Personality and Individual Differences 53 (2012) 574–579

Contents lists available at SciVerse ScienceDirect

Personality and Individual Differences

journal homepage: www.elsevier .com/locate /paid

Individual differences in attentional networks: Trait and state correlates of the ANT

Gerald Matthews a,⇑, Moshe Zeidner b

a University of Cincinnati, Cincinnati, USAb University of Haifa, Haifa, Israel

a r t i c l e i n f o

Article history:Received 3 January 2012Received in revised form 17 April 2012Accepted 20 April 2012Available online 19 May 2012

Keywords:AttentionAttention Networks TestExecutive controlFive Factor ModelExtraversionConscientiousnessMood

0191-8869/$ - see front matter � 2012 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.paid.2012.04.034

⇑ Corresponding author.E-mail addresses: matthegl@ucmail.uc.edu,

(G. Matthews).

a b s t r a c t

Several trait and state factors may predict enhanced attentional functioning, but the mechanisms under-lying these effects require elucidation. The current study aimed to test relationships between traits (FiveFactor Model), three state factors and the Attentional Network Test (ANT). 99 Israeli undergraduates par-ticipated. Trait and state factors were differentially related to the ANT indices. As predicted, superiorexecutive control was associated with extraversion, conscientiousness, state task engagement, and lowstate distress. Differing predictor sets were found for alertness and orienting. Findings are discussed inrelation to cognitive neuroscience models of personality and subjective state.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Cognitive neuroscience may help us to understand individualdifferences (IDs) in attention (Matthews, 2008). The present studyaimed to investigate associations between trait and state factorsand the Attention Networks Test (ANT: Fan, McCandliss, Sommer,Raz, & Posner, 2002). The ANT assesses three key brain circuitsfor attention; executive control (EC), spatial orienting and alert-ness. EC is supported by frontal areas associated with resolutionof cognitive conflict, including anterior cingulate cortex and lateralprefrontal cortex (PFC). The orienting network controls direction ofattention following presentation of a spatial cue, and is supportedby the parietal lobules, frontal eye fields, and several subcorticalareas. The alertness network is supported by thalamic, frontal,and parietal areas, and maintains readiness to respond to targetstimuli. Identification of trait and state effects on these threenetworks may afford a more precise account of IDs in attentionthan resource theory explanations. This study is the first to testfor associations between the ANT and the Five Factor Model ofpersonality (Saucier, 2002), as well as three broad state factors oftask engagement, distress and worry. Next, we review traits andstates that may relate to energization of attention and to executivefunctioning.

ll rights reserved.

Gerald.Matthews@uc.edu

1.1. State and trait factors in energization processes

Studies of vigilance, visual search and other attentionallydemanding tasks show reliable effects of stimulant and de-arousingagents, prompting the hypothesis that arousal elevates the avail-ability of a resource for sustained information transfer (Humphreys& Revelle, 1984). Matthews, Davies, and Lees (1990) confirmed thatsubjective energetic arousal predicted attentional task perfor-mance. Furthermore, associations between energy and perfor-mance were moderated by task demands as predicted by resourcetheory; the higher the workload, the greater the correlation(Matthews, Warm, Reinerman, Langheim, & Saxby, 2010a). Recentstudies investigated subjective task engagement (Matthews et al.,2002), a higher-order state factor associated with motivation andconcentration as well as with energy. Like energy, high task engage-ment predicts superior attention (Matthews et al., 2010a, 2010b).Task engagement correlates with a physiological index of resourcemobilization, cerebral bloodflow velocity (CBFV: Matthews et al.,2010a). It also correlates with cognitive appraisal and copingprocesses that may support attention (Matthews et al., 2002).

