This article was downloaded by: [NIH Library]On: 29 April 2015, At: 14:20Publisher: 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

Effect of attention control on sustainedattention during induced anxietyChristian Grillona, Oliver J. Robinsona, Ambika Mathura & Monique Ernsta

a Section on Neurobiology of Fear and Anxiety, National Institute of MentalHealth, National Institutes of Health, Bethesda, MD, USAPublished online: 22 Apr 2015.

To cite this article: Christian Grillon, Oliver J. Robinson, Ambika Mathur & Monique Ernst (2015):Effect of attention control on sustained attention during induced anxiety, Cognition and Emotion, DOI:10.1080/02699931.2015.1024614

To link to this article: http://dx.doi.org/10.1080/02699931.2015.1024614

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

Effect of attention control on sustained attention duringinduced anxiety

Christian Grillon, Oliver J. Robinson, Ambika Mathur, and Monique Ernst

Section on Neurobiology of Fear and Anxiety, National Institute of Mental Health, NationalInstitutes of Health, Bethesda, MD, USA

(Received 18 December 2014; accepted 24 February 2015)

Anxiety has wide-reaching and complex effects on cognitive performance. Although it can intrude oncognition and interfere with performance, it can also facilitate information processing and behaviouralresponses. In a previous study, we showed that anxiety induced by threat of shock facilitatesperformance on the Sustained Attention to Response Task, a vigilance test, which probes responseinhibition to infrequent nogo stimuli. The present study sought to identify factors that may havecontributed to such improved performance, including on- and off-task thinking (assessed with thoughtprobes) and individual differences in attention control, as measured with the Attention Control Scale.Replicating our prior finding, we showed that shock threat significantly reduced errors of commissionon the nogo trials. However, we extended this finding in demonstrating that this effect was driven bysubjects with low attention control. We therefore confirm that anxiety increases inhibitory control ofprepotent responses—a mechanism which is adaptive under threat—and show that this effect is greaterin those who rely more upon such prepotent responding, i.e., those with low attentional control.

Keywords: Anxiety; Vigilance; Stress; SART; Threat of shock.

Anxiety is a state of distress, tension and vigilance

in response to threat. Anxiety is adaptive, it

prepares organisms for action (Langner & Eickh-

off, 2013), increases arousal and facilitates bot-

tom-up sensory processing (Cornwell et al., 2007;

Pessoa, 2009), but it can also disrupt top-down

cognitive control caused by interference from

internal and external task-irrelevant stimuli (Bar-

Haim et al., 2007; Eysenck, Derakshan, Santos, &

Calvo, 2007; Forster, Nunez Elizalde, Castle, &

Bishop, 2015). Research on the cognitive effects of

anxiety has mostly focused on cognitive disruption

by discrete external threats (e.g., attentional/affect-

ive bias) (Bar-Haim et al., 2007), but little is

known about the effect of anxiety on sustained

attention tasks, when attention needs to be main-

tained over time (Forster et al., 2015; Righi,

Mecacci, & Viggiano, 2009). On the one hand,

Correspondence should be addressed to: Christian Grillon, Section on Neurobiology of Fear and Anxiety, National Institute of

Mental Health, National Institutes of Health, 15K North Drive, Building 15K, Room 203, MSC 2670, Bethesda, MD 20892-2670,

USA. E-mail: Christian.grillon@nih.gov

Oliver J. Robinson is now at the Institute of Cognitive Neuroscience, University College London, London, UK. Ambika Mathur

is now at the Department of Biobehavioral Health, The Pennsylvania State University, PA, USA.

This work was authored as part of the Contributor's official duties as an Employee of the United States Government and is therefore a work

of the United States Government. In accordance with 17 USC. 105, no copyright protection is available for such works under US Law.

1

COGNITION AND EMOTION, 2015

http://dx.doi.org/10.1080/02699931.2015.1024614

Dow

nloa

ded

by [

NIH

Lib

rary

] at

14:

20 2

9 A

pril

2015

vigilance tasks would seem to be especially vulner-able to lapses of attention, including disruption byanxious cognition (Forster et al., 2015; Horowitz& Becker, 1971; Smallwood, Fitzgerald, Miles, &Phillips, 2009; Su, Tran, Wirtz, Langteau, &Rothman, 2009; Vinski & Watter, 2013). Onthe other hand, vigilance tasks could benefit fromthe arousing and energising properties of anxiety(Easterbrook, 1959; Lang, Bradley, & Cuthbert,1992; Langner & Eickhoff, 2013).

