Available at:
http://hdl.handle.net/2078.1/156080
[Downloaded 2019/04/19 at 07:55:37 ]
"Does attention bias modification improve attentional
control? A double-blind randomized experiment
with individuals with social anxiety disorder"
Heeren, Alexandre ; Mogoase, Cristina ; McNally, Richard J. ; Schmitz, Anne ; Philippot, Pierre
Abstract
People with anxiety disorders often exhibit an attentional bias for threat. Attention
bias modification (ABM) procedure may reduce this bias, thereby diminishing
anxiety symptoms. In ABM, participants respond to probes that reliably follow
non-threatening stimuli (e.g., neutral faces) such that their attention is directed
away from concurrently presented threatening stimuli (e.g., disgust faces). Early
studies showed that ABM reduced anxiety more than control procedures lacking
any contingency between valenced stimuli and probes. However, recent work
suggests that no-contingency training and training toward threat cues can be as
effective as ABM in reducing anxiety, implying that any training may increase
executive control over attention, thereby helping people inhibit their anxious
thoughts. Extending this work, we randomly assigned participants with
DSM-IV diagnosed social anxiety disorder to either training toward non-threat (ABM),
training toward threat, or no-contingency conditi...
Document type : Article de périodique (Journal article)
Référence bibliographique
Heeren, Alexandre ; Mogoase, Cristina ; McNally, Richard J. ; Schmitz, Anne ; Philippot,
Pierre.
Does attention bias modification improve attentional control? A double-blind randomized
experiment with individuals with social anxiety disorder.
In:
Journal of Anxiety Disorders
, Vol. 29,
no.01, p. 35-42 (2015)
ContentslistsavailableatScienceDirect
Journal
of
Anxiety
Disorders
Does
attention
bias
modification
improve
attentional
control?
A
double-blind
randomized
experiment
with
individuals
with
social
anxiety
disorder
Alexandre
Heeren
a,∗,
Cristina
Mogoas¸
e
b,
Richard
J.
McNally
c,
Anne
Schmitz
a,
Pierre
Philippot
aaPsychologicalScienceResearchInstitute,UniversitéCatholiquedeLouvain,Louvain-la-Neuve,Belgium bDepartmentofClinicalPsychologyandPsychotherapy,Babes¸-BolyaiUniversity,Cluj-Napoca,Romania cDepartmentofPsychology,HarvardUniversity,Cambridge,MA,USA
a
r
t
i
c
l
e
i
n
f
o
Articlehistory:
Received7November2013
Receivedinrevisedform18October2014 Accepted25October2014
Availableonline24November2014
Keywords:
Attentionbiasmodification Attentioncontrol Socialanxietydisorder Speechperformance Cognitivebiasmodification Attentionalbias
a
b
s
t
r
a
c
t
Peoplewithanxietydisordersoftenexhibitanattentionalbiasforthreat.Attentionbiasmodification (ABM)proceduremayreducethisbias,therebydiminishinganxietysymptoms.InABM,participants respondtoprobesthatreliablyfollownon-threateningstimuli(e.g.,neutralfaces)suchthattheir atten-tionisdirectedawayfromconcurrentlypresentedthreateningstimuli(e.g.,disgustfaces).Earlystudies showedthatABMreducedanxietymorethancontrolprocedureslackinganycontingencybetween valencedstimuliandprobes.However,recentworksuggeststhatno-contingencytrainingand train-ingtowardthreatcuescanbeaseffectiveasABMinreducinganxiety,implyingthatanytrainingmay increaseexecutivecontroloverattention,therebyhelpingpeopleinhibittheiranxiousthoughts. Extend-ingthiswork,werandomlyassignedparticipantswithDSM-IVdiagnosedsocialanxietydisordertoeither trainingtowardnon-threat(ABM),trainingtowardthreat,orno-contingencycondition,andweusedthe attentionnetworktask(ANT)toassessallthreecomponentsofattention.Aftertwotrainingsessions, subjectsinallthreeconditionsexhibitedindistinguishablysignificantdeclinesfrombaselineto post-traininginself-reportandbehavioralmeasuresofanxietyonanimpromptuspeechtask.Moreover,all groupsexhibitedsimilarlysignificantimprovementsonthealertingandexecutive(butnotorienting) componentsofattention.ImplicationsforABMresearcharediscussed.
©2014ElsevierLtd.Allrightsreserved.
1. Introduction
Peoplewithsocialanxietydisorder(SAD)oftenexhibitan atten-tionalbiasforsocial-threatcues,suchasfacesexpressingdisgustor anger(e.g.,Mogg,Philippot,&Bradley,2004).Thisbiasmaycausally contributetoincreasinganxietyproneness(MacLeod,Rutherford, Campbell,Ebsworthy,&Holker,2002),andtherebyfigureinthe etiologyandmaintenanceofSADandotheranxietydisorders(for areview,seeVanBockstaeleetal.,2014).Accordingly,reducingit mayhaveyieldclinicalbenefits.
∗ Correspondingauthorat:LaboratoryofExperimentalPsychopathology, Insti-tuteofPsychologicalScience,UniversitéCatholiquedeLouvain,PlaceduCardinal Mercier10,B-1348Louvain-la-Neuve,Belgium.Tel.:+32497882315;
fax:+3210473774.
E-mailaddress:[email protected](A.Heeren).
Inspired by MacLeod et al. (2002), psychologists have used attentionbiasmodification(ABM)procedurestodiminishABand therebysymptomsofSAD(e.g.,Amir,Taylor,&Donohue,2011). TodevelopanABMprocedure,MacLeodetal.(2002)modifiedthe classicdot-probeparadigmthatmeasuresAB(MacLeod,Mathews, &Tata,1986).Intheoriginaldot-probetasks,participantsviewed two stimuli(e.g.,a threatening word/photographand a neutral word/photograph)simultaneouslypresented intwolocationsof acomputerscreenforapproximately500ms.Immediately there-after,aprobeappearedinthelocationpreviouslyoccupiedbyone ofthetwostimuli.Participantshavetorespondtotheprobeas quicklyaspossible.AnABwasdemonstratedwhenparticipants respondfastertotheprobewhenitreplacesathreatening stim-ulusthanwhenitreplacesanon-threateningstimulus,indicating thattheirattentionwasdirectedtothelocationoccupiedbythe threatening stimulus.In ABM, researcherstypically modifythe originaltasksuchsothattheprobenearlyalways(e.g.,95%ofthe trials)replacestheneutralstimulus,therebyredirectingsubjects’
http://dx.doi.org/10.1016/j.janxdis.2014.10.007
36 A.Heerenetal./JournalofAnxietyDisorders29(2015)35–42
attentiontonon-threateningcues.Inthecontrolcondition,there isnocontingencybetweencuesandprobes.Relativetothecontrol condition,ABM reducessymptomsin peoplewithSAD,as sev-eralstudieshaveshown(Amir,Weber,Beard,Bomyea,&Taylor, 2008;Amiretal.,2009;Li,Tan,Qian,&Liu,2008;Schmidt,Richey, Buckner,&Timpano,2009).ThesefindingssuggestthatABMcould haveimportantclinicalpotential,asitentailsaverysimple pro-tocol,littlecontactwithamentalhealthprofessional,andcanbe easilydisseminated.
