ContentslistsavailableatScienceDirect
Developmental
Cognitive
Neuroscience
j ou rn a l h om ep a ge :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
Normalization
of
EEG
activity
among
previously
institutionalized
children
placed
into
foster
care:
A
12-year
follow-up
of
the
Bucharest
Early
Intervention
Project
Ross
E.
Vanderwert
a,∗,
Charles
H.
Zeanah
b,
Nathan
A.
Fox
c,
Charles
A.
Nelson
III
d,e,faSchoolofPsychology,CardiffUniversity,70ParkPlace,TowerBuilding,CardiffCF103AT,UnitedKingdom bTulaneUniversity,NewOrleans,LA,UnitedStates
cUniversityofMaryland,CollegePark,MD,UnitedStates
dBostonChildren’sHospitalandHarvardMedicalSchool,Boston,MA,UnitedStates eHarvardGraduateSchoolofEducation,Cambridge,MA,UnitedStates
fHarvardCenterontheDevelopingChild,Cambridge,MA,UnitedStates
a
r
t
i
c
l
e
i
n
f
o
Articlehistory:Received30July2015
Receivedinrevisedform1December2015 Accepted9December2015
Availableonline13December2015
Keywords:
Earlydeprivation Fostercareintervention EEG
BEIP Alphaactivity Betaactivity
a
b
s
t
r
a
c
t
Extremesocialandcognitivedeprivationasaresultofinstitutionalcarehasprofoundeffectson devel-opmentaloutcomesacrossmultipledomainsformanyabandonedororphanedchildren.TheBucharest EarlyInterventionProject(BEIP)examinestheoutcomesforchildrenoriginallyplacedininstitutions whowereassessedcomprehensivelyandthenrandomizedtofostercare(FCG)orcareasusual(CAUG) andfollowedlongitudinally.Herewereportonthebrainelectricalactivity(electroencephalogram:EEG) of12-year-oldchildrenenrolledintheBEIP.PreviousreportssuggestedimprovementinrestingEEG activityforthegroupofchildrenplacedinthefostercareintervention,particularlythoseplacedbefore 24monthsofagecomparedtochildrenwhowererandomizedtoCAUGorthoseplacedintofamilies afterthisage.At12years,differencesbetweenthoseintheFCGandthoseintheCAUGpersistinthe alphaband(8–13Hz),butnotinhigherfrequencybands(i.e.inthebetaband;15–30Hz),exceptinthose childrenplacedintotheFCGwhoremainedinhighqualitycareenvironmentsoverthecourseofthe study.Thesefindingshighlighttheimportanceofmaintainingastablehighqualitycaregiving environ-ment,particularlyforchildrenexposedtoearlypsychosocialdeprivation,forpromotinghealthybrain development.
©2015TheAuthors.PublishedbyElsevierLtd.ThisisanopenaccessarticleundertheCCBYlicense (http://creativecommons.org/licenses/by/4.0/).
1. Introduction
Theuseofinstitutionalcareforabandonedororphanedinfants and children has remaineda commonpractice throughout the world(Browneetal.,2006;UNICEF,2015);however,institutional environmentsareharmfulforhealthyphysicalandpsychological developmentespeciallyforyoungchildren.Manyinstitutionshave highchild-to-caregiverratios,havehighlyregimentedschedules withextendedperiodswherechildrenareleftalone,andvery lit-tleengagementwiththecaregivers (Castle etal.,1999;Nelson,
2007;Smykeetal.,2007).Theconsequencesoftheserearing
con-ditionsoftenincludephysicalgrowthrestriction,awiderangeof
∗Correspondingauthor.Tel.:+4402920870467.
E-mailaddress:[email protected](R.E.Vanderwert).
behavioralproblems,anddeficitsincognitivefunctioncompared tochildrenraisedin families(Maclean,2003;Nelson, 2007; for review,seeNelsonetal.,2014).
Extremedeprivationdramaticallyaltersneuralarchitectureand functioning.A number of neuroimagingstudies have examined the impact of those early experiences in post-institutionalized adoptees. For example, structural magnetic resonance imaging (MRI)studieshaveshowndecreasedgreyandwhitematter
vol-umes(Mehtaetal.,2009;Sheridanetal.,2012),increasedamygdala
volume(Mehtaet al.,2009; Tottenhamet al.,2010), decreased cerebellarvolumes(Baueretal.,2009),anddisrupted connectiv-itybetweenthefrontaland temporal lobes(Eluvathingaletal., 2006)inpreviouslyinstitutionalizedchildrencomparedto com-munitycontrols.Togetherthesestudiesdemonstratethattypical neuraldevelopmentisalteredasaresultofearlydeprivation. How-ever,fewerstudieshaveexaminedtheimpactofinterventionand trackedchangesinbraindevelopmentasaresultofremovalfrom institutions.
http://dx.doi.org/10.1016/j.dcn.2015.12.004
TheBEIPisthefirststudytoexperimentallyexaminethe physi-cal,psychological,andneuralsequelaeofinstitutionalrearingand thedevelopmental trajectoriesof children removed and placed intoanovelfostercareintervention(Smykeetal.,2009;Zeanah etal.,2003).Agroupofinfantslivingininstitutionsaswellasan age-matchedcommunitycontrolsample(neverinstitutionalized group;NIG)alllivinginBucharest,Romaniawererecruitedinto thestudy.Afteraninitialscreeningtoexcludeforgenetic disor-dersorothermedicalconditionstheinfantscompletedabaseline assessment, including resting EEG. The infantswere then ran-domlyassignedtoeitherremainintheinstitution(careasusual; CAUG)orplacedwithafosterfamily(FCG).Randomassignment ofthesechildrenprovidesanopportunitytoexaminetheeffects oftheinterventionandrepeatedassessmentsoverthechildren’s lifehaveallowedexaminationofchangesin development asso-ciatedwithenrichmentoftheirearlycareenvironment.Greater detailregardingthestudydesignandethicalissuesaredescribed elsewhere(Miller,2009;MillumandEmanuel,2007;Zeanahetal.,
2012).
Previousreportsfromthisstudyfoundatthebaseline assess-ment, infantslivingin institutionsshowed markedly decreased powerinbothalphaandbetaactivityandincreasedthetapower compared to the never institutionalized community controls
(Marshall, Fox and the BEIP core group, 2004).The pattern of
higherthetaandloweralphapowerisonethatisassociatedwith attention-deficit/hyperactivitydisorder(ADHD)andotherlearning disorders(Barryetal.,2003,2009;Chabotetal.,2005;McLaughlin etal.,2010),theformerbeinghighlyprevalentamongpreviously institutionalizedchildren(Bosetal.,2011;Kreppneretal.,2001;
Wiiketal.,2011;Zeanahetal.,2009).