Task engagement may relate to executive functioning(Matthews et al., 2010a). The cortical sites for EC are influencedby the ventral tegmental dopamine system (Fan et al., 2002). Dopa-minergic pathways regulate active maintenance of task-relevantcontext in PFC (De Young & Gray, 2009), and support positive emo-tional states, especially states such as task engagement that reflectmotivated goal pursuit, rather than pleasure per se (Berridge &Kringelbach, 2008; Matthews et al., 2010a). Caffeine enhances both

G. Matthews, M. Zeidner / Personality and Individual Differences 53 (2012) 574–579 575

vigilance and EC (Bruny, Mahoney, Lieberman, & Taylor, 2010),whereas sleep deprivation has the opposite effect (Martella, Casa-grande, & Lupiáñez, 2011). Studies of executive functioning usingother tasks have generally also shown detrimental effects of fati-gue associated with sleep loss (Killgore, in press). Effects on otherattentional networks appear less robust. Bruny et al. (2010) foundthat their highest caffeine dosage enhanced alertness, but therewas no effect of sleep loss in the Martella et al. study. Orientingwas unaffected by caffeine (Bruny et al., 2010), and enhanced bysleep loss (Martella et al., 2011). Thus, associations betweenengagement and attention, as well as stressor effects on executivefunctioning, suggest that task engagement should correlate withEC.

Turning to trait energization factors, accounts of dopaminergicsupport of EC typically focus on the Behavioral Activation System,and related traits including extraversion (De Young & Gray, 2009).If extraversion is associated with dopaminergic activity, it shouldenhance EC. In addition, performance studies show bettermulti-tasking performance in extraverts in some task paradigms(Szymura, 2010). A difficulty for the hypothesis is that extraversionis rather weakly related to positive affect and task engagementduring performance (Matthews & Gilliland, 1999; Shaw et al.,2010).

Another relevant trait is conscientiousness, which relates to theplanning and impulse control functions critical for cognitive con-trol (De Young & Gray, 2009), and volume of the lateral PFC (DeYoung et al., 2010). Furthermore, conscientiousness is associatedwith task engagement during task performance, and with cognitivecorrelates of engagement such as task-focused coping (Matthewset al., 2006, 2010b).

1.2. Anxiety and negative affect

Traits and states associated with negative affect may also influ-ence attention and EC. Most relevant research has focused on traitanxiety. Eysenck and Derakshan’s (2011) Attentional Control The-ory (ACT) identifies anxiety with weakened EC over exogenous,stimulus-driven attention. Anxious individuals may maintainperformance effectiveness through compensatory effort, depend-ing on motivational factors. ACT differentiates multiple executiveprocesses, and links anxiety especially to weakness in inhibitorycontrol associated with fronto-parietal brain areas. ACT deals spe-cifically with anxiety, but comparable processes may operate inother negative emotional conditions (Wells & Matthews, 1994).As ACT predicts, anxiety and related traits typically impair inhibi-tory and other EC functions (Eysenck & Derakshan, 2011).

Two recent ANT studies provide conflicting findings. Moriya andTanno (2009) found that state and trait anxiety, depression and so-cial anxiety were all negatively associated with the ANT orientingindex. No significant correlates of EC and alerting were obtained.Pacheco-Unguetti, Acosta, Callejas, and Lupiáñez (2010; Study 1)failed to replicate these associations between anxiety and orient-ing, although, consistent with ACT, they found that trait (but notstate) anxiety was associated with poorer EC. A further study(Pacheco-Unguetti et al., 2010) found no effect of trait anxiety,but induced anxious mood enhanced orienting and alerting. Areview of ANT research (MacLeod et al., 2010) reported specificexecutive impairments in various clinical conditions, includingtwo associated with negative affectivity, borderline personalityand post-traumatic stress disorder.

1.3. Aims and hypotheses

The overall aim of the study was to assess associations betweenbroad trait and state measures and the ANT, in an Israeli sample.The primary focus was on correlates of EC, although we also

investigated orienting and alertness. The traits of the FFM (Saucier,2002) were assessed, as well as three broad subjective states oftask engagement, distress and worry (Matthews et al., 2002).Vocabulary was assessed as a potential control variable. Hypothe-ses were as follows:

State task engagement predicts superior performance ondemanding attentional tasks (Matthews et al., 2010a), Also,engagement may be associated with activity in mesolimbic dopa-minergic pathways that enhance EC functions of PFC (De Young& Gray, 2009). Thus, state task engagement should correlate withEC (H1).

Extraversion is related to dopaminergic afferents to frontal cor-tex, and to superior dual-task performance. Extraversion shouldthus correlate with EC (H2). Conscientiousness is associated withimpulse control, task engagement and task-focused coping. It tooshould correlate with EC (H3).