In a previous study, we reported that anxietyinduced by threat of shock improved performanceon a vigilance task, the Sustained Attention toResponse Task (SART) (Robinson, Krimsky, &Grillon, 2013a). The present study sought toextend these findings by examining the potentialrole that attentional lapses and individual differ-ences in attention control may play in suchanxiety-related improved performance.

During SART, subjects respond to high-fre-quency go trials and inhibit responses to rare nogotrials. Because of the tedious nature of SART,subjects progressively withdraw their attention,which promotes attentional lapses and “mindwandering” (also called task-unrelated thoughts,TUTs) (Robertson, Manly, Andrade, Baddeley, &Yiend, 1997; Smallwood et al., 2004), leading tonogo commission errors (i.e., failure to inhibitresponse to nogo trials) (McVay & Kane, 2012b;Robertson et al., 1997). Studies have foundinconsistent effects of anxiety on SART so far.While we reported that anxiety induced by shockthreat improved nogo accuracy (i.e., improvedinhibition of the prepotent go response) (Robin-son et al., 2013a), a finding we are replicating inseveral ongoing studies, including a study inpatients with anxiety disorders, a recent studyshowed an association between trait anxiety andslower go reaction time (RT) in periods withoutnogo errors (Forster et al., 2015). This latterfinding is consistent with cognitive theories ofanxiety, according to which anxiety-induced per-formance deficits are caused by worry and self-preoccupation, which take up limited processingresources necessary to perform the task (Eysencket al., 2007) or by poor attention control (Bishop,2009). However, another study reported no effect

of state or trait anxiety on RT or accuracy duringSART (Righi et al., 2009). A key differencebetween our study and these studies is that weexamined the effect of changes in state anxiety viaan experimental threat manipulation, whereas thetwo other studies examined individual differencesin trait anxiety. Hence, in the absence of anexperimental manipulation to increase state anxi-ety, state anxiety may have been too low to detectan effect in the study by Righi et al. (2009), andthe performance impairment associated with traitanxiety may reflect individual differences of per-sonality characteristics in the study by Forster et al.(2015). Furthermore, while high trait anxiety maybe associated with poor attention control (Forsteret al., 2015), an increase in state anxiety may,depending of the task, be associated with betterattention control (Robinson, Vytal, Cornwell, &Grillon, 2013b), as suggested by our previousfindings (Robinson et al., 2013a). In fact, theability to increase response inhibition during ananxious state may be adaptive (Birk, Dennis, Shin,& Urry, 2011). It is also consistent with theoret-ical models linking anxiety to activation of thebehavioural inhibition system (Gray, 1987; Quay,1997; Wright, Lipszyc, Dupuis, Thayapararajah,& Schachar, 2014).

In our initial study, we expected that anxietyinduced by shock threat would impair perform-ance (Robinson et al., 2013a). This was based ontwo assumptions. First, because SART promotesTUTs (Robertson et al., 1997) and TUTs areincreased by personal concerns, worries, stress andnegative mood (Horowitz & Becker, 1971; Klin-ger, 1971; McVay & Kane, 2010; Su et al., 2009),we assumed that threat would promote anxiousmentation and increase overall TUTs. Second, thisincrease in anxious TUTs would take up limitedresources dedicated to the task, hence impairingperformance (Eysenck et al., 2007). The fact thatshock anticipation did not impair SART perform-ance suggests that shock threat did not increaseTUTs. In other words, TUTS may not haveaffected the influence of state anxiety on perform-ance. One objective of this study was therefore toprobe subjects’ thoughts while performing the taskto examine the effect of shock threat on TUTs.

GRILLON ET AL.

2 COGNITION AND EMOTION, 2015

Dow

nloa

ded

by [

NIH

Lib

rary

] at

14:

20 2

9 A

pril

2015

Using real-time probe sampling (Gruberger, Ben-Simon, Levkovitz, Zangen, & Hendler, 2011;Teasdale et al., 1995), we assessed three types ofthoughts, task-relevant thoughts (TRTs) and twotypes of TUTs, threat-related TUTs (threatTUTs)and non–threat-related TUTs (nonthreatTUTs)(see Materials and Methods). Based on ourprevious findings that performance was notimpaired by induced anxiety (Robinson et al.,2013a), contrary to some theories (McVay &Kane, 2010), we did not expect that TUTs wouldbe affected by shock anticipation. Rather, weexpected increased threatTUTs (and decreasednonthreatTUTs) during shock anticipation com-pared to safe periods.

