Neural correlates of attention to emotional facial expressions in dysphoria

This article was downloaded by: [Universita di Padova]On: 17 June 2014, At: 06:43Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office:Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Click for updates

Cognition and EmotionPublication details, including instructions for authors and subscriptioninformation:http://www.tandfonline.com/loi/pcem20

Neural correlates of attention to emotionalfacial expressions in dysphoriaGiulia Buodoab, Giovanni Mentoa, Michela Sarloab & Daniela Palombaab

a Department of General Psychology, University of Padova, Padova, Italyb Center for Cognitive Neuroscience, University of Padova, Padova, ItalyPublished online: 12 Jun 2014.

To cite this article: Giulia Buodo, Giovanni Mento, Michela Sarlo & Daniela Palomba (2014): Neuralcorrelates of attention to emotional facial expressions in dysphoria, Cognition and Emotion, DOI:10.1080/02699931.2014.926862

To link to this article: http://dx.doi.org/10.1080/02699931.2014.926862

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”)contained in the publications on our platform. However, Taylor & Francis, our agents, and ourlicensors make no representations or warranties whatsoever as to the accuracy, completeness, orsuitability for any purpose of the Content. Any opinions and views expressed in this publicationare the opinions and views of the authors, and are not the views of or endorsed by Taylor &Francis. The accuracy of the Content should not be relied upon and should be independentlyverified with primary sources of information. Taylor and Francis shall not be liable for anylosses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilitieswhatsoever or howsoever caused arising directly or indirectly in connection with, in relation to orarising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantialor systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, ordistribution in any form to anyone is expressly forbidden. Terms & Conditions of access and usecan be found at http://www.tandfonline.com/page/terms-and-conditions

Neural correlates of attention to emotional facialexpressions in dysphoria

Giulia Buodo1,2, Giovanni Mento1, Michela Sarlo1,2, and Daniela Palomba1,2

1Department of General Psychology, University of Padova, Padova, Italy2Center for Cognitive Neuroscience, University of Padova, Padova, Italy

The present study investigated whether dysphoric individuals have a difficulty in disengagingattention from negative stimuli and/or reduced attention to positive information. Sad, neutral andhappy facial stimuli were presented in an attention-shifting task to 18 dysphoric and 18 controlparticipants. Reaction times to neutral shapes (squares and diamonds) and the event-related potentialsto emotional faces were recorded. Dysphoric individuals did not show impaired attentionaldisengagement from sad faces or facilitated disengagement from happy faces. Right occipitallateralisation of P100 was absent in dysphoric individuals, possibly indicating reduced attention-related sensory facilitation for faces. Frontal P200 was largest for sad faces among dysphoricindividuals, whereas controls showed larger amplitude to both sad and happy as compared withneutral expressions, suggesting that dysphoric individuals deployed early attention to sad, but nothappy, expressions. Importantly, the results were obtained controlling for the participants’ traitanxiety. We conclude that at least under some circumstances the presence of depressive symptoms canmodulate early, automatic stages of emotional processing.

Keywords: Dysphoria; Emotion; Attention; Facial expressions; Event-related potentials.

In contrast with the large and consistent pattern offindings in support of a memory bias in depression(see Gotlib & Joormann, 2010; Wisco, 2009), thebody of evidence regarding depression-relatedbiases of attention appears to be more varied andcomplex. The findings of behavioural studiesemploying sub-threshold stimulus exposure indifferent paradigms (e.g., the emotional Strooptask, the dot-probe task, the affective primingtask) have been fairly consistent in reporting alack of a mood-congruent attentional bias in

depression, thus suggesting that automatic, pre-attentive processes are unlikely to be involved (e.g.,Bradley, Mogg, Millar, & White, 1995; Mathews,Ridgeway, & Williamson, 1996; Mogg, Bradley,& Williams, 1995; Mogg, Bradley, Williams, &Mathews, 1993; but see Dannlowski et al., 2006,for an exception). The results obtained with longerstimulus exposure times (500–1500 ms) have beenmore mixed, with some studies reporting thatdepressed individuals attend to negative wordsor facial expressions to a greater degree than

Correspondence should be addressed to: Giulia Buodo, Department of General Psychology, University of Padova, Via Venezia,

8 - 35131 Padova, Italy. E-mail: [email protected]

© 2014 Taylor & Francis 1

COGNITION AND EMOTION, 2014

http://dx.doi.org/10.1080/02699931.2014.926862

Dow

nloa

ded

by [

Uni

vers

ita d

i Pad

ova]

at 0

6:43

17

June

201

4

non-depressed individuals (e.g., Gotlib & McCann,1984; Mathews et al., 1996), and others failing tofind such effect (e.g., Gotlib, McLachlan, & Katz,1988; McCabe & Gotlib, 1995; Mogg, Millar, &Bradley, 2000). Importantly, positive findings weremostly obtained when stimulus durations were onesec or longer (e.g., Bradley, Mogg, & Lee, 1997;Donaldson, Lam, & Mathews, 2007; Gotlib &Cane, 1987; Koster, De Raedt, Goeleven, Franck,& Crombez, 2005). Taken together, the findingson the attentional bias in depressed individuals havebeen interpreted to suggest that depression does notfacilitate early, automatic attentional capture to-wards depression-related stimuli but, rather, impairsattentional disengagement from negative informa-tion once it has come into the focus of attention (seeBradley et al., 1997; Mogg & Bradley, 1998, 2005).

More recently, the focus of research has movedfrom investigating whether an attentional biasdoes occur in depression to clarifying its exactcontent. In particular, it is debated whetherdepression is specifically characterised by atten-tional dwelling on negative information or, rather,by reduced attention to positive information, or acombination of both processing biases. Indeed,some studies have reported that depressed patientsshow a relative reduction in the processing ofpositive verbal and pictorial stimuli relative tonegative stimuli and/or relative to control partici-pants (Deldin, Keller, Gergen, & Miller, 2001;Nandrino, Dodin, Martin, & Henniaux, 2004;Shestyuk, Deldin, Brand, & Deveney, 2005; Sloan,Strauss, Quirk, & Sajatovica, 1997; Surguladzeet al., 2004). These findings lend support tothe hypothesis that one of the key mechanismsunderlying emotion dysregulation in depression isthe inability to use positive and rewarding stimulito regulate negative mood (Gotlib & Joormann,2010). It thus remains to be determined whetherthe attentional bias in depression can be bestunderstood as decreased attention towards positiveinformation or as increased attention towardsnegative information, or a combination of bothprocessing biases.

Another unresolved issue is whether attentionalbiases are only apparent in individuals with clinicaldepression (either current or remitted; see Gotlib

et al., 2004; Joormann & Gotlib, 2007), or can bereliably highlighted also in individuals with dys-phoria, that is, individuals who report elevateddepressive symptoms on a psychometric instru-ment designed to assess such symptoms, but arenot formally diagnosed with major depressivedisorder or dysthymia (Frewen & Dozois, 2005;Haaga & Solomon, 1993; Kendall, Hollon, Beck,Hammen, & Ingrain, 1987). Indeed, increasedattention to negative stimuli may serve to createthe persistent negative mood that, in turn, may putan individual at risk for a depressive episode(Beevers & Carver, 2003). Several studies onindividuals with dysphoria failed to observe anattentional bias for depression-related material,thus casting doubt on the existence of a reliableattentional bias in individuals with sub-clinicaldepressive symptoms (Koster, Leyman, De Raedt,& Crombez, 2006; Yovel & Mineka, 2004, 2005).However, a recent meta-analysis indicates thatbiased attention to negative stimuli is not exclusiveto patients with clinical depression, but is alsoreported in individuals with sub-clinical depressivesymptoms, as well as in subjects undergoingdepressive mood induction. Therefore, an atten-tional bias towards negative stimuli might beassociated with sub-clinical depression, ratherthan being a marker of clinical depression only(Peckham, McHugh, & Otto, 2010).

A relevant confounding factor in research oninformation processing biases in depression/dys-phoria is the high comorbidity between anxietyand both clinical and sub-threshold depression(Fava et al., 2000; Mineka, Watson, & Clark,1998; Preisig, Merikangas, & Angst, 2001).Depression and anxiety appear to bias attentionalprocessing differently. In anxious individuals, theinitial automatic orienting of attention towardsdanger cues is followed by difficulty in disengage-ment or by attentional avoidance during later,more strategic stages of processing (Cisler &Koster, 2010; Mogg & Bradley, 2005). In con-trast, depressed individuals do not initially attendto but tend to dwell on depression-related stimuli(see Mogg & Bradley, 2005). It is unclear howthese patterns of attentional bias interact inindividuals with co-occurring symptoms of

BUODO ET AL.