Existing studies of the ANT and negative affect have yieldedinconsistent results (Moriya & Tanno, 2009; Pacheco-Unguettiet al., 2010). However, the broader literature on anxiety, negativeaffect and attentional impairments (e.g., Eysenck & Derakshan,2011) suggests that emotional stability should correlate with supe-rior EC (H4).

Finally, we tested whether trait effects were mediated bytransient states. On the basis of known trait-state associations(Matthews et al., 2002), as well as the neuroscience of traits (DeYoung & Gray, 2009), we hypothesized that associations betweenextraversion and conscientiousness and EC would be mediatedby task engagement (H5a). Similarly, associations between neurot-icism and EC may be mediated by states of negative affect, opera-tionalized here as distress and worry (H5b).

2. Methods

2.1. Participants

99 undergraduates (66% female) at the University of Haifa par-ticipated. Mean age was 24. Forty-two participants self-identifiedas Jewish; the remainder reported various, predominantly Muslim,religious affiliations.

2.2. Task and measures

2.2.1. VocabularyIt was assessed with the Verbal subtest of the Level III Milta

Intelligence test (Ortar, 1966), which is comprised of 25 multiplechoice items. For each word, respondents were asked to choosethe best definition from five alternative single words.

2.2.2. PersonalityThe Mini-Modular Markers (3M40; Saucier, 2002) assessed the

FFM traits: Extraversion, Emotional Stability, Agreeableness, Con-scientiousness, and Openness. Participants rated 40 personalitydescriptors on 1–9 response scales, anchored by ‘‘extremely accu-rate’’ and ‘‘extremely inaccurate’’ as scale endpoints.

2.2.3. Subjective stateThe short version of the Dundee Stress State Questionnaire

(DSSQ; Guznov, Matthews, Funke, & Dukes, 2011) assessed threesubjective state factors: task engagement (energy, task motivation,concentration), distress (tension, unpleasant mood, lack of confi-dence), and worry (self-focused attention, low self-esteem, cogni-tive interference related to task and personal concerns). Thereare 21 items, with 5-point response scales. The short DSSQ is basedon the full-version DSSQ: see Matthews et al. (2002) for a descrip-tion of its psychometric properties.

Table 1Descriptive statistics for trait and state measures.

Variable Mean SD a

Vocabulary 9.33 5.08 .84Extraversion 43.00 9.93 .73Emotional stability 44.19 9.81 .70Conscientiousness 52.61 10.79 .81

576 G. Matthews, M. Zeidner / Personality and Individual Differences 53 (2012) 574–579

2.2.4. ANTThe ANT task version programmed in E-Prime by Fan et al.

(2002) was used. The essential features of the task are as follows(see Fan et al., 2002, for full description). On each trial, the partic-ipant must press one of two keys to discriminate whether a centralarrow stimulus (the target) points left or right. The target is flankedby distractor stimuli, and appears above or below a central fixationpoint. The target stimulus may be preceded by a cue stimulus thateither has a general alerting function, or indicates whether the tar-get will appear above or below fixation. Two attributes of the taskare manipulated across trials. The first is cue type (four levels),which may be absent (central fixation cross only), a central cue(asterisk), a double cue (asterisks above and below fixation cross),or a spatial cue (single asterisk above or below fixation cross). Thesecond attribute is the flanker stimulus type (three levels), whichmay be congruent with the target (arrow points in same direction),incongruent (arrow points in opposite direction), or neutral (linestimuli with no arrow head). In each case, two flankers arepresented on either side of the target. The temporal sequence ofstimulus presentation is as follows: (1) central fixation cross for400–1600 ms, (2) cue or no cue for 100 ms, (3) central fixationcross for 400 ms, (4) target until participant responds, and (5) cen-tral fixation cross until total trial duration of 4000 ms has elapsed.

The alerting index is calculated by subtracting the mean reac-tion time (RT) of all double-cue conditions from the mean RT ofall no-cue conditions. The double cue alerts the participant to theimminent onset of the target while keeping attention diffusedacross upper and lower screen locations. The orienting index is cal-culated by subtracting the mean RT of the spatial cue conditionsfrom the mean RT of the center cue conditions. All these cues arealerting, but only the spatial cues direct attention to the locationat which the target will appear. The executive control index is cal-culated by subtracting the mean RT of the congruent flanking con-ditions from the mean RT of incongruent flanking conditions. Highscores indicate delay in inhibiting response to competing stimuli,and hence poor EC.