One mechanism by which shock threat couldimprove SART performance is via increasedarousal or motivation (Chiles, 1958; Spence,Farber, & McFann, 1956), which could thenenhance the ability to cope with vigilance tasks(Singh, Dwivedi, & Sinha, 1979), or via activationof the behavioural inhibition system (Gray, 1987;Quay, 1997; Wright et al., 2014). Poor perform-ance during SART is more likely to be found inindividuals with a weakened ability to inhibitprepotent responses or to cope with the mono-tonous nature of the task, resulting in impairedgoal maintenance or sustained attention (Helton,Kern, & Walker, 2009; McVay and Kane, 2012a;Robertson et al., 1997). Such characteristics arelinked to attention control, the executive ability todirect attention (Judah, Grant, Mills, & Lechner,2013). Accordingly, individuals most likely tobenefit from shock threat are those with aweakened ability to cope with the monotonousnature of vigilance tasks, i.e., individuals with poorattention control (McVay & Kane, 2012a; Robert-son et al., 1997). One possibility, therefore, is thatindividuals with low attention control are thosewho contribute to the improved performanceduring shock threat. Indeed, because they shouldshow poor SART performance under safe condi-tion, they should be more likely to benefit fromthreat-related increased arousal or motivationor activation of the behavioural inhibition system.By contrast, individuals with high attention con-trol are unlikely to benefit from shock threat

because they should show high and stable per-formance independent of the testing context. Thishypothesis is consistent with results of a recentstudy that showed that shock threat improvedcognitive performance during a continuous per-formance task and a working memory task only insubjects with low attention control (Hansen,Johnsen, & Thayer, 2009). Based on these find-ings, we expected lower attention control subjectsto show lower performance in the absence ofthreat but that their performance would improveduring shock threat. Attention control was meas-ured using the Attentional Control Scale (ACS)(Derryberry & Reed, 2002; Rothbart & Derry-berry, 1981), a well-established questionnaire thatassesses self-regulatory control associated with goalmaintenance, attention shifting, effortful attentionand resistance to prepotent responses (Derryberry& Rothbart, 1997; Evans & Rothbart, 2007), suchas nogo responses (Herrmann, Jacob, Unterecker,& Fallgatter, 2003, Wiersema & Roeyers, 2009).

To summarise, this study examined the impactof shock threat on TUTs and assessed whetherindividual differences in attention control contrib-uted to the influence of state anxiety on SARTperformance. Analyses focused on nogo commis-sion errors (i.e., failure to inhibit prepotentresponses), as omission errors to go trials wereminimally affected by shock threat in our previousstudy (Robinson et al., 2013a).

MATERIALS AND METHODS

Participants

Sixty healthy volunteers (31 female) completed thestudy. Subjects were included based on: (1) no pastor current psychiatric disorders according toStructured Clinical Interview for DSM-IV-TRAxis I Disorders, Research Version, Patient Edi-tion (First, Spitzer, Gibbon, & Williams, 2002),(2) no medical condition that interfered with theobjectives of the study as established by a physicianand (3) no use of illicit drugs or psychoactivemedications according to history and confirmed bya negative urine screen. All subjects gave writteninformed consent approved by the National

ANXIETY AND SUSTAINED ATTENTION

COGNITION AND EMOTION, 2015 3

Dow

nloa

ded

by [

NIH

Lib

rary

] at

14:

20 2

9 A

pril

2015

Institute of Mental Health Human InvestigationReview Board and were compensated for theirparticipation.

Procedure

We used a similar procedure as in our prior study(Robinson et al., 2013a), except that thought probeswere added to the task. Briefly, subjects performedSART during alternating safe conditions, whenthey were safe from shock, and threat conditions,when they were at risk of receiving unpredictable(and performance-independent) unpleasant shocks.Subjects’ anxiety level was assessed with subjectivereports and the acoustic startle reflex (i.e., the ocularmotor response to a loud burst of noise). This reflexwas utilised because it is robustly increased byaversive states and constitutes a reliable measure ofanxiety (Grillon, 2008). One of the advantages ofthe startle methodology is that, because it is aresponse to an external stimulus under control of theexperimenter, it can be used to probe ongoingchanges in anxiety state.