However,overthepasttwoyears,otherstudieshavereported mixed findings (e.g., Boettcher et al., 2013; Bunnell, Beidel, & Mesa,2013;Carlbringetal.,2012).Severalexplanationsforthese mixedfindingshavebeenformulated(Emmelkamp,2012;Heeren, De Raedt,Koster,&Philippot,2013; Klumpp&Amir,2010).For example,trainingattention,regardlessofthedirectionofthe con-tingencybetweenprobesandcues,maybolstertop-downattention controlin waysthat strengthenone’sabilitytocontrolanxiety.
KlumppandAmir(2010)reporteddatacongruentwiththis hypoth-esis. In their experiment, they randomly allocated moderately sociallyanxiousindividualstooneofthreedifferentconditions: (1)trainingtoattendtonon-threat(i.e.,ABM),(2)attendtothreat, or (3) a control condition in which there was no contingency betweencuesandprobes.Afterasingle-session,individualswho weretrainedtoattendtothreataswellasthosereceivingABM reportedlessstateanxietyinresponsetoanimpromptuspeech comparedtoindividualsintheno-contingencycontrolcondition.
However,Heeren,Reese,McNally,andPhilippot(2012)didnot replicatethiseffectamongparticipantsdiagnosed with general-izedsocialphobia.Inthisexperiment,participantswererandomly assigned toreceivefoursessionsofone ofthethreeconditions mentionedabove.Theyfoundthat,incontrasttothetwoother con-ditions,thosewhoweretrainedtoattendtonon-threatreported less behavioralandphysiological (i.e.,skinconductance reactiv-ity)indicesofanxietyinresponsetoanimpromptuspeech,anda decreaseinAB.Thesestudiessuggestthattheprocessesmediating theimpactofABMonanxietymaybemorecomplicatedthan com-monlyassumed.However,itremainsunclearwhetherthebenefits apparentinthesetwostudiesresultfromincreasedexecutive con-troloverattentionasnonemeasuredit.Morerecently,McNally, Enock,Tsai,andTousian(2013)reportedanexperimentinwhich theyrandomlyassignedspeech-anxiousindividualstooneofthe three trainingconditions mentionedabovewhile alsoincluding self-reportandbehavioralmeasuresofexecutiveattentioncontrol beforeandafterthetraining.Afterfoursessionsoftraining, par-ticipants,irrespectiveofgroupassignment,exhibitedsignificant decreasesin self-report,behavioral,and physiological measures ofanxietyassociatedwithaspeakingtask.Moreimportantly,all threetrainingconditionsimprovedattentionalcontrol,asindexed throughtheexecutiveconflictscoreoftheattentionnetworktask (ANT;Fan,McCandliss,Sommer,Raz,&Posner,2002)andthe Atten-tionalControlScalequestionnaire(Derryberry&Reed,2002).
Consideringaplaceboeffectforthewidespreadclinicalbenefits,
McNallyetal.(2013)suggestedthatthehalooftechnology embod-iedinacomputerizedfixforone’spublicspeakingfearmightfoster positiveexpectanciesthataccountfortheobservedimprovement. This interpretation seems plausibleas McNally et al. informed participantsofthepotentialtherapeuticbenefitsoftraining.In con-trast, Heeren,Reese,etal.(2012)and KlumppandAmir(2010)
informedparticipantsthattheresearchconcernedprocesses asso-ciatedwithSAD;theydidnotmentionanypotentialtherapeutic benefits.
Further,incontrasttoKlumppandAmir’shypothesis,recent evi-dencesuggeststhatABinSADmayresultnotonlyfromimpairment in theexecutivecontrolofattention,butalsofromimpairment inorientingtowardnon-emotionalstimuli(e.g.,Moriya&Tanno, 2009).AccordingtoPosnerand Petersen(1990),therearethree
componentstoattention:alerting,orienting,andexecutive con-trol.However,eveniftheANTevaluatesthesethreeindependent attentionalnetworks,McNallyetal.(2013)onlyreporteddataon thechangeinexecutiveconflictcomponentoftheANTastheir hypothesisonlyconcernedexecutivecontrol.Todate,nopublished studyhasexploredtheimpactofABMonallthreecomponentsof attention.
Inthepresentdouble-blindexperiment,werandomlyassigned individualswithaDSM-IVdiagnosisofSADtooneofthree condi-tions:(1)attendtonon-threatstimuli,(2)attendtothreatstimuli, and(3)no-contingencycontrol.Subjectswerenottoldaboutthe possibletherapeuticbenefitsoftraining.Rather,theyweremerely informed that the study concerned the cognitive mechanisms underlyingsocialinteractionamongshypeople.Weassessedthe effects oftheseprocedures onchangein AB,onself-reportand behavioralmeasuresofanxietyduringaspeechperformance,and onallthreeattentionalnetworksoftheANT,incontrasttoMcNally etal.(2013)whosehypothesisconcernedonlytheexecutive con-flictmeasureoftheANT.
We addressedseveralissues. If,asKlumppand Amir(2010)
havesuggested,attentiontrainingiseffectivebecauseofincreased attentional control arising from any contingency-based proce-dure regardless of the direction of attention, then participants ineithertheattendtothreatorattendtonon-threatconditions shouldexhibitgreaterimprovementthanparticipantsinthe no-contingencycontrolconditionontheexecutivenetworkoftheANT aswellasmeasuresofanxiety.Byconstrast,ifattentiontrainingis effectivebecauseofattendingtonon-threat,asHeeren,Reese,etal. (2012)havesuggested,thenonlytheparticipantsintheattendto non-threatconditionshoulddemonstrateclinicalbenefits.Finally, ifattentiontrainingiseffectiveregardlessofthepresenceofa con-tingency,asMcNallyetal.(2013)havesuggested,allgroupsshould exhibitimprovementintheexecutivenetworkofANT.
2. Materialsandmethods 2.1. Participants
Werecruited61individualswithaprimaryDSM-IVdiagnosis ofSAD(AmericanPsychiatricAssociation,1994)fromthe Univer-sitéCatholiquedeLouvaincommunity.Atotalof445volunteers respondedtoourinvitationtotakepartinaninvestigationofthe mechanismsunderlyingsocialinteractionamongshypeople. Fol-lowing screening, 148individuals who scoredabove 56 onthe self-report version of theLiebowitz Social Anxiety Scale(LSAS;
Liebowitz,1987)wereselected.Aclinicalpsychologistthenused theFrenchversionoftheMiniInternationalNeuropsychiatric Inter-view(MINI;Lecrubier,Weiller,Bonora,Amorin,&Lépine,1998)to diagnoseDSM-IVAxisIdisorders.
