbetween12 and15yearswereselected,resultinginasampleof534(371boys, 163girls)adolescents.

Data were collected at intervals of 6–8 months acrossfour waves.Atbaseline,525adolescentsparticipatedinthestudy(98% of534,notethatsomeadolescentswereabsentduringtheﬁrst wavebutparticipatedinthefollowingwaves).Overall,415(78%) adolescentsparticipatedinwave2,399(75%)adolescents partici-patedinwave3,and425(80%)adolescentsparticipatedinwave4. Missingcaseswereimputedusingamultipleimputationprogram

(4)

(i.e.,multipleimputationbychainedequations(MICE),whichis explainedbelow).Thedataoverfourwaveswerecollectedeither viatheInternet (inthecase of mainstreameducation) orvia a trainedresearchassistant(inthecaseofSEeducation).Adolescents fromSEcompletedthequestionnairewithpenandpaperfollowed byacomputersessiontoassessthecomputertasks.FiveoftheSE studentscompletedonlythecomputertasksatbaselineanddidnot completethequestionnaire.Sinceweusedmultipleimputationto handlemissingcases,weincludedtheseadolescentsinthestudy forfurtheranalysis.

2.2. Measures

2.2.1. Demographicalmeasures

Sex,ageandexternalizingbehavioralproblemswereassessed usingaquestionnaire.TheStrengthandDifﬁcultiesQuestionnaire (Goodman,1997)wasusedtoassessexternalizingbehavior prob-lems(hyperactivityandconductproblems).Thescaleincludes25 items,withthreeansweringcategories(i.e.,nottrue,somewhat trueorcertainlytrue).Forthepurposeofthisstudyweusedthe externalizingsubscaleofthisquestionnaire,whichincludesitems suchas“Iamrestless,overactiveandcannotstaystillforlong”or “Ioftenloosetemper”.

2.2.2. Alcoholuse

Todeterminetheinitiatingofdrinkingandbingedrinking(five ormoreglassesonaday),weaskedparticipantstoindicate,for weekendand weekdays separately,the numberof glasses they consumedonanaveragesingleday(e.g.,“Whenyouconsume alco-holonaweek/weekenddayhowmanyglassesdoyoudrink).For thepurposeofthesurvivalanalysis,thiscontinuousmeasurewas recodedasdichotomousvariable.Theinitiatingoffirstalcoholic drinkwasdefinedasdrinkingatleastonestandardglassof alco-hol(containing12.5mlofpurealcohol)onaweekorweekendday. Participantsreceivedeithera“one”ora“zero”onbothvariables, whichservedasanindicatoroftheoccurrenceoftheevent(i.e., ini-tiatingofdrinkingorbingedrinking;“one”indicatedthattheevent occurred).

2.2.3. Workingmemory

Working memory performance was assessed with the Self OrderedPointingTask(SOPT;PetridesandMilner,1982). Partic-ipantscompleteda concreteversion oftheSOPT.Pictureswere simultaneouslyplaced at different positions onscreen, starting withapracticetrialof4picturesfollowedbyfourtrialswith6, 8,10, and12 pictures, respectively.Thepositionof pictures on screenwasrandomlyshufﬂedinanumberofvariationsequaltothe amountofpictures.Participantswereaskedtoselecteachpicture oncethroughoutthevariations,butnottoselectthesame posi-tiontwiceinarow.Participantswereencouragedtocompletethe taskwithoutanyerrors.Irrespectiveoftheerrorsmade,all par-ticipantsreceivedthefeedback“welldone”aftereachtrial.With respecttothesecondinstruction,thetaskwasprogrammednotto allowselectingthesamelocationtwiceinarow.Foreachtesttrial, aproportionalerrorscorewascomputedbydividingthenumberof correctresponsesbythenumberoftotalpictureswithineachtrial toaccountfortaskdifﬁculty(CraggandNation,2007).Thetotal SOPTscorewascalculatedbytakingthemeanoftheproportion ofcorrectscoresofthefourtrials.Higherscoresindicatedbetter workingmemoryfunctioning.

2.2.4. Responseinhibition

ResponseinhibitionwasassessedusingtheStrooptask(Stroop, 1935).TheStrooptask,asameasureofresponseinhibition,has beenusedsuccessfullyasamoderatorinpreviousstudies(Houben andWiers,2009;Peetersetal.,2012).Participantswereinstructed

toindicatethecoloroftheword(i.e.,red,green,blue,oryellow) thatappearedonthescreenbypressingthecorrespondingkeyon thekeyboardwhileignoringthemeaningoftheword.Participants startedwithapracticeblock,whichconsistedof40trialswith sym-bols(e.g.,@@@@or&&&&).Thepracticeblockwasfollowedbya testblockwith28trials.Trialscouldbecongruent(i.e.,meaningof thewordmatchesthecolor),neutral(i.e.,coloredsymbolsinstead ofwords),orincongruent(i.e.,meaningoftheworddiffersfromthe color),andtheywerepresentedinrandomorderbetween partici-pants.Eachtrialwasrepeateduntilacorrectanswerwasgiven.An errormessage,includingadescriptionofthekeysusedandtheir correspondingcolor,followedanincorrectresponse.Aresponse inhibitionscorewascalculatedbysubtractingtheRTsonneutral trialsfromtheRTsonincongruenttrials,withhigherRTs indicat-ingmoreproblemswithcontrol.Foranalyticalpurposes,RTswere dividedby1000.