Followingplacementintofostercare,EEGwasagaincollected whenthechildrenwere42months.Themeanageofplacementfor thefostercaregroupwas22months.Atthe42monthassessment therewasahintthatearlierplacementintofostercarewas ben-eficial,asinfantsplacedatyoungeragesshowedincreasingalpha powerrelativetoolderplacedinfants(Marshalletal.,2008).The effectoftimingofplacementinfostercarebecameclearbythetime thechildrenwere8-years-old;infantsplacedintofostercarebefore 24monthswereindistinguishablefromchildreninthecommunity whilethoseplaced after24 monthswereidenticaltotheCAUG
(Vanderwertetal.,2010).Further,ageofremovalfrominstitutions
alsosignificantlyimpactedthedevelopmentaltrajectoriesofalpha andbetapowerbetween42-monthsand8-years(Stamoulisetal., 2015)suggestingongoingplasticityinbetarhythmthatwasnot detectedinthefollow-upat8years.
Theaimofthisstudyistoexaminethecontinuingeffectsof fos-tercareinterventionontheneuralactivityofchildrenremoved frominstitutions in infancy.For the current study,restingEEG activitywasacquiredwhenchildrenintheBEIPstudywere12years ofage.Ourprimaryhypothesiswasthattheinterventioneffects observedinthealphabandwhenthechildrenwere8yearsold wouldpersisttothecurrentassessmentandthattheintervention wouldalsoimprovepowerinthebetabandintheFCGcompared
totheCAUG.Wealsowantedtoexaminewhethertimingeffects oftheintervention(i.e.,ageatplacementintofostercare)would affectthepatternofalphaandbetaactivity.
Weconductedtwoadditionalexploratoryanalyses.First, fol-lowinguparecentexaminationofthedevelopmentaltrajectoriesof theEEGfromchildrenintheBEIPstudy(Stamoulisetal.,2015),we alsowantedtoexaminecontinuingchangesinthethreefrequency bandsbetweenages8-and12-years.Second,overthedurationof theBEIPstudysomeofthechildrenintheFCGhaveexperienced disruptionstotheircaregivingenvironment.Recently,Humphreys
etal.(2015)foundthatchildreninstableplacementinfostercare
hadfewerinternalizingandexternalizingsymptomscomparedto thosewithunstableplacements.Therefore,wealsoexaminedthe effectsofstabilityofplacementontheEEG.
2. Methods
2.1. Sample
Atthe12-yearassessment,EEGwascollectedon50(26male) childrenfromtheFCG,49(27male)childreninCAUG,and48(22 male)childrenfromtheNIG.Thethreegroupsdidnotdifferon gender,2(2,N=147)=.863, p=.65;oronageatdatacollection (M=12.60,SD=.54;F(2,146)=.324,p=.724).Anumberofchildren wereexcludedfromanalysesduetoanexcessivenumberofbad channelsortoofewartifactfreeepochs(>6channels;N=3CAUG, 3FCG, 2NIG);thetasknotbeingcompleteddue toequipment failureorsubjectnon-compliance(N=1CAUG,1FCG);medication use(N=7CAUG,5FCG);orwithEEGdataexceeding±2SDfrom thegroupmeanineitherabsoluteorrelativepoweratmultiple electrodeclustersacrossthescalp,inmultiplefrequencybands,or meetingboththesecriteria(N=5CAUG,3FCG,5NIG).Ofthe origi-nalsample,19CAUG,17FCG,and40NIGchildrenwereunavailable fortesting(Table1).
The University Institutional Review Boards of the principal investigators(Fox,Nelson,andZeanah)andoftheUniversityof Bucharestapprovedthestudyprotocol.Romanianlawdictatesthat consentbegivenbythelegalguardianofeachchild,thereforein thecaseswhereconsentwasunavailable,assentwasobtainedfrom eachcaregiverwhoaccompaniedachildtothevisitandfromeach child.
2.2. EEGrecording&analysis
ContinuousEEG was recorded using a 64-channel HydroCel GeodesicSensorNet(ElectricalGeodesicInc.,Eugene,OR)thatwas connectedtoa NetAmps300amplifier(Electrical GeodesicInc., Eugene,OR)andreferencedonlinetoasinglevertexelectrode(Cz). Channelimpedanceswerekeptatorbelow50kandsignalswere sampledat500Hz.Thechild’sheadcircumferencewasmeasured andanappropriatelysizednetwasfitted.EEGwasrecordedwhile thechildrensatquietlyinachair,alternating1-minepochsofeyes openand eyesclosedfora totalof6min.Duringtheeyesopen
Table1
Continuityofparticipantsfromthe8-yearto12-yearwaveofassessments.
Assessedat8yr&12yr Assessedat8yrnot12yr Assessedat12yrnot8yr Missingat8yr&12yr Mean(SD)
Group Placementage
CAUG 42 6 7 13 –
FCG 45 8 5 9 22.62(7.01)
Stable 21 0 3 2 21.50(7.15)
Disrupted 24 2 1 1 23.81(6.95)
NIG 29 13 19 27 –
Note.Therearenodifferencesbetweengroupsinthenumberofchildrenassessedormissingat8-yearor12-yearwaves.Therearenodifferencesinageofplacementinto theinterventionbetweentheFCG-StableandFCG-Disrupted.
Fig.1. Exampleofthe64-channelEEGnetwithclustershighlightedingrey.Frontalclustersarecenteredonchannels12and60;centralon20and50;parietalon28and 42;andoccipitalonchannels35and39.
condition,childrenwereinstructedtofixateonasmallwhitecross onthecenterofacomputerscreen.EEGdatawerepreprocessed offlineusingNetStation4.5(ElectricalGeodesicInc.,Eugene,OR). TheEEGdatawerefirstbandpassfiltered(.3to100Hz),bad chan-nelswerethenidentifiedandreplaced,andanaveragereference wascomputed.TheEEGsignalswerethenexportedtoMatlabfor artifactrejection and poweranalysis.Blinks and otherartifacts wereautomaticallyidentified withathreshold of±100V and excludedfromanalyses.Thecleandatawerethen submittedto aFast Fouriertransform(FFT) witha 1-sHanningwindow and 50%overlap.TheCAUG(M=269.12,SD=46.75),FCG(M=259.39, SD=64.80),andNIG(M=282.44,SD=48.17)contributedanequal numberofwindowsforanalysis(F(2,111)=1.81,p=.17).Spectral power(V2)intheeyesopenconditionwascomputedforthe fol-lowingbands:theta(4–7Hz),alpha(8–13Hz),andbeta(15–30Hz). Theeyes open conditionwaschosen for consistency with pre-viousassessments(Marshall etal., 2004; Marshallet al., 2008;
Vanderwertetal.,2010).Thepowerineachfrequencybandwas
thenaveragedoverclustersofchannelsapproximatingthe inter-national10–20locationsF3,F4,C3,C4,P3,P4,O1,andO2(Fig.1).