InadditiontothepresenceofaDSM-IVdiagnosisofSocial Anxi-etyDisorder,allparticipantshadtofulfillseveralinclusioncriteria: (a)nocurrentsubstanceabuseordependence,(b)nocurrentheart, respiratory,neurologicalproblems,oruseofpsychotropic medica-tions,(c)nocurrentpsychologicalorpsychiatrictreatment,and(d) normalorcorrected-to-normalvision.Thesecriteriawereassessed viaquestionnaire.Sixty-twoparticipantsmettheDSM-IVdiagnosis ofSocialAnxietyDisorder;41hadthegeneralizedsubtype,and21 hadthespecificsubtype.Onedeclinedourinvitationtoparticipate, andso61participantsenrolledinthestudy;theircharacteristics aredisplayedinTable1.Weobtainedwritteninformedconsent fromeachparticipant.Eachparticipantwastestedindividuallyin aquietroomandallsessionsoccurredinthesamelaboratory.The studywasapprovedbytheEthicalCommitteeoftheUniversité CatholiquedeLouvain(UCL,Belgium),andconductedaccordingto theDeclarationofHelsinki.Participantsreceivedfinancial compen-sation(15Euros)fortheirparticipation.
A.Heerenetal./JournalofAnxietyDisorders29(2015)35–42 Table1
Participants’characteristicsasafunctionoftrainingallocation(SDinparentheses).
Attendtothreat(n=21) Attendtonon-threat(n=22) No-contingency(n=18) For!2 p
Age 28.48(11.08) 24.41(7.24) 24.72(8.68) 1.28a .29 %Female/%Male 81.00/19.00 88.90/11.10 72.70/27.30 1.64b .44 Yearsofeducation 16.81(2.80) 16.18(2.65) 15.39(2.25) 1.45a .24 BDI-II 13.33(8.72) 12.59(7.73) 14.78(7.01) .387a .68 STAI-T 47.43(9.75) 47.82(7.48) 42.89(7.40) 2.07a .14 LSAS 71.57(11.38) 73.41(10.21) 78.22(11.79) 1.83a .17
Note:Attendtothreat,trainingtoattendtothreateningmaterial;attendtonon-threat,trainingtoattendtoneutralmaterial;no-contingency,trainingwithoutcontingency betweencuesandprobes;BDI-II,BeckDepressionInventory-II;STAI-T,SpielbergerState-TraitAnxietyInventory-Trait;LSAS,LiebowitzSocialAnxietyScale.
aValueforF(2,58). bValuefor!2(2,N=61). 2.2. Materials
2.2.1. Attentiontrainingstimuli
Werandomlyselected70facepairswithouthairlines(35men, 35women)fromtheKarolinskaEmotionalDirectedFacesdatabase (Lundqvist,Flykt,&Öhman,1998),whichisastandardizedsetof emotionalexpressions.InaccordwithmostpreviousABMstudies, thefacesdisplayedeitherthreatening(i.e.,disgust)orneutralfacial expressions.
2.2.2. ABassessmentstimuli
ToassessAB,weusedeightFrenchsocialthreatwords(stupid, humiliation,embarrassed,shame,mockery,foolish,idiot,rejection) andeightneutralwords(book,radiator,spoon,tree,computer, pro-cession,piano,towel),matchedonfrequencyofusageinFrench (New,Pallier,Ferrand,&Matos,2001)andsimilartothosein previ-ousresearch(Heeren,Lievens,&Philippot,2011;Heeren,Peschard, &Philippot,2012).Thethreatandneutralwordsdidnotdifferin length,t(14)=0.44, p>.66, d=0.23.We usedwords,rather than faces,intheassessmenttrialstotestwhethertheeffectsoftraining withonetypeofstimulusgeneralizetoanothertype.
2.3. Measures 2.3.1. Questionnaires
Participants werescreened viatheself-reportversion ofthe LSAS(Liebowitz,1987),andtheyalsocompletedtheTraitAnxiety Inventory(STAI-T;Spielberger,Gorsuch,Lushene,Vagg,&Jacobs, 1983),andtheBeckDepressionInventory(BDI-II;Beck,Steer,& Brown,1996)atthebeginning ofthefirsttraining session. The LSAS isa24-itemscalethatmeasuresanxietyandavoidanceof socialinteractionandperformancesituations.TheSTAI-Tisa 20-item self-report questionnaireassessing anxietyproneness.The BDIisa21-itemself-reportmeasureofsymptomsofdepression.We usedthevalidatedFrenchversionsofthesescales(LSAS,Heeren, Maurage,etal.(2012);BDI-II,Becketal.,1996;STAI-T, Bruchon-Schweitzer&Paulhan,1993).
2.3.2. MeasureofAB
Forassessingeffectsoftrainingonattentiontothreatcues,we askedparticipantstocompleteanindependentmeasureofABat baselineandpost-training.Weusedamodifiedversionofthe spa-tialcueingtaskidenticaltothatusedbyothers(e.g.,Amiretal., 2008;Julian,Beard,Schmidt,Powers,&Smits,2012).Wordswere presentedinlowercasewhiteletters(5–8mminheight)againsta blackbackground,inthecenterofthescreen.Oneachtrial,theword appearedinarectangletotheleftorrightofthecentralfixation cross,therebydirectingattentiontotheleftorright.After600ms, thecueworddisappeared,andanasterisk(theprobe)appearedin oneofthetwolocations.
The probe remained on the screen until the participant responded,andthecomputerrecordedthisresponselatencyeach
trial.Theinter-trialinterval,fromtargetoffsettothenextfixation cross,was1650ms.Onsometrials,thecuewordwasvalid(i.e., theprobeappearedinthesamelocationasthecueword),whereas onothersthecuewordwasinvalid(i.e.,theprobeappearedinthe locationoppositetothecueword).
Participantswereexposedto192experimentaltrials,twothirds ofwhichwerevalidlycued(128=8words×2wordtypes×2word positions×4 repetitions),one sixth were invalidly cued (32=8 words×2 word types×2 word positions), and one sixth were
uncued (32=8 words ×2 word types×2 word positions; e.g.,
Stormark,Nordby,&Hugdahl,1995).Trialswerepresentedina dif-ferentrandomorderforeachparticipant.Amir,Elias,Klumpp,and Przeworski(2003)foundthatsociallyanxiousparticipantsshowed significantlylongerresponselatenciesoninvalidcuedsocialthreat trialscomparedtonon-anxiouscontrolsonthistask,suggesting thatABmayreflectdifficultydisengagingfromthreateningstimuli.