2.3. Missingdata

Becausewe wanted todeterminewhethera participant ini-tiateddrinkingand/orbingedrinkingduringstudyfollow-up,it wasnecessarytohaveadatasetthatwasascompleteas possi-ble.Missingdata,especiallywhenitinvolvesariskfactororevent, isparticularlyproblematicinsurvivalanalyses(vanBuurenetal., 1999).Multipleimputation(MI)isa commonlyusedmethodto handlemissingdata.PreviousresultsindicatedthatMIisareliable andvalidmethod(Sterneetal.,2009).However,tocheckwhether thiswasalsothecaseforthepresentsample,asimulationstudy wasperformed(performedonthehigh-risksample,whichhadthe largestnumberofmissingdata).Todeterminethebestmethodto handlethemissingdata,severaloptionswerecompared,including listwisedeletion,manualfillinmethods(e.g.,lookingatpreviousor subsequentwave/lookingatmajority/fillinginzerosateach miss-ing),andMI.Wecomparedthefivemethodsontwocriteriathat wererelevantforourstudy:(1)percentageofonsettersthatwould beidentifiedbyeachmethodand(2)thepredictiveeffectoftwo variablescommonlyassociatedwithalcoholuse,namelysensation seekingandimpulsivity.Theresultsrevealedthatmultiple impu-tationwastheoptimalmethodtohandlethemissingdatainthe currentstudy(papersubmittedforpublication,detailed descrip-tionofthesimulationstudycanberequestedfromthefirstauthor). We usedtheprogram MICE(R-packageinR)tocreate five dif-ferentimputationfiles,whichweretransportedtoMplusversion 7(MuthénandMuthén,1998–2010)forsurvivalanalyses.MICE replaces themissing information withplausible values byboth minimizingtheuncertaintyandthebiasofthemodel(vanBuuren etal.,1999).Plausiblevaluesareimputedbasedoncarefully cho-senconfoundingvariablesthatserveaspredictorsofthemissing values.Forthepresentstudy,wecheckedcorrelationmatricesto determinethevariablesthatshouldbeusedtopredictmissingdata. Bothcovariatesandtheoutcomemeasurewereincludedin the imputationmodel,asrecommendedbythedevelopers(vanBuuren andOudshoorn,2000).Forthebinaryvariables,wechosethe logis-ticregressionmethod.Forallothercontinuescovariates,weused thepredictivemeanmatchingmethod.Lifetimealcoholuse,daily smoking,druguse(cannabis,ecstasy,cocaine,hallucinogens), gen-der,age,andpersonality(impulsivity,sensationseeking,anxiety, andhopelessness)wereselectedaspredictorsofmissingness.

2.4. Analyticalprocedure

First,descriptivesfordemographicalinformation,alcoholuse andexecutivefunctioningwerepresentedforthetwosamples sep-arately(mainstreamversusSE,Table1)andfortheonsetandbinge sampleseparately (Table2).Second,theﬁveimputeddataﬁles werepreparedfor discrete-timesurvival analysis.Sincetheaim

(5)

Fig.1. Structuralsurvivalmodelwithageindicators(12–17years)andpredictors(i.e.,responseinhibitionandworkingmemory)oftheinitiatingof(binge)drinking.

Table1

Descriptivesforthetwosamples(mainstreamversusspecialeducation)separately.

Mainstream education Special education t-Test Age 13.84(.82) 13.70(.73) t(532)=−1.93 Sex(%boys) 39 88 Hyperactivityproblems(SDQ) 1.62(.46) 1.93(.48) t(501)=10.94* Conductproblems(SDQ) 1.26(.23) 1.60(.38) t(501)=6.72*

Alcoholquantitybaseline .28(1.52) 3.13(7.66) t(503)=4.81*

Workingmemory .80(.09) .73(.10) t(493)=−7.51*

Responseinhibition .07(.22) .21(.35) t(492)=4.89*

SDQ=StrengthandDifﬁcultiesQuestionnaire.

* _p_<_.05.

Table2

Descriptivestatisticsforthetotalsample,theonsetsampleandthebingedrinking sample.