Agreeableness 57.06 9.64 .81Openness 52.01 9.59 .73Engagement (baseline) 20.80 4.83 .68Distress (baseline) 9.34 5.42 .79Worry (baseline) 15.76 5.58 .69Engagement (post-task) 16.84 6.72 .77Distress (post-task) 8.73 5.45 .77Worry (post-task) 13.55 6.02 .76

550

600

650

700

RT

(ms)

No

2.3. Procedure

Initially, participants completed the personality and vocabularymeasures, and a baseline measure of subjective state, followed bypractice on the ANT. Next, 47 participants performed a 15 min vig-ilance task requiring participants to make line length judgments onbriefly presented, visually degraded line stimuli (Matthews et al.,2006), and 52 performed a control task (reading magazines). Thevigilance task was intended as a between-subjects manipulationof fatigue. However, the two groups did not differ in subjectivestates experienced during the ANT, or in performance, even duringthe first ANT trial block. Correlations between traits, states andANT also did not differ in the two groups. Because the manipula-tion was ineffective in producing any lasting state or performancechange, we report pooled data from the whole sample. The ANTwas administered as three blocks of 96 trials each, presented inrandom order, with a duration of about 15 min. Subjective statewas assessed following the ANT; participants were instructed toreport on their feelings during the task.

450

500

Neutral Congruent Incongruent

Flanker Type

CenterDoubleSpatial

Fig. 1. Mean RT for correct trials as a function of cue and flanker type.

3. Results

Means and SDs for trait and state measures are shown in Table1. Bonferroni-corrected t-tests comparing baseline with post-ANTstate factors, showed significant decreases in task engagement(t(98) = 5.64, p < .01) and in worry (t(98) = 3.47, p < .01), but nochange in distress. The task was tiring but not especially stressful.

We tested task parameter effects on mean RT using a (cue: nocue, center cue, double cue, spatial cue) � 3 (flanker type: neutral,congruent, incongruent) repeated-measures ANOVA, with Box’scorrection applied for violation of sphericity (see Fig. 1). Main ef-fects of cue (F(2.593,254.147) = 349.89, p < .01, g2

p = .781) and flan-ker (F(1.253,122.803) = 381.58, p < .01, g2

p = .796) were significant.So was the cue � flanker interaction (F(4.886,478.803) = 17.21,p < .01, g2

p = .149). The pattern of means was very similar to that re-ported by Fan et al. (2002; Fig. 2). Response was speeded by alltypes of cue, and slowed by incongruent flankers. Similar to Fanet al. (2002), the interaction results from the spatial cue having astronger effect with incongruent flankers than with the othertypes.

Means (and SDs) for the ANT indices were for alerting 47.9 ms(27.7), for orienting 52.9 ms (25.8) and for EC 116.4 ms (60.9).These means resemble those reported by Fan et al. (2002), exceptthat they reported a smaller mean for EC of 84 ms. Subsequentanalyses of EC used a log-transform of the index to reduce positiveskew. The only significant inter-index correlation was between ori-enting and EC (r = .25, p < .05). As in Fan et al. (2002), error rateswere low (<5% overall), and analysis of errors showed no evidencefor contamination of indices by speed-accuracy tradeoff. Vocabu-lary was not significantly correlated with any of the ANT indices.

Trait and state correlates of the three indices are shown in Table2. Post-task DSSQ scores were used because these are most repre-sentative of state during task performance. As hypothesized, extra-version and conscientiousness were associated with superior EC

E

C Engage.

Distress

EC .22

.34

-.22

-.44

-.29

-.25

Fig. 2. Path model for trait and state effects on EC.

Table 2Trait and state correlates of three ANT indices.

Alertness Orienting EC

Extraversion .10 �.08 �.31**

Emotional stability .01 �.23* �.16Conscientiousness �.11 �.18 �.24*

Agreeableness �.06 �.15 �.03Openness �.21* .03 �.11Engagement �.08 �.28** �.22*

Distress .09 .26** .22*

Worry �.07 .07 .07

* p < .05.** p < .01.

