How emotional arousal raises the stakes of mental competition Mara

21 Pages·2012·0.51 MB·English

How emotional arousal raises the stakes of mental competition

Mara Mather

Much research indicates that experiencing emotional arousal (such as when hearing a
loud gunshot) modifies perception, memory encoding and decision processes. But
effects of arousal range from enhancing to impairing across different paradigms and
stimuli and canonical models of emotional arousal cannot explain both the enhancing
and impairing effects. In this talk, I will make the case that arousal biases neural
competition to enhance high priority information and suppress low priority information.
PerspectivesonPsychologicalScience
Arousal-Biased Competition in 6(2)114–133
ªTheAuthor(s)2011
Reprintsandpermission:
Perception and Memory
sagepub.com/journalsPermissions.nav
DOI:10.1177/1745691611400234
http://pps.sagepub.com
Mara Mather1 and Matthew R. Sutherland2
1DavisSchoolofGerontologyandDepartmentofPsychology,UniversityofSouthernCalifornia,LosAngeles
2DepartmentofPsychology,UniversityofSouthernCalifornia,LosAngeles
Abstract
Oureverydaysurroundingsbesiegeuswithinformation.Thebattleisforashareofourlimitedattentionandmemory,withthebrain
selectingthewinnersanddiscardingthelosers.Previousresearchshowsthatbothbottom–upandtop–downfactorsbiascompetition
infavor ofhigh prioritystimuli. We proposethat arousal duringan eventincreases thisbias both in perceptionandin long-term
memoryoftheevent.Arousal-biasedcompetitiontheoryprovidesspecificpredictionsaboutwhenarousalwillenhancememoryfor
eventsandwhenitwillimpairit,whichaccountsforsomepuzzlingcontradictionsintheemotionalmemoryliterature.
Keywords
arousal, emotional memory, biasedcompetition, attention
Selectionistheverykeelonwhichourmentalshipisbuilt.And weareawareofcanaccountforthemall.Furthermore,evenat
inthiscaseofmemoryitsutilityisobvious.Ifweremembered thespecificlevel,eachseparatearousaleffectcharacterization
everything,weshouldonmostoccasionsbeasilloffasifwe (such as ‘‘arousal induces retrograde amnesia’’) faces contra-
rememberednothing. dictory findings in the literature (see Table 1). In this article,
—WilliamJames,ThePrinciplesofPsychology(1890,p.680) we propose a new theory of arousal-biased competition to
account for how arousal affects memory selectivity. We start
The brain’s ability to prioritize information allows us to
by outlining the theoryand how it operates in perception. We
think and take action without being overwhelmed by external
then turn to the domain of memory to explain how the theory
stimuli or internal thoughts and feelings. Attending to what is
accountsfortheeclecticandapparentlycontradictoryfindings
importantwhileignoringextraneousdetailcanenhanceperfor-
outlinedinTable1.
manceinchallengingsituations,suchasfacingadifficulttask
or a threat to one’s safety. A wide range of cognitive and
emotionalchallengesincreaseautonomicarousal,affectingheart
Arousal-Biased Competition
rate,galvanicskinresponse,andpupildilation.Evensomething
assimpleasanemotionalpictureshownforafewsecondscan What we call arousal-biased competition (ABC) is the notion
increase autonomic arousal (e.g., Bradley, Miccoli, Escrig, & that arousal (whether elicited by external stimuli, internal
Lang, 2008). Emotional stimuli and cognitive challenges also thoughts, or stress hormones) modulates the strength of com-
increase levels of stress hormones such as epinephrine and peting mental representations, enhancing memory for items
cortisol. thatdominatethecontestforselectiveattention.Thiscompeti-
Given the importance of focused attention in challenging tionforrepresentationbeginsduringperceptionandcontinues
situations, it would be helpful if arousal increased how selec- intolong-termconsolidation.Duringperception,arousalbiases
tive the brain is when processing information. Many studies competition in favor of perceptually conspicuous or goal-
haveexaminedtheroleofarousalintheselectivityofinforma- relevantstimuli.Arousalthenenhancesmemoryconsolidation
tionprocessing,elicitingarousalinavarietyofways,including
via emotional stimuli, stress, and administration of stress hor-
mones. Some prominent characterizations of how arousal CorrespondingAuthor:
Mara Mather, 3715 McClintock Ave, University of Southern California, Los
affects the selectivity of memory are listed in Table 1. These
Angeles,CA90089
arousaleffectsonmemoryarediverse,andnoprevioustheory
E-mail:[email protected]
Arousal-Biased Competition 115
Table1.ProminentCharacterizationsofHowArousalAffectsMemory,WithRelatedFindings
Arousaleffect Finding(s)illustratingtheeffect Inconsistentfinding(s)
1.Arousalleadsto Emotionalarousalenhancesmemoryforcentraldetailsat Thematicallyinducedemotionenhancesmemoryfor
memorynarrowing thecostofperipheraldetails.Thishasbeeninterpreted noncentralinformation(Laneyetal.,2004).Also,an
asevidencethatarousalcausesmemorynarrowing arousingpictureshowninonelocationonthescreen
(Burke,Heuer,&Reisberg,1992;Christiansonetal., impairsmemoryforbackgroundinformationmore
1991).Arelatedphenomenaistheweapon-focus thanforspatiallyperipheralbutforegroundinforma-
effect,inwhichpeopleremembertheweaponwhen tion(Matheretal.,2009).
witnessingarealorsimulatedcrimebutforgetother
scenedetails(Steblay,1992).