Total(N=534) Onset(N=336) Binge(N=458)

M SD M SD M SD

Agebaseline 13.75 .76 13.69 .77 13.73 .78

Alcoholuse 2.19 6.45 0 0 .35 .83

Responseinhibition .16 .32 .13 .30 .14 .30

Workingmemory .75 .11 .76 .10 .76 .11

Note:Thetotalsamplecontainsallparticipantsfrombothsampleorigins (main-streamandspecialeducation).Theonsetsamplecontainsonlythepartofthetotal samplethatdidnotdrinkatbaseline.Thebingesamplecontainsonlythepartofthe totalsamplethatdidnotbingeatbaseline.

ofthestudywastoexamineexecutivefunctioningasriskfactorfor

initiatingaﬁrstalcoholicdrink,weexcludedalladolescentswho

alreadyconsumedalcoholatbaseline(fortheonsetsample)andall

adolescentswhoalreadyengagedinbingedrinking(drinkingﬁve

ormoreglasses)atbaseline(forthebingesample).Foreach

ado-lescent,theageoftheonsetandtheageoftheﬁrstbingedrinking

episodewereassessedforeveryhalfayear(e.g.,12,12.5,13,...,17),

resultingin11possibletimepointsatwhichtheeventcouldhave

occurred(seeFig.1).UsingMplusforthediscrete-survivalanalysis,

alatentfactoroftheprobabilitythatanindividualwould experi-encetheeventatacertaintimepoint(i.e.,age)wasestimated.

Survival analyses were conducted using Mplus version 7 (MuthénandMuthén,1998–2010).Thefactorloadingsassessing theprobabilityoftheeventoccurrence(seeFig.1onset/binge)were ﬁxedatone(seeFig.1ageindicators)toascertainthattheeffect ofexecutivefunctioningontheonsetofalcoholuse/bingedrinking wasthesameforeachage.Residualvarianceswereﬁxedatzero. Wetestedforpossiblesexdifferencesusingmultigroupanalyses. WeusedtheclusteringoptioninMplustoaccountfortheeffectof setting(homeversusschoolassessment).

3. Results

3.1. Descriptivesforseparategroups(specialeducationversus mainstream)

Table1representsthedescriptivestatisticsforadolescentsfrom mainstreamandspecialeducationseparately.Ascanbeseenin Table 1, students in SE education signiﬁcantly differed onsex, hyperactivity and conduct problems,executive functioning and alcoholusefrommainstreamstudents.Effectsizesfordifference inworkingmemoryandresponseinhibitionwereofmediumsize (Cohen’sdworkingmemory=.53,responseinhibition=.46).

3.2. Descriptivestotalgroup

Table 1 includes descriptive statistics of age and executive functioning at baseline for the total sample and the two ana-lyticalsamples(i.e.,onset andbingedrinking)separately.Small differencesforsumscoresonthetwoexecutivefunctioningtask emerged. We used t-tests to examine whetherpossible differ-ences in executive functioning were found between thegroup whoalreadyengagedindrinkingatbaseline(excludedfromthe study)comparedtoalcoholnaiveadolescentsatbaseline(included inthestudy).Signiﬁcantdifferencesonexecutivefunctioningat baselineemergedbetweenthetwogroupswithlowerscoresfor workingmemory(M=.72fordrinkersandM=.76foralcoholnaïve adolescents)andhigherscoresforresponseinhibition(indicating relativelyweakerexecutivefunctioningforthedrinkers(working memory:t=3.26,p=.01;responseinhibition:t=−2.36,p=.02).

Figs.2and3representthecummulativeincidenceratesofthe initiating ofﬁrst drink(Fig.2)and ﬁrst bingedrinkingepisode (Fig.3)forthetotalgroupandthenon-(binge)drinkinggroupat baseline.

3.3. Predictionofﬁrstdrinkandﬁrstbingedrinkingepisode

Table3showstheresultsofthesurvivalanalysiswithworking memoryandresponse inhibition aspredictors ofthe ﬁrst alco-holicdrinkandbingedrinkingepisode.Multigroupanalysisfavored themodelin whichwe constrainedsex differencestobeequal

Table3

Oddsratio’sandstandardizedregressioncoefﬁcientsforresponseinhibitionand workingmemorypredictingonsetofdrinkingandbingedrinking.

Onset(N=336) Binge(N=458)

Odds ˇ p Odds ˇ p

Responseinhibition 2.20 .80 .01 1.45 .37 .57

(6)

Fig.2. Cumulativeincidenceratesoftheageofonsetforthetotalsampleandtheonsetsampleseparately.

(BICconstrained onset=1338, BIC unconstrained onset=1347;

BICconstrained binge=1645, BICunconstrained binge=1696),

suggesting no signiﬁcantdifferences in thepredictive effect of

working memory and response inhibition on the initiating of

(binge)drinking.Bothexecutivefunctionswereanalyzedina

mul-tivariatemodeltoexaminetheuniqueinﬂuenceofeachfunction.

Bothworkingmemoryandresponseinhibitionpredictedthe

initi-atingoftheﬁrstalcoholicdrink.Weakerworkingmemory(ˇ=−.51,

OR=.15,p=.01)andpoorerresponseinhibition(ˇ=.80,OR=3.12,

p=.01,higherscoresindicatepoorerresponseinhibition)increased

thelikelihoodthatadolescentswouldinitiatedrinkingduringstudy

follow-up,indicating anearlier initiating ofﬁrstdrinkand ﬁrst

bingedrinkingepisodeforthoseadolescentswithrelativelyweaker

executivefunctions.

Workingmemory,butnotresponseinhibitionpredictedthe

ini-tiatingofaﬁrstbingedrinkingepisode.Lowerscoresonworking

memory(ˇ=−.08,OR=.33,p=.04)increasedthelikelihoodthat

adolescentswouldinitiatebingedrinkingduringstudyfollow-up.