Foreachclusterandband,weexaminedbothabsolutepower (thenaturallogarithmofthespectralpower)andrelativepower (theproportionofpowerineachbandrelativetothetotalpower between4and30Hz).Developmentally,thereisashiftfrom domi-nanceinlowerfrequencybands(e.g.greaterdeltaandthetapower) tohigherfrequencies(i.e.alphapower);relativepowerisanideal indextomeasurethesedevelopmentalchangeswhileminimizing inter-individualdifferencesinabsolutepowerduetomaturational factors,suchasskull thickness(Somsenetal.,1997).However, becauserelative power is a proportion score, changes in abso-lutepowerinonebandaffectsrelativepowervaluesintheother
bands,therefore,absolutepowerishelpfulinidentifyingthe spe-cificband(s)thatresultinchangesinrelativepower.Wepresent theresultsforbothabsoluteandrelativepowerfollowingthe pro-ceduresatpreviousassessments(Marshalletal.,2004;Marshall
etal.,2008;Vanderwertetal.,2010).
2.3. Analysisapproach
Inordertotestourhypothesisthattherewasanintervention effectontheEEG,ourfirstsetofanalysesemployanintent-to-treat approachbasedonthedesignoftheBEIPprojectasarandomized, controlledtrialoffostercareasaninterventionforinstitutionalized children.Intheinterveningyearssincetheoutsetofthestudy,a numberofculturalandpoliticalchangesinRomaniahaveresulted intheformationofanationalfostercareprogramwiththeaimof decreasingthenumberofinfantsandchildrenlivingininstitutions. Asaresultofthesepolicies,manyoftheCAUGandFCGchildren havehadchangesintheirlivingsituations;beingremovedfrom institutionsortheirstudyfosterfamiliesandplacedingovernment fostercarehomesorreunitedwiththeirbiologicalfamilies(see
Fig.2).
Intent-to-treatanalysistreatseachchild’sdataasiftheyhad remainedwithin theirrandomlyassigned groups. These effects are,therefore,aconservativeestimateoftheinterventioneffect, andweconsidered˛<.10ofinterestforfollow-upexamination. DifferencesinEEGpowerwereexaminedwithseparaterepeated measuresanalysesofvariance(RM-ANOVAs)foreachbandwith 4regions(frontal,central,parietal,occipital)and2hemispheres (left,right)aswithin-subjectsfactorsandplacementgroup(CAUG, FCG)asthebetweensubjectsfactor.Greenhouse–Geisser correc-tionwasusedforviolations ofsphericity.Independentsamples
Fig.2.Groupstatusat12-yearsforthecareasusual,fostercare,andnever-institutionalizedgroupsintheBEIPstudy.Threechildreninfostercareweremovedfromone placementtoanotherandwerethereforeplacedinthedisruptedfoster-caregroup.
t-testswereusedforanyposthoccomparisons.Notethat inter-actionsnotinvolvinggrouparenotreported.Forthosefrequency bandswhereaninterventioneffectwasfoundwethenexamined thescalpactivityoftheinterventiongroupsastheycomparedto thenever-institutionalizedchildren.
Wenextconsideredage-at-placementeffectsintheFCGonthe EEGmeasuresusingcorrelationanalysestoidentifyanyfrequency bandsthatweresensitivetotheagethechildwasremovedfrom theinstitution.
AsubsampleofchildrenprovidedEEGdatafrombothwhenthey were8-years-oldandatthecurrentassessment.EEGwascollected usingidenticalproceduresatbothassessments(Vanderwertetal., 2010),however,therewasachangeintheEEGacquisition sys-tems.Therefore,toexaminechangesin theEEGbetween8and 12years, weusedtherelativepowermetric,which better con-trolsforacquisitionsystem,analysissoftware,andfrequencyband differences.
A recent study foundthat children who remained in stable fostercarehad improvedmentalhealth outcomescomparedto thosewhohadmultipledisruptionsintheircare(Humphreysetal., 2015).Thesedisruptionsincaremayimpactneuraldevelopment; therefore,wealsoexaminedtheeffectsofplacementstabilityin eachofthethreefrequencybands.Justoverhalfofthechildren intheFCGremainedwiththeiroriginalfosterparentatage12 years (seeTable1).Therefore,we setasidetheoriginal intent-to-treatgroupingsandexaminedtheeffectsofthestabilityfoster careplacementsforchildrenintheFCGgrouponly,separatingthe FCGintothosechildrenwhoremainedintheiroriginalBEIPfoster placement(FCG-Stable)andthosewhoexperienceddisruptionsin theirplacements(FCG-Disrupted;Humphreysetal.,2015;Rutter,
2015).
3. Results
3.1. Intent-to-treatanalyses
Fortheintent-to-treatanalyses,eachbandwasanalyzed sep-arately with a 4 region (frontal, central, parietal, occipital)×2 hemisphere(left,right)×2group(CAUG,FCG)RM-ANOVA.
3.1.1. Absolutepower
Examination of absolutepower revealed no main effects or interactionsinvolvingtheintent-to-treatgroupingsinanyofthe threebands.
3.1.2. Relativepower
Thetaband.Examinationofinterventioneffectsinrelativetheta
powerrevealedamarginallysignificantmaineffectofgroup(F(1, 69)=3.32, p=.073,2
p=.063).Examinationof thegroup means
revealedthattheCAUG(M=.41,SD=.09)hadgreaterpower com-paredtotheFCG(M=.37,SD=.09).
To examine the intervention effect of relative theta power withinthecontextofthenever-institutionalizedgroupwe com-putedaveragepoweracrossthescalpandemployedanunivariate ANOVAwiththeaveragedrelativethetapowerasthedependent variableandthethreegroups(CAUG,FCG,andNIG)asthebetween subjectsvariable.Resultsoftheanalysisrevealedasignificantmain effect ofgroup (F(3,109)=741.96, p<.001, 2
p=.953).Post hoc
examinationofthemaineffectrevealedthattheCAUGhadgreater power compared totheFCG(t(69)=1.95, p=.055)and the NIG (M=.35,SD=.07;t(72)=2.60,p=.011).Therewasnodifferencein powerbetweentheFCGandNIG(t(77)=.63,n.s.).
Fig.3.Meanabsolute(top)andrelative(bottom)powerfromthetheta(4–7Hz), alpha(8–13Hz),andbeta(15–30Hz)bandsforthecare-as-usualgroup(CAUG), fos-tercareintervention(FCG),andnever-institutionalizedcommunitycontrols(NIG). *p<.05,†p<.10.
Alphaband. Examinationof relative alphapower revealed a
marginallysignificantmaineffectofgroup(F(1,69)=3.88,p=.053,
2
p=.053).ExaminationofthegroupmeansrevealedthattheCAUG
(M=.37,SD=.10)hadless powercompared totheFCG(M=.42, SD=.09).