Following attachment of the electrodes, ninestartle stimuli were delivered every 18–25 s duringthe habituation phase. This was followed by a shockwork-up procedure that sets the shock intensity at alevel that was uncomfortable but not painful (a 4 ona scale of 1–5, where 1 is barely perceptible and 5 ispainful) by gradually increasing the intensity of theshock. Next, subjects performed a variant of SART(Robertson et al., 1997) in alternating safe andshock threat conditions.

Sustained Attention to Response Task

The stimuli were presented on a monitor. Parti-cipants were asked to respond to frequent “go”stimuli (“=”) and to withhold their response tooccasional “nogo” stimuli (“O”). These stimuliwere randomly distributed and were presented for250 ms at a rate of one every 2000 ms. There weretwo runs with a 5-min rest between runs. Eachrun consisted of a total of six continuous 110-sSART blocks: three safe blocks alternating withthree threat blocks. In each block, the go stimuliwere presented on 50 occasions while the nogo

stimulus occurred five times per block for a total of150 go and 15 nogo trials (adding up to 10% oftotal trials) per safe or threat condition. The firstSART block was counterbalanced between runsand across subjects such that half of the subjectsstarted with a safe and a threat condition in thefirst and second runs, respectively, and the otherhalf with the reverse order. Subjects were asked tofocus equally on speed and accuracy.

Startle stimuli, shocks and threat condition

In each run, SART started with three startlestimuli (before the first go/nogo block). Subse-quently, three startle stimuli were delivered in eachblock to assess subjects’ anxiety. Startle stimulialways occurred between two go trials. A singleshock was delivered at the end of one of the threatblock in each run just prior to a go trial that wasnot included in the analysis (for a total of twoshocks during the experiment). Subjects wereinformed that shock could be administered onlyin the threat condition and never in the safecondition. They were told that the computerdecided the number of shocks and time of shockdelivery and that the probability of shocks did notdepend on their performance. The onset of thesafe and threat conditions was signalled by a texton the monitor that indicated “you are safe fromshock” and “you are at-risk of shock”, respectively.After each run, subjects rated retrospectively theiranxiety in the safe and threat condition on ananalogue scale ranging from 0 (not at all anxious)to 10 (extremely anxious).

Thought probes

Immediately after each startle stimulus in the go/nogo blocks, questions on the monitor promptedsubjects to report their thoughts on a response keypad with choices of: TRT, threat-related (anxious)task-unrelated thoughts (threatTUTs) or threat-unrelated (non-anxious) task-unrelated thoughts(nonthreatTUTs). Thus, subjects were asked toreport their thoughts 18 times in the safe condi-tions (3 thought probes × 3 blocks × 2 runs) and

GRILLON ET AL.

4 COGNITION AND EMOTION, 2015

Dow

nloa

ded

by [

NIH

Lib

rary

] at

14:

20 2

9 A

pril

2015

18 times in the threat condition. Subjects were

given the following instructions:

Occasionally during the task you will be asked torate whether your thought was task-related, task-unrelated but threat-related, or task-unrelated andthreat-unrelated. If you were thinking about thetask, then we’ll say you were having a task-relatedthought. If you were thinking about the threat,then we’ll say you were having a threat-relatedthought. If you were not thinking about the taskand not thinking about the threat, then you werehaving a task-unrelated and threat-unrelatedthought. The question you need to answer aftereach loud noise is, “Was I thinking about doing thetask,” “Was I thinking about the threat,” or “Was Ithinking about something other than the task orthe threat?” Let’s define what I mean by “task-related thoughts,” “threat-related thoughts,” and“task-unrelated and threat-unrelated thoughts.”

A task-related thought is thinking about per-forming the task, paying attention to the stimuliand your responses.

A threat-related thought is any thought that isfocused solely on the threat experiment. Forexample, you may be thinking about how un-comfortable is the shock, when it will beadministered, how you feel, or when the currentcondition will stop.

A task-unrelated and threat-unrelated thought isnot thinking about the task or the threat of shock.For example, you may think of something youdid this morning or last night, a friend, teacher,or family member, or what you will do laterduring the day. Your mind may wander and youmay have thoughts and images about things thathave nothing to do with the threat, such aswhether you are hungry or tired, etc.

Questionnaires

During the screening procedure, which occurred

1–3 weeks prior to testing, subjects were asked to

fill out Spielberger trait anxiety inventory (Spiel-

berger, 1983) and the ACS (Derryberry & Reed,

2002). The ACS is a 20-item self-report scale that

measures attentional focusing (9 items) and atten-

tional shifting (11 items) (Derryberry & Rothbart,

1997). Higher scores on ACS reflect better ability

to direct and maintain attention.