4. Discussion

Thepresentstudywasoneoftheﬁrsttoexaminethepredictive

effectofexecutivefunctioningontheinitiatingoftheﬁrstalcoholic

drinkandbingedrinkingepisodeamongyoungadolescents,and

theﬁrsttoexamineadolescentswhowerealcohol-naïveat

base-line.Theresultsindicatedthatrelativelyweakworkingmemory

predictedboththeinitiatingoftheﬁrstalcoholicdrinkandtheﬁrst

bingedrinkingepisode,beyondtheeffectofresponseinhibition.In

addition,relativelypoorresponseinhibitionpredictedthe

initiat-ingofdrinkingbeyondtheeffectofworkingmemory,however,no

effectwasfoundforthepredictionofﬁrstbingedrinkingepisode.

ParticularlyadolescentsfromSEeducationrevealedrelativelypoor

workingmemoryfunctioningandresponse inhibition,a ﬁnding

supportedbypreviousstudies (Barkley, 1997;Pihlet al.,1990;

Niggetal.,2006).Itseemslikelythattheapoorerlevelof exec-utivefunctioning,observedintheseadolescents,placethematrisk foranearlyinitiatingof(binge)drinkingwhicheventuallymight increasetheriskforproblematicalcoholuse,laterinadolescence (KingandChassin,2007).

The findings of this study are in agreement with those of Khuranaand colleagues(2012)whofoundthat executive func-tioning,andmorespecificallyworkingmemory,predictedincrease in frequency of drinkingamong young adolescents.Fernie and colleagues(2013)foundsimilarresults, revealing thatresponse inhibition(andtwoothermeasuresofimpulsivity)predicted alco-holusesixmonthslater.Niggand colleagues(2006)foundthat executivefunctioningpredictedproblemdrinkingbehaviorinhigh risk adolescents (adolescents from alcoholic families). In their study,poorresponseinhibitionatage12–14predictedtheonsetof alcoholrelatedproblemsatage15–17.However,unlikeNiggetal. (2006),wedidnotfindapredictiveeffectofresponseinhibition onbingedrinking.Adifferentsample,adifferenttask,adifferent outcomevariableanddifferentanalyzingtechniqueswereusedin thecurrentstudy,whichmightexplainthedivergentresults.Itis possiblethattheStrooptask,usedinthisstudytoassessresponse inhibition, wasless sensitiveindetectingdifferences compared

(7)

totheworkingmemorytask,perhapsexplainingwhywedidnot ﬁndauniqueeffectofresponseinhibitionontheinitiatingofthe ﬁrstbingedrinkingepisode.Inaddition,resultsindicatethat deci-sionmakingisimpairedafteralcoholconsumption.Goodworking memoryfunctioning allowsbetterdecisionmaking.It therefore mightbepossiblethatpoorerworkingmemoryfunctioningimpairs decisionmakingresultinginmoredrinking/bingedrinking(George etal.,2005).

Khuranaand colleagues (2012) found that impulsivity fully mediatedtheeffectofworkingmemoryonthedrinkingbehaviorof adolescents,suggestingthatweaknessesinworkingmemoryaffect behaviorthroughdisinhibitionandimpulsivity.Contrarytowhat Khuranaandcolleagues(2012)found,theresultsofthepresent studyindicatedthatatleasttwoexecutivefunctionsuniquely con-tributedtothepredictionoftheinitiatingofdrinkingbehavior.In addition,inthepresentstudy,workingmemorywasaunique pre-dictorofbingedrinkinginadolescentswhiledisinhibitionwasnot. Thelatterﬁndingisinagreementwithseveralstudiesthathave foundassociationsbetweenbingedrinkingandworkingmemory functioning(StephensandDuka,2008a;Cregoetal.,2009;Squeglia etal.,2011).However,thesecross-sectionalstudies(Cregoetal., 2009;Squegliaetal.,2011)donotshedlightonthedirectionofthe relation.Althoughtheresultsofthepresentstudydonotruleout areverseeffectofalcoholuseoncognitivefunctioning(e.g., bidi-rectionaleffectcf.Peetersetal.,2014),theresultsdorevealthat someweaknessesinexecutivefunctioningprecedetheinitiating ofdrinkingbehavior.Thecontinueduseofalcoholmayaggravate thepre-existingdeﬁcits(Khuranaetal.,2012;Peetersetal.,2014), causingdelayindevelopmentorevendeclineintheperformance onexecutivefunctioningtasks.

