This article was downloaded by: [Stony Brook University]On: 22 October 2014, At: 17:23Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Social NeurosciencePublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/psns20

Functional connectivity with the anterior cingulateis associated with extraversion during the emotionalStroop taskBrian W. Haas a , Kazufumi Omura a , Zenab Amin a , R. Todd Constable b & Turhan Canli ca Stony Brook University , Stony Brook, NY, USAb Yale University School of Medicine , New Haven, CT, USAc Stony Brook University , Stony Brook, NY, USAPublished online: 20 Jan 2011.

To cite this article: Brian W. Haas , Kazufumi Omura , Zenab Amin , R. Todd Constable & Turhan Canli (2006) Functionalconnectivity with the anterior cingulate is associated with extraversion during the emotional Stroop task, SocialNeuroscience, 1:1, 16-24, DOI: 10.1080/17470910600650753

To link to this article: http://dx.doi.org/10.1080/17470910600650753

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Functional connectivity with the anterior cingulate isassociated with extraversion during the emotional

Stroop task

Brian W. Haas, Kazufumi Omura and Zenab Amin

Stony Brook University, Stony Brook, NY, USA

R. Todd Constable

Yale University School of Medicine, New Haven, CT, USA

Turhan Canli

Stony Brook University, Stony Brook, NY, USA

Previous research has investigated the association of personality traits with brain activation in responseto emotional stimuli. Our current research efforts are directed at understanding the temporal dynamicsof networks of structures associated with particular personality traits, and gain insights into the functionalcontributions of more narrowly defined trait-facets that comprise these personality traits. To begin thisprocess, we conducted a functional magnetic resonance imaging (fMRI) study using an emotionalattention task (emotional Stroop paradigm) and addressed the question whether individual differences inextraversion and its lower-level facets were associated with differences in activation, and in functionalconnectivity, of the anterior cingulate (AC) cortex. We replicated our earlier finding that extraversionwas associated with increased AC activation to positive, relative to neutral, word stimuli, but now showthat distinct facets of extraversion can account for this association. When analyzing for functionalconnectivity, we found that greater extraversion across individuals was associated with greater functionalconnectivity between the AC and the inferior parietal lobule, and that this association could also beaccounted for by distinct facets of extraversion. Our data suggest that extraversion and some of its lower-level facets are associated with individual differences across a network of structures believed to be criticalin cognitive and affective processing.

INTRODUCTION

Extraversion is posited to be a higher-order trait

that is central in contemporary ‘‘Big Five’’ and

earlier influential models of personality (Costa &

McCrae, 1992; Eysenck, 1990; John & Srivastava,

1999). Underlying extraversion is a ‘‘positive

emotional core’’ (Watson & Clark, 1997), which

is expressed in the predisposition of extraverted

individuals to experience positive mood states

more frequently (Costa & McCrae, 1980) and to

attend to positively valenced stimuli more

strongly (Amin, Constable, & Canli, 2004; Derry-

berry, 1994) than less extraverted individuals. As

a higher-order trait, extraversion can be concep-

tualized as comprised of lower-level elements

or facets. For example, Depue & Morrone-

Strupinsky (2005) hypothesize two forms of

Correspondence should be addressed to: Brian W. Haas, Department of Psychology, Stony Brook University, Stony Brook, NY

11794�2500, USA. E-mail: bhaas@ic.sunysb.edu

This material is based upon work supported by the National Science Foundation under Grant No. 0224221 and by Stony Brook

University.

# 2006 Psychology Press Ltd

SOCIAL NEUROSCIENCE, 2006, 1 (1), 16�24

www.psypress.com/socialneuroscience DOI:10.1080/17470910600650753

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 1

7:23

22

Oct

ober

201

4

extraversion, affiliative and agentic, that areproposed to map onto distinct neural networks.Another example is an influential five-factormodel of personality (Costa & McCrae, 1992)that conceptualizes six facets of extraversion toinclude Warmth, Gregariousness, Assertiveness,Activity, Excitement Seeking, and Positive Emo-tions.

Previous studies have begun to map extraver-sion onto neural circuits processing a variety ofemotional stimuli, such as pictures (Canli, Zhao,Desmond, Kang, Gross, & Gabrieli, 2001), faces(Canli, Sivers, Whitfield, Gotlib, & Gabrieli,2002), and words (Canli, Amin, Haas, Omura, &Constable, 2004). However, it is not knownwhether activation in these circuits is differen-tially driven by the lower-level facets of extraver-sion, nor is it known whether the functionalconnectivity between brain regions varies acrossindividuals as a function of extraversion or itslower-level facets.