2.Emotionalarousal Interspersingemotionalslidesthroughoutaslideshow Memoryisbetterforthespecificdetailsofemotional
enhancesmemoryfor increasesthedegreetowhichparticipantsrecall objectsthanforthedetailsofnonemotionalobjects
gistbutnotdetail thegistratherthanthedetailsofcertaintarget (Kensingeretal.,2006),andwhetheremotion
(nonmanipulated)slides,indicatingthatemotional enhancesorimpairsmemoryforthegistanddetails
arousalenhancesmemoryforthegistbutnotdetailsof dependsonhowattentionwasdirectedduring
events(Adolphsetal.,2005). encoding(Kensinger,Garoff-Eaton,&Schacter,2007).
3.Arousalenhances Emotionalarousalenhancesmemoryofintrinsicfeatures Participantsaskedtolearnwordpairsandthengiventhe
within-objectmemory ofanobjectbutdoesnotenhancememoryforasso- firstwordinthepairasacuetoretrievethesecond
binding ciationsbetweenitems(Kensinger,2009;Mather, wordwerebetteratrecallingneutralwordspaired
2007). withemotionallyarousingwords(Guillet&Arndt,
2009).
4.Arousalcreatesret- Arousingwordsorpicturescanleadtoretrograde Arousingpicturescanleadtoretrogradeenhancement,in
rogradeamnesia amnesia,inwhichprecedingneutralwordsorpictures whichprecedingpicturesaremorelikelytoberecalled
aremorelikelytobeforgotten(Knight&Mather,2009; (Andersonetal.,2006;Knight&Mather,2009).
Strangeetal.,2003).
5.Arousalenhances Postencodingarousalenhancesmemoryforemotional Arousalexperiencedafterexposuretoneutralinforma-
consolidationfor stimulimorethanforneutralstimuli,indicatingthat tioncanenhancememoryforthatneutralinformation
emotionalitems arousalenhancesmemoryconsolidationofemotional, whentestedacoupleofdaysoraweeklater(Aber-
butnotneutral,information(Buchanan&Lovallo, crombie,Kalin,Thurow,Rosenkranz,&Davidson,
2001;Cahilletal.,2003). 2003;Andersonetal.,2006;Andreano&Cahill,2006;
Knight&Mather,2009;Nielson&Powless,2007;
Nielsonetal.,2005;seealsoMaheuetal.,2004).
Note.Seetextforhowarousal-biasedcompetitionaccountsforconsistentandinconsistentfindings.
forthemostconspicuousorgoal-relevantstimuli,regardlessof features targeted for selective attention cannot always be
whetherthosestimuliwerearousingornot.ABCtheorybuilds defined based on feature locations. Instead, attention is often
on existing biased-competition models of attention that we objectbased(forareview,seeScholl,2001).So,forinstance,
describe inthenext section. when two transparent objects overlap, attention may be
directedtooneobject,impairingprocessingoftheotherobject
(e.g., O’Craven, Downing, & Kanwisher, 1999; Valdes-Sosa,
Biased Competition in Nonarousing
Cobo,&Pinilla, 2000).
Situations
The biased-competition theory of attention accounts for
Imagine scanning a crowd of faces to find a friend. How can theseobject-basedeffectsbyassumingthatobjectsinthevisual
youfindaspecificface?Bothbottom–upcuesfromthescene field compete for neural representation (Bundesen, 1990;
(suchasahandwaving)andtop–downgoals(suchasfindinga Bundesen, Habekost, & Kyllingsbaek, 2005; Deco & Rolls,
friend with blonde hair) can bias attention toward particular 2005; Desimone, 1998; Desimone & Duncan, 1995; Kastner
faces,increasingtheprocessingresourcesdevotedtothatface & Ungerleider, 2001; Miller & Cohen, 2001). Biased-
(Beck & Kastner, 2009). Once the friend’s face is located, competitiontheoryproposesthreebasicideas(Beck&Kastner,
anotherchallengeistomaintainthatface’srepresentationand 2009; Duncan, 2006). First, the competitive nature of visual
relative location despite potential distraction from the rest of processing means that a stronger neural response to any one
thescene.Inthissection,wereviewresearchonselectiveatten- visualobjectcomesattheexpenseofweakerresponsestooth-
tion andbiased competitioninnonarousing situations. ers.Second,top–downgoalsorsignalsbiascompetition.Third,
One intuitive metaphor of selective attention is that of a competitionis integrated across brainregions, so thata visual
spotlight, in which attention is focused on one region of the objectthatdominatesinthevisualcortexwilllikelydominate
visualfield,withstimulioutsidethatregionhavinglittleinflu- inotherregionssuch as theprefrontal andparietalcortices.
ence(Posner, Snyder,&Davidson,1980).However,there are Biasedcompetitionhasbeenobservedinsinglecellrecord-
many ways in which selective attention does not conform to ings from the inferotemporal cortex of monkeys. This high-
this metaphor (Cave & Bichot, 1999). For instance, the set of level visual area has neurons that selectively respond to
116 Mather and Sutherland
complex object properties, contributing to object recognition. Routes to Stimuli Priority
Asingle-cellrecordingstudyidentified,foreachrecordedcell,
InABC,aparticularitem’sprioritydetermineswhetherarou-
apicturethatproducedastrongresponse(PictureA)andapic-
sal will enhance or impair perception of that item. Arousal
turethatproducedaweakornoresponse(PictureB;Chelazzi,
amplifies the effects of competition, improving perception
Duncan, Miller, & Desimone, 1998). When Picture A was
ofhighpriorityinformationandweakeningperceptionoflow
shown alone, the cell increased its firing rate dramatically.