Stimulation and physiological responses

Stimulation and recording were controlled by acommercial system (Contact Precision Instruments,UK). The acoustic startle stimulus was a 40-msduration, 103-dB (A) burst of white noise presentedvia headphones. The eye blink reflex was recordedwith two electrodes placed under the left eye and aground electrode placed on the left arm. Theelectromyographic (EMG) eye blink signal wasamplified with bandwidth set to 30–500 Hz anddigitised at a rate of 1000 Hz. The shock wasadministered either on the left wrist or on the leftmiddle and ring fingers, depending on where thesubject reached the desired intensity.

Data analysis

For the go trials, correct responses consisted of anygo trial in which there was a response, and for thenogo trials, correct trials were those in which noresponse was provided. Performance was deter-mined for each condition (threat, safe) and trialtype (go, nogo) by dividing the number of correcttrials by the total number of each trial type. Onetrial following a shock was excluded from analyses.Three measures of RT were also computedbecause they inform on the nature of informationprocessing during SART. These included meancorrect-go RT, to assess speed–accuracy trade-off(Peebles & Bothell, 2004), go RT variability, ameasure of endogenous attention (Hu, He, & Xu,2012; McVay & Kane, 2009) and pre-nogo RT.This latter measure provides an assessment of thedevelopment of habitual responses and attentionallapses because nogo errors of commission areusually preceded by faster RT (Robertson et al.,1997). Response variability was determined bycalculating the standard deviation of the mean RTfor go trials for each subject. Pre-error RTs wereaveraged across the four go trials before nogotrials, averaged separately for each condition andstratified into correct and error of commissionnogo trials (Robertson et al., 1997).

After full-wave rectification and smoothing theEMG signal, peak startle/eye blink magnitude wasdetermined in the 20- to 100-ms timeframe

ANXIETY AND SUSTAINED ATTENTION

COGNITION AND EMOTION, 2015 5

Dow

nloa

ded

by [

NIH

Lib

rary

] at

14:

20 2

9 A

pril

2015

following stimulus onset relative to a 50-ms pre-stimulus baseline. The startle responses from eachparticipant were transformed to z scores andconverted to T scores and then averaged separatelywithin the safe and the threat conditions. Thesubjective anxiety scores were also averaged withineach condition.

Because the thought (TRTs, threatTUTs andnonthreatTUTs) scores were not normally distrib-uted, they were normalised with a square roottransformation. They were analysed in two steps.First, we examined whether the rate of TRTschanged between conditions. Second, we assessedthe change in the rates of TUTs (threatTUTs vs.nonthreatTUTs) between conditions. Data wereanalysed with mixed-model repeated measuresanalyses of variance (ANOVAs) and t-tests. Mul-tiple regression analyses were used to identifyunique predictors of nogo accuracy. Partial corre-lations (rs) are also reported.

RESULTS

Demographics

Subjects’ mean (standard error of the mean[SEM]) trait anxiety and ACS scores were 28.0(.8) and 59.9 (.9), respectively. Trait anxiety andACS scores were negatively correlated (r = −.33,p < .01).

Anxiety measures

As expected, startle magnitude was larger in thethreat (mean = 52.0, SEM = .28) compared to thesafe (mean = 46.4, SEM = .28) condition (t(59) =8.7, p < .0001), and retrospective ratings ofanxiety were also higher in the threat (mean =2.6, SEM = .2) compared to the safe (mean = 5.2,SEM = .3) condition (t(59) = 12.1, p < .0001).

Performance

Nogo trial accuracy and go trial accuracy, meanRT and RT variability were analysed usingpaired t-tests between safe and threat conditions(Table 1). Replicating our previous finding

(Robinson et al., 2013a), accuracy to nogo trials(t(59) = 3.6, p = .0006) improved from the safe tothe threat condition. This was paralleled by adecrease in RT variability to go trials from thethreat to the safe condition (t(59) = 2.5, p = .01).There was also a trend for better accuracy to gotrials (t(59) = 1.7, p = .08) in the threat comparedto the safe condition with no significant change inRT to go trials between conditions (t(59) = 1.3,not significant [ns]).

Comparison of RT that preceded correct andincorrect nogo trials (Table 1) was conducted witha condition (safe, threat) × nogo trial type (correctrejection, error of commission) ANOVA. Twosubjects had no nogo trials error and were notincluded in the analysis. The nogo trial type maineffect was significant due to faster RT precedingerrors compared to correct trials (F(1,57) = 17.1,p = .0001). The condition effect was also signific-ant due to faster response in the threat comparedto the safe condition (F(1,57) = 7.3, p = .009).There was no significant interaction effect.