To address these questions, we used functionalmagnetic resonance imaging (fMRI) and animplicit emotional attention task known as theemotional Stroop paradigm (Whalen et al., 1998).In this task, participants are asked to make colorjudgments of words that are of varying emotionalvalence (e.g., positive valence: the word joywritten in blue). Prior imaging studies have foundactivation differences in the rostral anteriorcingulate (AC; Whalen et al., 1998) and theinferior parietal lobule (IPL; Compton et al.,2003) in response to emotional relative to neutralwords during this task. In addition, this task haspreviously been used to asses individual differ-ences in AC function in posttraumatic stressdisorder (PTSD; Shin et al., 2001) and attentiondeficit hyperactivity disorder (ADHD; Bush etal., 1999) patients relative to controls. Further-more, we have previously shown that AC activa-tion during the emotional Stroop task varies withextraversion, even when controlling for positivemood (Canli et al., 2004). Thus, we used theemotional Stroop task to elicit activation in theAC as an a priori region of interest, and plannedto investigate individual differences in AC activa-tion and functional connectivity as a function ofextraversion and its lower-level elements.

Based on findings that have associated extra-version with attention to positive stimuli (Derry-berry, 1994), we predicted that individuals withhigher levels of extraversion would show beha-vioral evidence for greater interference by posi-tive words, compared to neutral words, as

measured by longer reaction times. Based onneuroimaging studies implicating both AC andIPL in the emotional Stroop task (Compton et al.,2003), we predicted that individuals with higherlevels of extraversion would also show increasedlevels of functional connectivity between the ACand IPL in response to positive relative to neutralwords. Finally, we were interested in determiningthe individual contributions of the lower-levelfacets of extraversion to observed behavioral,activation and functional connectivity effects.

METHODS

Participants

Twenty-six healthy right-handed participants (12males) were recruited from the campus of StonyBrook University or Yale University. Subjects’mean age was 22.0 years (SD�/2.7, range�/

18�28). Subjects had no history of brain injury,reported no substance abuse within the past sixmonths, were not on any mood-altering medica-tion, and had no physical limitations that prohib-ited them from participating in an fMRI study.The study was conducted with approval from theInstitutional Review Boards of both Stony BrookUniversity and Yale University. Informed consentwas obtained from all participants.

Personality and mood questionnaires

Prior to scanning, all participants completed theRevised NEO Personality Inventory (NEO PI-R;Costa & McCrae, 1992). The NEO PI-R coverseach of the ‘‘Big Five’’ personality traits (neuroti-cism, extraversion, openness, agreeableness, andconscientiousness) and measures the six lower-level facets of extraversion (E1: Warmth, E2:Gregariousness, E3: Assertiveness, E4: Activity,E5: Excitement Seeking, and E6: Positive Emo-tions). Personality data were scored to representT values, with the population mean defined asT�/50 and one standard deviation of T�/10.Participants also completed a measure of theirmood state prior to the scan, the Profile of MoodStates (POMS; McNair, Lorr, & Droppleman,1971), which quantifies separate subscalesrelated to positive mood (vigor) and negativemood (tension, depression, anger, fatigue, andconfusion).

EXTRAVERSION, ANTERIOR CINGULATE CONNECTIVITY 17

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 1

7:23

22

Oct

ober

201

4

Stimuli

Word stimuli were selected on the basis of theiraffective rating from the Affective Norms forEnglish Words (ANEW; Bradley & Lang, 1999)stimulus library. Based on normed valence ratingson a scale from 1 to 9, with 1 indicating the mostnegative rating and 9 indicating the most positiverating, three sets of 24 word stimuli were con-structed: positive (M�/8.03, SD�/0.25), neutral(M�/5.32, SD�/0.29) and negative (M�/2.00,SD�/0.28). These three words categories wereincluded in order to maintain an analogous designrelative to our previous findings (Canli et al.,2004), though for the current research on E, thepositive and neutral word conditions are ofparticular interest. The three sets differed sig-nificantly in normed valence ratings, F(2, 69)�/

2944.15, pB/.0001, but did not differ significantlyin arousal, F(2, 69)�/0.39, p�/.96, word length,F(2, 69)�/ 2.24, p�/ .16, or word frequency,F(2, 69)�/0.01, p�/.99.