2 COGNITION AND EMOTION, 2014

Dow

nloa

ded

by [

Uni

vers

ita d

i Pad

ova]

at 0

6:43

17

June

201

4

depression and anxiety. In fact, in several studiesof the attentional bias in depression/dysphoria theparticipants’ trait anxiety was not measured (e.g.,Beevers & Carver, 2003) or was not controlled forin data analysis (e.g., Koster et al., 2005), thusmaking it impossible to unambiguously attributethe obtained results to the effect of depressivesymptoms alone. Overall, the evidence in this areaof research is far from being conclusive. In order toobtain a clearer understanding of the effects ofdepression and anxiety on attention, both condi-tions should be assessed even when distinguishingtheir separate effects is not the primary researchgoal (Beuke, Fischer, & McDowall, 2003).

Despite the large number of studies usingbehavioural measures in the relevant literature,there have been limited attempts to explore theneural correlates of the attentional bias in depres-sion/dysphoria by using the event-related poten-tials (ERPs). Furthermore, these studies differalong a wide array of procedural and population-related variables. For instance, different experi-mental tasks have been employed: from delayedmatch-to-sample (Shestyuk et al., 2005) to wordcounting (Nandrino et al., 2004), affective go/no-go (Krompinger & Simons, 2009), visual oddball(Cavanagh & Geisler, 2006; Ilardi, Atchley,Enloe, Kwasny, & Garratt, 2007), recognitionmemory (Deldin et al., 2000, 2001) and affectivepriming (Dai, Feng, & Koster, 2011; Yao et al.,2010). Although attentional processes are impli-cated in all these tasks, the underlying attentionalmechanisms are different. It should also be notedthat each of the ERP components can be elicitedby tasks that tap different cognitive processes(Key, Dove, & Maguire, 2005), and this makesit further difficult to provide general interpreta-tions. In addition, different stimulus types (emo-tional words, faces with emotional expressions andaffective pictures) and different exposure times(from 100 ms to >1 sec) have been employed. Asfor participants, the large majority of studies hasfocused on individuals with depressive symptomsof clinical relevance. With such high heterogen-eity, it is difficult to summarise and evaluate theavailable evidence.

Most studies in this area focused specifically onthe amplitude and/or latency of the P300, an ERPcomponent related to the allocation of attentionalresources to stimulus evaluation (Picton, 1992),and did not systematically analyse other ERPcomponents (see Cavanagh & Geisler, 2006;Deldin et al., 2001; Ilardi et al., 2007; Krompinger& Simons, 2009; Nandrino et al., 2004). Overall,depressed individuals appear to respond withlarger P300 amplitude to negative as comparedwith positive stimuli or healthy controls (Bistricky,Atchley, Ingram, & O’Hare, 2014; Dai et al.,2011; Ilardi et al., 2007; Krompinger & Simons,2009; Nandrino et al., 2004), and/or show smallerP300 to positive as compared with negative andneutral stimuli, and relative to controls (Nandrinoet al., 2004; Shestyuk et al., 2005). Such findingsare thought to be consistent with behaviouralevidence to indicate that depressed individualsallocate more attentional resources to negativematerial and/or have reduced attention to positivestimuli. Using a cue-target task, Dai and Feng(2009) specifically investigated visuo-spatial atten-tion by measuring the P300 to face cues precedinga neutral target located on the same (valid trial) oron the opposite side (invalid trial) of the facelocation. Currently depressed participants showedlarger P300 amplitude for happy faces in theinvalid than in the valid condition as comparedwith never-depressed and remitted individuals,and larger P300 amplitude for sad faces in thevalid than in the invalid condition as comparedwith healthy controls. Overall, these findings wereinterpreted as evidence that depressed participantsboth shifted attention away from positive stimuliand had difficulties disengaging attention fromnegative stimuli.

Fewer studies have examined earlier stages ofinformation processing by measuring the ERPcomponents preceding the P300. Dai et al. (2011)measured the P100 as an index of early visuo-spatial processing and discrimination within thefocus of attention in an affective priming task.Compared with remitted patients and healthycontrols, participants with current major depres-sion had larger P100 to sad face target-probespreceded by sad face target-primes (positive

EMOTIONAL FACE PROCESSING IN DYSPHORIA

COGNITION AND EMOTION, 2014 3

Dow

nloa

ded

by [

Uni

vers

ita d

i Pad

ova]

at 0

6:43

17

June

201

4

priming condition), suggesting facilitation of earlyperceptual analysis. No difference among groupswas found for happy faces in the early processingstages. Using a similar paradigm, Yao et al. (2010)observed a reduction of P200 amplitude fornegative target words preceded by negative dis-tractor words in depressed patients compared withcontrols. This finding suggests that individualswith depression have a deficit in the inhibition ofnegative prime distractor stimuli that results inreduced early attentional capture by subsequentsalient targets (Ashley, Vuilleumier, & Swick,2004; Carretié, Hinojosa, Martín-Loeches, Mer-cado, & Tapia, 2004; Thomas, Johnstone, &Gonsalvez, 2007). Lastly, using pairs of emotionalpictures (negative-neutral, positive-neutral andneutral-neutral) in a dot-probe task, Mingtian,Xiongzhao, Jinyao, Shuqiao, and Atchley (2011)found that depressed individuals did not showlarger P100 amplitude to the probe in the validthan in the invalid positive-neutral pair as controlsdid. Because P100 amplitude reflects sensoryvisual processing of stimuli at an attended location(Woldorff et al., 2002), this finding indicates thatdepressed individuals did not shift attentiontowards positive pictures relative to neutral onesbefore probe presentation. Overall, the pattern ofERP findings related to the early attentionalprocessing suggests that the attentional bias indepressed individuals is characterised by a lack ofattention orienting to positive information and/orby facilitated processing of negative stimuli.

The present study was aimed at further invest-igating the extent to which dysphoria is charac-terised by an attentional bias towards negativeinformation, away from positive information or acombination of both processing biases. Faces withemotional expressions, either negative (sad), neut-ral or positive (happy), were used because of theirhigher ecological validity compared to word stim-uli. An attention-shifting paradigm was employed(Fox, Russo, Bowles, & Dutton, 2001, experiment5, as modified by Bar-Haim, Lamy, & Glickman,2005) that allows directly assessing the disengage-ment component of attention. In this task, facestimuli were presented singly at fixation, andparticipants were required to respond by pressing

a key to a neutral target (a square or a diamond)that appeared randomly above, below, to the left orto the right of the face stimulus. Reaction times(RTs) to target stimuli were recorded during thetask. Impaired disengagement from negative stim-uli should be indexed by longer RTs to targetspreceded by sad faces as compared with RTs totargets preceded by neutral faces. A bias away frompositive stimuli (i.e., facilitated disengagement)should be indexed by shorter RTs to targetspreceded by happy faces as compared with RTsto targets preceded by neutral faces.

We recorded the ERPs to face stimuli, and wemeasured the amplitude and latency of the P100,N100, N170 and P200 components. More spe-cifically, the amplitude and latency of the P100and N100 were considered as measures of theefficiency and speed of early modulation of spatialattention over perceptual processing (Luck, 2005).

The face-specific N170 was measured as theindex of structural encoding processes precedingface identification (Bentin, Allison, Puce, Perez,& McCarthy, 1996; Bentin & Deouell, 2000).As it is unclear whether the N170 is modulatedby emotional facial expression (Blau, Maurer,Tottenham, & McCandliss, 2007; Eimer &Holmes, 2002; Eimer, Holmes, & McGlone,2003), we intended to explore whether an influ-ence of facial effect on N170 amplitude, if present,could be modified by the occurrence of depressivesymptoms.

The amplitude and latency of P200 weremeasured as indices of the efficiency and speedof early, automatic detection of stimulus salience,preceding slower, strategic recruitment of proces-sing resources (e.g., Ashley et al., 2004; Carretiéet al., 2004; Thomas et al., 2007).

Since we did not expect visible peaks after theP200 (see Sarlo & Munafò, 2010), we measuredthe mean amplitude in the time windows of theP300 (300–400 ms) and of the late positivecomplex (LPC; 400–500 and 500–600 ms). TheP300/LPC reflect the allocation of attentionalresources for stimulus evaluation and categorisa-tion (Picton, 1992).

Based on the extant ERP literature, it waspossible to hypothesise that an attentional bias

BUODO ET AL.

4 COGNITION AND EMOTION, 2014

Dow

nloa

ded

by [

Uni

vers

ita d

i Pad

ova]

at 0

6:43

17

June

201

4

towards negative stimuli and/or away from positivestimuli in dysphoric individuals might be mani-fested not only at relatively late, elaborative stagesof information processing (beyond 300 ms post-stimulus), but also at earlier, more automaticstages (within the first 300 ms). Specifically,increased processing (possibly impairing atten-tional disengagement) of negative faces could bemanifested as greater amplitude and/or shorterlatency of P100, N100 and P200, and largerP300/LPC as compared with positive and neutralstimuli and/or with control participants. On theother hand, reduced processing of positive faces(which could facilitate attentional disengagement)could be indexed by smaller amplitude and/orlonger latency of P100, N100 and P200, andsmaller P300/LPC as compared with negative andneutral stimuli and/or with control participants.