Thought probes

Figure 1 shows the frequency of each type ofthought. The TRT and TUT (threatTUTs +nonthreatTUTs) scores were analysed separatelybecause of colinearity (TRTs + threatTUTs +nonthreatTUTs = 1). Two main results emerged.First, the rates of TRTs did not significantly differbetween the safe and threat conditions (t(59) =

Table 1. Performance to go and nogo trials in the safe andthreat condition

Safe Threat

Nogo accuracy .70 (.02) .75 (.02)**Go accuracy .89 (.01) .90 (0.1)Go RT, ms 342.3 (6.1) 339.2 (6.1)Go variability, ms 133.0 (8.0) 120.1 (8.0)*RT to 4 go trials before nogo trials, msCorrect omission 341.5 (8.0) 338.2 (7.3)Commission errors 324.1 (10.6)*** 298.5 (7.7)***

*p = .015; **p = .0006. ***Main effect of false alarm versus correct

omission: p = .0001.

GRILLON ET AL.

6 COGNITION AND EMOTION, 2015

Dow

nloa

ded

by [

NIH

Lib

rary

] at

14:

20 2

9 A

pril

2015

1.0, p = .32). Second, within the off-task thoughtcategories (TUT), there was a differential effect ofthe safe/threat condition on thought types (con-dition × thought types: F(1,59) = 49.6, p <

.00009). There were more nonthreatTUTs thanthreatTUTs in the safe condition (t(59) = 2.6,p = .009), and more threatTUTs than nonthreat-TUTs in the threat condition (t(59) = 5.1, p <

.00009). Taken together, these results indicatethat the rate of anxious thoughts (threatTUTs)increased in the threat compared to the safecondition, but this increase was at the expense ofnon-anxious mind wandering (nonthreatTUTs),not at the expense of TRTs.

To examine the validity of the thought probemethodology, we examined the associationbetween thought types and our behavioural andphysiological measures of performance and anxi-ety. We examined whether (1) TRTs correlatedwith an objective measure of endogenous atten-tion, go RT variability (Hu et al., 2012; McVay &Kane, 2009), and (2) threatTUTs correlated withmeasures of state anxiety (subjective anxiety andpotentiated startle). The TRTs correlated nega-tively with RT variability only in the safe condi-tion (r = −.25, p < .05), such that the more on-task the subjects reported to be, the less variabletheir RT was (or, conversely, the less on-task, themore RT variability). The threatTUTs correlatedpositively with subjective anxiety in the threatcondition (r = .31, p < .03). In addition, thechange in the rate of threatTUTs from safe to

threat correlated positively with the differencestartle magnitudes (i.e., fear-potentiated startle;r = .30, p = .03) and with the difference retro-spective anxiety scores (r = .27, p = .04).

Correlation and regression

To investigate potential individual predictors ofperformance, correlations between nogo accuracyand ACS, trait anxiety, TRTs, go RT, and go RTvariability were first calculated separately for thesafe and threat conditions to determine variablesto be used in the multiple regression analysis.Nogo accuracy in the safe condition was positivelycorrelated with ACS (r = .39, p < .01) andnegatively correlated with go RT variability (r =−.39, p < .01). Nogo accuracy in the threatcondition was positively correlated with go meanRT (r = .32, p < .01) and negatively correlatedwith go RT variability (r = .36, p < .01).

To test which of the variables that significantlycorrelated with nogo accuracy explained independ-ent variance, separate stepwise multiple regressionanalyses were used with the nogo accuracy as thedependent variable for the safe and threat condi-tion data. For the safe condition, the final modelcomprised three independent predictors, whichtogether explained 44% of the variance; nogoaccuracy was correlated positively with ACS (beta= .31, F = 9.1, p < .01; partial correlation rs = .33,p < .01) and go mean RT (beta = .42, F =15.0,p < .001; rs = 2.4, p = .06), and negatively with RTvariability (beta = −.56, F = 25.9, p < .0001; rs =−.42, p < .001). For the threat condition, the finalmodel included two independent predictors,which together explained 41% of the variance;nogo accuracy was correlated positively with themean go RT (beta = .62, F = 29.7, p < .0001; rs =−.41, p < .001) and negatively with RT variability(beta = −.63, F = 30.1, p < .0001; rs = .33, p <

.01). Thus, ACS did not contribute to the modelin the threat condition.