Behavioral procedures andexperimental design

Participants were placed in the scanner as theyresponded to positive, neutral and negative wordsand as they passively viewed a fixation crossduring a baseline resting condition. Participantswere asked to respond as quickly and as accu-rately as possible to words that were presented inblue, green or red font. Each trial lasted 3000milliseconds (ms), of which 1500 ms consisted ofstimulus presentation and 1500 ms consisted of afixation cross. Response times were recordedfollowing the onset of each stimulus. Each parti-cipant was presented with 12 word blocks (4 foreach condition) and 4 fixation blocks. Each blockconsisted of 6 trials. Thus, each participantresponded to a total of 72 novel words (24 foreach condition).

fMRI procedures

Whole-brain imaging data were acquired on a 3TSiemens Trio Scanner. For structural whole-brainimages, a three-dimensional high resolutionspoiled gradient scan and a T1 scan (24 slices, 5mm thickness; oriented parallel to the linebetween the anterior and posterior commissure)

were conducted. Functional images were acquiredusing a gradient echo T2*-weighted echoplanarimaging (EPI) scan that was conducted with a flipangle of 808, repetition time (TR) of 1.5 s, echotime (TE) of 30 ms, and a field of view (FOV) of a220�/220 mm matrix.

Functional data were preprocessed and statis-tically analyzed using SPM2 (Wellcome Depart-ment of Cognitive Neurology). The images weretemporally realigned to the middle slice, spatiallyrealigned to the first in the time series, coregis-tered to the T1 volume image, which was seg-mented and normalized to the gray mattertemplate. Spatial transformations derived fromnormalizing the segmented gray matter were thenapplied to all functional volumes, which werethen spatially smoothed with an 8 mm full width�half maximum isotropic Gaussian filter.

Fixed-effects models (Friston, 1994) were usedat the individual subject level of analysis andrandom effects models (Holmes & Friston, 1998)were used for group-level analyses. At the in-dividual level, models were created (generallinear model) in order to represent all conditionsand all data were then high-pass filtered. Forfunctional analyses restricted to the a prioriregion of interest (AC), we used an automatedmethod for generating ROI masks based on theTalairach Daemon database (Maldjian, Lavrienti,Kraft, & Burdette, 2003) and applied a signifi-cance threshold of pB/.05 and an extent thresholdof 50 voxels (uncorrected) within this ROIcorresponding to a per-pixel false positive rateof p�/.0002 (Forman, Cohen, Fitzgerald, Eddy,Mintun, & Noll, 1995).

Functional connectivity analysis

In order to determine areas exhibiting differentialconnectivity with the AC during the positiverelative to the neutral word conditions, a psycho-physiological interaction (PPI) analysis was per-formed (Friston, Buechel, Fink, Morris, Rolls, &Dolan, 1997). This analysis identifies regions thatare significantly more positively correlated withthe AC during the positive word condition,relative to the neutral word condition.

For each subject, we computed the localmaximum within the rostral area of the ACresponse to positive relative to neutral wordconditions. The rostral region of the AC wasselected due to its proposed role in affectiveprocessing (Bush, Luu, & Posner, 2000) and the

18 HAAS ET AL.

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 1

7:23

22

Oct

ober

201

4

emotional Stroop task (Whalen et al., 1998). Thisarea was defined by using a ROI mask based onthe Talairach Daemon database (Maldjian et al.,2003). The individual time course from an 8 mmradius sphere for each subject was then extracted.These time courses were representative of thefirst principal component in the time series of allthe raw voxels within the sphere. In addition, thetime courses were high-pass filtered and meancorrected. These data were then entered into aPPI analysis for each subject modeled such thatthe positive word condition was weighted as 1 andthe neutral word condition was weighted as �/1.Therefore, this analysis assesses which brain lociare significantly more positively correlated withthe AC during the positive word conditionrelative to the neutral word condition.

At the group level, all of individual t-contrasts(representative of the above stated interaction)were entered into a random effects regressionanalysis with higher scores of extraversion pre-dicting increased functional connectivity withthe AC. We employed a significance thresholdof pB/.005 and an extent threshold of 20 voxels(uncorrected) for this analysis. In order to in-vestigate the individual contributions of eachfacet, we used a multiple regression analysis andrestricted the search space to those voxels identi-fied to be associated with E alone.

RESULTS

Reaction time

We investigated differences in reaction time bymeans of a 3�/1 analysis of variance (ANOVA).This analysis revealed that reaction times during

the positive (M�/702.12, SD�/119.33), neutral(M�/683.42, SD�/113.38), and negative (M�/

693.69, SD�/92.93) word conditions were notsignificantly different from one another, F(2,75)�/0.19, p�/.826. In addition, there were nosignificant differences between the positive andneutral word conditions, F(1, 50)�/0.34, p�/.565,or between the negative and neutral word condi-tions, F(1, 50)�/0.13, p�/.722).