2.3.3. Attentionalnetworktask(ANT)
TheANTwasadministeredtodeterminetheefficiencyofthree independentattentionalnetworks:alerting,orienting,and exec-utivecontrol(Fanetal.,2002).Participantshadtodetermineas quicklyandasaccuratelypossiblethedirectionofacentralarrow (thetarget)locatedinthemiddleofa horizontallineprojected eitheratthetoporatthebottomofthescreen.Theyresponded bypressingthecorrespondingbutton(leftorright) onthe key-board.Eachtargetwasprecededbyeithernocue,acentercue(an asteriskreplacingthefixationcross),adoublecue(twoasterisks, oneappearingaboveandonebelowthefixationcross),oraspatial cue(anasteriskappearingaboveorbelowthefixationcrossand indicatingthelocationoftheupcomingtarget).Moreover,flankers appearedhorizontallyoneachsideofthetarget.Therewerethree possibleflanker types: eithertwo arrows pointing in thesame directionasthetarget(congruentcondition),twoarrowspointing intheoppositedirectionofthetarget(incongruentcondition),or twodashes(neutralcondition).Eachtrialhadthefollowing struc-ture:(1)acentralfixationcross(randomdurationbetween400and 1600ms);(2)acue(100ms);(3)acentralfixationcross(400ms); (4)atargetanditsflankers,appearing aboveorbelowthe fixa-tioncross(thetargetremainedonthescreenuntiltheparticipant respondedorfor1700msifnoresponseoccurred);(5)acentral fix-ationcross[lastingfor3500msminusthesumofthefirstfixation period’sdurationandthereactiontime(RT)].RT(inmilliseconds) andaccuracy(percentageofcorrectresponses)wererecordedfor eachtrial.
TheANTtaskcomprised288trials,dividedinthreeblocksof96 trialseach(withashortbreakbetweenblocks).Therewere48 pos-sibletrials,basedonthecombinationoffourcues(nocue,center cue,doublecue,spatialcue),threeflankers(congruent, incongru-ent,neutral),twodirectionsofthetargetarrow(left,right)andtwo localizations(upperorlowerpartofthescreen).Trialswere pre-sentedinarandomorderandeach possibletrialwaspresented twicewithinablock.
38 A.Heerenetal./JournalofAnxietyDisorders29(2015)35–42
2.3.4. Speechtask
Weadministeredaspeechtasktoassessself-reportand behav-ioral measures of anxiety at baseline and post-training. Each participantbegan thetasksittingin achair30cmfroma com-puter screen.Asetofinstructions thenappeared onthescreen andinformedparticipantsthattheywouldhavetomakea2-min speechconcerningcontroversialtopicswidelydiscussedinthe Bel-gianmedia,andthattheirperformancewouldbevideorecorded. Twotopics(abortionandlegalizationofcannabis)wererandomly counterbalancedbetweentimesofassessment.Theyweregiven 2mintoprepareandasheetofpapertowritedowntheirnotes; however,theyweretoldthattheywouldnotbeallowedtouse thesenotesduringthespeech.Afterparticipantshadpreparedtheir speech, theyweredirected tostandin frontof avideo camera. Justbeforethespeech,theexperimenteraskedparticipantstorate, usingSubjectiveUnitsofDiscomfortScale(SUDS;Wolpe,1958),their levelofsituationalanxietyfrom0(notanxious)to100(extremely anxious).Theparticipantthenperformedthespeechwhilebeing videorecorded.Twoclinicalpsychologists,blindtotraining condi-tion,usedtheBehavioralAssessmentofSpeechAnxiety(BASA;Mulac &Sherman,1974)methodtoratethespeechoftheparticipant.The BASAincludes18molecularcategories(e.g.,havingaclearvoice, searchingforthewords),andthemeanscoreofthesecategories hasexcellentconcurrentvaliditywithexperts’ratingsofspeech anxiety(Mulac&Sherman,1974).Interraterreliabilityofthetotal scorewashigh(r=.65,p<.001atbaseline;r=.75,p<.001at post-training).Accordingly,weaveragedthescoresofthetworaters. Thesametworatersassessedboththebaselineandpost-training speechesofaparticipant.
2.4. Attentiontraining
Attentiontraining consistedof a standard probe discrimina-tiontask,modifiedtotrainparticipantseithertoattendprimarily tonon-threatcues,tothreatcues,ortono-contingencytraining (controlcondition).Forallconditions,afixationcrossappearedfor 500msinthecenterofthescreen,followedbytwofacial expres-sionsofthesameperson,adisgustexpressionandaneutralone, presented for500ms.A probe appeared(i.e.,theletter Eor F), replacingoneofthefaces.Itremainedonthescreenuntilthe par-ticipantindicateditsidentitybypressingthecorrespondingkey. Theinter-trialintervalwas1500ms.
Intheattend-to-non-threatcondition,theprobereplacedthe non-threatening(neutral)faceon95%ofthetrials.Inthe attend-to-threatcondition,theprobereplacedthethreatening(disgust)face on95%ofthetrials.Inthecontrolcondition,theprobereplaced theeachfaceon50%ofthetrials(i.e.,therewerenocontingency betweencuesandprobes).
Participants completed 560 trials in one block. Each of the 70 threatening faces appeared four times, paired with a non-threateningfaceofthesameindividual,inpositionsthat rep-resentedallcombinationsofthelocationsandprobetypes.This procedure wasrepeated2 times (i.e.,560=70 stimuli×2
posi-tions×2arrowdirections×2repetitions).Theinstructionswere presentedonthecomputerandwereidenticalforalltheconditions. Faceswerepositioned4cmfromthetop/bottomofthescreen,8cm fromtheipsilateraledge,22.5cmfromthecontralateraledge,and centeredvertically.Eachfacewas7.5cmhighand7.5cmwide.
2.5. Generalprocedure
The procedure consisted of two sessions of ABM, separated by oneday.Participants wererandomlyassigned tooneof the three conditionsviaacomputerizedrandomizationsystem.The participants and the experimenters were blind to condition. Participants first completed a demographic questionnaire, the
STAI (Trait version), and theBDI-II. We then administered the modifiedspatialcueingtask,whichprovidedabaselineindexof attention bias,theANT,and thespeechtask. Next,participants completedthetwotraining sessions,eachlasting about30min. Aftercompletingthesecondsession,participantscompletedthe second ANTaswellasthesecondmodified spatialcueingtask. Finally,participantswereinvitedtocompletethesecondspeech task.Participantsweredebriefedattheendoftheexperiment.