G. Matthews, M. Zeidner / Personality and Individual Differences 53 (2012) 574–579 577

(high EC scores indicate poor functioning). Associations betweenbetter EC and high engagement and low distress were also consis-tent with prediction. Perhaps surprisingly, higher orienting wasassociated with lower emotional stability, lower engagement,and lower distress. Alertness was associated only with loweropenness.

Also, several significant associations between the personalitytraits and the post-task state assessments were found. Taskengagement was related only to higher conscientiousness (r = .33,p < .01), whereas distress was associated with lower extraversion(r = �.27, p < .05), lower emotional stability (r = �.21, p < .05), andlower conscientiousness (r = �.21, p < .05). Higher worry also cor-related with lower extraversion (r = �.26, p < .05) and lower emo-tional stability (r = �.31, p < .01), and with higher openness(r = .24, p < .05). Thus, the two trait predictors of EC, extraversionand conscientiousness, were also associated with states that corre-lated with EC: engagement and distress in the case of conscien-tiousness, and distress in the case of extraversion. This set ofassociations suggests possible mediation hypotheses. For example,

Table 3Fit statistics for models tested using path analysis.

Model Description

1. Independent effects Each trait and state has a single path to EC. No trai2. Trait effects only Both traits have a path to EC and paths to states. N3. Full mediation Both states have a path to EC. C and E have paths t

traits to EC4. Mediated effect of C Both states have a path to EC. C has paths to states

path to EC only5. Mediated effect of E Both states have a path to EC. E has paths to states

path to EC only6. Modified partial mediation See text

Note: CFI = Comparative Fit Index, RMSEA = Root Mean Square Error of Approximation, SR* p < .05.** p < .01.

effects of traits on EC might be mediated by states, or the influenceof traits on states might be incidental to their influence on EC.

To explore these issues further a series of path analyses wererun, each representing a different model for EC. Models 1–5 werespecified a priori, and fitted using maximum likelihood estimationwith EQS. All models included two paths necessary for fit: (1) a cor-relation between extraversion and conscientiousness, and (2) acorrelation or path from engagement to distress. Additional pathsfor each model are specified in Table 3.

The first two models tested whether good fit could be obtainedwith no mediation of any kind. Model 1 assumed that each traitand state predictor influenced EC independently. Model 2 assumedthat traits influenced both states and EC independently, with nopaths from states to EC. Both models can be rejected on the basisof poor fit. Models 3–5 explored mediation hypotheses. All modelspermitted the two state variables to influence EC. With two traitvariables, either or both might have an indirect effect, such thatthe effect of the trait was mediated by states. Model 3 testedH5a, i.e., that both trait effects were state-mediated. Fit remainedpoor. Models 4 and 5 specified a direct effect of one trait, and anindirect, state-mediated effect of the other. The best fit for the apriori models was found for Model 4, in which extraversion di-rectly influenced EC, and conscientiousness had an indirect effect,mediated by engagement and distress.

Values for the Model 4 RMSEA and standardized RMR fit coeffi-cients (see Table 3) suggest that fit might be improved. To con-struct the final path model (Model 6), the Wald test was used todrop paths which did not improve fit, i.e., conscientiousness ? dis-tress, distress ? EC, and extraversion ? engagement, leading toimproved fit statistics (see Fig. 2). The path coefficients shownare standardized. The limitation of this model is that it was fine-tuned post hoc, but it may suggest hypotheses for futureinvestigation.

4. Discussion

The study identified several trait and state correlates of IDs inthe functioning of attentional networks. Hypotheses H1–H3broadly addressed IDs in the energization of attention. Thesehypotheses were confirmed: state task engagement and the extra-version and conscientiousness traits were all associated withsuperior executive functioning. Path analyses suggested that extra-version might have a direct effect on EC, and conscientiousness anindirect effect, mediated by task engagement, providing partialsupport for H5a. Contrary to prediction, there was no effect of emo-tional stability on EC (H4). Thus, the mediation hypothesis (H5b)could not be tested, although the distress state was associated withpoorer EC. The study also showed several correlates of the ANTorienting index; participants low in emotional stability andengagement, and high in distress, showed a greater facilitation of