We also investigated the relationship betweenRT with extraversion and the lower-level facetscores. Table 1 displays the correlations betweenextraversion and facet scores and the RT differ-ence between positive and neutral word condi-tions. Higher scores of this difference score reflectgreater interference in response to positivelyvalenced words. As shown in the table, extraver-sion and four of the facets (E1: warmth, E2:Gregariousness, E5: Excitement Seeking and E6:Positive Emotions) were significantly and posi-tively correlated with the RT difference score,while two of the facets (E3: Assertiveness and E4:Activity) were not. The null finding for E3 isconsistent with the insignificant intercorrelationsof the E3 with the other facets scores in oursample. No significant relationships were foundbetween any other personality or mood measureand the RT difference between positive andneutral or negative and neutral words.

AC activation as a function ofextraversion

Figure 1 displays AC activation in response topositive relative to neutral words as a function ofextraversion. We found that extraversion (whencontrolling for positive mood and sex) was

TABLE 1

Reaction time (ms) during each emotional condition

P-N RT Et E1 E2 E3 E4 E5

Et .72** *E1 .70** .82** *E2 .62* .70** .59* *E3 �/.02 .27 .01 �/.11 *E4 .37 .70** .43* .15 .20 *E5 .51* .75** .40* .56* .19 .43* *E6 .68** .90** .82** .46* .16 .66** .53*

Note : P-N RT: Reaction Time difference between the Positive and Neutral word conditions; Et: Extraversion total scored to

represent T values; E1: Warmth facet of Extraversion; E2: Gregariousness facet of Extraversion; E3: Assertiveness facet of

Extraversion; E4: Activity facet of Extraversion; E5: Excitement Seeking facet of Extraversion; E6: Positive Emotions facet of

Extraversion. *p B/.05; **p B/.001.

EXTRAVERSION, ANTERIOR CINGULATE CONNECTIVITY 19

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 1

7:23

22

Oct

ober

201

4

associated with increased AC activation in re-

sponse to positive relative to neutral words.

Consistent with our previous findings (Canli et

al., 2004), the peak voxel (MNI: �/16, 50, 2; p�/

.002) in response to this contrast was localized in

the rostral AC. This activation was not driven by a

motor response, in that no significant AC activa-

tion was found to be associated with each

subjects’ RT difference score between positive

and neutral words.In order to investigate the individual contribu-

tions of the lower-level facet scores to this

observed effect, we used a multiple regression

analysis, which tested for the relationship of each

facet score with AC activation observed to be

associated with extraversion alone while control-

ling for all other facet scores. Specifically, we

restricted our search space to an area defined as

the voxels found to be significantly correlated

with extraversion in the AC (Figure 1). Figures 2

and 3 display AC activation significantly asso-

ciated with the facet scores of extraversion. This

analysis revealed that two of the six facet scores

(E1: Warmth, E5: Excitement Seeking) signifi-

cantly contributed to this observed activation (at

the p�/.05 level). E1 (Warmth) was associated

with two clusters localized in the left AC and one

in the right AC (Figure 2), while E5 (Excitement

Seeking) was associated with two clusters loca-

lized in the right AC (Figure 3).

Functional connectivity with AC as afunction of E

Figure 4 displays and Table 2 lists areas where

higher scores in extraversion were associated with

greater functional connectivity to the AC during

the positive relative to neutral word conditions.

Figure 1. Sagital and coronal view of significant anterior

cingulate (AC) activation correlating with extraversion (con-

trolling for positive mood and sex). Three significant clusters

were found displaying this pattern (Left AC. 494 voxels; peak

voxel MNI: �/16, 50, 2; t�/3.26, p�/.002, L. AC. 58 voxels;

MNI: �/8, 34, 32; t�/2.68, p�/.007, R. AC. 54 voxels; MNI: 6,

46, 12; t�/2.58, p�/.009).

Figure 2. Sagital and coronal view of significant anterior

cingulate (AC) activation correlating with E1 (Warmth) within

the restricted search space found to correlate with extraver-

sion (Figure 1). Three significant clusters were found display-

ing this pattern (Left AC. 69 voxels; peak voxel MNI: �/16,

44, �/2; t�/3.37, p�/.002, R. AC. 71 voxels; MNI: 8, 26,

�/4; t�/2.94, p�/.004, L. AC. 23 voxels; MNI: �/2, 26,

�/4; t�/2.42, p�/.013).

Figure 3. Sagital and coronal view of significant anterior

cingulate (AC) activation correlating with E5 (Excitement

Seeking) within the restricted search space found to correlate

with extraversion (Figure 1). Two significant clusters were

found displaying this pattern (Right AC. 37 voxels; peak voxel

MNI: 4, 48, 10; t�/2.65, p�/.008, R. AC. 51 voxels; MNI: 8, 26,

�/4; t�/2.35, p�/.009).