priorityinformation.Thisleadstothequestionofwhatdeter-
However,whenAandBwereshowntogether,thecellinitially
minespriority.Asdescribedbelow,bottom-upsensoryinflu-
showedafiringrateinbetweenthatseenforAandBbythem-
ences(thewavinghandinthecrowd)andtop–downcognitive
selves.Ashorttime(300ms)later,thefiringratedivergedfor
factorssuchasgoalrelevanceandexpectations(thesearchfor
the2-picturepresentationsdependingonthemonkey’scurrent
a blond-haired friend) can independently determine priority
goal.IfPictureAwasthetarget,thefiringrateforthatcellrose
(forareview,seeFecteau&Munoz,2006).Otherfactorsthat
tosamerateaswhenAwaspresentedbyitself.IfPictureBwas
predictpriority,suchasunexpectedness,emotionalrelevance,
thetarget,thefiringratedecreasedtothesamerateaswhenB
and social relevance, reflect interactions of sensory and cog-
was presented byitself.
nitive signals.
These data illustrate two important points. The first is that
competitionfromPictureBinterfereswiththeresponsetothe
preferred picture, as indicated by the lower initial response to Bottom–up
Picture A in the two-picture presentation than in the A-only
presentation. The second important point is that the competi- The bottom–up route to priority is via perceptual contrast.
tion in the two-picture presentation conditions is biased by Targetsinanarray‘‘popout’’whentheydifferfromtheircon-
top–down goals, so that more behaviorally relevant informa- text, whether that difference is in orientation, motion, lumi-
tion winsthecompetition for representation. nance, or color (Nothdurft, 2000). Using this fundamental
fMRI studies alsoreveal neural response patterns consistent ruleaboutwhatdrawsattention,computationalmodelscanpre-
with biased competition (e.g., Kastner, De Weerd, Desimone, dictwhereeyefixationswilllandonanimagebyanalyzingthe
& Ungerleider, 1998; Reddy, Kanwisher, & VanRullen,2009). image’s center-surround contrast at every location (Berg,
As in the single-cell recording studies (Chelazzi et al., 1998; Boehnke, Marino, Munoz, & Itti, 2009; Itti & Koch, 2000).
Chelazzi, Miller, Duncan, & Desimone, 2001), directing atten- In these salience map models, the center-surround contrast is
tion to one of multiple objects reduces the suppressive effects computedfor avarietyofdifferentfeatures(e.g.,color,inten-
of the competing objects, bringing activation closer to levels sity,orientation,flicker,andmotion)atdifferentspatialscales
seen with just one object presented alone (Kastner et al., and all the separate estimates of within-feature contrast are
1998).Duringdirectedattention,afrontoparietalattentionalnet- assembled into a global saliency map. Similar contrast algo-
work (including the anterior cingulate cortex; ACC) generates rithmscanpredictattentiontoauditorystimuli(Kayser,Petkov,
top–down biasing signals that modulate activity in sensory Lippert,&Logothetis,2005).
regions (Dehaene, Kerszberg, & Changeux, 1998; Hampshire, In these models, a center-surround differentiation process
Duncan, & Owen, 2007; Hon, Epstein, Owen, & Duncan, mimics properties oflocalcortical inhibition. When the con-
2006;Hung,Driver,&Walsh,2005;Miller&Cohen,2001). trast is high between the center and the surround, activation
Competition for mental representation continues after sti- for the center is increased and activation for the surround is
muli disappear. For instance, in one study, participants saw a inhibited. The result of several iterations of this competitive
face and a scene next to each other on the screen (Johnson & processisthatonlyafewlocationsonthefeaturemapsremain
Johnson, 2009a). After the pictures disappeared, participants active (Fig. 1A). However, if the image consists of a set of
were cued to think about one of the two pictures. When com- itemswithequalcontrast(Fig.1B),thecenter-surroundcom-
pared with a control condition in which they were not asked petitionprocessleadsallpeakstoinhibiteachothersuppres-
to refresh either picture, thinking about the scene increased sing the entire map.
activity in the parahippocampal place area, a scene-selective ABC predicts that arousal will amplify the effects of
area,whereasthinkingaboutthefacesuppressedactivityinthe contrast by boosting excitation for the highest contrast item
parahippocampal place area (see also K.J. Mitchell, Johnson, and inhibition of the surrounding context, leading to an even
Higgins,&Johnson,2010).Thus,top–downgoalsbiascompe- sparsersetofactivelocationsinthesaliencymapthanunder
tition among mental representations even when there are no nonarousing conditions. But those that do survive will have
external perceptual stimuli. greater activation levels.
The selectivity supported by biased competition should be
particularlyimportantduringstressfulorthreateningsituations,
Top–down
whendistractionsmaybedangerous.Indeed,anumberofstud-
ies provide evidence that the arousal from stress or a threat Priorityisalsodeterminedbytop–downfactorssuchasgoals,
biases competition in perception and attention. However, knowledge,andexpectations.Inparticular,stimulithatarerel-
beforeturningtotheevidenceforABC,wefirstdiscussprior- evant to current goals gain priority over irrelevant stimuli.
ity, akeyconcept for ourtheory. Many theoretical models posit that goal-directed attention
Arousal-Biased Competition 117
Fig.1.OutputfromIttiandKoch’s(2000)computationalmodeltoderiveasaliencymapfromfea-
turecontrast.InPanelA,onestronglyactivatedlocationissurroundedbyweakerones.Iterative
spatial competition leads this location to gain further strength while suppressing surrounding
regions.InPanelB,numerousstronglyactivatedlocationsinhibiteachother,leadingtosuppression
ofalllocations.FigureadaptedfromIttiandKoch(2000).