4.1. Limitations

Besidesthestrengthsofthestudy,suchasthelargesample sizeofyoungadolescentsandtheinclusionof ahigh-risk sam-ple,somelimitations shouldbeconsidered.Althoughtheexact samemeasureswiththesameinstructionswereusedinthetwo differentsamples(mainstreamandSEeducation),theassessment procedurewasslightlydifferent.Adolescentsinthehigh-risk sam-plecompletedtasksunder theguidanceof a researchassistant whileadolescentsin themainstreamsample completed assess-mentathome(i.e.,onlineassessment).Forthatreasonweadjusted for clustering effects due to differences in assessment setting. Furthermore,itis possiblethattheexclusion ofdrinkersatthe studybaselinelimitstherepresentativenessofthecurrent sam-plecomparedtonationalaverages.Though,adolescentswhowere excluded fromthe analysis revealed significantworse working memoryfunctioningandpoorerresponseinhibition,whichisin linewithourfindings(anearlyinitiatingofalcoholuseispredicted byrelativelyweakexecutivefunctioning).Thegeneralizabilityof theseresultsis restricteddue tothespecificcharacteristicsthe specialeducationsample(e.g.,externalizingbehavioralproblems). Unfortunately,we couldnot controlfor behavioral disorders in ouranalysisbecauseinformationaboutDSMIVbehavioral disor-derswasnotcollected(onlygeneralandlimitedinformationon externalizingbehaviorwasavailablebymeansoftheSDQ). Nev-ertheless,intheDutcheducationalsetting,approximately5%of thestudentsin secondaryeducationattenda schoolfor special education.Therefore,thesestudentsreflectanimportantpartof theDutchstudentpopulation.Furthermore,multipleimputation hasbeencriticizedasmethodofhandlingmissingdatabecauseit couldaddnoisetothedata(Rubin,1996).Thiscriticismappears toloseitsvaluebecauserecentsophisticatedimputationmethods areabletohandlemissingdatabettercomparedtoolder meth-ods,suchaslistwisedeletion(Rubin,1996;Sterneetal.,2009).In addition,thesimulationstudyinthepresentstudysupportedthis

critique.Moreover,inthepresentstudybingedrinkingwasdefined asdrinkingfiveormoreglassesonaday.Generally,bingedrinking referstodrinkingfiveormoreglassesatoneoccasion(Wechsler andNelson,2001).However,itseemsunlikelythatadolescentsof thisage(12–16years)regularlyhavetheopportunitytodrinkin theafternoonaswellasintheeveningonweekenddays. More-over,bingedrinkersconsumedonaveragealmostnineglasseson abingedrinkingday,whichcanbeinterpretedasheavydrinking forsuchyoungadolescents.Nevertheless,itispossiblethatsome oftheadolescentbingedrinkersinthisstudydidnotconsumeall fiveglassesinaroworatoneoccasion.Arelatedpossiblelimitation ofthestudyisthatalcoholusewasassessedbasedonthe partici-pants’ownindication,withnoexternalcorroborationfromparents orfriends.Thispotentiallyallowedparticipantstooverreportor underreporttheiralcoholuse.Nevertheless,astudybyKoningand colleagues(2010)revealedthatself-reportofalcoholuseisa reli-ablemethodtoassessalcoholconsumptionofadolescents.Lastly, recentstudiessuggestthatrepeatedcyclesofbingedrinkingare associatedwithexecutivedysfunction,andpoorworkingmemory performanceinparticular(Dukaetal.,2004;StephensandDuka, 2008b).Sinceweonlyexaminedthefirstepisodeofbinge drink-ingwewerenotabletolookatthisrelation.Futureresearchcould includenumberofbingeepisodesasimportantcovariateinrelation toexecutivefunctioning.

4.2. Implicationsandconclusion

Thefindingsofthecurrentstudyhaveimplicationsfortheory andforprevention.First,tothebestofourknowledgethisstudyis oneofthefirsttodemonstratethatdeficitsinexecutive function-ingprecedetheinitiatingofdrinkingamongadolescents.Previous studieshave foundstrongindications for executivefunctioning beingariskfactor foralcoholuseamongadolescents,however, thedesignofthesestudiespreventedtoderivestronginferences aboutdeficitsinexecutivefunctioningprecedingtheactualonsetof drinkingbehavior(seePeetersetal.,2014,foranoverview).Second, severalclinicalstudieshavefoundcognitivedeficitsafterheavyand prolongedalcoholuse(Hansonetal.,2011;Tapertetal.,2002).It ispossiblethatsomeofthesefounddeficitswerealreadypresent beforedrinkingbehaviorwasinitiated,andamagnifiedandbiased effectofalcoholonadolescents’cognitivefunctioningmayarise. Itshouldbenoted,however,thatthesestudiesexcluded adoles-centswithanexternalizingbehavioraldisorder(conductdisorder, ADHD),whilethecurrentstudyincludedadolescentswith exter-nalizingbehavioralproblems.Inaddition,deficitshavebeenfound forawiderangeofcognitivefunctions(e.g.,attention,visuospatial functioning),andalcoholmightaffectspecificpartsofthe adoles-centbraindifferently.Furtherresearchisneededtospecifywhich cognitivefunctionsaresusceptibletoheavydrinkingandwhich functionsaremainlyofinterestinpredictingthedrinkingonsetof adolescents.