20 HAAS ET AL.

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 1

7:23

22

Oct

ober

201

4

This analysis revealed a large cluster on the rightinferior parietal lobule (MNI: 50, �/36, 36, 218voxels; peak voxel: pB/.001). In addition, clusterswere found localized on the right middle frontalgyrus (MFG; pB/.001), right fusiform gyrus (pB/

.001), and right orbital frontal gyrus (p�/.001).We then investigated the individual contribu-

tions of the lower-level facet scores to thisobserved effect. We conducted a multiple regres-sion analysis for each facet score while controllingfor all others and restricted our search space to anROI defined as the functional connectivity foundto be associated with extraversion (Figure 4). Thisanalysis revealed that three of the six lower-levelfacet scores (E1: Warmth, E2: Gregariousness,E6: Positive Emotions) significantly contributedto the observed functional connectivity associatedwith extraversion. E1 (warmth) predicted con-nectivity with the right MFG (MNI: 36, 38, 0, 11voxels; p�/.009) and the right postcentral gyrus(MNI: 12, �/28, 58, 16 voxels; p�/.01). E2(Gregariousness) and E6 (Positive Emotions)

also predicted connectivity with the right MFG(E2: MNI: 24, 20, 30; 88 voxels; p�/.005. E6: MNI:34, 26, 28, 16 voxels; p�/.017).

DISCUSSION

Several recent studies have investigated theassociation of personality with brain activationwith respect to localized brain activation (Canliet al., 2002; Eisenberger, Lieberman, & Satpute,2005; Gray & Braver, 2002; Reuter et al., 2004)but the association of personality with brainfunctional connectivity is unknown. In this study,we demonstrated that higher extraversion scoreswere associated with greater functional connec-tivity between the AC and a number of otherbrain regions during the processing of positiverelative to neutral word stimuli. Furthermore, wealso extended our previous findings (Canli et al.,2004) to show that AC activation associated withextraversion can be further dissociated by taking

Figure 4. Functional connectivity with AC predicted by extraversion. Areas displaying increased functional connectivity with the

AC as a function of extraversion (E) overlaid upon a standardized template brain. The largest cluster was localized on the right

inferior parietal lobule (IPL; 218 voxels; peak voxel MNI: 50, �/36, 36; t�/3.84, p B/.001). Complete list of locations, number of

voxels, and significance values for each cluster are reported in Table 2.

TABLE 2

Areas displaying increased functional connectivity with the anterior cingulate associated with extraversion

MNI Co-ordinates

Area x y z size t-value p-value

R. MFG 30 24 28 148 4.39 B/.001

R. postcentral gyrus 14 �/32 58 84 3.98 B/.001

R. fusiform gyrus 42 �/46 �/14 28 3.93 B/.001

R. IFG 38 38 2 59 3.89 B/.001

R. IPL 50 �/36 30 218 3.84 B/.001

R. OFC 16 48 �/6 30 3.64 .001

L. calcarine sulcus �/16 �/64 12 24 3.42 .001

Note : MFG: Middle frontal gyrus; IFG; Inferior frontal gyrus; IPL Inferior parietal lobule; OFC: Orbital frontal cortex.

EXTRAVERSION, ANTERIOR CINGULATE CONNECTIVITY 21

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 1

7:23

22

Oct

ober

201

4

into account the contributions of the lower-levelfacet scores of extraversion.

These results demonstrate that changes inrostral AC activation are associated with thepersonality trait of extraversion. This is consistentwith our previous work (Canli et al., 2004) as wellas recent study using a cognitive oddball para-digm (Eisenberger et al., 2005). In particular, therostral affective division of the AC has beenimplicated in assessing the salience of emotionaland motivational information and in the regula-tion of emotional responses (Bush et al., 2000).We observed increased rostral AC activation insubjects who scored higher on E and E1(Warmth) and E5 (Excitement Seeking) in re-sponse to positive relative to neutral words. Thisfinding is consistent with research demonstratingthe AC to be responsive to happy mood induction(Habel, Klein, Kellermann, Shah, & Schneider,2005) and increased reward (Rogers et al., 2004).A higher degree of extraversion may be asso-ciated with a tendency to be more sensitive topositive stimuli and therefore to experience morepositive mood states. This is supported by beha-vioral (Rusting, 1999) and fMRI (Canli et al.,2004) data demonstrating the interaction of moodstates and personality traits.