v2 (df) CFI RMSEA SRMR AIC

t-state paths 13.0 (4)* .74 .182 .132 5.00o state-EC paths 18.6 (3)** .56 .276 .122 5.95o states; no paths from 5.0 (2) .92 .148 .063 .10

but not to EC. E has direct 4.1 (3) .97 .071 .063 �1.96

but not to EC. C has direct 12.4 (3)** .73 .214 .111 6.35

1.2 (4) 1.00 .000 .032 �6.76

MR = Standardized Root Mean Square Residual, AIC = Akaike Information Criterion.

578 G. Matthews, M. Zeidner / Personality and Individual Differences 53 (2012) 574–579

response speed from the spatial cue. ANT alertness was the indexleast sensitive to the trait and state factors.

The association between higher engagement and superior EC isconsistent with experimental studies showing effects of stimulantand fatiguing agents on various executive tasks (Bruny et al., 2010;Killgore, in press), and with evidence linking positive emotions todopaminergic pathways that support cognitive control (De Young& Gray, 2009). It is also consistent with the reliable associationsfound between subjective task engagement and performance of avariety of demanding attentional tasks (Matthews et al., 2010a,2010b; Shaw et al., 2010). Thus, the association between taskengagement and performance on these tasks may be mediatedby EC. On certain tasks, vigilance decrement may reflect impair-ments in supervisory attentional control (Helton, Head, & Russell,2011). Other processes, such as working memory impairment(Matthews & Campbell, 2010), may also contribute to task engage-ment effects on vigilance.

Path models suggested that engagement may mediate theassociation between conscientiousness and EC. Conscientiousnessrelates to both engagement and coping strategies such as task-focus that may enhance task-directed effort (Matthews et al.,2006). Most studies of conscientiousness and performance haveaddressed work performance. It may be timely to focus also oninformation-processing mechanisms (De Young & Gray, 2009);superior EC is likely to support performance in a range of appliedsettings.

The correlation between extraversion and EC is consistent withevidence for superior executive control of inhibition processes inextraverts. Such evidence comes from studies of multitasking(Szymura, 2010) and of resistance to distraction (Dobbs, Furnham,& McClelland, 2011). Previously, the superior cognitive control ofextraverts has been attributed to dopaminergic pathways to PFC(De Young & Gray, 2009). If so, higher dopaminergic activity inextraverts should produce positive, approach-related affectivestates (Berridge & Kringelbach, 2008), including high task engage-ment. However, as in other recent studies of attention (Shawet al., 2010), extraversion and engagement were unrelated. Indeed,while high engagement predicts enhanced sustained attention,extraversion tends to be negatively associated with vigilance (Shawet al., 2010). Therefore, the effects of extraversion and of taskengagement on EC and attention may not, in fact, reflect a commonmechanism. A possible explanation, suggested by the cognitive-adaptive theory of personality (Matthews, 2008), is that extraver-sion relates to biases in multiple processing modules thatcollectively support adaptation to socially demanding environ-ments. Perhaps functioning of the prefrontal sites associated withEC is superior in extraverts irrespective of activity in mesolimbicdopaminergic pathways.

We failed to confirm the hypothesized association betweenneuroticism and impaired EC, although state distress was associ-ated with poorer EC. This finding adds to evidence that negativeemotional states may impair executive functions, especially inhibi-tion (Eysenck & Derakshan, 2011). It is surprising that emotionalstability did not correlate with EC. Perhaps it is trait anxiety ratherthan general negative emotionality which is associated withimpaired inhibition. However, studies of anxiety and the ANT alsoobtained inconsistent results for EC, as previously described (Mor-iya & Tanno, 2009; Pacheco-Unguetti et al., 2010). Alternatively,the ANT may not be optimal for assessment of the inhibitory func-tions linked to anxiety by ACT (Eysenck & Derakshan, 2011). Also,individuals high in negative emotionality may compensate forinefficient inhibitory processing through increasing effort.