Withrespecttoprevention,theﬁndingsofthepresentstudy conﬁrmtheimportanceofearlyinterventions.Ithasbeen demon-strated thatexecutive functionsmoderate theeffectof implicit cognitivemotivationalprocessesonalcoholandsubstanceusein adolescents(Grenardetal.,2008;Peetersetal.,2012;Thushetal., 2008).Astheseprocesses getstronger withincreaseddrinking, delayoftheonsetofdrinkingappearstobeagoodgeneralstrategy topreventproblematicalcoholuseinadolescents(Koningetal., 2009,2011).Assomeadolescentsappeartobeatagreaterriskfor anearlyonsetofdrinking,additionaltargetedpreventionmight bewarrantedas well(cf.Conrodet al.,2008).Recently several promisinginterventionshavebeenintroducedtoincreaseresponse inhibition(Houbenetal.,2011)andworkingmemory(Klingberg etal.,2005;Dovisetal.,2012)inadolescentsandyoungadults.

(8)

Thesetrainingmethodscontributetoanincreaseinexecutive con-trol,andtheymighteventuallypostponetheageoftheonset.

Thefindingsofthisstudydemonstratethatpre-existingdeficits inexecutivefunctioningpredicttheinitiatingofthefirstalcoholic drinkandfirstbingedrinkingepisodeamongyoungadolescents. The results suggest that relatively weakworking memory and disinhibitionarerisk factorsfor theearlyinitiating of drinking. Moreover,weakworkingmemoryfunctioningisariskfactorfor bingedrinking.Acontinuedheavydrinkingpatternmightfurther worsenexecutivefunctioning(Squegliaetal.,2009;Tapertetal., 2002), beyondthese pre-existingeffects. Futureresearch could benefitfromdisentanglingcognitivedeficitsinducedbyalcoholuse fromcognitivedeficitsthatpredictalcoholuse,withthecurrent findingsprovidingafirstmodeststep.

Conﬂictofinterest

Theauthorsdeclarenoconﬂictofinterest.

Acknowledgements

ThisstudywassupportedbytheDutchHealthCareResearch Organization(ZON-Mw;60-600-97-172)andaVicigrant 453-08-001fromtheDutchNationalScienceFoundation(N.W.O.)awarded toprof.Dr.ReinoutW.Wiers.

References

Baddeley,A.,1983.Workingmemory.Philos.Trans.R.Soc.B302,311–324.

Barkley,R.A.,1997.Behavioralinhibition,sustainedattention,andexecutive func-tions:constructingaunifyingtheoryofADHD.Psychol.Bull.121,65–94.

Blakemore,S.J.,Choudhury,S.,2006.Developmentoftheadolescentbrain: implica-tionsforexecutivefunctionandsocialcognition.J.ChildPsychol.Psychiatry47, 296–312.

Bunge,S.A.,Wright,S.B.,2007.Neurodevelopmentalchangesinworkingmemory andcognitivecontrol.Curr.Opin.Neurobiol.17,243–250.

Casey,B.,Jones,R.M.,Hare,T.A.,2008.Theadolescentbrain.Ann.N.Y.Acad.Sci. 1124,111–126.

Conrod,P.J.,Castellanos,N.,Mackie,C.,2008.Personality-targetedinterventions delaythegrowthofadolescentdrinkingandbingedrinking.J.ChildPsychol. Psychiatry49,181–190.

Cragg,L.,Nation,K.,2007.Self-orderedpointingasatestofworkingmemoryin typicallydevelopingchildren.Memory15,526–535.

Crego,A.,Holguín,S.R.,Parada,M.,Mota,N.,Corral,M.,Cadaveira,F.,2009.Binge drinkingaffectsattentionalandvisualworkingmemoryprocessinginyoung universitystudents.Alcohol.Clin.Exp.Res.33,1870–1879.

Crone, E.A., Dahl, R.E., 2012. Understanding adolescence as a period of social–affective engagement and goal ﬂexibility. Nat. Rev. Neurosci. 13, 636–650.

Duka,T.,Gentry,J.,Malcolm,R.,2004.Consequencesofmultiplewithdrawalsfrom alcohol.Alcohol.Clin.Exp.Res.280,233–246.

Dovis, S., Van der Oord, S., Wiers, R.W., Prins, P.J., 2012. Can motivation normalizeworkingmemoryandtaskpersistenceinchildrenwith attention-deﬁcit/hyperactivitydisorder?Theeffectsofmoneyandcomputer-gaming.J. Abnorm.ChildPsychol.40,669–681.

Fernie,G.,Peeters,M.,Gullo,M.J.,Christiansen,P.,Cole,J.C.,Sumnall,H.,Field, M.,2013.Multiplebehaviouralimpulsivitytaskspredictprospectivealcohol involvementinadolescents.Addiction108,1916–1923.

George,S.,Rogers,R.D.,Duka,T.,2005.Theacuteeffectofalcoholondecisionmaking insocialdrinkers.Psychopharmacology(Berl)182,160–169.

Goodman,R.,1997.TheStrengthsandDifﬁcultiesQuestionnaire:aresearchnote.J. ChildPsychol.Psychiatry38,581–586.

Grenard,J.L.,Ames,S.L.,Wiers,R.W.,Thush,C.,Sussman,S.,Stacy,A.W.,2008.