Our finding that the extraversion facets ofWarmth and Excitement Seeking contributed toAC activation during the processing of positiveword stimuli suggests that these facets may beparticularly related to the positive emotionalattention mechanisms assessed by the emotionalStroop. Depue & Morrone-Strupinsky (2005)have recently suggested that extraversion can bebroken down into two neurobiologically indepen-dent trait elements of affiliation and agency.Affiliation is characterized by being warm, affec-tionate, and valuing close interpersonal bonds,whereas agency is characterized by a tendency tobe socially dominant and the enjoyment ofleadership roles (Morrone-Strupinsky & Depue,2004). The NEO PI-R facets of Warmth (E1) andGregariousness (E2) are thought to load stronglyon the affiliation trait (Costa & McCrae, 1992;Morrone-Strupinsky & Depue, 2004). In ourstudy high scores of the Warmth (E1) subscalewere found to be associated primarily with leftAC activation. This may suggest that individualsscoring high on this facet are particularly skilledin communicative processes and display a heigh-tened sensitivity to positive stimuli presentedsemantically. This is consist with research demon-strating left hemisphere function associated with

language (Buxhoeveden & Casanova, 2000) andwith research demonstrating a correlation be-tween AC activation and subjects’ capacity toexperience emotion in oneself and in others(Lane, Reiman, Axelrod, Yun, Holmes, &Schwartz, 1998).

The finding of several structures displayingincreased connectivity with the AC is consistentwith our understanding of the functional anatomyof this structure (Bush et al., 2000). The AC isimplicated in several models of emotion (Beau-regard, Levesque, & Bourgouin, 2001; Killgore &Yurgelun-Todd, 2004) and cognition (Botvinick,Cohen, & Carter, 2004; Garavan, Ross, Kaufman,& Stein, 2003) and likely mediates a wide varietyof attentional processes (Bush et al., 2000). Inparticular, we observed that extraversion wasassociated with increased AC connectivity withthe inferior parietal lobule (IPL). The IPL is astructure implicated in a many aspects of atten-tion (Behrmann, Geng, & Shomstein, 2004; Sha-piro, Hillstrom, & Husain, 2002) and haspreviously been shown to be activated in responseto emotional relative to neutral words during theemotional Stroop task (Compton et al., 2003).Our finding suggests that extraversion is linked toincreases in coupling between the AC and IPLthat correspond to greater attention to positivelyvalenced stimuli. This is consist with other re-search that has demonstrated this pattern bothbehaviorally (Derryberry, 1994) and neurally(Amin et al., 2004).

Three of the six lower-level facets of extraver-sion (E1: Warmth, E2: Gregariousness, and E6:Positive Emotions) were associated with in-creased AC connectivity with the right middlefrontal gyrus (MFG). This finding may indicate amodulatory role for personality in the cognitiveaspects of this task. Previous research has foundan association between personality and brainfunction during working memory (Gray &Braver, 2002) and cognitive control (Gray, Bur-gess, Schaeffer, Yarkoni, Larsen, & Braver, 2005)tasks and that MFG activation is greater duringconditions of increased cognitive control (Derr-fuss, Brass, & von Cramon, 2004; Egner & Hirsch,2005). These mechanisms of cognitive controlmay explain why Warmth and Gregariousness(forms of affiliative extraversion, according toDepue) are associated with increased MFGconnectivity with the AC. Individuals with ele-vated levels of affiliative extraversion may haveto exert greater effort to maintain attention to the

22 HAAS ET AL.

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 1

7:23

22

Oct

ober

201

4

cognitive task (word color) while ignoring emo-tionally affiliative information.

Our analysis strategy revealed that individualdifferences in extraversion and some of its facetsare associated with changes in functional con-nectivity, but is silent on the directionality of thisconnectivity. In addition, we did not investigateany negative correlations that may have impli-cated inhibitory mechanisms. More sophisticatedanalysis approaches, such as structural equationmodeling or dynamic causal modeling will need tobe conducted that can extract additional informa-tion about the direction and specific function ofthis connectivity. The data reported here are alsolimited to verbal stimuli of positive valence.Future studies investigating the association ofextraversion and its lower-level facets with brainfunction should therefore include non-verbalstimuli, and present stimuli at varying levels ofarousal and of differing valences.

This research demonstrated that extraversionis associated with individual differences in brainactivation and functional connectivity during anemotional attention task, and that lower-levelfacets of extraversion can be identified thatmediate some of these associations. These dataprovide evidence that subfacets of extraversioncan be dissociated both in terms of neuralactivation and in terms of functional connectivitybetween the AC and a number of other brainregions. This opens the door for future studies tofurther differentiate neural networks that sub-serve specific forms of extraversion or otherfundamental human personality traits.

In conclusion, we expect that this form of‘‘functional dissection’’ of higher-order personal-ity traits at the neural level of analysis willcontribute to the development of biologically-based dynamic models of personality in whichtraits represent causal mechanisms of brainfunction.

Manuscript received 30 December 2005

Manuscript accepted 22 February 2006

REFERENCES

Amin, Z., Constable, R. T., & Canli, T. (2004).Attentional bias for valenced stimuli as a functionof personality in the dot-probe task. Journal ofResearch in Personality, 38(1), 15�23.