Emotional stability correlated with weaker orienting, and statedistress correlated positively with orienting. Consistent with thislatter finding Pacheco-Unguetti et al. (2010; Study 2) found thatanxious mood enhanced orienting, Pacheco-Unguetti et al. (2010)

suggested that anxious mood increases sensitivity to exogenous,bottom-up processing, consistent anxiety shifting the balance be-tween exogenous and endogenous processing (Eysenck & Derak-shan, 2011). The present findings conflict with those of Moriyaand Tanno (2009), who found that various scales for negative affectrelated to weaker orienting.

Suprisingly, there was a significant correlation of .25 betweenorienting and EC, suggesting that stronger EC is associated withweaker orienting. Typically, the correlation between these indicesis close to zero (Fan et al., 2002; MacLeod et al., 2010). Thus, thecorrelation here was anomalous, although effects of the cue andflanker parameters on response time were similar to those foundby Fan et al. (2002). A tentative suggestion is that orienting maybe sensitive to strategic control. Possibly, in this study, participantsviewed the spatial cues as distracting, and those high in taskengagement and low in distress were able to apply their superiorEC to suppressing the orienting response.

The inter-relationship of EC and orienting raises difficulties ininterpreting correlations between states and the ANT, given thatengagement and distress were not uniquely related to a singleattentional network. There is previous evidence for both energeticstates (Killgore, in press) and negative affect (Eysenck & Derakshan,2011) impacting executive functions. Thus, the present data onstate factors in EC are consistent with existing findings. Studiesof state factors and spatial orienting are inconsistent, and so thefindings for orienting here may be less robust than those for EC.However, the lack of discrimination of state correlates is a limita-tion of the study, and further work might especially focus on IDsin spatial attention.

In conclusion, the study showed facilitative effects of taskengagement, conscientiousness and extraversion on the ANT EC in-dex that are broadly consistent with prior research. The associationbetween task engagement and EC is consistent with the hypothe-sized dopaminergic basis for positive, approach-related emotions(Berridge & Kringelbach, 2008), but the mechanism for the extra-version effect requires further investigation. Although distresswas associated with poorer EC, trait emotional stability was not.The inconsistencies across studies of negative emotion and theANT call into question whether it is the optimal tool for testing pre-dictions from leading theories in the area, such as ACT (Eysenck &Derakshan, 2011). Theories may require testing across a range ofmeasures of executive functioning, especially as multiple functionsmay be discriminated psychometrically.

Acknowledgement

We are grateful to Nirit Zwang for her assistance in conductingthis study.

References

Berridge, K. C., & Kringelbach, L. (2008). Affective neuroscience of pleasure: Rewardin humans and animals. Psychopharmacology, 199, 457–480.

Bruny, T. T., Mahoney, C. R., Lieberman, H. R., & Taylor, H. A. (2010). Caffeinemodulates attention network function. Brain and Cognition, 72, 181–188.

De Young, C. G., & Gray, J. R. (2009). Personality neuroscience: Explaining individualdifferences in affect, behavior, and cognition. In P. J. Corr & G. Matthews (Eds.),The Cambridge handbook of personality psychology (pp. 323–346). New York:Cambridge University Press.

De Young, C. G., Hirsh, J. B., Shane, M. S., Papademetris, X., Rajeevan, N., & Gray, J. R.(2010). Testing predictions from personality neuroscience. Brain structure andthe big five. Psychological Science, 21, 820–828.

Dobbs, S., Furnham, A., & McClelland, A. (2011). The effect of background music andnoise on the cognitive test performance of introverts and extraverts. AppliedCognitive Psychology, 25, 307–313.

Eysenck, M. W., & Derakshan, N. (2011). New perspectives in attentional controltheory. Personality and Individual Differences, 50, 955–960.

Fan, J., McCandliss, B. D., Sommer, T., Raz, A., & Posner, M. I. (2002). Testing theefficiency and independence of attentional networks. Journal of CognitiveNeuroscience, 14, 340–347.

G. Matthews, M. Zeidner / Personality and Individual Differences 53 (2012) 574–579 579

Guznov, S., Matthews, G., Funke, G., & Dukes, A. (2011). Use of the RoboFlagsynthetic task environment to investigate workload and stress responses inUAV operation. Behavior Research Methods, 43, 771–780.

Helton, W. S., Head, J., & Russell, P. N. (2011). Reliable and unreliable-warning cuesin the sustained attention to response task. Experimental Brain Research, 209,401–407.