Workingmemorycapacitymoderatesthepredictiveeffectsofdrug-related associationsonsubstanceuse.Psychol.Addict.Behav.22,426–432.

Hanson,K.L.,Medina,K.L.,Padula,C.B.,Tapert,S.F.,Brown,S.A.,2011.Impactof adolescentalcoholanddruguseonneuropsychologicalfunctioninginyoung adulthood:10-yearoutcomes.J.ChildAdolesc.Subst.Abuse20,135–154.

Hermens,D.F.,Lagopoulos,J.,Tobias-Webb,J.,DeRegt,T.,Dore,G.,Juckes,L.,Latt, N.,Hickie,I.B.,2013.Pathwaystoalcohol-inducedbrainimpairmentinyoung people:areview.Cortex49,3–17.

Houben,K.,Wiers,R.W.,Jansen,A.,2011.GettingaGriponDrinkingBehavior: trainingworkingmemorytoreducealcoholabuse.Psychol.Sci.22,968–975.

Houben,K.,Wiers,R.W.,2009.Responseinhibitionmoderatestherelationship betweenimplicitassociationsanddrinkingbehavior.Alcohol.Clin.Exp.Res. 33,626–633.

Kepper,A.,Monshouwer,K.,VanDorsselaer,S.,Vollebergh,W.,2011.Substanceuse byadolescentsinspecialeducationandresidentialyouthcareinstitutions.Eur. Child.Adoles.Psych.20,311–319.

Khurana,A.,Romer,D.,Betancourt,L.M.,Brodsky,N.L.,Giannetta,J.M.,Hurt,H.,2012.

Workingmemoryabilitypredictstrajectoriesofearlyalcoholuseinadolescents: themediationalroleofimpulsivity.Addiction108,506–515.

King,K.M.,Chassin,L.,2007.Aprospectivestudyoftheeffectsofageofinitiation ofalcoholanddruguseonyoungadultsubstancedependence.J.Stud.Alcohol Drugs68(2),256–265.

Klingberg,T.,Fernell,E.,Olesen,P.J.,Johnson,M.,Gustafsson,P.,Dahlström,K., Gill-berg,C.G.,Forssberg,H.,Westerberg,H.,2005.Computerizedtrainingofworking memoryinchildrenwithADHD–arandomized,controlledtrial.J.Am.Acad. ChildPsychiatry44,177–186.

Koning,I.M.,Harakeh,Z.,Engels,R.C.M.E.,Vollebergh,W.A.M.,2010.Acomparisonof self-reportedalcoholusemeasuresbyearlyadolescents:questionnairesversus diary.J.Subst.Use15(3),166–173.

Koning,I.M.,VandenEijnden,R.J.,Verdurmen,J.E.,Engels,R.C.,Vollebergh,W.A., 2011.Long-termeffectsofaparentandstudentinterventiononalcoholusein adolescents:aclusterrandomizedcontrolledtrial.Am.J.Prev.Med.40,541–547.

Koning,I.M.,Vollebergh,W.A.M.,Smit,F.,Verdurmen,J.E.E.,VanDenEijnden,R.J.J.M., TerBogt,T.F.M.,Stattin,H.,Engels,R.C.M.E.,2009.Preventingheavyalcoholuse inadolescents(PAS):clusterrandomizedtrialofaparentandstudent interven-tionofferedseparatelyandsimultaneously.Addiction104,1669–1678.

Luna, B.,Garver, K.E.,Urban,T.A., Lazar,N.A.,Sweeney, J.A.,2004. Maturation of cognitive processes from late childhood to adulthood. Child Dev. 75, 1357–1372.

Miller,J.W.,Naimi,T.S.,Brewer,R.D.,Jones,S.E.,2007.Bingedrinkingandassociated healthriskbehaviorsamonghighschoolstudents.Pediatrics119,76–85.

Miyake,A.,Friedman,N.P.,Emerson,M.J.,Witzki,A.H.,Howerter,A.,Wager,T.D., 2000.Theunityanddiversityofexecutivefunctionsandtheircontributions tocomplex“frontallobe”tasks:alatentvariableanalysis.Cogn.Psychol.41, 49–100.

Muthén,L.K.,Muthén,B.O.,1998–2010.MplusUser’sGuide,sixthed.Muthén& Muthén,LosAngeles,CA.

Nigg,J.T.,Wong,M.M.,Martel,M.M.,Jester,J.M.,Puttler,L.I.,Glass,J.M.,Adams, K.M.,Fitzgerald,H.E.,Zucker,R.A.,2006.Poorresponseinhibitionasa predic-torofproblemdrinkingandillicitdruguseinadolescentsatriskforalcoholism andothersubstanceusedisorders.J.Am.Acad.ChildAdolesc.Psychiatry45, 468–475.

Oswald,D.P.,Best,A.M.,Coutinho,M.J.,Nagle,H.A.L.,2003.Trendsinthespecial educationidentiﬁcationratesofboysandgirls:acallforresearchandchange. Exceptionality11,223–237.

Pihl,R.O.,Peterson,J.,Finn,P.R.,1990.Inheritedpredispositiontoalcoholism: char-acteristicsofsonsofmalealcoholics.J.Abnorm.Psychol.99,291–301.