Beauregard, M., Levesque, J., & Bourgouin, P. (2001).Neural correlates of conscious self-regulation of

emotion. Journal of Neuroscience, 21(18), RC165,1�6.

Behrmann, M., Geng, J. J., & Shomstein, S. (2004).Parietal cortex and attention. Current Opinions inNeurobiology, 14(2), 212�217.

Botvinick, M. M., Cohen, J. D., & Carter, C. S. (2004).Conflict monitoring and anterior cingulate cortex:An update. Trends in Cognitive Science, 8(12), 539�546.

Bradley, M. M., & Lang, P. J. (1999). Affective norms forEnglish words (ANEW). Gainesville, FL: TheNIMH Center for the Study of Emotion andAttention, University of Florida, USA.

Bush, G., Frazier, J. A., Rauch, S. L., Seidman, L. J.,Whalen, P. J., Jenike, M. A., et al. (1999). Anteriorcingulate cortex dysfunction in attention-deficit/hyperactivity disorder revealed by fMRI and theCounting Stroop. Biological Psychiatry, 45(12),1542�1552.

Bush, G., Luu, P., & Posner, M. I. (2000). Cognitive andemotional influences in anterior cingulate cortex.Trends in Cognitive Science, 4(6), 215�222.

Buxhoeveden, D., & Casanova, M. (2000). Compara-tive lateralisation patterns in the language area ofhuman, chimpanzee, and rhesus monkey brains.Laterality, 5(4), 315�330.

Canli, T., Amin, Z., Haas, B., Omura, K., & Constable,R. T. (2004). A double dissociation between moodstates and personality traits in the anterior cingulate.Behavioral Neuroscience, 118(5), 897�904.

Canli, T., Sivers, H., Whitfield, S. L., Gotlib, I. H., &Gabrieli, J. D. (2002). Amygdala response to happyfaces as a function of extraversion. Science,296(5576), 2191.

Canli, T., Zhao, Z., Desmond, J. E., Kang, E., Gross, J.,& Gabrieli, J. D. E. (2001). An fMRI study ofpersonality influences on brain reactivity to emo-tional stimuli. Behavioral Neuroscience, 115(1), 33�42.

Compton, R. J., Banich, M. T., Mohanty, A., Milham,M. P., Herrington, J., Miller, G. A., et al. (2003).Paying attention to emotion: An fMRI investigationof cognitive and emotional Stroop tasks. Cognitive,Affective & Behavioral Neuroscience, 3(2), 81�96.

Costa, P. T., Jr., & McCrae, R. R. (1980). Influence ofextraversion and neuroticism on subjective well-being: Happy and unhappy people. Journal ofPersonality and Social Psychology, 38, 668�678.

Costa, P. T., & McCrae, R. R. (1992). Professionalmanual of the Revised NEO Personality Inventoryand NEO Five-Factor Inventory. Odessa, FL: PARInc.

Depue, R. A., & Morrone-Strupinsky, J. V. (2005). Aneurobehavioral model of affiliative bonding: Im-plications for conceptualizing a human trait ofaffiliation. Behavioral Brain Science, 28(3), 313�350; discussion 350�395.

Derrfuss, J., Brass, M., & von Cramon, D. Y. (2004).Cognitive control in the posterior frontolateralcortex: Evidence from common activations in taskcoordination, interference control, and workingmemory. Neuroimage, 23(2), 604�612.

Derryberry, D. (1994). Temperament and attention:Orienting towards and away from positive and

EXTRAVERSION, ANTERIOR CINGULATE CONNECTIVITY 23

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 1

7:23

22

Oct

ober

201

4

negative signals. Journal of Personality and SocialPsychology, 66, 1128�1139.

Egner, T., & Hirsch, J. (2005). The neural correlatesand functional integration of cognitive control in aStroop task. Neuroimage, 24(2), 539�547.

Eisenberger, N. I., Lieberman, M. D., & Satpute, A. B.(2005). Personality from a controlled processingperspective: An fMRI study of neuroticism, extra-version, and self-consciousness. Cognitive, Affective& Behavioral Neuroscience, 5(2), 169�181.

Eysenck, H. J. (1990). Biological dimensions of person-ality. In L. A. Pervin (Ed.), Handbook of person-ality: Theory and research (pp. 244�276). New York:Guilford Press.

Forman, S. D., Cohen, J. D., Fitzgerald, M., Eddy, W. F.,Mintun, M. A., & Noll, D. C. (1995). Improvedassessment of significant activation in functionalmagnetic resonance imaging (fMRI): Use of acluster-size threshold. Magnetic Resonance in Med-icine, 33(5), 636�647.