Each component was also analysed in terms oftopographic distribution (caudality and/or lateral-ity), in order to explore whether possible ERPdifferences between dysphoric and non-dysphoricindividuals could be region-specific. Indeed,region-specific ERP effects in addition to val-ence-specific effects have been previously reportedin depression, regarding in particular impairedright posterior functioning (see Deldin et al.,2000; Kayser, Bruder, Tenke, Stewart, & Quitkin,2000).

Lastly, and importantly, we measured theparticipants’ trait anxiety and statistically con-trolled for a possible role of this variable inmodulating behavioural and electrophysiologicalresponses.

METHOD

Participants

In order to preliminarily identify individuals withhigh and low symptoms of depression, the BeckDepression Inventory II (BDI II; Beck, Steer, &Brown, 1996; Italian version by Ghisi, Flebus,Montano, Sanavio, & Sica, 2006) was adminis-tered to 157 undergraduate students at the Uni-versity of Padova. They were informed that theywould be contacted within one week for

participation in a study on electrocortical responsesto emotional visual stimuli, but they were unawarethat they would be selected on the basis of theirquestionnaire scores.

The BDI II is a widely used, reliable and validself-report that assesses the severity of recent (i.e.,past two weeks) cognitive-affective and somaticsymptoms of depression. It consists of 21 itemsanswered on a 0–3 scale, and the total score rangesfrom 0 to 63. Several studies have shown that theBDI II is a reliable and valid measure of severity ofdepressive symptomatology in college populations(Sprinkle et al., 2002) and community samples(Lasa, Ayuso-Mateos, Vázquez-Barquero, Díez-Manrique, & Dowrick, 2000). The Italian versionof the BDI II has proved to have good psycho-metric properties in terms of validity and reliability(Ghisi et al., 2006). A score of 12, correspondingto the 90th percentile of score distribution amongmales and to the 80th percentile among females, isindicated as the optimal cut-off score to discrim-inate individuals with and without depressivesymptomatology (Ghisi et al., 2006).

In line with recommendations for the use of theBDI in non-clinical samples, and for categorisa-tion of research participants (see Frewen &Dozois, 2005; Haaga & Solomon, 1993; Kendallet al., 1987; Tennen, Hall, & Affleck, 1995), welabelled male and female undergraduates whoscored equal or above 12 on the BDI II asdysphoric individuals, and invited them to parti-cipate in the study.

Among eligible undergraduates classified asdysphoric individuals (N = 28), 18 volunteered toparticipate (15 females and 3 males; mean age =22.66 ± 2.11 years; range = 19–29 years; meanBDI II score = 18.38 ± 3.89; range = 14–29).According to the criteria reported by Beck et al.(1996) in the manual of the original version of theBDI II, “mild depressive symptoms” (BDI scoresbetween 14 and 19) were present in 12 partici-pants, “moderate depressive symptoms” (BDIscores between 20 and 28) in 5 participants and“severe depressive symptoms” (BDI scoresbetween 29 and 63) in 1 participant.

In order to ensure separation between groups,for the non-dysphoric control group we selected

EMOTIONAL FACE PROCESSING IN DYSPHORIA

COGNITION AND EMOTION, 2014 5

Dow

nloa

ded

by [

Uni

vers

ita d

i Pad

ova]

at 0

6:43

17

June

201

4

18 participants (14 females and 4 males; mean age= 24.05 ± 2.20 years; range = 21–30 years) withBDI II scores ≤7, corresponding to the median ofthe obtained score distribution (mean BDI IIscore = 3.22 ± 2.18; range = 0–7). Individualswith mid-range BDI II scores (i.e., between 8 and11) were not recruited.

Dysphoric individuals and non-dysphoric indi-viduals did not differ with regard to age, t(34) =−1.92; p = .85, two-tailed.

Participants had no history of psychiatric orneurological disorders and were not taking med-ication, as ascertained by means of a general healthquestionnaire.

The study was approved by the Ethical Com-mittee of the Department of General Psychology,University of Padova (Protocol No. 36), and allvolunteers gave written informed consent prior toparticipation, according to the Declaration ofHelsinki. Dysphoric participants were informedthat, if interested, psychological treatment wasavailable.

Participants were not paid for participation anddid not receive course credit.

Attention-shifting task

A computerised attention-shifting experimentalparadigm (Bar-Haim et al., 2005; Fox et al.,2001; Sarlo & Munafò, 2010) was administered,consisting in the identification of two differenttarget stimuli surrounding the face stimuli pre-sented at the centre of a 19-inch computer screenat a viewing distance of 135 cm. The face stimuliwere 48 different digitised colour pictures depict-ing eight male and eight female Caucasian modelswith sad, happy and neutral facial expressions,selected from the NimStim Set of Facial Expres-sions (Tottenham et al., 2009). The pictures werepresented on a grey background at the centre ofthe screen inside a black outline frame measuring13.76 × 13.76 cm. The target stimuli consisted ofblack filled shapes, either a square or a diamond,measuring 1 × 1 cm.

Each trial began with the presentation of afixation cross at the centre of the black outlineframe for 1000 ms. Next, a face was presented

inside the frame for 1850 ms. Six hundredmilliseconds after the onset of the face stimulus,a target was randomly presented for 50 ms, 8.38cm above, below, to the left or to the right of thecentre of the computer screen. Targets werepresented within the same visual fixation area ofthe face, and therefore participants did not need tomove their eyes to perform the task. After facestimulus offset, there was a variable intertrialinterval (500–1000 ms). The participants wereexplicitly instructed to keep their gaze at the centreof the screen, and to identify the target shape bypressing one of two alternative keys as fast andaccurately as possible. The pictures were presentedin two consecutive blocks, each consisting of atotal of 48 stimuli (16 for each facial expression),and in random order within each block.

Picture presentation was accomplished usingE-prime software (Psychology Software Tools,Pittsburgh, PA, USA).

Procedure

Prior to the experiment, participants were askedto complete the trait version of the State-TraitAnxiety Inventory (Pedrabissi & Santinello, 1989;STAI; Spielberger, 1983). They were then seatedon a comfortable chair in a dimly lit, sound-attenuated room and an elastic cap embedded withelectrodes was applied for EEG recording. Fol-lowing a 10-min adaptation period, instructionsfor the task were given.

Electrophysiological recordings and dataanalyses

The electroencephalogram (EEG) was recordedwith tin electrodes mounted in an elastic cap from19 scalp sites (Fp1, Fp2, F3, Fz, F4, F7, F8, C3,Cz, C4, P3, Pz, P4, T3, T4, T5, T6, O1 and O2)referenced to linked mastoids, according to theInternational 10–20 System. For the purpose ofartefact scoring, vertical and horizontal electroo-culograms (EOGs) were recorded from electrodepairs (bipolar) placed above and below the righteye and at the external canthi of both eyes. Allelectrode impedances were kept below 10 kΩ.

BUODO ET AL.

6 COGNITION AND EMOTION, 2014

Dow

nloa

ded

by [

Uni

vers

ita d

i Pad

ova]

at 0

6:43

17

June

201

4

Signals were bandpass filtered (0.1–40 Hz), digi-tised at 250 Hz (16-bit A/D converter) and storedon to a Pentium IV computer. The EEG and theEOG signals were filtered and amplified by aSynAmps unit amplifier (Neuroscan, Inc., Com-pumedics, Ltd, El Paso, TX, USA).

Continuous EEG data were corrected for eye-blinks using a regression-based weighting coeffi-cients technique, as implemented in the SCAN4.1 software (Edit module; Neurosoft, Inc.). TheEEG was then segmented offline into 700-msepochs from 100 ms before to 600 ms after facestimulus onset. The EEG epochs were baselinecorrected against the mean voltage during the100-ms prestimulus period. All EEG epochs werevisually scored for eye movements and otherartefacts, and each portion of data containingartefacts greater than ±70 µV in any channel wasrejected for all the recorded channels prior tofurther analysis. Artefact-free trials with correctbehavioural responses were separately averaged foreach subject and each emotional condition. On thebasis of the inspection of grand-average ERPwaveforms at occipitotemporal electrodes (T5,T6, O1 and O2), the P100 was computed as themost positive peak between 80 and 140 ms, andthe N170 as the most negative peak between 140and 200 ms from stimulus onset. On the basis ofwaveform inspection at frontal, central and parietalelectrodes (F3, Fz, F4, C3, Cz, C4, P3, Pz, P4),the N100 was computed as the most negative peakbetween 90 and 150 ms from stimulus onset, andthe P200 as the most positive peak between 140and 200 ms from stimulus onset. Furthermore,since no other clear peaks were visible, the meanamplitude values were computed within threesuccessive post-stimulus time windows, approxi-mately covering the time range of the P300 (300–400 ms) and the LPC (400–500 and 500–600 ms).