3. Results 3.1. Datareduction 3.1.1. Spatialcueingtask
Following Ratcliff’s (1993) recommendations, we addressed outliersanderrorsintheRTtasksasfollows.First,trialswith incor-rectresponseswereexcluded(0.77%oftrialsatbaseline;1.02%of trialspost-training).Second,RTslowerthan200msorgreaterthan 2000mswereremovedfromanalyses(0.27%oftrialsatbaseline; 0.22%oftrialsatpost-training).Third,RTsofmorethan1.96SD
beloworaboveeachparticipant’smeanforeachexperimental con-ditionwereexcludedasoutliers(0.80%oftrialsatbaseline;0.83% oftrialsofthedataatpost-training).
3.1.2. ANT
Weexcludeddatafromtrialswithincorrectresponses(0.80% of trials atbaseline;0.75% of trialsat post-training),RTs lower than200msorgreaterthan2000ms(0.41%oftrialsatbaseline; 0.37%oftrialsatpost-training),andRTsexceeding1.96SDbelow or above each participant’smean for each experimental condi-tion(0.26%ofremainingtrialsatbaseline;0.23%atpost-training). FollowingFanetal.(2002),wecomputedthealertingeffectby sub-tractingthemean(i.e.,RToraccuracyscore)fordoublecuetrials fromthemeanfornocuetrials(Nocue–Doublecue);the orien-tingeffectbysubtractingthemeanforspatialcuetrialsfromthe meanresultforcentercuetrials(Centercue–Spatialcue);andthe
executiveconflicteffectbysubtractingthemeanforcongruent tri-als(summedacrosscuetypes)fromthemeanforincongruenttrials (Incongruent–Congruent).Forbothalertingandorientingeffects, greatersubtractionscoresforRT(andlowerforaccuracy)indicated greaterefficiency.Incontrast,greatersubtractionscoresforRT(and lowerforaccuracy)onexecutiveconflictindicatedincreased diffi-cultywithexecutivecontrolofattention(Fan,McCandliss,Fossella, Flombaum,&Posner,2005).
3.2. Groupequivalence
AsshowninTable1,thegroupsdidnotdifferatbaselineonthe STAI-trait,BDI-II,orLSAS,andwereindistinguishableintermsof age,gender,andyearsofeducation.
3.3. ChangeinAB
WesubjectedRTstoa3(Condition)×2(Time:Baseline, post-training)×2(Validity:valid,invalid)×2(WordType:Socialthreat, neutral)analysisofvariance(ANOVA)withrepeatedmeasurement onthelastthreefactors.TheANOVArevealedasignificant Condi-tion×Time×Validityinteraction,F(2,58)=3.10,p<.05,"2p=.10. TherewasnoCondition×Time×Validity×WordTypeinteraction, F(2,58)=.65,p>.52,"2p=.02,noraCondition×Time×WordType
interaction,F(2,58)=.72,p>.49,"2p=.02.
To probe the significant interaction, we computed separate Condition×TimeANOVAsforvalidandinvalid trialsseparately. For validtrials,thisanalysisrevealedamaineffectofTime,F(1, 58)=9.08, p<.005, "2p=.13, but nosignificantTime×Condition
A.Heerenetal./JournalofAnxietyDisorders29(2015)35–42
interaction, F(2, 58)=2.00, p>.14, "2p=.06. For invalid trials,
there was noTime effect, F(1, 58)=.91, p>.34, "2p=.01, nora Time×Conditioninteraction,F(2, 58)=2.15, p>.12, "2p=.07. As depictedinTable2,theseresultssuggestadecreaseinRTforvalid trialsfrombaselinetopost-training,regardlessoftheexperimental condition.
3.4. Changeinattentionnetworkcomponents
WesubjectedRTstoa3(Condition)×2(Time:Baseline,
post-training)×3(AttentionalNetwork:Alerting,Orienting,Executive control)ANOVAwithrepeatedmeasurementonthelasttwo fac-tors. TheANOVArevealedamain effectofTime,F(1,58)=4.40,
p<.05, "2p=.07, and a significant Time×Attentional Network
interaction,F(1,58)=16.19,p<.0001,"2p=.22.Therewasno Condi-tion×Time×AttentionalNetworkinteraction,F(2,58)=.26,p>.77,
"2p=.02. To explore this significant interaction, we computed
separate paired t-tests for each attention network separately, regardlessoftheexperimentalcondition.Theseanalysesrevealed thatparticipantsexhibitedsignificantimprovement,frombaseline topost-training,intheAlertingnetwork,t(60)=2.21,p<.05,and intheExecutiveControlnetwork,t(60)=3.88,p<.05.Therewasno significantchangefortheOrientingnetwork,t(60)=1.89,p>.07. ResultsareshowninTable2.
3.5. Changeinemotionalreactivitytospeechtask
For theSUDSandBASA data,wecomputedseparate3 (Con-dition)×2(Time:Baseline,post-training)ANOVAswithrepeated measurementonthelastfactors.FortheSUDSratings,theANOVA revealed a main effect of Time, F(1, 58)=29.03,p<.0001, "2p= .33,butnosignificantTime×Conditioninteraction,F(2,58)=0.29,
p>.75,"2p=.01.FortheBASAscores,again,theANOVArevealeda maineffectofTime,F(1,58)=19.76,p<.01,"2
p=.27,butno
signif-icantTime×Conditioninteraction,F(2,58)=2.13,p>.13,"2p=.08. AsdepictedinFig.1,allgroups exhibitedasignificantdecrease in bothself-reportedandbehavioralmeasuresofanxietytothe impromptuspeechfrombaselinetopost-training.
3.6. Additionalanalyses
Asseveralstudiessuggest, improvementinattentioncontrol mayberelatedtothechangeinemotionalreactivitytothespeech task, wecomputedPearsoncorrelationcoefficients betweenthe former (i.e.,post-trainingminusbaseline score)andlatter vari-ables(i.e.,post-trainingminusbaselinescore)forboththeSUDS andBASAmeasuresandboththeexecutiveandthealerting compo-nentsoftheANT.However,thecorrelationswereneithersignificant fortheexecutivecomponent[rs(61)<.20,ps>.29]northealerting
componentofattention[rs(61)<.30,ps>.23].
4. Discussion
Wehadtwopurposesinthisstudy.First,wesoughtto investi-gatetheinfluenceofthepresenceandthedirectionofacontingency betweencuesandprobeduringtrainingonchangesinABforthreat aswellasonself-reportedandbehavioralmeasuresofanxietyin responsetoanimpromptuspeechamongsubjectswithaDSM-IV diagnosisofSAD.Second,weaimedtoexploretheimpactofABM onallthreecomponentsofattention,notonlyfortheexecutive component(cf.McNallyetal.,2013).