enhances and suppresses representations of stimuli during Emotional relevance
perception (forareview,see Walther&Koch,2007).Consis-
Emotionalrelevance isanother contributortopriority thathas
tent with such models, researchers have found that directing
bottom–up and top–down components. Many studies demon-
people to remember scenes and ignore faces while they look
strate that emotional stimuli stand out more than neutral sti-
atpicturesofeachleadstoabove-baselineactivityinthepara-
muli. For instance, when a neutral and an arousing picture
hippocampal place area, a scene-selective brain region,
aresimultaneouslypresented,one’seyesaremorelikelytofirst
whereas directing them to ignore scenes and remember faces
fixate on the arousing picture and then fixate more frequently
leads to below-baseline activity in the parahippocampal place
on it (Knight et al., 2007; LaBar, Mesulam, Gitelman, &
area (Gazzaley, Cooney, McEvoy, Knight, & D’Esposito,
Weintraub, 2000; Rosler et al., 2005). This bias favoring the
2005).Thisrevealsbothtop–downenhancement andsuppres-
arousing picture in a neutral–arousing pair even occurs when
sionofmentalrepresentationsoftheperceivedscenes.Further-
participants are directly instructed to ignore the arousing pic-
more, internal representations of previously perceived stimuli
tures (Nummenmaa, Hyona, & Calvo, 2006). Viewing emo-
arealsoenhancediftheyaregoalrelevantorsuppressedifthey
tional stimuli also increases activity in visual brain areas
aregoalirrelevant(e.g.,Johnson&Johnson,2009b).ABCpre-
associated with object recognition, such as the fusiform and
dictsthatarousalwillincreasepriorityforgoal-relevantstimuli
inferotemporal cortices (Sabatinelli, Flaisch, Bradley,
anddecrease priority forgoal-irrelevant stimuli.
Fitzsimmons, & Lang, 2004; Taylor, Liberzon, & Koeppe,
2000), and it leads to an early posterior negativity in event-
Surprise
relatedbrainpotentialstudies(Schupp,Flaisch,Stockburger,&
Although perceptual contrast and goal relevance are indepen- Junghofer, 2006), suggesting prioritized visual processing of
dent factors, other predictors of priority are determined by emotional stimuli. Similar results have been observed in other
interactionsofbottom–upperceptionandtop–downreflection. sensorysystems:Emotionallyevocativesoundsincreaseactivity
Forinstance,novelandunexpected stimuliareprioritizeddue in the auditory cortex (superior temporal gyrus; Zald & Pardo,
tothemismatchbetweenperceptualinputandpriorknowledge 2002)andtastingpleasantoraversiveliquidsolutionsincreases
(Itti&Baldi,2009;Ranganath&Rainer,2003).Analogousto activity inthe taste cortex(fronto-opercular insula; O’Doherty,
salience maps for perceptual contrast, a computational model Rolls,Francis,Bowtell,&McGlone,2001).
can quantifyhow surprising each location on a feature map is Debate about why emotional stimuli have enhanced prior-
bycomparingbeliefsaboutwhatislikelytobeinthatlocation ity has yielded evidence for both bottom–up and top–down
before andafter seeing the information (Baldi&Itti, 2010). mechanisms. For instance, a study in which participants
118 Mather and Sutherland
Fig.2.A:ThesequenceforeachtrialinSutherlandandMather(2011).B:Arousingsounds
increasedtheproportionofdarklettersreportedanddecreasedtheproportionoflight
lettersreported.
detectedoneemotionalfaceinanarrayofneutralfacesfound Evidence for ABC in Perception
that computationally modeled bottom–up visual saliency
Inthissection,wefocusonABCinperception,reviewingevi-
(usingthemodelshowninFig.1;Itti&Koch,2000)predicted
dence that arousal amplifies the bottom–up salience and the
face detection speed (Calvo & Nummenmaa, 2008). A study
top–down competitive advantage of high and low priority sti-
examining event-related brain potentials in response to
muli.We discuss how ABCcan account for arousal effects in
difficult-to-see faces reveals top–down influences (Lee
binocular rivalry and tracking multiple items, as well as the
et al., 2010). When participants looking at a neutral face
brain mechanisms of how arousal modulates top–down
thought they saw an emotional face, activity in posterior
priority.
visualregionsshowedthesametypeofenhancedactivityseen
during actual emotional face perception. Thus, due to a mix-
ture of factors, including bottom–up perceptual contrast and ABC amplifies perceptual contrast
top–down cognitive factors, emotional stimuli are likely to
Totestwhetherarousalincreasesthecompetitiveadvantageof
have high priority. Furthermore, arousal (evoked either by
high contrast (and therefore salient) over low contrast (and
those same stimuli or by another source) should increase the
thereforelesssalient)stimuli,weplayedarousingornonarous-
competitive advantage of the emotional stimuli over lower
ing sounds before briefly presenting a circular array of eight
priority stimuli.
letters(Fig.2;Sutherland&Mather,2011).Threeoftheletters
were printed in high contrast dark gray on the white back-
ground, whereas the other five letters were printed in lower
Social relevance
contrastgray.Participantsthenreportedwhichletterstheysaw.