The reason why ACS contributed to the modelin safe but not threat was probably because onlythe low-ACS subjects improved their performancefrom safe to threat (i.e., low ACS is associatedwith lower performance in safe but not in threat).

Figure 1. Proportion of each thought type in the safe and threat

conditions. Error bars are standard errors of the mean.

***p < .0009.

ANXIETY AND SUSTAINED ATTENTION

COGNITION AND EMOTION, 2015 7

Dow

nloa

ded

by [

NIH

Lib

rary

] at

14:

20 2

9 A

pril

2015

To examine this possibility, an additional analysis

examined potential predictors of the accuracy

improvement from safe to threat. We conducted

a stepwise regression analysis using the difference

scores (threat minus safe) for nogo accuracy as the

dependent variables. The potential predictors were

ACS and the difference scores (threat minus safe)

for the variables that correlated with performance

in the safe or threat condition, i.e., threatTUTs,

go RT mean and go RT variability. The final

model explained 12% of the variance with only

one variable, ACS, correlating negatively with the

difference nogo accuracy score (beta = −.27, F =

4.8, p < .03). Thus, the lower ACS was, the more

go accuracy improved from safe to threat condi-

tion. For illustration purpose, we calculated nogo

accuracy in the safe and threat conditions in low-

and high-ACS groups based on a median split of

ACS scores (Figure 2). Accuracy in the high-ACS

group increased slightly and non-significantly

from the safe to the threat condition (t(28) = .7,

ns). In contrast, in the low-ACS group, nogo

accuracy improved significantly from the safe to

the threat condition (t(30) = 4.2, p = .0002).

Because a prior study reported slower RT in

high trait anxious subjects, in a post hoc analysis,

we examined the association between trait anxiety

and performance. Consistent with the study by

Righi et al. (2009), we found no significant

correlation between trait anxiety and accuracy in

the safe or threat conditions (all p > .2).

DISCUSSION

In a prior study, we reported improved nogoaccuracy during shock threat (Robinson et al.,2013a). The present study replicates this finding,and, importantly, provides clues as to the variablesthat may have contributed to this performanceimprovement. Specifically, we identified individualdifferences in attention control as an importantfactor influencing performance distinctly duringthreat and safety. Other potential mechanisms willbe discussed.

Individual differences in attention controlemerged as a powerful modulator of performanceon sustained attention during threat. Specifically,during the safe condition, low attention control, asassessed with ACS, was associated with lowerperformance (increased nogo errors of commis-sion) compared to high attention control. Thisresult is consistent with the role of attentioncontrol in sustained attention tests such as SART(Herrmann et al., 2003; Wiersema & Roeyers,2009). However, during shock threat, performanceimprovement was associated with low attentioncontrol, such that the threat condition permittedindividuals with low attention control to performat the level of individuals with high attentionalcontrol. Finally, in contrast to individuals with lowattention control, those with high attention con-trol performed similarly in the safe and threatconditions. It is unlikely that failure to improveduring shock threat in this latter group was due toceiling performance since responses were correctlyinhibited on only 74% of the nogo trials. Thesefindings are reminiscent of results of a shockthreat study, in which low heart rate variability,indicating poor attention control (Healy, 2010;Nigg, 2006; Stifter & Jain, 1996; Thayer & Lane,2000), was associated with improved cognitiveperformance during a threat relative to safe context(Hansen et al., 2009). Thus, in the study byHansen et al. and in our study, shock threatpromoted an attentional state that benefited sub-jects with poor attention control, i.e., subjects whowere most likely to be in a relative inattentivestate. Conceivably, the impaired performanceassociated with low attention control in a safe

Figure 2. Accuracy rates on nogo trials during the safe and threat

conditions in the low- and high-ACS groups. Error bars are

standard errors of the mean. ***p < .0009.

GRILLON ET AL.

8 COGNITION AND EMOTION, 2015

Dow

nloa

ded

by [

NIH

Lib

rary

] at

14:

20 2

9 A

pril

2015

environment could be due to reduced ability tocope with the tedious nature of the vigilance task,leading to poor effortful attention.