To examine the intervention effect of relative alpha power withinthecontextofthenever-institutionalizedgroupwe com-putedaveragepoweracrossthescalpandemployedanunivariate ANOVAwiththeaveragedrelativealphapowerasthedependent variableandthethreegroups(CAUG,FCG,andNIG)asthebetween subjectsvariable.Resultsoftheanalysisrevealedasignificantmain effectof group(F(3,109)=789.04,p<.001, 2
p=.956). Posthoc
examinationofthemaineffectrevealedthattheCAUG hadless powercompared tothe FCG(t(69)=2.10, p=.040) and theNIG (M=.44,SD=.08;t(72)=3.05,p=.003).Therewasnodifferencein powerbetweentheFCGandNIG(t(77)=1.00,n.s.;Fig.3).
Betaband.Examinationofinterventioneffectsinrelativebeta
powerfoundnomaineffectsorsignificantinteractionswithgroup.
3.2. Age-at-placement
Toexaminepossiblesensitiveperiodsincorticaldevelopment, wecomputedcorrelationsbetweentheaverageabsoluteand rel-ativepowerineachfrequencybandandtheageatplacementfor theFCG.Nosignificantassociationsweredetected.
3.3. ContinuityofEEGpower
Continuityofrelativepowerineachbandwasexaminedusing Pearsonproduct-momentcorrelationswithineachgroupbetween relativepowerat8yearsandrelativepowerat12yearsineach band. Relative powerin each bandwas significantlycorrelated betweenthetwotimepointsinboththeCAUGandFCGgroupsin allthreebands.Relationsbetween8-yearand12-yearassessments fortheNIGwerelimitedtothealphaandbetabands(Table2). Toexaminewhetherthechangeinrelativepowerwasrelatedto ageatplacementintofostercare,wecalculatedthedifferencein
Table2
Correlationsbetweenrelativepowerat8yearand12yearassessments.
Group n Theta Alpha Beta
CAUG 29 .728*** .796*** .504**
FCG 35 .732*** .616*** .769***
Ageatplacement −.272 .333* −.182
NIG 24 .367 .562** .500*
Note.CorrelationsreportedforageatplacementintheFCGarebasedonthechange inrelativepowerfrom8to12years.
***p<.001. **p<.01.
*p=.051.
relativepower(12–8years)foreachbandandranadditional cor-relationswithAge-at-Placement.Thechangeinrelativepowerin thealphabandwastheonlysignificantcorrelationwiththeFCG Age-at-Placement(r(35)=.333,p=.051;Fig.4).
3.3.1. Groupdifferencesinchangesinrelativepower
Wenextanalyzedgroupdifferencesinrelativepowerchange between8-and12-years.Becausetherewerenogroupdifferences inregionorhemisphereateitherassessmenttimepoint,we aver-agedtherelativepoweracrossthescalptoexaminescalp-wide changesinbandpower.Weemployeda2Time(8-year,12-year)×3 Band(Theta,Alpha,Beta)×3Group(CAUG,FCG,NIG)RM-ANOVA. The analysis revealed significant main effects of Time (F(1, 85)=607.75, p<.001, 2
p=.88) and Band (F(2, 170)=208.84, p<.001, 2
p=.71) qualified by Band×Time (F(2, 170)=20.50, p<.001, 2
p=.194) and Band×Group (F(4, 170)=5.27, p=.001,
2
p=.110)interactions.PosthocanalysesoftheBandxTime
inter-actionrevealedasignificantdecreaseinpowerbetweenTime1 (M=.42,SD=.08)andTime2(M=.37,SD=.09;t(85)=7.04,p<.001) inthethetabandandasignificantincreasebetweenTime1(M=.17, SD=.05)andTime2(M=.19,SD=.05;t(85)=3.05,p=.003)inthe betaband.Therewerenosignificantchangesinthealphaband.
Posthocanalysesfor theBandxGroupinteractionrevealed that in thetheta band, the CAUG (M=.43, SD=.08) had higher relativepowercomparedtotheFCG(M=.39,SD=.08;t(62)=1.89,
p=.062)andtheNIG(M=.36,SD=.07;t(51)=3.30,p=.002)andno differencebetweentheFCGandNIG(t(57)=1.50,p=.14).Inthe alphaband, theCAUG(M=.37,SD=.08)had significantlylower relativepowercomparedtotheFCG(M=.42,SD=.08;t(62)=2.47,
p=.017)andtheNIG(M=.45,SD=.08;t(51)=3.71,p<.001)andno
Fig.4. Relationbetweenchangeinrelativealphapowerandageatplacementinto fostercarefortheFCG.Greybarsarethepooledstandarderroracrossthescalp acrosstimepoints.
differencebetweentheFCGandNIG(t(57)=1.55,p=.127).There werenodifferencesbetweenthegroupsin thebetaband.Plots oftheindividual-levelchangesinrelativepowerineachbandare availableinSupplementarymaterials.
3.4. Effectsofplacementstability
Toexamineplacementstability,eachbandwasanalyzed sep-arately with 4 region (frontal, central, parietal, occipital)×2 hemisphere (left, right)×4 placement stability (CAUG, FCG-Disrupted,FCG-Stable,NIG)RM-ANOVAs.
3.4.1. Absolutepower
Thetaband.Therewerenomaineffectsorsignificant
interac-tionswithplacementstabilityinabsolutethetapower.
Alphaband.Therewerenomaineffectsorsignificant
interac-tionswithplacementstabilityinabsolutealphapower.
Betaband.Examinationofabsolutebetapowerrevealedamain effectofplacement stability(F(3,107)=4.23,p=.028,2
p=.081)
qualifiedbyahemisphere×stabilityinteraction(F(3,107)=3.52,
p=.018,2
p=.090).Follow-upanalysesofthemaineffectrevealed
theFCG-Disrupted (M=1.10, SD=.41)hadsignificantlylessbeta powerthantheFCG-Stable(M=1.50,SD=.41;t(35)=2.97,p=.005) andNIG(M=1.35,SD=.41;t(59)=2.27,p=.027).Thereweretrend differencesbetweentheCAUG(M=1.29,SD=.41)andboththe FCG-Disrupted(t(51)=1.68,p=.099)andtheFCG-Stable(t(48)=1.69,
p=.097).Allothercontrastswerenotsignificant.
Post hoc analysesof theinteraction revealed that the FCG-Stablehadsignificantlygreaterbetapowerinbothhemispheres compared to theFCG-Disrupted (Left: t(35)=3.61, p<.001; and Right:t(35)=2.08, p=.045);theFCG-Disrupted hadsignificantly lessbetapowerinthelefthemisphereandmarginallylesspower in the right compared to the NIG (Left: t(59)=2.45, p=.017; and Right: t(59)=1.93, p=.058); and in the left hemisphere, theCAUGhadsignificantlymorepowerthantheFCG-Disrupted (t(51)=2.08, p<.042) and significantly less power than the FCG-Stable(t(48)=2.02,p=.049).Allothercontrastswerenot sig-nificant.