Statistical analyses

Because dysphoric individuals scored significantlyhigher than non-dysphoric individuals in self-reported trait anxiety as measured by the STAI(49.38 ± 9.64 and 34.33 ± 5.41, respectively;

t(34) = 5.77; p < .001, two-tailed), and in order tocontrol for the effect of trait anxiety on attentionalperformance, behavioural and ERP data wereanalysed using analyses of covariance (ANCOVAs)with the STAI trait score as the covariate.

For the amplitude and latency of P100 andN170, measured at T5, T6, O1 and O2, theANCOVA designs included emotional valence(sad, neutral and happy), area (temporal andoccipital) and laterality (left and right) as within-subjects factors, and group (dysphoric individualsand non-dysphoric individuals) as between-sub-jects factor. The amplitude and latency of N100and P200, and the mean amplitude in the 300–400, 400–500 and 500–600 ms time windows,measured at F3, Fz, F4, C3, Cz, C4, P3, Pz andP4, were analysed using an ANCOVAs withemotional valence, area (frontal, central and pari-etal) and laterality (left, midline and right) aswithin-subjects factors, and group as between-subjects factor.

Mean RTs to targets and percent accuracy inthe attention-shifting task were entered intoseparate ANCOVAs with emotional valence aswithin-subjects factor and group as between-subjects factor.

Post hoc comparisons (Tukey’s honest signific-ant difference, p < .05) were employed to furtherexamine significant effects.

RESULTS

Behavioural data

A significant main effect of emotional valence wasfound for RTs, F(2, 66) = 4.23; p = .019. MeanRTs were 551 ms (SD = 13.7) for targetsfollowing happy faces, 544 ms (SD = 13.2) fortargets following neutral faces and 539 ms (SD =14.8) for targets following sad faces. Post hoccomparisons only showed a tendency for RTs tobe faster following sad than happy expressions(p = .06). The group main effect and the group ×emotional valence interaction were not significant.

Overall, the response accuracy was very high(95.7% for targets following neutral faces, 95.5%for targets following happy faces and 94.7% for

EMOTIONAL FACE PROCESSING IN DYSPHORIA

COGNITION AND EMOTION, 2014 7

Dow

nloa

ded

by [

Uni

vers

ita d

i Pad

ova]

at 0

6:43

17

June

201

4

targets following sad faces). The main effects ofgroup and emotional valence, and their interac-tion, were not significant (all F’s < 1; all ps > .3).

Event-related potentials

The grand-average ERPs elicited by sad, neutraland happy facial expressions from Fz, Cz and Pz,as well as from O1 and O2 electrodes in dysphoricindividuals and non-dysphoric individuals areshown in Figures 1 and 2, respectively.

Results are summarised with a focus on signi-ficant main effects and interactions which arerelevant to the research questions and includegroup and/or emotional valence among factors.

P100

For P100 amplitude, a significant group × area ×laterality interaction was found, F(1, 33) = 4.85;p = .035. There were no between-group differ-ences, but the post hoc tests identified significantwithin-group differences. Whereas in the temporalarea both dysphoric individuals and non-dysphoricindividuals showed larger P100 amplitude on theright than on the left side (p = .0007 and .027,respectively), in the occipital area dysphoric indi-viduals failed to show the right asymmetry evi-denced by non-dysphoric individuals (p = .99 andp = .008, respectively).

For P100 latency, no significant effects invol-ving group and/or emotional valence were found.

N100

For N100 amplitude, a significant group × later-ality interaction emerged, F(2, 66) = 4.08,p = .021. Post hoc analysis did not highlight sig‐nificant group differences. However, the within-group pattern was different in dysphoric individualsand non-dysphoric individuals. In non-dysphoricindividuals, amplitude was larger over the midline than over both the left (p = .035) and the right(p = .0001) side, whereas in dysphoric individualsthe N100 was larger over the midline and the leftside than over the right side (ps < .0003).

For N100 latency, no significant effects invol-ving group were found. The significant emotionalvalence × area interaction, F(4, 132) = 3.41;p = .01, indicated that latency was longer in theparietal as compared with the frontal area forneutral (p = .01) and happy (p = .005), but not sad(p = .12), facial expressions.

N170

No significant main effects or interactions invol-ving group and/or emotional valence were foundeither for the amplitude or the latency of thiscomponent.

P200

For P200 amplitude, a significant emotionalvalence × area interaction was found, F(4, 132) =3.6; p = .008, that was further specified by thethree-way emotional valence × area × lateralityinteraction, F(8, 264) = 2.15; p = .031. In thefrontal area, P200 was larger for sad than forhappy and neutral facial expressions at left,midline and right sites (all ps < .0001). In thecentral area, P200 was larger for sad than neutraland happy facial expressions on the left side(p < .0001), whereas it was larger for both sadand happy as compared with neutral faces on themidline and right sites (all ps < .0001). In theparietal area, P200 was larger for sad than neutraland happy facial expressions on the left side, andlarger for sad than neutral facial expressions atmidline and right sites (all ps < .001).

The group × emotional valence × area interac-tion was also significant, F(4, 132) = 2.52;p = .044. Post hoc analysis failed to reveal signific-ant group differences. However, the within-grouppattern was different in dysphoric individuals andnon-dysphoric individuals. In the frontal area,dysphoric individuals showed larger P200 to sadthan happy and neutral expressions (ps < .0001),that did not differ from each other (p = .1). Innon-dysphoric individuals, P200 was larger forsad and happy expressions, that did not differfrom each other (p = .1), as compared with neutralexpressions (ps < .046). In the central area,

BUODO ET AL.

8 COGNITION AND EMOTION, 2014

Dow

nloa

ded

by [

Uni

vers

ita d

i Pad

ova]

at 0

6:43

17

June

201

4

Figure 1. Grand-average event-related potentials waveforms to happy, neutral and sad facial expressions recorded at midline electrodes in

dysphoric and non-dysphoric individuals. The N100 and P200 components are marked by black arrows. LPC = late positive complex.

EMOTIONAL FACE PROCESSING IN DYSPHORIA

COGNITION AND EMOTION, 2014 9

Dow

nloa

ded

by [

Uni

vers

ita d

i Pad

ova]

at 0

6:43

17

June

201

4

differences among facial expressions were notsignificant in dysphoric individuals, whereas innon-dysphoric individuals P200 was larger forhappy than for neutral expressions (p = .016). Atparietal electrodes, no differences among facialexpressions were found in either group.

No significant effects involving group and/oremotional valence were observed concerning P200latency.

Mean amplitude of the P300/LPC

No significant effects involving group and/oremotional valence were found for the time rangesof the P300 and the LPC (P300: 300–400 ms;LPC: 400–500; and 500–600 ms).

DISCUSSION

The present study aimed at investigating theneural correlates of attentional processing of emo-tional facial expressions in dysphoric individuals.Specifically, it was sought to determine (1)whether dysphoria is characterised by an atten-tional bias towards negative information, awayfrom positive information, or a combination ofboth biases; (2) whether the disengagement com-ponent of visuo-spatial attention is specifically

involved in the attentional bias. In view of thefact that the neural correlates of the attentionalbias in dysphoric individuals have not beensystematically investigated, the ERPs to emotionalfacial stimuli were recorded, and both early andlate components were examined. Facial stimulipresented singly at fixation were entirely irrelevantto the task of discriminating two target shapesappearing at one of four locations around the face.The task irrelevance of facial stimuli allowed tospecifically investigate spontaneous, implicit pro-cessing of emotional facial expressions, whileavoiding the confound arising from the activationof other processes such as facial affect identifica-tion and categorisation, that are explicitly requiredduring tasks that require participants to judgeemotional facial expressions. Indeed, some of theinconsistencies in the literature about emotionalinformation processing in depression/dysphoriamight be related to the use of tasks involvingimplicit or explicit facial emotion processing.

The most relevant result to be noticed is thatdifferent response patterns to facial expressionswere found in dysphoric and non-dysphoric indi-viduals during the early, but not late, stages ofprocessing. At frontal leads, the P200 was largerfor sad than neutral and happy facial expressionsamong dysphoric individuals, whereas controls

Figure 2. Grand-average event-related potentials waveforms at O1 and O2 in dysphoric and non-dysphoric individuals. The P100

component is marked by a black arrow.

BUODO ET AL.