Peeters,M.,Wiers,R.,Monshouwer,K.,vandeSchoot,R.A.J.G.,Janssen,T.,Vollebergh, W.,2012. Automaticprocessesinat-riskadolescents: theroleof alcohol-approachtendenciesandresponseinhibitionindrinkingbehavior.Addiction 107,1339–1346.

Peeters,M.,Monshouwer,K.,Schoot,R.A.,Janssen,T.,Vollebergh,W.A.M.,Wiers, R.W.,2013.Automaticprocessesandthedrinkingbehaviorinearlyadolescence: aprospectivestudy.Alcohol.Clin.Exp.Res.37,1737–1744.

Peeters,M.,Vollebergh,W.A.M.,Wiers,R.W.,Field,M.,2014.Psychologicalchanges andcognitiveimpairmentsinadolescentheavydrinkers.AlcoholAlcohol.49, 182–186.

Petrides,M.,Milner,B.,1982.Deﬁcitsonsubject-orderedtasksafterfrontal-and temporal-lobelesionsinman.Neuropsychologia20,249–262.

Rubin,D.B.,1996.Multipleimputationafter18years.JAmStatAssoc91,473–489.

Stephens,D.N.,Duka,T.,2008a.Cognitiveandemotionalconsequencesofbinge drinking:roleofamygdalaandprefrontalcortex.Philos.Trans.R.Soc.363, 3169–3179.

Squeglia, L.M., Spadoni, A.D., Infante, M.A., Myers, M.G., Tapert, S.F., 2009.

Initiatingmoderatetoheavyalcoholusepredictschangesin neuropsychol-ogicalfunctioningforadolescentgirlsandboys.PsycholAddict.Behav.23, 715–722.

Squeglia,L.M.,Schweinsburg,A.D.,Pulido,C.,Tapert,S.F.,2011.Adolescentbinge drinkinglinkedtoabnormalspatialworkingmemorybrainactivation: differen-tialgendereffects.Alcohol.Clin.Exp.Res.35,1831–1841.

Steinberg,L.,2007.Risktakinginadolescencenewperspectivesfrombrainand behavioralscience.Curr.DirectPsychol.Sci.16,55–59.

Stephens,D.N.,Duka,T.,2008b.Cognitiveandemotionalconsequencesofbinge drinking:roleofamygdalaandprefrontalcortex.Philos.Trans.R.Soc.B363, 3169–3179.

Sterne,J.A.,White,I.R.,Carlin,J.B.,Spratt,M.,Royston,P.,Kenward,M.G.,Wood,A.M., Carpenter,J.R.,2009.Multipleimputationformissingdatainepidemiological andclinicalresearch:potentialandpitfalls.Br.Med.J.338,b2393.

Stroop,J.R.,1935.Studiesofinterferenceinverbalreactions.J.Exp.Psychol.18, 643–662.

Tapert, S.F., Granholm, E., Leedy, N.G., Brown,S.A., 2002. Substance useand withdrawal:neuropsychologicalfunctioningover8yearsinyouth.J.Int. Neu-ropsychol.Soc.8,873–883.

Thush,C.,Wiers,R.W.,Ames,S.L.,Grenard,J.L.,Sussman,S.,Stacy,A.W.,2008. Inter-actionsbetweenimplicitandexplicitcognitionandworkingmemorycapacity inthepredictionofalcoholuseinat-riskadolescents.DrugAlcoholDepend.94, 116–124.

vanBuuren,S.,Boshuizen,H.C.,Knook,D.L.,1999.Multipleimputationofmissing bloodpressurecovariatesinsurvivalanalysis.Stat.Med.18,681–694.

(9)

vanBuuren,S.,Oudshoorn,C.,2000.MultivariateImputationbyChainedEquations. MICEV1.0User’sManual.TNOPreventieenGezondheid,Leiden.

Verdejo-García,A.,Lawrence,A.J.,Clark,L.,2008.Impulsivityasavulnerability markerforsubstance-usedisorders:reviewofﬁndingsfromhigh-riskresearch, problemgamblersandgeneticassociationstudies.Neurosci.Biobehav.Rev.32, 777–810.

Wechsler,H.,Nelson,T.F.,2001.BingedrinkingandtheAmericancollegestudents: What’sﬁvedrinks?Psychol.Addict.Beh.15,287–291.

Wiers,R.W.,Bartholow,B.D.,vandenWildenberg,E.,Thush,C.,Engels,R.C.,Sher,K.J., Grenard,J.,Ames,S.L.,Stacy,A.W.,2007.Automaticandcontrolledprocessesand thedevelopmentofaddictivebehaviorsinadolescents:areviewandamodel. Pharmacol.Biochem.Behav.86,263–283.

Zanotti,C.,Morgan,A.,Currie,D.,Looze,M.D.,Roberts,C.,Samdal,O.,Smith,O.R.F., Barnekow,V.,2012.SocialDeterminantsofHealthandWell-BeingAmongYoung People.WorldHealthOrganizationRegionalOfﬁceforEurope.