3.4.2. Relativepower
Thetaband.Examinationofrelativethetapowerrevealeda
sig-nificantmaineffectofplacementstability(F(3,107)=2.77,p=.045,
2
p=.072).Posthoct-testsrevealedtheCAUGhadgreater
rela-tivepowercomparedtotheFCG-StableandtheNIG(t(48)=2.16,
p=.036andt(72)=2.43,p=.017,respectively).Allothercontrasts werenotsignificant.
Alphaband.Therewerenomaineffectsorsignificant
interac-tionswithplacementstabilityinrelativealphapower.
Beta band. Examination of relative beta power revealed a significant hemisphere by placement stability interaction (F(3, 107)=3.80,p=.012,2
p=.096;Fig.5).Posthoct-testsrevealedthe
FCG-Stablegrouphadsignificantlygreaterrelativepowerintheleft hemispherecomparedtotheFCG-DisruptedandNIG(t(35)=2.61,
p=.013andt(56)=2.73,p=.008,respectively).Allothercontrasts werenotsignificant.
4. Discussion
Theaimofthisstudywastoexaminethecontinuingimpactof afostercareinterventiononneuralactivityfollowingearly psy-chosocialdeprivation. At12years, children whowereremoved frominstitutionsandplacedintofostercaredemonstrated contin-uingbenefitsasaresultofinterventionasindexedbytheirresting EEGactivity; specifically,children placed intohigh quality fos-tercaredisplayedmarginallydecreasedthetapowerandelevated alphapowercompared tochildren who receivedcareas usual.
Fig.5. Relativebetapowerintheleftandrighthemispheresbetweenplacement stabilitygroups.*p<.05;**p<.01.
Thisfindingisillustratedbythemeasuresofrelativepowerwhere theFCGchildren resembledtheNIGchildrenwithanincreasing proportionofpowerinthealphabandwhiletheCAUGshowed animmaturepatternofbrainactivitywithgreaterdistributionof powerinthethetaband(Benningeretal.,1984;Matthisetal.,1980;
Raineetal.,2001).
Thedistributionofgreaterpowerinslowerfrequenciesofthe EEGisnotonlycharacteristicofextremedeprivationorsocial iso-lation(Gendreauetal.,1972;Marshalletal.,2004;Zubeketal., 1963),buthasbeenobservedintypicallydevelopingpopulations of children living in impoverishedenvironments. A number of studieshave shownthat children livingin low resourcehomes (e.g.,poornutrition,unsanitarylivingconditions,lessresponsive orsensitivecaregiving)showasimilar“immature”patternofEEG activityas we have observedin the BEIP sample.For example, Oteroandcolleagues(Otero,1994,1997;Oteroetal.,2003),have followeda sampleofmiddle-lowandlowsocioeconomicstatus infantsfrom18-monthsto6yearsandmeasuredEEGatthree dif-ferenttimepointsandfoundthatchildrenwithlowerresources– comparedtochildrenwithmorecognitiveandsocialstimulation throughparentalinvolvement–demonstratedgreaterEEGactivity inslowerfrequencies(deltaandthetabands)andlowerpowerin higherfrequenciesoftenassociatedwithbetterattentionand cog-nitiveperformance(alphaandbetabands;Fernandezetal.,2000). Indeed,anenrichednurseryschoolinterventionhasbeenshownto altertheneuralactivityofimpoverishedchildren,increasingpower infastEEGactivityandimprovedattentionalorientingduringa cog-nitivetask(Raineetal.,2001);suggestingthatearlyintervention mayimproveoutcomesforallchildren.
ThereappearstobeahighdegreeofstabilityintheEEG tra-jectoriesbetweenthe8-and12-yearassessments.Arecentstudy examinedthedevelopmentaltrajectoriesofEEGactivityintheBEIP samplefrombaselinethrough8years(Stamoulisetal.,2015).They foundthatinterventiongroupassignmentwassignificantlyrelated tothetrajectoriesofthealphaandbetabands.Thereare differ-encesinthemethodsthatprecludedirectcomparison,however, examinationofthechangesintheEEGbetween8 and12years suggestthatallgroupscontinuetoshowmaturationalchangesin theEEGthatremaindependentuponearlyexperiencesconsistent withthosereportedbyStamoulisandcolleagues.Moreover,the decreasesinrelativetheta,increasesinrelativebeta,andstability ofrelativealphapowerwithinthisagerangeareconsistentwith otherlongitudinalstudiesofEEGintypicallydevelopingchildren
(Benningeretal.,1984;Clarkeetal.,2001;Matthisetal.,1980).
The results of the current 12-year follow up study demon-stratecontinuityofinterventioneffectsthatwerefoundat8years. Whenchildrenwereobservedatage8thoseplacedintofostercare
before24monthswereindistinguishabletoneverinstitutionalized childreninalphapower(Vanderwertetal.,2010).Thelackofa significantdifferencebetweentheFCGandNIGineitherabsolute orrelativepowerinanyofthefrequencybandsandnorelations betweenEEGandtheageatplacementintofostercaresuggests thatcontinuedhighqualitycarelikelyimprovesthe neurodevel-opmentaloutcomesofchildrenwhohaveexperiencedearlysocial andcognitivedeprivation.
Thereisonecaveattothisinterpretation.Whenwesetaside intent-to-treatandexaminedtherelationsbetweenEEGand stabil-ityofcareversusanyfostercareexperience,theresultssuggestthat maintainingastablecaregivingenvironmenthadagreaterimpact onEEGactivitythanforchildrenwhoexperienceddisruptionsin caregiving.WhenwesplittheFCGintothosewhoremainedwith theirinitialcaregiversandthosewhochangedcaregivers,through, forexample,adoptionorreintegrationwiththeirbiologicalfamily, ourpredictedimprovementsinbetaactivitywereobservedonly inthosechildrenwhoremainedinstable,high-qualitycare.These improvementsinbrainactivitymirrorrecentevidencethat stabil-ityofcaredecreasedinternalizingandexternalizingsymptomsin thissampleat12years(Humphreysetal.,2015).
TheheightenedbetabandactivityfortheFCG-Stablechildren isencouraging,asbetaactivityhasbeenassociated withbetter performanceontasksthatrequireincreasedcognitiveloadsuch asverbalandspatialworkingmemory(Fernandezet al.,2000), motorperformance(Pfurtschelleretal.,1996),languageprocessing
(Papanicolaouetal.,1986),andprocessingemotionallyvalenced
stimuli,includingfacesorIAPSstimuli(GuntekinandBasar,2014;
RayandCole,1985).Arecentreviewonthefunctionofbeta
activ-ity suggestsits role in cognitivecontrol and maintenance of a currentbehavioralstate(EngelandFries,2010).Thus,increased restingbetabandactivitymayreflectcompensatorytop-down con-trolmechanismsrequiredtofollowtheexperimenter’sinstructions tositstill andremainquietduringthetestingsession. Alterna-tively,thegreaterpowerinthebetabandintheFCG-Stablechildren maybeindicativeoffunctionalpotentialforimprovementsinother domains.