Finally, social relevance can also increase stimulus priority. Asexpected,participantsweremorelikelytoreportthehigh
Whenaskedjusttolookatavisualscenethatincludeshuman contrast letters than the low contrast letters. More interesting,
faces, people frequently fixate on the eyes (Birmingham, however, was that hearing an arousing sound before viewing
Bischof, & Kingstone, 2009a, 2009b)—a tendency that is not theletterssignificantlyincreasedreportingofhighcontrastlet-
accounted for by computationally modeled bottom–up visual tersanddecreasedreportingoflowcontrastletters.Thus,arou-
saliency (Itti & Koch, 2000). People’s default interest in the salamplifiedthecompetitiveadvantageofperceptuallysalient
social information other people’s eyes convey may be the stimuliandthecompetitivedisadvantageoflesssalientstimuli.
mechanismunderlyingthistypeofpriority(e.g.,Birmingham, Inasecondexperiment,participantswererandomlyassignedto
Bischof, & Kingstone, 2008). Information-seeking goals may haveafixationintervalbetweenthearousingsoundandthelet-
lead eyes to attract more attention than would be predicted terpresentationthatrangedfrom750to3000ms(areplication
fromtheir perceptual contrast. ofthefirstexperiment)ortohaveanintervalthatrangedfrom
Arousal-Biased Competition 119
Fig.3.A:TrialsequencefromPhelpsetal.(2006).Participantsindicatedwhetheroneofthefour
gratings was tilted or none were tilted. B: A preceding cue varied in its spatial location and in
whether it was a fearful or neutral face. ISI ¼ interstimulus interval. Figure adapted from Phelps
etal.(2006).
4000 to 6000 ms. In the short-interval condition, the ABC to the right. The gratings were shown for only a brief period
effectreplicated,buttherewasnosignificanteffectofarousal (40 ms; see Fig. 3A). Before the gratings, Phelps et al. dis-
inthelong-intervalcondition.Thus,foronlyabriefperiodafter playedacueconsistingeitherofafearfuloraneutralfacepre-
theirpresentation,arousingauditorystimulibiascompetitionin sented in one location or in all four potential grating locations
favor ofhighcontrast visualstimuli. (Fig.3B).Ofcourse,therewasaneffectofspatialcueing:Show-
Inthatstudy,higherandlowercontraststimulicompetedfor ingjustoneface(eitherneutralorfearful)inthesamelocationas
mental representation. However, if a low contrast stimulus is theupcomingtiltedgratingincreasedaccuracy.Therewasalso
shownaloneandisthetargetofattention,ABCshouldincrease aneffectofemotion:Whenafearfulface,ratherthananeutral
its priority, as it is the most strongly activated stimulus in the face, was used to cue the relevant grating, participants could
display. Indeed, several studies indicate that presenting an identify the tilt direction at lower levels of contrast, consistent
arousing cue can make subsequent low contrast stimuli pre- with the studies reviewed above.1 But an additional intriguing
sented alone on the screen more easily perceived (Laretzaki, aspectofthisstudywasthefindingthattheemotionalenhance-
Plainis, Argyropoulos, Pallikaris, & Bitsios, 2010; Padmala mentwasgreaterwhenthecueconsistedofjustonefaceinthe
&Pessoa,2008;Phelps,Ling,&Carrasco,2006).Forinstance, samelocationastheupcomingtiltedgratingthanwhenthecue
whenparticipantswithlowtraitanxietyweretoldtoanticipate wasfourfacesofthesametype.Thustheeffectsofemotionand
receiving 1 to 3 electric shocks during a 2-min experiment attention were not only independent effects, they interacted to
block,theyshowedashorterP100latencyofthevisualevoked enhanceperceptionmorethaneithereffectdidonitsown.
responsewhenshownan8%or12%contrastgrating(Laretzaki
etal.,2010).TheP100latencywasshorterundershockantic-
ABC amplifies the effects of goal relevance
ipation for both gratings, but even shorter for the higher con-
trast (12%) grating than for the lower contrast (8%) one. It is ConsistentwithABCtheory,astudyusingevent-relatedbrain
interesting to note that participants with high trait anxiety did potentialmeasuresrevealedthatarousalamplifiestheeffectof
not show any significant effect of anticipating shock. Future goalrelevance,especiallyduringstimulusevaluationphasesof
studiesareneededtoseeiftheseparticipantswerealreadytoo processing (Schupp et al., 2007). In this study, participants
aroused at baseline to see any further effect or whether they viewed a rapid slide show of pictures consisting of erotic
might have been more internally focused rather than task pictures of people, neutral pictures of people, and scenes of
focused(theyratedtheiralertnesssignificantlylowerthandid mutilationorinjury.Participantswereeitheraskedtocountthe
thelow-trait-anxietyparticipants).AnfMRIstudyrevealedthat erotica, neutral people, or mutilation pictures. This counting
thistypeofarousal-enhancedperceptionoftargetitemsisasso- taskmanipulatedwhichtypeofpicturewasthefocusofatten-
ciated with increased activation in areas V1–V4 of the visual tion. An ERP component thought to reflect the process of sti-
cortex(Padmala&Pessoa,2008).Inthisstudy,anauditorycue mulus evaluation called the P3 (appearing about 200–350 ms
oneachtrialindicatedwhetherornotthetrialmightinvolvea after stimulus onset) was more pronounced for each stimulus
shock.Anticipatingashockenhanceddetectionoflow-contrast typewhenitwasthefocusofattention,butthisattentioneffect
gratings, an effect that was associated with greater activity in wasgreaterfortheeroticaandmutilationpicturesthanforthe
the early visual cortex. neutralpictures.
Inasimilarstudy(Phelpsetal.,2006),participantstriedto Ofparticularinterestarestudiesthatseparatethearousalor
identifywhetheroneofthefourgratingswastiltedtotheleftor stress-inducing stimulus from the competition for attention.