ConsistentwithMcNallyetal.’s(2013)findings,allthreegroups exhibitedstatisticallyindistinguishablereductionsfrombaseline topost-traininginself-reportedandbehavioralmeasuresof
anx-ietyassociatedwiththeimpromptuspeech.Moreover,ourstudy Table
2 Mean performances for the spatial cueing task (in millisecond) and the attention network task (differential index in millisecond) as a function of training conditions and time (SD in parentheses). Outcome Attend to threat Attend to non-threat No-contingency Cue validity Word type Baseline Post-training Baseline Post-training Baseline Post-training Spatial cueing task Valid Neutral 421.11 (60.48) 405.54 (80.40) 402.73 (72.35) 363.33 (55.92) 411.64 (60.87) 400.78 (60.53) Threat 425.93 (62.22) 405.22 (80.35) 413.82 (88.58) 365.24 (52.91) 412.12 (72.95) 407.97 (79.29) Invalid Neutral 441.66 (85.33) 464.01 (87.89) 433.45 (130.50) 403.89 (50.04) 476.70 (153.11) 450.04 (84.93) Threat 443.05 (82.50) 475.22 (92.82) 445.34 (142.93) 398.45 (72.56) 481.47 (158.97) 448.01 (105.69) Attention network task Network Alerting 40.18 (25.23) 49.46 (25.80) 47.81 (24.55) 53.17 (23.51) 41.26 (19.63) 47.24 (24.49) Orienting 48.21 (28.60) 46.64 (28.99) 51.00 (21.99) 46.56 (17.76) 37.72 (21.23) 36.57 (22.87) Executive 97.00 (32.16) 91.54 (31.87) 113.02 (38.19) 95.59 (32.84) 100.72 (35.57) 83.70 (30.62) Notes : Attend to threat, training to attend to threatening material; attend to non-threat, training to attend to neutral material; no-contingency, training without contingency between cues and probes; alerting, alerting network; orienting, orienting network; executive, executive network.
40 A.Heerenetal./JournalofAnxietyDisorders29(2015)35–42
Fig.1. Changeinanxietyreactivitytotheimpromptuspeechchallengeasafunctionoftrainingconditionandtime.Note:scoresfortheSubjectiveUnitsofDiscomfortScale appearonthelowerpart(b).Theupperpart(a)depictsthescoresfortheBehavioralAssessmentofSpeechAnxiety(meanofthetworaters).Errorbarsrepresentsstandard errorsofthemean.*p<.05;**p<.01.
replicates this effect in participants with a DSM-IV diagnosis of SAD who were unaware of the potentialanxiolytic benefits of training. Our results are also consistent with randomized controlled trials that reported significant reductions in anxiety symptomsinSADsubjectsinthenon-contingencygroupthatwere just asgreatasthoseintheABMgroup(e.g.,Boettcher,Berger, & Renneberg, 2012; Julian et al., 2012). Interestingly, another studyrevealedthatsociallyanxioussubjectsrandomizedtoeither ABM or no-contingency conditions exhibited indistinguishably largereductionsinanxietysymptoms,andlargerreductionsthan exhibitedbyparticipantsrandomizedtoawait-listgroup(Enock, Hofmann,&McNally,2014).Wealsoobservedimprovementson boththeexecutiveconflictandthealertingcomponentsof atten-tion,regardlessoftrainingcondition,consistentwithKlumppand Amir’s(2010)suggestionthattrainingincreasesattentional con-trol,irrespectiveofacontingency,andthismayenableincreased controloveranxiousthoughts.
However,wefailedtoreplicatethefindingsofAmiretal.(2008, 2009)onthespatialcueingtask.Thatis,allgroupshadfasterRTs post-trainingirrespectiveofthevalence ofcues,suggestiveofa practiceeffectalone.Therearevariouspotentialexplanationsfor ourfailuretoreplicateAmiretal.’sfindings.First,ourparticipants failedtoexhibitanABatbaseline,andAmiretal.(2011)reported foundthatABMismostpotentwithanxiouspeoplewhoexhibitthis bias.AcomparisonbetweenRTsforthreateningwordsoninvalid
versusvalidtrialsindicatednodifficultydisengagingfromthreat cuesatbaseline,andhencenoAB,t(60)=1.04,p=.30.Moreover, participantswerenotfastertorespondtothreatthannon-threat cuesonvalidtrialst(60)=1.28,p=.20, therebyexhibitingnoAB indexedbyfacilitatedattentiontothreat.We thenreranallthe analyseswithABatbaselineasacovariate,buttheresultswere essentiallyunchanged.Second,improvementintop-down regula-tionofexecutiveattentionmayproduceclinicalbenefits(Bomyea &Amir, 2011).Allgroupsexhibited thisimprovement, suggest-ing that higher-order corticalstructures,suchas theprefrontal cortex and itsfunctionally relatedstructures (e.g.,anterior cin-gulatecortex),down-regulateemotion-relevantlimbicstructures (Miller&Cohen,2001).Forinstance,prefrontalcortexandrelated structures are critically involved in down-regulating amygdala processingduringextinctionlearning(e.g.,Myers&Davis,2007; Quirk,Garcia,&Gonzalez-Lima,2006).Moreover,merelytraining executiveattention(withoutemotionalstimuli)decreasedanxiety insub-clinicalparticipants(Bomyea&Amir,2011).Ontheother hand,nonsignificantcorrelationsbetweenchangeinanxietyand executivecontrolruncountertothisinterpretation.Third, expo-suretothreatfacesmayhaveanxiolyticeffectsinallgroups.Yetthis cannotexplainwhyABMhasreducedanxietymorethancontrol proceduresdidinpreviousstudies.
Thepresentfindingsdonotsupportpreviousstudiesshowing thatABMoutperformscontrolconditions(e.g.,Amiretal.,2009;
A.Heerenetal./JournalofAnxietyDisorders29(2015)35–42
Li etal.,2008).Incontrast, theysuggest thatattentiontraining mayhelpalleviateSADregardlessofthepresenceofacontingency. Takentogether,theconclusionsofthepresentstudycombinedwith thosefrompreviousstudiesimplythatABMisnotyetreadyfor wide-scaledisseminationasamainstreamtreatment,norshouldit yetbedirectlymarketedtomentalhealthconsumers.Aspointed outbyClarke,Notebaert,andMacLeod(2014),asisthecaseforany newpsychologicalintervention,anearlyobjectiveforresearchers priortodisseminatingit,mustbetooptimizecapacitytomodify thetargetprocess,namely,ABforthreat.