Theneuralmechanismsaffectedbytheinterventionareunclear; however,a recentstructuralMRIstudyofthesesameBEIP chil-drenidentifiedaplausibleanatomicalexplanation.Sheridanetal.
(2012)foundthatwhitemattervolumemediatedtheBEIP
inter-ventioneffectonalphapowerinchildrenat8yearsofage.They concludedthatinstitutionalcaremayhaveresultedinthedelay ofwhitemattermaturationandearlierinterventionmayprevent lossofplasticity.Onepossibilityforthisfindingisthatageof inter-ventionmaynotaffectplasticitypersebutaffecttherateatwhich whitemattermatures.
Analternativeinterpretation,inlightoftheplacementstability findings,isthatstable,high-qualitycaregivingmaysupport con-tinuedwhitematterdevelopment.Thereisverylittleevidencein supportofthis interpretationas onlya handfulof studieshave examinedcaregivingandcorticaldevelopment.InastructuralMRI studyof10-year-oldchildren,thefamily’sincome-to-needsratio predictedbothwhiteandgraymattervolume(Lubyetal.,2013). Income-to-needsisaproxyfortheclusterofexperiencesassociated withpoverty;amongthesearestressedcaregivingandmultiple disruptionstohousingandcaregivers,however,interpretationof thesedatashouldbeapproachedwithcautionastherearestilla numberofconcomitantfactors(e.g.educationalresources, nutri-tion,etc)thatlikelyalsocontributetothesefindings.Moredirectly relatingcaregivingtocorticaldevelopment,aninterventionstudy trainingsensitiveandresponsivecaregivingbehaviorsinparents ofpreterm infantsfoundincreasedwhite mattervolume, com-paredtothecontrolinfants, withinthefirstfewmonthsofthe intervention(Alsetal.,2004).Whilethisdemonstratesremarkable plasticityincorticaldevelopment,thisinterventionfocusedonthe
firstfewmonthsoflife,likelypriortotheclosureofanysensitive periodforwhitematterdevelopment.Thefindingofstablefoster careenhancingbrainactivityaddstootherresearch document-ingtheimportanceofstabilityforimprovingemotionregulation, stressreactivity,inhibitorycontrol,andinternalizingand external-izingsymptomology(Blair andRaver,2012;Dozieretal.,2006;
Lewisetal.,2007;Humphreysetal.,2015)andwillbeanimportant
directionforfuturestudiesofneuraldevelopment.
OnelimitationofthisstudyistheuseofrestingEEGasanindex ofneuralactivity.Withoutspecifictasksitisdifficulttoderiveany conclusionsregardingthefunctionalbenefitsoftheimprovedEEG powerasaresultoftheintervention.For example,Eluvathingal
etal.(2006)haveidentifiedincreasedcortico-cortical
connectiv-ityinpreviouslyinstitutionalizedchildrenthatmaybestructurally compensatorybutprovidenoimprovedcognitivefunctioning.It ispossiblethatthedifferencesweobservedatage12maybea resultofsimilarcompensatorycircuits,whichcouldaccountforthe improvedEEGactivitybutmaynotimpartanysocialorcognitive advantages.
Thedatafromthiscurrentstudysuggestthatafamilybased fostercareinterventionadministeredupthroughwhenchildren were54monthsofagehadcontinuedeffectsonEEGactivitysome 8years later. At12years, thechildrenin thefostercaregroup showdecreasedthetapowerandincreasedalphapowerthatis comparabletonever-institutionalizedchildrencomparedto insti-tutionalizedchildren who didnotreceive theintervention.The expectedinterventioneffectsinthebetabandwerenotobserved outright,however,whenwebrokeintent-to-treat,childreninthe interventiongroupwhoremainedintheiroriginalplacementhad significantlygreaterbetabandpower,comparedtothosechildren thatchangedcaregiversordidnotreceivetheintervention.These improvementsinneuralactivitymaybeimportantforthe chil-dren’scontinuingdevelopmentastheyenterpubertyorschooling transitions.
AppendixA. Supplementarydata
Supplementarydataassociatedwiththisarticlecanbefound,in theonlineversion,athttp://dx.doi.org/10.1016/j.dcn.2015.12.004.
References
Als,H.,Duffy,F.H.,McAnulty,G.B.,Rivkin,M.J.,Vajapeyam,S.,Mulkern,R.V.,etal., 2004.Earlyexperiencealtersbrainfunctionandstructure.Pediatrics113, 846–857,http://dx.doi.org/10.1542/peds.113.4.846.
Barry,R.J.,Clarke,A.R.,Johnstone,S.J.,2003.Areviewofelectrophysiologyin attention-deficit/hyperactivitydisorder:I.Qualitativeandquantitative electroencephalography.Clin.Neurophysiol.114,171–183,http://dx.doi.org/ 10.1016/S1388-2457(02)00362-0.
Barry,R.J.,Clarke,A.R.,Johnstone,S.J.,McCarthy,R.,Selikowitz,M.,2009. Electroencephalogram/ˇratioandarousalinattention-deficit/hyperactivity disorder:evidenceofindependentprocesses.Biol.Psychiatry66,398–401,
http://dx.doi.org/10.1016/j.biopsych.2009.04.027.
Bauer,P.M.,Hanson,J.L.,Pierson,R.K.,Davidson,R.J.,Pollak,S.D.,2009.Cerebellar volumeandcognitivefunctioninginchildrenwhoexperiencedearly deprivation.Biol.Psychiatry66,1100–1106,http://dx.doi.org/10.1016/j. biopsych.2009.06.014.
Benninger,C.,Matthis,P.,Scheffner,D.,1984.EEGdevelopmentofhealthyboys andgirls:resultsofalongitudinalstudy.Electroencephalogr.Clin. Neurophysiol.57,1–12,http://dx.doi.org/10.1016/0013-4694(84)90002-6. Blair,C.,Raver,C.C.,2012.Childdevelopmentinthecontextofadversity:
experientialcanalizationofbrainandbehavior.Am.Psychol.67,309–318,
http://dx.doi.org/10.1037/a0027493.
Bos,K.,Zeanah,C.H.,Fox,N.A.,Drury,S.S.,McLaughlin,K.A.,Nelson,C.A.,2011. Psychiatricoutcomesinyoungchildrenwithahistoryofinstitutionalization. Harv.Rev.Psychiatry19,15–24,http://dx.doi.org/10.3109/10673229.2011. 549773.
Browne,K.,Hamilton-Giachritsis,C.,Johnson,R.,Ostergren,M.,2006.Overuseof institutionalcareforchildreninEurope.BMJ332,485–487,http://dx.doi.org/ 10.1136/bmj.332.7539.485.
Castle,J.,Groothues,C.,Bredenkamp,D.,Beckett,C.,O’Connor,T.,Rutter,M.,the E.R.A.StudyTeam,1999.Effectsofqualitiesofearlyinstitutionalcareon
cognitiveattainment.Am.J.Orthopsychiatry69,424–437,http://dx.doi.org/10. 1037/h0080391.