Humphreys, M. S., & Revelle, W. (1984). Personality, motivation and performance. Atheory of the relationship between individual differences and informationprocessing. Psychological Review, 91, 153–184.

Killgore, W. D. S. (in press). Socio-emotional and neurocognitive effects of sleep loss.In G. Matthews, P. A. Desmond, C. Neubauer, & P. A. Hancock (Eds.), Handbook ofoperator fatigue. Aldershot, UK: Ashgate Press.

MacLeod, J. W., Lawrence, M. A., McConnell, M. M., Eskes, G. A., Klein, R. M., & Shore,D. I. (2010). Appraising the ANT: Psychometric and theoretical considerations ofthe Attention Network Test. Neuropsychology, 24, 637–651.

Martella, D., Casagrande, M., & Lupiáñez, J. (2011). Alerting, orienting and executivecontrol: The effects of sleep deprivation on attentional networks. ExperimentalBrain Research, 210, 81–89.

Matthews, G. (2008). Personality and information processing: A cognitive-adaptivetheory. In G. J. Boyle, G. Matthews, & D. H. Saklofske (Eds.), Sage handbook ofpersonality theory and testing. Personality theories and models (Vol. 1, pp. 56–79).Thousand Oaks, CA: Sage.

Matthews, G., & Campbell, S. E. (2010). Dynamic relationships between stress statesand working memory. Cognition and Emotion, 24, 357–373.

Matthews, G., Campbell, S. E., Falconer, S., Joyner, L., Huggins, J., Gilliland, K., et al.(2002). Fundamental dimensions of subjective state in performance settings:Task engagement, distress and worry. Emotion, 2, 315–340.

Matthews, G., Davies, D. R., & Lees, J. L. (1990). Arousal, extraversion, and individualdifferences in resource availability. Journal of Personality and Social Psychology,59, 150–168.

Matthews, G., Emo, A. K., Funke, G., Zeidner, M., Roberts, R. D., Costa, P. T. Jr.,, et al.(2006). Emotional intelligence, personality, and task-induced stress. Journal ofExperimental Psychology: Applied, 12, 96–107.

Matthews, G., & Gilliland, K. (1999). The personality theories of H.J. Eysenck and J.A.Gray: A comparative review. Personality and Individual Differences, 26, 583–626.

Matthews, G., Warm, J. S., Reinerman, L. E., Langheim, L. K., & Saxby, D. J. (2010a).Task engagement, attention and executive control. In A. Gruszka, G. Matthews,& B. Szymura (Eds.), Handbook of individual differences in cognition: Attention,memory and executive control (pp. 205–230). New York: Springer.

Matthews, G., Warm, J. S., Reinerman, L. E., Langheim, L., Washburn, D. A., & Tripp, L.(2010b). Task engagement, cerebral blood flow velocity, and diagnosticmonitoring for sustained attention. Journal of Experimental Psychology: Applied,16, 187–203.

Moriya, J., & Tanno, Y. (2009). Dysfunction of attentional networks for non-emotional processing in negative affect. Cognition and Emotion, 23, 1090–1105.

Ortar, G. (1966). MILTA intelligence test. Jerusalem: Hebrew University.Pacheco-Unguetti, A. P., Acosta, A., Callejas, A., & Lupiáñez, J. (2010). Attention and

anxiety: Different attentional functioning under state and trait anxiety.Psychological Science, 21, 298–304.

Saucier, G. (2002). Orthogonal markers for orthogonal factors: The case of the BigFive. Journal of Research in Personality, 36, 1–31.

Shaw, T. H., Matthews, G., Warm, J. S., Finomore, V., Silverman, L., & Costa, P. T. Jr.,(2010). Individual differences in vigilance. Personality, ability and states ofstress. Journal of Research in Personality, 44, 297–308.

Szymura, B. (2010). Individual differences in resource allocation policy. In A.Gruszka, G. Matthews, & B. Szymura (Eds.), Handbook of individual differences incognition: Attention, memory and executive control (pp. 205–230). New York:Springer.

Wells, A., & Matthews, G. (1994). Attention and emotion: A clinical perspective. Hove:Erlbaum.