10 COGNITION AND EMOTION, 2014

Dow

nloa

ded

by [

Uni

vers

ita d

i Pad

ova]

at 0

6:43

17

June

201

4

showed larger amplitude to both sad and happy ascompared with neutral expressions. The ERPP200 elicited by emotional cues is viewed as anindex of early, automatic detection of stimulussalience (a stimulus characteristic with strongaffective value), preceding a slower, strategicrecruitment of processing resources (e.g., Ashleyet al., 2004; Carretié et al., 2004; Thomas et al.,2007). On this basis, our findings indicate that incontrols, emotional facial expressions, both un-pleasant (sad) and pleasant (happy), capturedautomatic visual attention more than neutral faces,whereas in dysphoric individuals only sad facesappeared to be automatically detected, while happyfaces did not capture more attention than neutralfaces. It is thus possible that in individuals withself-reported elevated depressive symptoms theaffective salience of pleasant stimuli is abnormallyreduced, so that these stimuli fail to captureattention. On the other hand, unpleasant stimuliappear to be salient enough to draw attentionautomatically. Interestingly, these findings fit wellwith the behavioural data obtained with a dot-probe task by Shane and Peterson (2007), whoreported significantly reduced attention towardspositive stimuli in dysphoric than non-dysphoricindividuals. Because in dysphoric individuals thebias away from positive stimuli emerged morerobustly when the interval between word onsetand probe presentation was 200 ms, these findingsconverge with ours in suggesting that the reducedprocessing of positive stimuli in dysphoria may bemanifested at the early stages of processing relatedto initial orienting of attention.

Another group difference involving early pro-cessing stages concerned the P100 component.Irrespective of facial expression, P100 amplitudewas larger over the right than the left occipital areain controls, but not in dysphoric individuals. Somestudies in the literature do provide evidence of aright occipital lateralisation of P100 elicited byemotional facial stimuli among healthy individuals(Batty & Taylor, 2003, 2006; Holmes, Winston,& Eimer, 2005; Proverbio, Brignone, Matarazzo,Del Zotto, & Zani, 2006; Rousselet, Macé, &Fabre-Thorpe, 2004). Importantly, though, theP100 right occipital lateralisation cannot be

considered as a specific index of a hemisphericdominance for face processing, as it is often foundwhen visual tasks have been employed with non-face stimuli, such as illusory contours (Proverbio &Zani, 2002), and low spatial frequency patterns(Kenemans, Baas, Mangun, Lijffijt, & Verbaten,2000). Thus, in the context of our study, the rightlateralisation of P100 amplitude (or the lackthereof) might more generally be related to earlymodulation of spatial attention over sensory-evoked cortical activity in response to visualstimuli (Di Russo, Martínez, & Hillyard, 2003;Martínez et al. 2001). In a recent study on healthyindividuals, P100 amplitude was found to be largerover the right than over the left occipital sites inresponse to to-be-ignored, non-target emotionalfaces presented at an attended location (Wijers &Banis, 2012). In our study, participants wererequired to shift their attention from task-irrelev-ant emotional facial stimuli presented centrally atfixation, i.e., within the focus of spatial attention,towards a peripheral neutral target. Therefore, wecould reason that the right occipital lateralisationof P100 in controls might indicate enhancedattention to face stimuli and early sensory facilita-tion through rapid top-down effect on visualcortex, i.e., visual information falling within thespotlight of spatial attention is facilitated andpassed along to higher levels of processing (DiRusso et al., 2003; Hillyard & Anllo-Vento, 1998;Martínez et al. 2001). Indeed, there is evidencethat faces are hard to ignore, and that facedetection may be, at least in part, obligatory (e.g.,Lavie, Ro, & Russell, 2003). In contrast, atten-tion-related sensory facilitation for faces appearedto be reduced in dysphoric individuals, possiblybecause the target detection task might have beenmore attentionally demanding for dysphoric thanfor non-dysphoric individuals, and this may havenegatively impacted the initial stages of processingof the facial stimuli.

Lastly, a different lateralisation in dysphoricindividuals and controls also emerged for N100,with a more left-lateralised pattern in dysphoricindividuals with respect to controls. The func-tional significance of this previously unreportedeffect is difficult to interpret at present. Future

EMOTIONAL FACE PROCESSING IN DYSPHORIA

COGNITION AND EMOTION, 2014 11

Dow

nloa

ded

by [

Uni

vers

ita d

i Pad

ova]

at 0

6:43

17

June

201

4

research should further explore this issue bysystematically including the laterality factor in theanalysis of early ERP components.

Importantly, our findings do not appear to beaccounted for by the participants’ trait anxiety. Asexpected, based on the well-documented highcomorbidity of anxiety and depressive sympto-matology (e.g., Preisig et al., 2001), trait anxietywas significantly higher in the dysphoric than inthe control group. However, the potentially con-founding effect of trait anxiety was controlled forin our analyses, and therefore it is unlikely thatthis variable per se can explain the ERP findingsobtained in the present study. As also suggestedby studies where anxiety levels in depressedindividuals were measured and controlled forin data analysis (e.g., Donaldson et al., 2007;Oehlberg, Revelle, & Mineka, 2012), the influ-ence of depressive symptoms over emotionalprocessing appears to be genuine and independ-ent of co-occurring anxiety (see also Peckhamet al., 2010).

In contrast with early ERP components, RTsto neutral targets following the onset of emotionalfaces, and the P300/LPC did not highlight anysignificant difference between dysphoric and non-dysphoric individuals. Specifically, dysphoric indi-viduals did not respond slower to targets presentedafter the onset of sad than after the onset ofneutral faces, which would have indicatedimpaired disengagement from sad faces. Dys-phoric individuals also did not show larger latepositivity of the ERPs to sad faces, which wouldhave indicated increased allocation of attentionalresources to negative stimuli. Moreover, dysphoricindividuals did not respond faster to targetspresented after the onset of happy than after theonset of neutral faces, which would have indicatedfacilitated disengagement from happy faces, anddid not show smaller late positivity of the ERPs tohappy faces, which would be expected if reducedattentional allocation to positive stimuli was tooccur. Overall, then, these findings are consistentwith that part of the literature indicating that theallocation of attentional resources is not increasedfor negative cues among individuals with sub-clinical depressive symptoms (see Baert, De Raedt,

& Koster, 2010; Karparova, Kersting, & Suslow,2005; Koster et al., 2006; Mogg et al., 2000; Yovel& Mineka, 2004, 2005). It is possible that the taskwe employed was too simple for possible atten-tional biases to fully emerge in our sample ofdysphoric individuals. Also, it is to note that themajority of the participants in our sample had milddepressive symptoms, and it is possible thatmaintained attention to negative informationonly emerges when symptom severity is higher(i.e., moderate to severe; see Baert et al., 2010). Asproposed by some authors, individuals with sub-clinical symptoms, unlike those with a full-blownemotional disorder, may be able to override theirtendency to be distracted by emotional stimuli byusing top-down control strategies over attentionalresource allocation (Derryberry & Reed, 2002;Mathews & Mackintosh, 1998; Williams, Math-ews, & MacLeod, 1996; Williams, Watts,MacLeod, & Mathews, 1997). However, thereare other studies where individuals classified asdysphoric were indeed found to exhibit sustainedattention towards negative stimuli, either dueto impaired inhibitory processes (e.g., Gotlib,Neubauer Yue, & Joormann, 2005; Joormann,2004) or to impaired attentional disengagement(e.g., Caseras, Garner, Bradley, & Mogg, 2007;Koster, De Raedt, Leyman, & De Lissnyder,2010; Koster et al., 2005; Sears, Thomas,LeHuquet, & Johnson, 2010). Significant hetero-geneity across studies in terms of sample selectioncriteria and task designs may account for theobserved inconsistencies.

The analyses on the face-specific N170, acomponent related to structural encoding pro-cesses prior to face identification (Bentin &Deouell, 2000; Bentin et al., 1996), did not revealsignificant differences between dysphoric indivi-duals and controls. This finding is relevant in thatit suggests that the obtained emotion-relatedgroup differences on the P200 component werenot related to more general effects of facialconfiguration (see Ashley et al., 2004).

To summarise, group differences were found inearly, automatic ERP components suggesting thatat least under some circumstances the presence ofdepressive symptoms can modulate the early stages

BUODO ET AL.

12 COGNITION AND EMOTION, 2014

Dow

nloa

ded

by [

Uni

vers

ita d

i Pad

ova]

at 0

6:43

17

June

201

4

of emotional processing, rather than later, con-trolled processes related to the maintenance ofattention. Based on behavioural indices like RTs,most studies on the attentional bias in depression/dysphoria have instead supported the view thatautomatic, pre-attentive processes are unlikely tobe involved in the attentional bias in depression(Bradley et al., 1995; Mathews et al., 1996; Mogget al., 1993, 1995). However, RTs reflect the totaltime needed for information processing andresponse selection and execution, whereas electro-physiological and neuroimaging data can berelated to more specific stages of informationprocessing (see Suslow et al., 2010).

We acknowledge that the present study hassome limitations. In particular, the fact that thesample of participants was relatively small andconsisted of predominantly female undergraduatestudents may limit the generalisability of ourfindings to other samples (e.g., older individuals,community samples). Also, we did not computethe ERPs for neutral target shapes. These datawould provide critical information to clarify theprocesses of attentional engagement to, and dis-engagement from, emotional face stimuli, e.g.,reduced amplitude or delayed latency of ERPcomponents to targets following sad faces indysphoric individuals might indicate difficultydisengaging from sad faces to re-engage attentionwith the targets.