Thepresentstudyhasseverallimitations.First,wedidnot col-lectfollow-updata.Assuch,onecannotdeterminewhetherthe effects we observedweremorethan transient.Second,thefact that ourparticipantsdidnotexhibit anABatbaselinecouldbe consideredasastronglimitationthatmayhavehamperedthe pos-sibilitytodetectABchanges.Indeed,ifreductioninABmediates theanxiolyticeffectsofABM,thensuchclinicalbenefits presup-poseelevatedABatbaseline(Eldaretal.,2012).Ontheotherhand, if ABM worksvia heightenedattentionalcontrol, notreduction inABperse,thentheabsenceofABatbaselinewouldnot pre-cludeanxiolyticeffectsofABM.Ourfindingsalsoraisequestions abouthowcommonABisamongSADindividuals.It ispossible thatpeopledevelopSADviapathwaysotherthanthroughAB.Itis alsopossiblethatextantproceduresforassessingABare insuffi-cientlyreliabletodetectattentionalbiasesforthreat(seeMcNally et al., 2013; Schmukle, 2005).Finally, we did not evaluatethe participants’viewsonthetwotopics forthespeechtask (abor-tion; marijuanalegalization). Accordingly,wecannotdetermine whethertheperformanceofthosewhofeltstronglyaboutthese issuesmighthavebeeninfluencedbytheintensityoftheirviews. However,theabsenceofasignificantdifferencebetweenthetopic chosenforthespeechtaskonbothBASAandSUDSatbaselineand post-training(ts<1.61,ps>.11)tendstoruleoutthishypothesis. Futurestudiesshouldfurtherexplorethisissue.
In conclusion, thepresent findings adds tosmall but grow-ingliteratureindicatingthatthemechanismsdrivingABMinSAD maybemorecomplicatedthatinitiallyassumed.AlthoughABM often reduces SADsymptoms, we foundsimilar improvements among participantsexposed to a reverse contingency and toa no-contingencycondition.Further,thepresentstudywasthefirst toassesstheimpactofattentiontraining onthethreeattention networksassessedbytheANT.Participantsexhibited,regardless ofthetrainingtheyreceived,animprovementonthealertingand executivecomponentsofattention.Futurestudiesareneededto identifythemoderatorsinfluencingwhenABMoutperforms con-trolconditions.
Acknowledgments
ThisworkwassupportedinpartbyaJointResearchGrantfrom theBelgianFrenchCommunity(awardedtoPierrePhilippot)and bypost-doctoralresearchfellowgrantfromtheBelgianNational FundsforScientificResearch“F.R.S.–FNRS”(awardedto Alexan-dreHeeren).Thewritingofthispaperalsoreceivedthesupport fromtheBelgianFoundation“Vocatio”andtheBelgianFrench Com-munityGrantforScientificExcellence(bothawardedtoAlexandre Heeren).Thesefoundationsdidnotexertanyeditorialinfluence overthisarticle.
References
AmericanPsychiatricAssociation.(1994).Diagnosticandstatisticalmanualofmental disorders(4thed.).Washington,DC:Author.
Amir,N.,Beard,C.,Taylor,C.T.,Klumpp,H.,Elias,J.,Burns,M.,etal.(2009).Attention traininginindividualswithgeneralizedsocialphobia:arandomizedcontrolled trial.JournalofConsultingandClinicalPsychology,77,961–973.
Amir,N.,Elias,J.,Klumpp,H.,&Przeworski,A.(2003).Attentionalbiastothreatin socialphobia:facilitatedprocessingofthreatordifficultydisengagingattention fromthreat?BehaviourResearchandTherapy,41,1325–1335.
Amir,N.,Taylor,C.T.,&Donohue,M.C.(2011).Predictorsofresponsetoan atten-tionmodificationprogramingeneralizedsocialphobia.JournalofConsultingand ClinicalPsychology,79,533–541.
Amir,N.,Weber,G.,Beard,C.,Bomyea,J.,&Taylor,C.T.(2008).Theeffectsofa single-sessionattentionmodificationprogramonresponsetoapublic-speaking challengeinsociallyanxiousindividuals.JournalofAbnormalPsychology,117, 860–868.
Beck,A.T.,Steer,R.A.,&Brown,G.K.(1996).BeckDepressionInventorymanual(2nd ed.).SanAntonio,TX:PsychologicalCorporation(Frenchadaptation,1998.Paris, France:EditionsduCentredePsychologieAppliquée).
Boettcher,J.,Berger,T.,&Renneberg,B.(2012).Internet-basedattentiontrainingfor socialanxiety:arandomizedcontrolledtrial.CognitiveTherapyandResearch,36, 1–15.
Boettcher,J.,Leek,L.,Matson,L.,Holmes,E.A.,Browning,M.,MacLeod,C.,etal. (2013).Internet-basedattentionbiasmodificationforsocialanxiety:a random-izedcontrolledcomparisonoftowardsnegativeandtrainingtowardspositive cues.PLoSONE,8,e71760.
Bomyea,J.,&Amir,N.(2011).Theeffectsofanexecutivefunctioningtraining pro-gramonworkingmemorycapacityandintrusivethoughts.CognitiveTherapy andResearch,35,529–535.
Bruchon-Schweitzer,M.,&Paulhan,I.(1993).Adaptationfrancophonedel’inventaire d’anxiétéTrait-Etat(FormeY)deSpielberger.Paris,France:EditionsduCentre PsychologieAppliquée.
Bunnell,B.E.,Beidel,D.C.,&Mesa,F.(2013).Arandomizedtrialofattentiontraining forGeneralizedSocialPhobia:doesattentiontrainingchangesocialbehavior?
BehaviorTherapy,44,662–673.
Carlbring,P.,Apelstrand,M.,Sehlin,H.,Amir,N.,Rousseau,A.,Hofmann,S.G.,etal. (2012).Internet-deliveredattentionbiasmodificationtraininginindividuals withsocialanxietydisorder–adoubleblindrandomizedcontrolledtrial.BMC Psychiatry,12,66.
Clarke,P.J.F.,Notebaert,L.,&MacLeod,C.(2014).Absenceofevidenceor evi-denceofabsence:reflectingontherapeuticimplementationsofattentionalbias modification.BMCPsychiatry,14,8.
Derryberry,D.,&Reed,M.A.(2002).Anxiety-relatedattentionalbiasesandtheir regulationbyattentionalcontrol.JournalofAbnormalPsychology,111,225–236.
Eldar,S.,Apter,A.,Lotan,D.,Edgar,K.P.,Naim,R.,Fox,N.A.,etal.(2012). Atten-tionbiasmodificationtreatmentforpediatricanxietydisorders:arandomized controlledtrial.AmericanJournalofPsychiatry,169(2),213–220.
Emmelkamp,P.M.G.(2012).Attentionbiasmodification:theEmperor’snewsuit?
BMCMedicine,10,63.
Enock,P.M.,Hofmann,S.G.,&McNally,R.J.(2014).Attentionbiasmodification treatmentviasmartphonetoreducesocialanxiety:arandomizedcontrolled trial.CognitiveTherapyandResearch,38,200–216.