Friston, K. J. (1994). Statistical parametric mapping. InR. Thatcher (Ed.), Functional neuroimaging: Tech-nical foundations (pp. 79�93). San Diego, CA:Academic Press.

Friston, K. J., Buechel, C., Fink, G. R., Morris, J., Rolls,E., & Dolan, R. J. (1997). Psychophysiological andmodulatory interactions in neuroimaging. Neuro-image, 6(3), 218�229.

Garavan, H., Ross, T. J., Kaufman, J., & Stein, E. A.(2003). A midline dissociation between error-pro-cessing and response-conflict monitoring. Neuro-image, 20(2), 1132�1139.

Gray, J. R., & Braver, T. S. (2002). Personality predictsworking-memory-related activation in the caudalanterior cingulate cortex. Cognitive, Affective &Behavioral Neuroscience, 2(1), 64�75.

Gray, J. R., Burgess, G. C., Schaefer, A., Yarkoni, T.,Larsen, R. J., & Braver, T. S. (2005). Affectivepersonality differences in neural processing effi-ciency confirmed using fMRI. Cognitive, Affective& Behavioral Neuroscience, 5(2), 182�190.

Habel, U., Klein, M., Kellermann, T., Shah, N. J., &Schneider, F. (2005). Same or different? Neuralcorrelates of happy and sad mood in healthy males.Neuroimage, 26(1), 206�214.

Holmes, A. P., & Friston, K. J. (1998). Generalisability,random effects & population inference. Neuro-image, 7, S754.

John, O. P., & Srivastava, S. (1999). The Big Five traittaxonomy: History, measurement, and theoreticalperspectives. In L. A. Pervin & O. P. John (Eds.),Handbook of personality: Theory and research (2nded.) (pp. 102�138). New York: Guilford Press.

Killgore, W. D., & Yurgelun-Todd, D. A. (2004).Activation of the amygdala and anterior cingulate

during nonconscious processing of sad versus happyfaces. Neuroimage, 21(4), 1215�1223.

Lane, R. D., Reiman, E. M., Axelrod, B., Yun, L. S.,Holmes, A., & Schwartz, G. E. (1998). Neuralcorrelates of levels of emotional awareness. Evi-dence of an interaction between emotion andattention in the anterior cingulate cortex. Journalof Cognitive Neuroscience, 10(4), 525�535.

Maldjian, J. A., Lavrienti, P. J., Kraft, R. A., &Burdette, J. B. (2003). An automated method forneuroanatomic and cytoarchitectonic atlas-basedinterrogation of fMRI data sets. Neuroimage, 19,1233�1239.

McNair, M. L., Lorr, M., & Droppleman, L. F. (1971).Profile of mood states. San Diego, CA: EdITS.

Morrone-Strupinsky, J. V., & Depue, R. A. (2004).Differential relation of two distinct, film-inducedpositive emotional states to affiliative and agenticextraversion. Personality and Individual Differences,36(5), 1109�1126.

Reuter, M., Stark, R., Hennig, J., Walter, B., Kirsch, P.,Schienle, A., et al. (2004). Personality and emotion:Test of Gray’s personality theory by means of anfMRI study. Behavioral Neuroscience, 118(3), 462�469.

Rogers, R. D., Ramnani, N., Mackay, C., Wilson, J. L.,Jezzard, P., Carter, C. S., et al. (2004). Distinctportions of anterior cingulate cortex and medialprefrontal cortex are activated by reward processingin separable phases of decision-making cognition.Biological Psychiatry, 55(6), 594�602.

Rusting, C. L. (1999). Interactive effects of personalityand mood on emotion-congruent memory andjudgment. Journal of Personality & Social Psychol-ogy, 77(5), 1073�1086.

Shapiro, K., Hillstrom, A. P., & Husain, M. (2002).Control of visuotemporal attention by inferiorparietal and superior temporal cortex. CurrentBiology, 12(15), 1320�1325.

Shin, L. M., Whalen, P. J., Pitman, R. K., Bush, G.,Macklin, M. L., Lasko, N. B., et al. (2001). An fMRIstudy of anterior cingulate function in posttraumaticstress disorder. Biological Psychiatry, 50(12), 932�942.

Watson, D., & Clark, L. A. (1997). Extraversion and itspositive emotional core. In P. Hogan, J. Johnson, &S. R. Briggs (Eds.), Handbook of personality psy-chology. New York: Academic Press.

Whalen, P. J., Bush, G., McNally, R. J., Wilhelm, S.,McInerney, S. C., Jenike, M. A., et al. (1998). Theemotional counting Stroop paradigm: a functionalmagnetic resonance imaging probe of the anteriorcingulate affective division. Biological Psychiatry,44(12), 1219�1228.

24 HAAS ET AL.

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 1

7:23

22

Oct

ober

201

4