Future studies need to further explore, at theneural level, the modulation of depressive symp-toms over attention to emotional stimuli in non-clinical, dysphoric samples. Also, the possibilitythat transient depressed mood drives biases inemotional information processing deserves furtherinvestigation using mood-induction tasks.Increased knowledge of these issues would haverelevant clinical implications, both for the devel-opment of effective trainings and for the under-standing of the neural processes underlyingattentional re-training. For instance, computer-based trainings might be devised to reduce orprevent depressive symptoms by modifyingdeployment of attention to emotional information(Baert, De Raedt, Schacht, & Koster, 2010;Browning, Holmes, & Harmer, 2010).

Manuscript received 2 August 2013

Revised manuscript received 26 March 2014

Manuscript accepted 18 May 2014

First published online 11 June 2014

REFERENCES

Ashley, V., Vuilleumier, P., & Swick, D. (2004). Timecourse and specificity of event-related potentials to

emotional expressions. Neuroreport, 15, 211–216.

doi:10.1097/00001756-200401190-00041Baert, S., De Raedt, R., & Koster, E. H. W. (2010).

Depression-related attentional bias: The influence

of symptom severity and symptom specificity. Cog-

nition & Emotion, 24, 1044–1052. doi:10.1080/

02699930903043461Baert, S., De Raedt, R., Schacht, R., & Koster, E. H.

W. (2010). Attentional bias training in depression:

Therapeutic effects depend on depression severity.

Journal of Behavior Therapy & Experimental Psychi-

atry, 41, 265–274. doi:10.1016/j.jbtep.2010.02.004Bar-Haim, Y., Lamy, D., & Glickman, S. (2005).

Attentional bias in anxiety: A behavioral and ERPstudy. Brain and Cognition, 59(1), 11–22.doi:10.1016/j.bandc.2005.03.005

Batty, M., & Taylor, M. J. (2003). Early processing ofthe six basic facial emotional expressions. Cognitive

Brain Research, 17, 613–620. doi:10.1016/S0926-

6410(03)00174-5Batty, M., & Taylor, M. J. (2006). The development of

emotional face processing during childhood. Devel-

opmental Science, 9, 207–220. doi:10.1111/j.1467-7687.2006.00480.x

Beck, A.T., Steer, R. A., & Brown, G. K. (1996).Manual for the Beck Depression Inventory (2nd ed.).San Antonio, TX: The Psychological Association.

Beevers, C. G., & Carver, C. S. (2003). Attentional biasand mood persistence as prospective predictors ofdysphoria. Cognitive Therapy and Research, 27, 619–637. doi:10.1023/A:1026347610928

Bentin, S., Allison, T., Puce, A., Perez, E., & McCarthy,G. (1996). Electrophysiological studies of face per-ception in humans. Journal of Cognitive Neuroscience,8, 551–565. doi:10.1126/science.1598577

Bentin, S., & Deouell, L. Y. (2000). Structural encod-ing and identification in face processing: ERPevidence for separate mechanisms. Cognitive Neurop-sychology, 17, 35–54. doi:10.1080/026432900380472

Beuke, C. J., Fischer, R., & McDowall, J. (2003).Anxiety and depression: Why and how to measure

EMOTIONAL FACE PROCESSING IN DYSPHORIA

COGNITION AND EMOTION, 2014 13

Dow

nloa

ded

by [

Uni

vers

ita d

i Pad

ova]

at 0

6:43

17

June

201

4

their separate effects. Clinical Psychology Review, 23,831–848. doi:10.1016/S02727358(03)00074-6

Bistricky, S. L., Atchley, R. A., Ingram, R., & O’Hare,A. (2014). Biased processing of sad faces: An ERPmarker candidate for depression susceptibility. Cog-nition and Emotion, 28, 470–492. doi:10.1080/02699931.2013.837815

Blau, V. C., Maurer, U., Tottenham, N., & McCan-dliss, B. D. (2007). The face-specific N170 com-ponent is modulated by emotional facial expression.Behavioral and Brain Functions, 3, 7. doi:10.1186/1744-9081-3-7

Bradley, B. P., Mogg, K., & Lee, S. C. (1997).Attentional biases for negative information ininduced and naturally occurring dysphoria. BehaviourResearch and Therapy, 35, 911–927. doi:10.1016/S0005-7967(97)00053-3

Bradley, B. P., Mogg, K., Millar, N., & White, J.(1995). Selective processing of negative information:Effects of clinical anxiety, concurrent depression, andawareness. Journal of Abnormal Psychology, 104, 532–536. doi:10.1037/0021-843X.104.3.532

Browning, M., Holmes, E. A., & Harmer, C. J. (2010).The modification of attentional bias to emotionalinformation: A review of the techniques, mechan-isms, and relevance to emotional disorders. Cognit-ive, Affective, & Behavioral Neuroscience, 10, 8–20.doi:10.3758/CABN.10.1.8

Carretié, L., Hinojosa, J. A., Martín-Loeches, M.,Mercado, F., & Tapia, M. (2004). Automaticattention to emotional stimuli: Neural correlates.Human Brain Mapping, 22, 290–299. doi:10.1002/hbm.20037

Caseras, X., Garner, M., Bradley, B. P., & Mogg, K.(2007). Biases in visual orienting to negative andpositive scenes in dysphoria: An eye movementstudy. Journal of Abnormal Psychology, 116, 491–497. doi:10.1037/0021-843X.116.3.491

Cavanagh, J., & Geisler, M. W. (2006). Mood effectson the ERP processing of emotional intensity infaces: A P3 investigation with depressed students.International Journal of Psychophysiology, 60(1), 27–33. doi:10.1016/j.ijpsycho.2005.04.005

Cisler, J. M., & Koster, E. H. W. (2010). Mechanismsof attentional biases towards threat in anxietydisorders: An integrative review. Clinical PsychologyReview, 30, 203–216. doi:10.1016/j.cpr.2009.11.003

Dai, Q., & Feng, Z. (2009). Deficient inhibition ofreturn for emotional faces in depression. Progress inNeuro-Psychopharmacology and Biological Psychiatry,33, 921–932. doi:10.1016/j.pnpbp.2009.04.012

Dai, Q., Feng, Z., & Koster, E. H. W. (2011).Deficient distracter inhibition and enhanced facil-itation for emotional stimuli in depression: AnERP study. International Journal of Psychophysi-

ology, 79, 249–258. doi:10.1016/j.ijpsycho.2010.10.016

Dannlowski, U., Kersting, A., Donges, U.-S., Lalee-Mentzel, J., Arolt, V., & Suslow, Th. (2006).Masked facial affect priming is associated withtherapy response in clinical depression. European

Archives of Psychiatry and Clinical Neuroscience, 256,215–221. doi:10.1007/s00406-005-0628-0

Deldin, P. J., Keller, J., Gergen, J. A., & Miller, G. A.(2000). Right-posterior face processing anomaly indepression. Journal of Abnormal Psychology, 109(1),116–121. doi:10.1037/0021-843X.109.1.116

Deldin, P. J., Keller, J., Gergen, J. A., & Miller, G. A.(2001). Cognitive bias and emotion in neuropsycho-logical models of depression. Cognition & Emotion,15, 787–802. doi:10.1080/02699930143000248

Derryberry, D., & Reed, M. A. (2002). Anxiety-relatedattentional biases and their regulation by attentionalcontrol. Journal of Abnormal Psychology, 111, 225–236. doi:10.1037/0021-843X.111.2.225

Di Russo, F., Martínez, A., & Hillyard, S. A. (2003).Source analysis of event-related cortical activityduring visuo-spatial attention. Cerebral Cortex, 13,486–499. doi:10.1093/cercor/13.5.486

Donaldson, C., Lam, D., & Mathews, A. (2007).Rumination and attention in major depression.Behaviour Research and Therapy, 45, 2664–2678.doi:10.1016/j.brat.2007.07.002

Eimer, M., & Holmes, A. (2002). An ERP studyon the time course of emotional face processing.Neuroreport, 13, 427–431. doi:10.1097/00001756-200203250-00013

Eimer, M., Holmes, A., & McGlone, F. P. (2003). Therole of spatial attention in the processing of facialexpression: An ERP study of rapid brain responses tosix basic emotions. Cognitive, Affective, and Beha-

vioral Neuroscience, 3, 97–110.Fava, M., Rankin, M. A., Wright, E. C., Alpert, J. E.,

Nierenberg, A. A., Pava, J., & Rosenbaum, J. F.(2000). Anxiety disorders in major depression. Com-prehensive Psychiatry, 41, 97–102. doi:10.1016/S0010-440X(00)90140-8

Fox, E., Russo, R., Bowles, R., & Dutton, K. (2001).Do threatening stimuli draw or hold visual attentionin subclinical anxiety? Journal of Experimental Psy-

chology: General, 130, 681–700. doi:10.1037/0096-3445.130.4.681

BUODO ET AL.