120 Mather and Sutherland
Thesestudiesindicatethatarousalenhancesperceptionforthe Consistent with this hypothesis, hearing an arousing word
highest priority stimulus and reduces perception for what has enhances one’sability toidentify briefly presented wordsthat
lower priority—even when the high priority stimuli are not follow, whereas seeing an arousing word impairs subsequent
themselves arousal inducing. For example, Hockey (1970b) visual word identification (Zeelenberg & Bocanegra, 2010).
had participants perform a central pursuit-tracking task, while When presented visually, the arousing stimuli are in direct
also indicating whenever one of six lamps signaled. The competition for visual processing with the target words,
lampsflankedbothsidesofthecentraltrackingtaskdisplay. whereaspresentingthestimuliintwomodalitiesreducescom-
Lamps closest to the display were considered ‘‘central,’’ petition.Whenpresentedinthesamemodality,thecompetition
whereas lamps furthest from the display were considered betweenthearousingcueandthetargetwordcanbereducedby
‘‘peripheral.’’ Some participants heard constant loud noise increasing the interval between them or by making the cue
(100 dB), whereas others heard less intense noise (70 dB). presentation extremely brief (Bocanegra & Zeelenberg,
Intheunequalcondition,thecentrallampsweremorelikely 2009a).Suchmanipulationscanfliptheeffectsofanemotional
tosignal—andwerethereforemoregoal-relevant—thanwere cue on subsequent word identification from impairment to
the peripheral lamps. Performance in the two noise condi- enhancement—allowing a demonstration of both emotion-
tions did not differ across the central and peripheral lamps induced ‘‘blindness’’and‘‘hypervision’’ inthesame study.
intheequalcondition,butintheunequalconditioninwhich Taken together, these results conflict with theoretical
centrallampsweremorelikelytosignal,noise-inducedarou- accounts that emotional arousal narrows the attentional spot-
sal led to faster responses to the central lamps and slower light, leading to enhanced processing of spatially central and
responses to the peripheral lamps. In other words, the impairedprocessingofspatiallyperipheralinformation.Instead,
noise-induced arousal increased existing biases toward the they indicate that arousal biases attention toward goal-relevant
goal-relevant lamps at the cost of the goal-irrelevant lamps or perceptually salient stimuli, regardless of their spatial loca-
but did not increase the bias toward spatially central cues tion. Unfortunately, observing this attentional bias is compli-
when spatial layout was not linked with goal relevance (see cated by the fact that the source of the arousal is often also a
also Hockey & Hamilton, 1970). prioritizedstimulusthatcompetesformentalrepresentation.
ABCtheoryindicatesthatacentralspatiallocationdoesnot
necessarily give a stimulus priority. Instead, goal relevance
ABC in binocular rivalry
interacts with the perceptual features of competing stimuli to
determine stimulus priority. Consistent with this idea, when When two different images are separately presented to each
arousalismanipulatedviaelectricshock,arousedsubjectsshow eye, one does not experience a constant blurred combination
greaterprocessingofgoal-relevantbutspatiallyperipheralcues ofthetwoimages.One’ssensoryexperienceinsteadoscillates
than do control subjects (Cornsweet, 1969).2 Moreover, when between coherent representations of each image, a phenom-
participantsareaskedtoclassifystimuliononedimension,they enon known as binocular rivalry (Blake & Logothetis, 2002).
arebetterabletoignoreirrelevantstimuluscharacteristicswhen Thedurationoftimethateitherimageoccupiesawarenesscan
stressedthanwhennotstressed(Chajut&Algom,2003).These becontrolledbyfocusingattentiontoparticularfeaturesofthe
findings suggest that arousal and stress enhance the priority of image(Chong&Blake,2006)oritsglobalconfiguration(J.F.
goal-relevant information, while diminishing the priority of Mitchell,Stoner,&Reynolds,2004).Likewise,perceptualfea-
goal-irrelevant information, which should have consequences tures,suchasvisualcontrast,alsoinfluencethelengthoftimea
forlatermemory(formoreaboutthepossibleroleofgoalrele- stimulusdominates awareness(Chong &Blake, 2006).
vance in emotional memory narrowing effects, see Levine & Emotionalarousalinfluencesthistypeofvisualcompetition.
Edelstein,2009). During rivalrous trials, emotional pictures dominate awareness
At first glance, other findings appear to contradict the idea longer thandoneutral pictures(Alpers & Pauli, 2006).Studies
thatarousalenhancesthepriorityofgoal-relevantinformation. that control for low-level perceptual differences suggest that
For instance, when participants compare the similarity of two arousal is the factor driving this perceptual bias, as grating
houses in the presence of two distracter faces, their perfor- patterns paired with electric shock (Alpers, Ruhleder, Walz,
mance drops when the facial expressions are emotional Muhlberger,&Pauli,2005)andschematicfacesdisplayingneg-
(Vuilleumier,Armony,Driver,&Dolan,2001).Butsuchresults ativeexpressions(Alpers&Gerdes,2007)dominateawareness
areexpectedwhenanarousal-elicitingstimulusisindirectcom- for longer than do equivalent nonarousing stimuli. Moreover,
petition with the task-relevant stimulus, because emotionally people are aware of grating patterns superimposed on faces
arousing stimuli themselves have high priority. ABC predicts expressingpositiveornegativeexpressionsforlongerthangrat-
thatifarousalisinducedinawayinwhichthearousingstimulus ings superimposed on neutral faces (Bannerman, Milders, De
is not in direct competition with the task-relevant stimuli, then Gelder,&Sahraie,2008,Fig.4).Theseeffectscanbeobserved
the processing of neutral goal-relevant stimuli should be foremotionalstimuliofeithervalence,andadirectcomparison
enhanced (e.g., Chajut & Algom, 2003; Cornsweet, 1969; of valence and arousal suggests that arousal is the emotional
Hockey, 1970b), and the processing of less relevant stimuli componentdrivingthisperceptualbias(Sheth&Pham,2008).