Fan,J.,McCandliss,B.D.,Fossella,J.,Flombaum,J.I.,&Posner,M.I.(2005).The activationofattentionalnetworks.NeuroImage,26,471–479.
Fan,J.,McCandliss,B.D.,Sommer,T.,Raz,A.,&Posner,M.I.(2002).Testingthe efficiencyandindependenceofattentionalnetworks.JournalofCognitive Neu-roscience,14,340–347.
Heeren,A.,DeRaedt,R.,Koster,E.H.W.,&Philippot,P.(2013).The(neuro)cognitive mechanismsbehindattentionbiasmodificationinanxiety:proposalsbasedon theoreticalaccountsofattentionalbias.FrontiersinHumanNeuroscience,7,119.
Heeren,A.,Lievens,L.,&Philippot,P.(2011).Howdoesattentiontrainingwork insocialphobia:disengagementfromthreatorreengagementtononthreat?
JournalofAnxietyDisorders,25,1108–1115.
Heeren,A.,Maurage,P.,Rossignol,M.,Vanhaelen,M.,Peschard,V.,Eeckhout,C.,etal. (2012).Theself-reportversionoftheLiebowitzSocialAnxietyScale: psychomet-ricpropertiesoftheFrenchversion.CanadianJournalofBehaviouralScience,44, 99–107.
Heeren,A.,Peschard,V.,&Philippot,P.(2012).Thecausalroleofattentionalbiasfor threatcuesinsocialanxiety:atestonacyber-ostracismtask.CognitiveTherapy andResearch,36,512–521.
Heeren,A.,Reese,H.E.,McNally,R.J.,&Philippot,P.(2012).Attentiontraining towardandawayfromtreatinsocialphobia:effectsonbehavioural, subjec-tive,andphysiologicalmeasuresofanxiety.BehaviourResearchandTherapy,50, 30–39.
Julian,K.,Beard,C.,Schmidt,N.B.,Powers,M.B.,&Smits,J.A.J.(2012). Atten-tiontrainingtoreduceattentionbiasandsocialstressorreactivity:anattempt toreplicateandextendpreviousfindings.BehaviourResearch&Therapy,50, 350–358.
Klumpp,H.,&Amir,N.(2010).Preliminarystudyofattentiontrainingtothreatand neutralfacesonanxiousreactivitytoasocialstressorinsocialanxiety.Cognitive TherapyandResearch,34,263–271.
Lecrubier,Y.,Weiller,E.,Bonora,L.I.,Amorin,P.,&Lépine,J.-P.(1998).French adap-tationoftheMiniInternationalNeuropsychiatricInterview(MINI5.0.0).Internal report.Paris,France:UnitéINSERM302,HôpitaldelaSalpétrière.
Li,S.W.,Tan,J.Q.,Qian,M.Y.,&Liu,X.H.(2008).Continualtrainingofattentional biasinsocialanxiety.BehaviourResearch&Therapy,46,905–912.
Liebowitz,M.R.(1987).Socialphobia.ModernProblemsofPharmacopsychiatry,22, 141–173.
Lundqvist,D.,Flykt,A.,&Öhman,A.(1998).TheKarolinskaDirectedEmotionalFaces (KDEF).Stockholm:DepartmentofNeurosciences,KarolinskaHospital.
MacLeod,C.,Mathews,A.,&Tata,P.(1986).Attentionalbiasinemotionaldisorders.
42 A.Heerenetal./JournalofAnxietyDisorders29(2015)35–42
MacLeod,C.,Rutherford,E.,Campbell,L.,Ebsworthy,G.,&Holker,L.(2002). Selec-tiveattentionandemotionalvulnerability:assessingthecausalbasisoftheir associationthroughtheexperimentalmanipulationofattentionalbias.Journal ofAbnormalPsychology,111,107–123.
McNally,R.J.,Enock,P.M.,Tsai,C.,&Tousian,M.(2013).Attentionbias mod-ification for reducing speechanxiety. Behaviour ResearchandTherapy, 51, 882–888.
Miller,E.K.,&Cohen,J.D.(2001).Anintegrativetheoryofprefrontalcortexfunction.
AnnualReviewofNeuroscience,24,167–202.
Mogg, K., Philippot, P., & Bradley, B. P. (2004). Selective attention to angry facesin clinicalsocial phobia.Journal of AbnormalPsychology, 113, 160–165.
Moriya,J.,&Tanno,Y.(2009).Dysfunctionofattentionalnetworksfornon-emotional processinginnegativeaffect.Cognition&Emotion,23,1090–1105.
Mulac,A.,&Sherman,R.(1974).Behaviouralassessmentofspeechanxiety.Quarterly JournalofSpeech,60,134–143.
Myers,K.M.,&Davis,M.(2007).Mechanismsoffearextinction.MolecularPsychiatry,
12,120–150.
New,B.,Pallier,C.,Ferrand,L.,&Matos,R.(2001).Unebasededonnéeslexicales dufranc¸aiscontemporainsurinternet:LEXIQUE.L’AnnéePsychologique,101, 447–462.
Posner,M.I.,&Petersen,S.E.(1990).Theattentionsystemsofthehumanbrain.
AnnualReviewofNeuroscience,13,25–42.
Quirk,G.J.,Garcia,R.,&Gonzalez-Lima,F.(2006).Prefrontalmechanismsin extinc-tionofconditionedfear.BiologicalPsychiatry,60,337–343.
Ratcliff,R.(1993).Methodsfordealingwithreactiontimeoutliers.Psychological Bulletin,114,510–532.
Schmidt,N.B.,Richey,J.A.,Buckner,J.D.,&Timpano,K.R.(2009).Attention train-ingforgeneralizedsocialanxietydisorder.JournalofAbnormalPsychology,118, 5–14.
Schmukle,S.C.(2005).Unreliabilityofthedotprobetask.EuropeanJournalof Per-sonality,19,595–605.
Spielberger,C.D.,Gorsuch,R.L.,Lushene,R.,Vagg,P.R.,&Jacobs,G.(1983).Manual forthestate-traitanxietyinventory.PaloAlto,CA:ConsultingPsychologistsPress.
Stormark, K.M.,Nordby, H.,&Hugdahl, K.(1995).Attentionalshiftsto emo-tionally chargedcues: behavioural andERP data. Cognition &Emotion, 9, 507–623.
VanBockstaele,B.,Verschuere,B.,Tibboel,H.,DeHouwer,J.,Crombez,G.,&Koster, E.H.W.(2014).Areviewofcurrentevidenceforthecausalimpactofattentional biasonfearandanxiety.PsychologicalBulletin,140,682–721.
Wolpe,J.(1958).Psychotherapybyreciprocalinhibition.Stanford,CA:Stanford Uni-versityPress.