14 COGNITION AND EMOTION, 2014

Dow

nloa

ded

by [

Uni

vers

ita d

i Pad

ova]

at 0

6:43

17

June

201

4

Frewen, P. A., & Dozois, D. J. A. (2005). Recognitionand interpretation of facial expressions in dysphoricwomen. Journal of Psychopathology and Behavioral

Assessment, 27, 305–315. doi:10.1007/s10862-005-2410-z

Ghisi, M., Flebus, G. B., Montano, A., Sanavio, E., &Sica, C. (2006). Beck depression inventory-second

edition. Adattamento Italiano: Manuale. Firenze:Organizzazioni Speciali.

Gotlib, I. H., & Cane, D. B. (1987). Constructaccessibility and clinical depression: A longitudinalinvestigation. Journal of Abnormal Psychology, 96,199–204. doi:10.1037/0021-843X.96.3.199

Gotlib, I. H., & Joormann, J. (2010). Cognition anddepression: Current status and future directions.Annual Review of Clinical Psychology, 6, 285–312.doi:10.1146/annurev.clinpsy.121208.131305

Gotlib, I. H., Kasch, K. L., Traill, S., Joormann, J.,Arnow, B. A., & Johnson, S. L. (2004). Coherenceand specificity of information-processing biases indepression and social phobia. Journal of Abnormal

Psychology, 113, 386–398. doi:10.1037/0021-843X.113.3.386

Gotlib, I. H., & McCann, C. D. (1984). Constructaccessibility and depression: An examination ofcognitive and affective factors. Journal of Personalityand Social Psychology, 47, 427–439. doi:10.1037/0022-3514.47.2.427

Gotlib, I. H., McLachlan, A. L., & Katz, A. N. (1988).Biases in visual attention in depressed and nonde-pressed individuals. Cognition & Emotion, 2, 185–200. doi:10.1080/02699938808410923

Gotlib, I. H., Neubauer Yue, D., & Joormann, J.(2005). Selective attention in dysphoric individuals:The role of affective interference and inhibition.Cognitive Therapy and Research, 29, 417–432.doi:10.1007/s10608-005-5753-6

Haaga, D. A. F., & Solomon, A. (1993). Impact ofKendall, Hollon, Beck, Hammen, and Ingram(1987) on treatment of the continuity issue in“depression” research. Cognitive Therapy and

Research, 17, 313–324. doi:10.1007/BF01177657Hillyard, S. A., & Anllo-Vento, L. (1998). Event-

related brain potentials in the study of visual selectiveattention. Proceedings of the National Academy of

Sciences, 95, 781–787. doi:10.1073/pnas.95.3.781Holmes, A., Winston, J. S., & Eimer, M. (2005). The

role of spatial frequency information for ERPcomponents sensitive to faces and emotional facialexpression. Cognitive Brain Research, 25, 508–520.doi:10.1016/j.cogbrainres.2005.08.003

Ilardi, S. S., Atchley, R. A., Enloe, A., Kwasny, K., &Garratt, G. (2007). Disentangling attentional biasesand attentional deficits in depression: An event-related potential P300 analysis. Cognitive Therapy

and Research, 31, 175–187. doi:10.1007/s10608-006-9113-y

Joormann, J. (2004). Attentional bias in dysphoria: Therole of inhibitory processes. Cognition & Emotion,18, 125–147. doi:10.1080/02699930244000480

Joormann, J., & Gotlib, I. H. (2007). Selective attentionto emotional faces following recovery from depres-sion. Journal of Abnormal Psychology, 116, 80–85.doi:10.1037/0021-843X.116.1.80

Karparova, S. P., Kersting, A., & Suslow, T. (2005).Disengagement of attention from facial emotion inunipolar depression. Psychiatry and Clinical Neuros-

ciences, 59, 723–729. doi:10.1111/j.1440-1819.2005.01443.x

Kayser, J., Bruder, G. E., Tenke, C. E., Stewart, J. E.,& Quitkin, F. M. (2000). Event-related potentials(ERPs) to hemifield presentations of emotionalstimuli: Differences between depressed patients andhealthy adults in P3 amplitude and asymmetry.International Journal of Psychophysiology, 36, 211–236. doi:10.1016/S0167-8760(00)00078-7

Kendall, P. C., Hollon, S. D., Beck, A. T., Hammen,C. L., & Ingrain, R. E. (1987). Issues and recom-mendations regarding use of the Beck DepressionInventory. Cognitive Therapy and Research, 11, 289–299. doi:10.1007/BF01186280

Kenemans, J. L., Baas, J. M. P., Mangun, G. R., Lijffijt,M., & Verbaten, M. N. (2000). On the processing ofspatial frequencies as revealed by evoked-potentialsource modeling. Clinical Neurophysiology, 111,1113–1123. doi:10.1016/S1388-2457(00)00270-4

Key, A. P. F., Dove, G. O., & Maguire, M. J. (2005).Linking brainwaves to the brain: An ERP primer.Developmental Neuropsychology, 27, 183–215.doi:10.1207/s15326942dn2702_1

Koster, E. H. W., De Raedt, R., Goeleven, E., Franck,E., & Crombez, G. (2005). Mood-congruent atten-tional bias in dysphoria: Maintained attention to andimpaired disengagement from negative information.Emotion, 5, 446–455. doi:10.1037/1528-3542.5.4.446

Koster, E. H. W., De Raedt, R., Leyman, L., & DeLissnyder, E. (2010). Mood-congruent attention andmemory bias in dysphoria: Exploring the coherenceamong information-processing biases. Behaviour

Research and Therapy, 48, 219–225. doi:10.1016/j.brat.2009.11.004

EMOTIONAL FACE PROCESSING IN DYSPHORIA

COGNITION AND EMOTION, 2014 15

Dow

nloa

ded

by [

Uni

vers

ita d

i Pad

ova]

at 0

6:43

17

June

201

4

Koster, E. H. W., Leyman, L., De Raedt, R., &Crombez, G. (2006). Cueing of visual attention byemotional facial expressions: The influence of indi-vidual differences in anxiety and depression. Person-ality and Individual Differences, 41, 329–339.doi:10.1016/j.paid.2005.12.022

Krompinger, J. W., & Simons, R. F. (2009). Electro-physiological indicators of emotion processing biasesin depressed undergraduates. Biological Psychology,81, 153–163. doi:10.1016/j.biopsycho.2009.03.007

Lasa, L., Ayuso-Mateos, J. L., Vázquez-Barquero, J. L,Díez-Manrique, F. J., & Dowrick, C. F. (2000). Theuse of the Beck depression inventory to screen fordepression in the general population: A preliminaryanalysis. Journal of Affective Disorders, 57, 261–265.doi:10.1016/S0165-0327(99)00088-9

Lavie, N., Ro, T., & Russell, C. (2003). The role ofperceptual load in processing distractor faces. Psycho-logical Science, 14, 510–515. doi:10.1111/1467-9280.03453

Luck, S. J. (2005). An introduction to the event-related

potential technique. Cambridge, MA: The MIT Press.Martinez, A., DiRusso, F., Anllo-Vento, L., Sereno,

M. I., Buxton, R. B., & Hillyard, S. A. (2001).Putting spatial attention on the map: Timing andlocalization of stimulus selection processes in striateand extrastriate visual areas. Vision Research, 41,1437–1457. doi:10.1016/S0042-6989(00)00267-4

Mathews, A., & Mackintosh, B. (1998). A cognitivemodel of selective processing in anxiety. CognitiveTherapy and Research, 22, 539–560. doi:10.1023/A:1018738019346

Mathews, A., Ridgeway, V., & Williamson, D. A.(1996). Evidence for attention to threatening stimuliin depression. Behaviour Research and Therapy, 34,695–705. doi:10.1016/0005-7967(96)00046-0

McCabe, S. B., & Gotlib, I. H. (1995). Selectiveattention and clinical depression: Performance on adeployment of attention task. Journal of Abnormal

Psychology, 104, 241–245. doi:10.1037/0021-843X.104.1.241

Mineka, S., Watson, D., & Clark, L. A. (1998).Comorbidity of anxiety and unipolar mood disorders.Annual Review of Psychology, 49, 377–412.doi:10.1146/annurev.psych.49.1.377

Mingtian, Z., Xiongzhao, Z., Jinyao, Y., Shuqiao, Y., &Atchley, R. A. (2011). Do the early attentionalcomponents of ERPs reflect attentional bias indepression? It depends on the stimulus presentationtime. Clinical Neurophysiology, 122, 1371–1381.doi:10.1016/j.clinph.2010.09.016