shouldbereduced(e.g.,Callaway&Thompson,1953;Hockey, HowdoesABCtheoryaccountfortheseeffects?Inthepro-
1970a;Nobata,Hakoda,&Ninose,2010). cedure depicted in Fig. 4, participants were told to focus only
Arousal-Biased Competition 121
Fig.4.ExamplestimuliusedinBannermanetal.(2008).Stimulus(a)waspresentedtooneeyewhile
stimulus(b),(c),or(d)waspresentedtotheothereye.
on the gratings appearing in the white circle and to indicate are top–down attentional goals represented? Much research
the grating orientations. Thus, participants in that study had suggests that a frontoparietal attention system comprised of
atop–downgoalthatgavethegratingspriority.Whenfocus- parts of the intraparietal cortex and superior frontal cortex
ing on a particular grating, if the background face evoked plays a key role in anticipating and implementing goal-
arousal(i.e.,hadanemotionalexpression),thatincreasedthe directed selection both in attention (Corbetta & Shulman,
competitiveadvantageofthatgratingovertheonebeingpre- 2002) and in memory retrieval (Cabeza, Ciaramelli, Olson,
sented to the other eye. & Moscovitch, 2008).
As previously stated in this article, emotional arousal
amplifies the effects of top–down attentional goals (Phelps
Arousal makes it more difficult to keep track of
et al., 2006; Schupp et al., 2007). One possibility is that the
multiple items
amygdaladrivesthesearousaleffectsbyenhancingtheimpact
Inthestudiesreviewedsofar,stimulieithervariedinpriorityor of top–down attentional goals in the frontoparietal network.
justonestimuluswasthefocusofattentionatatime,allowing Some initial evidence consistent with this possibility comes
someinformationtobenefitfromABC.However,havingmul- fromanfMRIstudyinwhichparticipantswereaskedtodetect
tiple high priority representations competing with each other afaceandascenepresentedneareachotherinarapidstreamof
can lead to overall suppression, as illustrated in Figure 1B. visual images (Lim, Padmala, & Pessoa, 2009). Some of the
ABCtheorypredictsthatarousalincreasessuchmutualinhibi- target faces and scenes were previously conditioned to be
tory effects. To maintain multiple representations in working arousing and some to be neutral. As in previous studies
memory,onemustcontinuouslycycle throughthem toensure (Anderson & Phelps, 2001; De Martino, Kalisch, Rees, &
thateachoneremainsactive.Whenexperiencingarousal,each Dolan, 2009), participants were less likely to show an atten-
time a stimulus representation is refreshed, there is an addi- tional blink (inability to detect the second target) when the
tionalcostforcompetingstimuli.Ifallofthestimulihaveequal secondtargetwasarousingthanwhenitwasnot,eventhough
priority, the result will be mutual inhibition and an overall inthisstudy,thearousingstimuliwereperceptuallyidenticalto
decrement inworkingmemory performance. the nonarousing stimuli. Furthermore, when participants
Consistent with this possibility, after viewing a slide show correctly identified the targets, they showed greater amygdala
ofnegativearousingpictures,participantswereworseattrack- activation for those arousing targets than for nonarousing
ingthelocationofmultiplemovingdotsthanwereparticipants targets. Most interesting, however, was that a trial-by-trial
whoviewedaslideshowofneutralpictures(Morelli&Burton, analysisofactivityrevealedthattheinfluenceoftheamygdala
2009).Likewise,whenparticipantssawasequenceoffourpic- onvisualcorticalresponseswaspartiallymediatedbythemed-
tures, eachappearinginadifferentlocation,theirmemoryfor ialfrontalgyrus—apartofthefrontoparietalattentionnetwork.
the picture-location pairs over a short delay was worse if the Thus,theamygdalamaymodulateactivityinthefrontoparietal
four pictures were arousing (either positive or negative) than attentionnetworktobiasattentiontowardhighprioritystimuli
iftheywereneutral(Matheretal.,2006;K.J.Mitchell,Mather, (seealsoMohanty,Egner,Monti,&Mesulam,2009).Thispos-
Johnson, Raye, & Greene, 2006). These findings suggest that sibilityfitswithevidencethattheamygdalaactivateswhensti-
arousalmakesitmoredifficulttomaintainmultiplerepresenta- muliarepotentiallygoalrelevant(e.g.,Cunningham,Raye,&
tionsofequal priority inworkingmemory. Johnson,2005;Sander,Grafman,&Zalla,2003)orhavehigh
perceptualsaliencewithinanemotionalcontext(Attar,Muller,
Andersen, Buchel,&Rose,2010).