Chabot,R.J.,diMichele,F.,Prichep,L.,2005.Theroleofquantitative
electroencephalographyinchildandadolescentpsychiatricdisorders.Child Adolesc.Psychiatr.Clin.N.Am.14,21–53,http://dx.doi.org/10.1016/j.chc. 2004.07.005.
Clarke,A.R.,Barry,R.J.,McCarthy,R.,Selikowitz,M.,2001.Ageandsexeffectsinthe EEG:developmentofthenormalchild.Clin.Neurophysiol.112,806–814,
http://dx.doi.org/10.1016/s1388-2457(01)00488-6.
Dozier,M.,Peloso,E.,Lindhiem,O.,Gordon,M.K.,Manni,M.,Sepulveda,S., Ackerman,J.,2006.Developingevidence-basedinterventionsforfoster children:anexampleofarandomizedclinicaltrialwithinfantsandtoddlers.J. Soc.Issues62,765–783,http://dx.doi.org/10.1111/j.1540-4560.2006.00486.x. Eluvathingal,T.J.,Chugani,H.T.,Behen,M.E.,Juhasz,C.,Muzik,O.,Maqbool,M.,
etal.,2006.Abnormalbrainconnectivityinchildrenafterearlysevere socioemotionaldeprivation:adiffusiontensorimagingstudy.Pediatrics117, 2093–2100,http://dx.doi.org/10.1542/peds.2005-1727.
Engel,A.K.,Fries,P.,2010.Beta-bandoscillations—signalingthestatusquo?Curr. Opin.Neurobiol.20,156–165,http://dx.doi.org/10.1016/j.conb.2010.02.015. Fernandez,T.,Harmony,T.,Silva-Pereyra,J.,Fernandez-Bouzas,A.,Gersenowies,J.,
Galan,L.,etal.,2000.SpecificEEGfrequenciesatspecificbrainareasand performance.NeuroReport11,2663–2668,http://dx.doi.org/10.1097/ 00001756-200008210-00012.
Gendreau,P.,Freedman,N.L.,Wilde,G.J.S.,Scott,G.D.,1972.ChangesinEEGalpha frequencyandevokedresponselatencyduringsolitaryconfinement.J. Abnorm.Psychol.79,54–59,http://dx.doi.org/10.1037/h0032339. Guntekin,B.,Basar,E.,2014.Areviewofbrainoscillationsinperceptionoffaces
andemotionalpictures.Neuropsychologia58,33–51,http://dx.doi.org/10. 1016/j.neuropsychologia.2014.03.014.
Humphreys,K.L.,Gleason,M.M.,Drury,S.S.,Miron,D.,Nelson,C.A.,Fox,N.A., Zeanah,C.H.,2015.Effectsofearlydeprivationonpsychopathologyatage12 years:follow-upofarandomizedcontrolledtrial.LancetPsychiatry2, 625–634,http://dx.doi.org/10.1016/S2215-0366(15)00095-4.
Kreppner,J.M.,O’Connor,T.G.,Rutter,M.,theEnglish&RomanianAdopteesStudy Team,2001.Caninattention/overactivitybeaninstitutionaldeprivation syndrome?J.Abnorm.ChildPsychol.29,513–528,http://dx.doi.org/10.1023/ A:1012229209190.
Lewis,E.E.,Dozier,M.,Ackerman,J.,Sepulveda-Kozakowski,S.,2007.Theeffectof placementinstabilityonadoptedchildren’sinhibitorycontrolabilitiesand oppositionalbehavior.Dev.Psychol.43,1415–1427,http://dx.doi.org/10.1037/ 0012-1649.43.6.1415.
Luby,J.,Belden,A.,Botteron,K.,Marrus,N.,Harms,M.P.,Babb,C.,etal.,2013.The effectsofpovertyonchildhoodbraindevelopment:themediatingeffectof caregivingandstressfullifeevents.JAMAPediatr.167,1135–1142,http://dx. doi.org/10.1001/jamapediatrics.2013.3139.
Maclean,K.,2003.Theimpactofinstitutionalizationonchilddevelopment.Dev. Psychopathol.15,853–884,http://dx.doi.org/10.1017/S0954579403000415. Marshall,P.J.,Fox,N.A.,theBEIPCoreGroup,2004.Acomparisonofthe
electroencephalogrambetweeninstitutionalizedandcommunitychildrenin Romania.J.Cogn.Neurosci.16,1327–1338,http://dx.doi.org/10.1162/ 0898929042304723.
Marshall,P.J.,Reeb,B.C.,Fox,N.A.,NelsonIII,C.A.,Zeanah,C.H.,2008.Effectsof earlyinterventiononEEGpowerandcoherenceinpreviouslyinstitutionalized childreninRomania.Dev.Psychopathol.20,861–880,http://dx.doi.org/10. 1017/S0954579408000412.
Matthis,P.,Scheffner,D.,Benninger,Chr.,Lipinski,Chr.,Stolzis,L.,1980.Changesin thebackgroundactivityoftheelectroencephalogramaccordingtoage. Electroencephalogr.Clin.Neurophysiol.49,626–635,http://dx.doi.org/10. 1016/0013-4694(80)90403-4.
McLaughlin,K.A.,Fox,N.A.,Zeanah,C.H.,Sheridan,M.A.,Marshall,P.,Nelson,C.A., 2010.Delayedmaturationinbrainelectricalactivitypartiallyexplainsthe associationbetweenearlyenvironmentaldeprivation&symptomsof attention-deficit/hyperactivitydisorder.Biol.Psychiatry68,329–336,http:// dx.doi.org/10.1016/j.biopsych.2010.04.005.
Mehta,M.A.,Golembo,N.I.,Nosarti,C.,Colvert,E.,Mota,A.,Williams,S.C.R.,etal., 2009.Amygdala,hippocampalandcorpuscallosumsizefollowingsevereearly institutionaldeprivation:theEnglishandRomanianAdopteesstudypilot.J. ChildPsychol.Psychiatry50,943–951,http://dx.doi.org/10.1111/j.1469-7610. 2009.02084.x.
Miller,F.G.,2009.Therandomizedcontrolledtrialasademonstrationproject:an ethicalperspective.Am.J.Psychiatry166,743–745,http://dx.doi.org/10.1176/ appi.ajp.2009.09040538.
Millum,J.,Emanuel,E.J.,2007.Theethicsofinternationalresearchwithabandoned children.Science318,1874–1875,http://dx.doi.org/10.1126/science.1153822. Nelson,C.A.,2007.Aneurobiologicalperspectiveonearlyhumandevelopment.
ChildDev.Perspect.1,13–18,http://dx.doi.org/10.1111/j.1750-8606.2007. 00004.x.