Mogg, K., & Bradley, B. P. (1998). A cognitive-motivational analysis of anxiety. Behaviour Research

and Therapy, 36, 809–848. doi:10.1016/S0005-7967(98)00063-1

Mogg, K., & Bradley, B. P. (2005). Attentional bias ingeneralized anxiety disorder versus depressive dis-order. Cognitive Therapy and Research, 29(1), 29–45.doi:10.1007/s10608-005-1646-y

Mogg, K., Bradley, B. P., & Williams, R. (1995).Attentional bias in anxiety and depression: The roleof awareness. British Journal of Clinical Psychology, 34(1), 17–36. doi:10.1111/j.2044-8260.1995.tb01434.x

Mogg, K., Bradley, B. P., Williams, R., & Mathews, A.(1993). Subliminal processing of emotional informationin anxiety and depression. Journal of Abnormal Psycho-logy, 102, 304–311. doi:10.1037/0021-843X.102.2.304

Mogg, K., Millar, N., & Bradley, B. P. (2000). Biases ineye movements to threatening facial expressions ingeneralized anxiety disorder and depressive disorder.Journal of Abnormal Psychology, 109, 695–704.doi:10.1037/0021-843X.109.4.695

Nandrino, J.-L., Dodin, V., Martin, P., & Henniaux,M. (2004). Emotional information processing in firstand recurrent major depressive episodes. Journal ofPsychiatric Research, 38, 475–484. doi:10.1016/j.jpsychires.2004.03.002

Oehlberg, K. A., Revelle, W., & Mineka, S. (2012).Time-course of attention to negative stimuli: Negat-ive affectivity, anxiety, or dysphoria? Emotion, 12,943–959. doi:10.1037/a0027227

Peckham, A. D., McHugh, R. K., & Otto, M. W.(2010). A meta-analysis of the magnitude of biasedattention in depression. Depression and Anxiety, 27,1135–1142. doi:10.1002/da.20755

Pedrabissi, L., & Santinello, M. (1989). Inventario

per l ’Ansia di ‘stato’ e di ‘tratto’: Nuova versione

Italiana dello STAI-Forma Y: Manuale [“State” and“Trait” Anxiety Inventory: New Italian version of theSTAI-form Y: Manual]. Firenze: OrganizzazioniSpeciali.

Picton, T. W. (1992). The P300 wave of the humanevent-related potential. Journal of Clinical Neuro-

physiology, 9, 456–479. doi:10.1097/00004691-199210000-00002

Preisig, M., Merikangas, K. R., & Angst, J. (2001).Clinical significance and comorbidity of subthresholddepression and anxiety in the community. Acta

Psychiatrica Scandinavica, 104, 96–103. doi:10.1034/j.1600-0447.2001.00284.x

Proverbio, A. M., Brignone, V., Matarazzo, S., DelZotto, M., & Zani, A. (2006). Gender differences in

BUODO ET AL.

16 COGNITION AND EMOTION, 2014

Dow

nloa

ded

by [

Uni

vers

ita d

i Pad

ova]

at 0

6:43

17

June

201

4

hemispheric asymmetry for face processing. BMC

Neuroscience, 7(1), 44. doi:10.1186/1471-2202-7-44Proverbio, A. M., & Zani, A. (2002). Electrophysiolo-

gical indexes of illusory contours perception inhumans. Neuropsychologia, 40, 479–491. doi:10.1016/S0028-3932(01)00135-X

Rousselet, G. A., Macé, M. J.-M., & Fabre-Thorpe, M.(2004). Animal and human faces in natural scenes: Howspecific to human faces is the N170 ERP component?Journal of Vision, 4, 13–21. doi:10.1167/4.1.2

Sarlo, M., & Munafò, M. (2010). When faces signaldanger: Event-related potentials to emotional facialexpressions in animal phobics. Neuropsychobiology,62, 235–244. doi:10.1159/000319950

Sears, C. R., Thomas, C. L., LeHuquet, J. M., &Johnson, J. C. S. (2010). Attentional biases indysphoria: An eye-tracking study of the allocationand disengagement of attention. Cognition & Emotion,24, 1349–1368. doi:10.1080/02699930903399319

Shane, M. S., & Peterson, J. B. (2007). An evaluationof early and late stage attentional processing ofpositive and negative information in dysphoria.Cognition & Emotion, 21, 789–815. doi:10.1080/02699930600843197

Shestyuk, A. Y., Deldin, P. J., Brand, J. E., & Deveney,C. M. (2005). Reduced sustained brain activityduring processing of positive emotional stimuli inmajor depression. Biological Psychiatry, 57, 1089–1096. doi:10.1016/j.biopsych.2005.02.013

Sloan, D. M., Strauss, M. E., Quirk, S. W., & Sajatovica,M. (1997). Subjective and expressive emotionalresponses in depression. Journal of Affective Disorders,46, 135–141. doi:10.1016/S0165-0327(97)00097-9

Spielberger, C. D. (1983). Manual for the state-trait

anxiety inventory (Form Y ). Palo Alto, CA: Con-sulting Psychologist Press.

Sprinkle, S. D., Lurie, D., Insko, S. L., Atkinson, G.,Jones, G. L., Logan, A. R., & Bissada, N. N. (2002).Criterion validity, severity cut scores, and test–retestreliability of the Beck Depression Inventory-II in auniversity counseling center sample. Journal of Coun-seling Psychology, 49, 381–385. doi:10.1037/0022-0167.49.3.381

Surguladze, S. A., Young, A. W., Senior, C., Brébion, G.,Travis, M. J., & Phillips, M. L. (2004). Recognitionaccuracy and response bias to happy and sad facialexpressions in patients with major depression.Neuropsy-chology, 18, 212–218. doi:10.1037/0894-4105.18.2.212

Suslow, T., Konrad, C., Kugel, H., Rumstadt, D.,Zwitserlood, P., Schöning, S., … Dannlowski, U.(2010). Automatic mood-congruent amygdala

responses to masked facial expressions in majordepression. Biological Psychiatry, 67, 155–160.doi:10.1016/j.biopsych.2009.07.023

Tennen, H., Hall, J. A., & Affleck, G. (1995).Depression research methodologies in the Journalof Personality and Social Psychology: A review andcritique. Journal of Personality and Social Psychology,68, 870–884. doi:10.1037/0022-3514.68.5.870

Thomas, S. J., Johnstone, S. J., & Gonsalvez, C. J.(2007). Event-related potentials during an emotionalStroop task. International Journal of Psychophysiology,63, 221–231. doi:10.1016/j.ijpsycho.2006.10.002

Tottenham, N., Tanaka, J. W., Leon, A. C., McCarry,T., Nurse, M., Hare, T. A., …Casey, B., & Nelson,C. (2009). The NimStim set of facial expressions:Judgments from untrained research participants.Psychiatry Research, 168, 242–249. doi:10.1016/j.psychres.2008.05.006

Wijers, A. A., & Banis, S. (2012). Foveal and parafovealspatial attention and its impact on the processing offacial expression: An ERP study. Clinical Neurophysi-ology, 123, 513–526. doi:10.1016/j.clinph.2011.07.040

Williams, J. M. G., Mathews, A., & MacLeod, C.(1996). The emotional Stroop task and psychopatho-logy. Psychological Bulletin, 120(1), 3–24. doi:10.1037/0033-2909.120.1.3

Williams, J. M. G., Watts, F. N., MacLeod, C., &Mathews, A. (1997). Cognitive psychology and emo-

tional disorders (2nd ed.). Chichester: Wiley.Wisco, B. E. (2009). Depressive cognition: Self-reference

and depth of processing. Clinical Psychology Review,29, 382–392. doi:10.1016/j.cpr.2009.03.003

Woldorff, M. G., Liotti, M., Seabolt, M., Busse, L.,Lancaster, J. L., & Fox, P. T. (2002). The temporaldynamics of the effects in occipital cortex of visual-spatial selective attention. Cognitive Brain Research,15, 1–15. doi:10.1016/S0926-6410(02)00212-4

Yao, S., Liu, M., Liu, J., Hu, Z., Yi, J., & Huang, R.(2010). Inhibition dysfunction in depression: Event-related potentials during negative affective priming.Psychiatry Research: Neuroimaging, 182, 172–179.doi:10.1016/j.pscychresns.2010.01.010

Yovel, I., & Mineka, S. (2004). Hierarchical models ofemotional disorders and emotion-congruent cognit-ive biases. Personality and Individual Differences, 36,679–694. doi:10.1016/S0191–8869(03)00125-9

Yovel, I., & Mineka, S. (2005). Emotion-congruentattentional biases: The perspective of hierarchicalmodels of emotional disorders. Personality and Indi-

vidual Differences, 38, 785–795. doi:10.1016/j.paid.2004.06.003

EMOTIONAL FACE PROCESSING IN DYSPHORIA

COGNITION AND EMOTION, 2014 17

Dow

nloa

ded

by [

Uni

vers

ita d

i Pad

ova]

at 0

6:43

17

June

201

4