Brain mechanisms of ABC and top–down goals
InLimetal.’s(2009)attentionalblinkparadigm,havingtar-
Asdiscussedearlier,asignatureofbiasedcompetitionisthat gets be emotionally arousing enhanced participants’ ability to
thesameneuronshowsdifferentresponsepropertiesdepend- carryouttheirgoalofdetectingthetarget.However,emotion-
ingonhowrelevantthestimulusistocurrentgoals.Buthow allyarousingstimulicanalsodistractpeoplefromtheircurrent
122 Mather and Sutherland
goal, as arousing stimuli attract attention themselves. In that of the images alternates between them, but the emotionally
case,onemightexpecttheamygdalatostillmodulateactivity arousingimage tendstowinthe binocular rivalrycompetition
in the frontoparietal attention network, but rather than enhan- for longer intervals (Alpers & Gerdes, 2007; Alpers & Pauli,
cinggoal-directedselection,theamygdalashoulddiminishthe 2006;Bannerman etal., 2008; Sheth&Pham, 2008).
impact of goal-directed selection by prioritizing the emotion- Arousal also amplifies the effects of directed attention
ally relevant distractor. Consistent with this possibility, when toward one target stimulus or location (Phelps et al., 2006;
focusing ontask-relevant stimuliinthepresence ofemotional Schupp et al., 2007), consistent with the ABC hypothesis that
distractors, amygdala activity increases while activity in the arousalenhancestheeffectsoftop–downrelevance.However,
frontoparietal network decreases (Dolcos & McCarthy, 2006; when there are multiple stimuli competing for dominance,
D.G.V.Mitchelletal.,2008).ThesefindingsandthoseofLim arousal interferes with distributing attention across multiple
etal.suggestthattheamygdalainfluencesactivityinthefron- stimuli (Morelli & Burton, 2009) and maintaining multiple
toparietal attention network—either increasing or decreasing representations in working memory (Mather et al., 2006; K.J.
how much attention is guided by top–down goals, which Mitchell et al., 2006). This arousal-based impairment when
depends on whether attending to the source of the arousal is multiplestimuliareequallythetargetofattentionisconsistent
consistent withcurrent taskgoals ornot. with the idea that competition between adjacent strong repre-
When arousing stimuli distract from current task goals, sentationsleads tomutualinterference (Fig. 1B).
interactions betweentheamygdalaandbrainregionsinvolved The ABC effects described in this section are likely
in resolving interference or conflict (such as the ACC or left mediated by interactions between the amygdala and attention
inferior frontal cortex) may help to bias attention toward the networks in the brain. For instance, there is evidence that the
emotionallyarousingstimuli,ormayhelptocountertheemo- amygdala interacts with the frontoparietal attention network
tional distraction. In general, the ACC activates in situations toenhance thepriority ofemotionalstimulithatarethetarget
involving conflict in information processing (Botvinick, of attentional goals (Lim et al., 2009). However, further
Cohen, & Carter, 2004; Carter & van Veen, 2007). In one research is needed to determine whether the amygdala helps
study, amygdala activity correlated with anterior and poster- to bias competition to favor nonarousing but high priority sti-
iorcingulatecorticesinthepresenceofdistractingemotional muliduringepisodes ofarousal.
images (D.G.V. Mitchell et al., 2008). Likewise, when
maskedfearfulfacesareusedtoorientattentiontothespatial
ABC During Encoding Shapes Memory
locationofdotprobes,amygdalaactivitywascorrelatedwith
ACC activity and the amygdala–ACC correlation predicted
In this section, we turn to the emotional memory effects out-
attentional orienting to the masked fearful faces (Carlson,
linedinTable1andmakethecasethatthefirstthreeofthese
Reinke,&Habib,2009).Thus,theACCmayhelpdisengage
effectsandtheirapparentcontradictionsintheliteraturecanbe
attentionfromotherstimuliandredirectittowardemotionally
explained by ABC during encoding affecting what is later
arousing stimuli. remembered.
Theleftinferiorfrontalcortexplaysakeyroleinresolving
interference (Aron, Robbins, & Poldrack, 2004; Jonides &
Memory narrowing
Nee, 2006), and one study found that its activity correlated
with the amygdala when distracting emotional stimuli were Inthe1950s,Callawayandcolleaguesintroducedtheideathat
present (Dolcos, Kragel, Wang, & McCarthy, 2006). How- stress narrows attention, arguing that, ‘‘a narrowed focus of
ever, unlike the ACC activity in other studies, the inferior attention (i.e., a decreased influence of peripheral factors)
frontal cortex activity was associated with greater success in seemsrelatedtosomeneurophysiologicalcomponentofacute
avoiding emotional distraction, rather than with orienting stress’’ (Callaway & Dembo, 1958, p. 74). Easterbrook’s
toward the emotional stimuli—thus, the left inferior frontal (1959) review of the literature argued that emotional arousal
cortex may provide top–down modulation of the amygdala, reduces the number of cues utilized in a task, contending that
reducing the degree to which distracting emotional stimuli arousalandstressdidnotuniformlyreducecueutilization,but
dominate attention. instead limited the use of peripheral (temporarily irrelevant)
cues in favor of central (immediately relevant) cues. Subse-
quent researchers suggested that arousal also has opposite
Summary of evidence for ABC in perception
effectsonmemoryforcentralandperipheraldetails,enhancing
Thestudiesreviewedaboverevealthatemotionalarousalleads memory for central detail at the cost of peripheral detail (for
toa‘‘winner-takes-more’’effect,inwhichalreadyconspicuous reviews, see Christianson, 1992; Levine & Edelstein, 2009;
stimuli gain priority while lower priority stimuli are further Reisberg&Heuer,2004).Aclassicexampleofarousal’seffect
suppressed.Forinstance,hearingarousingsoundsbeforeview- on central versus peripheral details is the ‘‘weapon focus
ing a set of letters makes the high contrast letters even more effect,’’inwhichthepresenceofaweaponreduceseyewitness
conspicuous than the low contrast letters (Sutherland & identification of the perpetrator (Loftus, Loftus, & Messo,
Mather, 2011). When oneeye is shown anemotionally arous- 1987;Steblay,1992).Otherexamplesconsistentwithmemory
ing image and the other is shown a neutral image, perception narrowing come from studies examining memory for