NelsonIII,C.A.,Fox,N.A.,ZeanahJr.,C.H.,2014.Anguishoftheabandonedchild. Sci.Am.308,62–67,http://dx.doi.org/10.1038/scientificamerican0413-62. Otero,G.A.,1994.EEGspectralanalysisinchildrenwithsocioculturalhandicaps.
Int.J.Neurosci.79,213–220,http://dx.doi.org/10.3109/00207459408986082. Otero,G.A.,1997.Poverty,culturaldisadvantageandbraindevelopment:astudy ofpre-schoolchildreninMexico.Electroencephalogr.Clin.Neurophysiol.102, 512–516,http://dx.doi.org/10.1016/S0013-4694(97)95213-9.
Otero,G.A.,Pliego-Rivero,F.B.,Fernandez,T.,Ricardo,J.,2003.EEGdevelopmentin childrenwithsocioculturaldisadvantages:afollow-upstudy.Clin.
Neurophysiol.114,1918–1925, http://dx.doi.org/10.1016/S1388-2457(03)00173-1.
Papanicolaou,A.C.,Loring,D.W.,Deutsch,G.,Eisenberg,H.M.,1986.Task-related eegasymmetries:acomparisonofalphablockingandbetaenhancement.Int.J. Neurosci.30,81–85,http://dx.doi.org/10.3109/00207458608985658. Pfurtscheller,G.,StancakJr.,A.,Neuper,C.,1996.Post-movementbeta
synchronization:acorrelateofanidlingmotorarea?Electroencephalogr. Clin.Neurophysiol.98,281–293, http://dx.doi.org/10.1016/0013-4694(95)00258-8.
Raine,A.,Venables,P.H.,Dalais,C.,Mellingen,K.,Reynolds,C.,Mednick,S.A.,2001. Earlyeducationalandhealthenrichmentatage3–5yearsisassociatedwith increasedautonomicandcentralnervoussystemarousalandorientingatage 11years:evidencefromtheMauritiusChildHealthProject.Psychophysiology 38,254–266,http://dx.doi.org/10.1111/1469-8986.3820254.
Ray,W.J.,Cole,H.W.,1985.EEGalphaactivityreflectsattentionaldemands,and betaactivityreflectsemotionalandcognitiveprocesses.Science228,750–752,
http://dx.doi.org/10.1126/science.3992243.
Rutter,M.,2015.Psychopathologyafterfostercareorinstitutionalrearing.Lancet Psychiatry2,576–577,http://dx.doi.org/10.1016/S2215-0366(15)00117-0. Sheridan,M.A.,Fox,N.A.,Zeanah,C.H.,McLaughlin,K.A.,NelsonIII,C.A.,2012.
Variationinneuraldevelopmentasaresultofexposuretoinstitutionalization earlyinchildhood.Proc.Natl.Acad.Sci.U.S.A.109,12927–12932,http://dx.doi. org/10.1073/pnas.1200041109.
Smyke,A.T.,Koga,S.F.,Johnson,D.E.,Fox,N.A.,Marshall,P.J.,Nelson,C.A.,etal.,the BEIPCoreGroup,2007.Thecaregivingcontextininstitution-rearedand family-rearedinfantsandtoddlersinRomania.J.ChildPsychol.Psychiatry48, 210–218,http://dx.doi.org/10.1111/j.1469-7610.2006.01694.x.
Smyke,A.T.,Zeanah,C.H.,Fox,N.A.,Nelson,C.A.,2009.Anewmodeloffostercare foryoungchildren:theBucharestearlyinterventionproject.ChildAdolesc. Psychiatr.Clin.N.Am.18,721–734,http://dx.doi.org/10.1016/j.chc.2009.03. 003.
Somsen,R.J.,van’tKlooster,B.J.,vanderMolen,M.W.,vanLeeuwen,H.M.,Licht,R., 1997.Growthspurtsinbrainmaturationduringmiddlechildhoodasindexed byEEGpowerspectra.Biol.Psychol.44,187–209,http://dx.doi.org/10.1016/ S0301-0511(96)05218-0.
Stamoulis,C.,Vanderwert,R.E.,Zeanah,C.H.,Fox,N.A.,Nelson,C.A.,2015.Early psychosocialneglectadverselyimpactsdevelopmentaltrajectoriesofbrain oscillationsandtheirinteractions.J.Cogn.Neurosci.27,2512–2528,http://dx. doi.org/10.1162/jocna00877.
Tottenham,N.,Hare,T.A.,Quinn,B.T.,McCarry,T.W.,Nurse,M.,Gilhooly,T.,etal., 2010.Prolongedinstitutionalrearingisassociatedwithatypicallarge amygdalavolumeanddifficultiesinemotionregulation.Dev.Sci.13,46–61,
http://dx.doi.org/10.1111/j.1467-7687.2009.00852.x.
UNICEF,2015.StateoftheWorld’sChildren2015:ExecutiveSummary,Retrieved fromhttp://www.unicef.org/publications/files/SOWC2015Summaryand Tables.pdf.
Vanderwert,R.E.,Marshall,P.J.,NelsonIII,C.A.,Zeanah,C.H.,Fox,N.A.,2010.Timing ofinterventionaffectsbrainelectricalactivityinchildrenexposedtosevere psychosocialneglect.PLoSONE5,e11415,http://dx.doi.org/10.1371/journal. pone.0011415.
Wiik,K.L.,Loman,M.M.,VanRyzin,M.J.,Armstrong,J.M.,Essex,M.J.,Pollak,S.D., Gunnar,M.R.,2011.Behavioralandemotionalsymptomsof
post-institutionalizedchildreninmiddlechildhood.J.ChildPsychol. Psychiatry52,56–63,http://dx.doi.org/10.1111/j.1469-7610.2010.02294.x. Zeanah,C.H.,Egger,H.L.,Smyke,A.T.,Nelson,C.A.,Fox,N.A.,Marshall,P.J.,Guthrie,
D.,2009.InstitutionalrearingandpsychiatricdisordersinRomanianpreschool children.Am.J.Psychiatry166,777–785,http://dx.doi.org/10.1176/appi.ajp. 2009.08091438.
Zeanah,C.H.,Fox,N.A.,Nelson,C.A.,2012.TheBucharestearlyintervention project:casestudyintheethicsofmentalhealthresearch.J.Nerv.Ment.Dis. 200,243–247,http://dx.doi.org/10.1097/NMD.0b013e318247d275.
Zeanah,C.H.,Nelson,C.A.,Fox,N.A.,Smyke,A.T.,Marshall,P.,Parker,S.W.,Koga,S., 2003.Designingresearchtostudytheeffectsofinstitutionalizationonbrain andbehavioraldevelopment:theBucharestEarlyInterventionProject.Dev. Psychopathol.15,885–907,http://dx.doi.org/10.1017/S0954579403000452. Zubek,J.P.,Welch,G.,Saunders,M.G.,1963.Electroencephalographicchanges
duringandafter14daysofperceptualdeprivation.Science139,490–492,