It is noteworthy that go RT variability wasreduced during threat condition, suggestingincreased endogenous attention (Hu et al., 2012;Manly, Robertson, Galloway, & Hawkins, 1999),an effect that may be related to increased pre-frontal cortex activity (Forster et al., 2015).Indeed, RT variability during cognitive tasks hasbeen associated to prefrontal top-down control ofattention (Bellgrove, Hester, & Garavan, 2004;Johnson et al., 2008). The threat condition mighthave raised arousal to an optimum attentioncontrol level, or it may have provided the motiva-tion to increase attentional effort towards the taskas a coping strategy to prevent distraction fromthreat processing and to reduce anxiety (Baumeis-ter, 1991; Van Dillen & Koole, 2007). Theincreased performance (decreased nogo errors ofcommission) in the threat condition may havebeen an indirect benefit of this coping strategy.

These results, together with previous findings(Forster et al., 2015; Righi et al., 2009) suggestdifferential effects of trait anxiety and state anxietyon SART performance, with trait anxiety poten-tially impairing attention control, and state anxietyimproving attention control. That trait anxiousindividuals perform poorly on SART is consistentwith reports that such individuals exhibit poorattention control. In the present study, forexample, trait anxiety correlated negatively withACS. However, the post hoc analysis showed thattrait anxiety was not associated with poor per-formance. Because we did not attempt to recruithigh trait anxious subjects, we do not knowwhether the present results would extend to hightrait anxious individuals. In addition, it is possiblethat increased arousal due to shock threat com-pensated for poor attention control associated withhigh trait anxiety. However, in an ongoing study(data in preparation), we show no difference inSART performance between psychiatricallyhealthy controls and individuals with anxietydisorders (who, by definition, have high traitanxiety). Thus, in this regard, our results areconsistent with those of Righi et al. (2009) who

reported no effect of trait anxiety on SARTperformance. Finally, in our ongoing study, theanxious patients also show improved accuracyduring shock threat. Hence, whether trait anxietyis associated with poor performance remain to bedemonstrated. Of course, this does not mean thatthe underlying neural mechanisms supportingperformance are similar in low- and high-traitindividuals (Righi et al., 2009). In fact, traitanxiety-related differences may be easier to identi-fy at the neural level than at the level of behaviour(Righi et al., 2009).

Other potential mechanisms of improved per-formance (nogo accuracy) in the threat versus safecondition can be proposed. SART performance isinfluenced by strategic speed–accuracy trade-off(Peebles & Bothell, 2004). Eysenck’s AttentionControl Theory stipulates that anxious individualsincrease their efforts to keep performance to a levelcomparable to low-trait anxious individuals(Eysenck et al., 2007). This is reflected in slowerRT to maintain accuracy (Eysenck et al., 2007). Infact, a recent study reported that high trait anxiousindividuals slowed down their go-RTs duringSART, probably to avoid making nogo commis-sion errors (Forster et al. [2015] but see Righiet al. [2009]). This RT slowdown was associatedwith reduced prefrontal cortex activation, suggest-ing poor prefrontal control (Forster et al., 2015).The present study showed no effect of shockthreat on go RT, indicating that the improvedaccuracy was not caused by a speed–accuracytrade-off.

Low attention control could also be associatedwith poor inhibitory control of prepotentresponses, and shock threat might have helped totransiently improve inhibitory control. This latterhypothesis is consistent with studies showing alink between anxiety and enhanced motor inhibi-tion to nogo trials (Righi et al., 2009; Sehlmeyeret al., 2010). It is also consistent with theoreticalmodels that propose that anxiety activates thebehavioural inhibition system (Gray, 1987; Quay,1997; Wright et al., 2014). At a neural level, bothSART (Grahn & Manly, 2012; Pardo, Fox, &Raichle, 1991) and anxious anticipation (Mechias,Etkin, & Kalisch, 2010) activate the premotor

ANXIETY AND SUSTAINED ATTENTION

COGNITION AND EMOTION, 2015 9

Dow

nloa

ded

by [

NIH

Lib

rary

] at

14:

20 2

9 A

pril

2015

cortex (Brodmann area 6), which is part of thedorsal attention network of the study by Corbetta,Patel, and Shulman (2008) and is closely linked tothe generation of action. It is therefore possiblethat in the present study, response inhibition wasfacilitated by threat-related potentiation of thisregion’s activity via input from the septo-hippo-campal system, the region implicated in thebehavioural inhibition system (Gray, 1987).

Because it has been argued that mind wander-ing depends on the extent to which the currentcontext primes personally relevant concerns (Klin-ger, 1971; McVay & Kane, 2010), one wouldexpect that a monotonous task such as SARTwould be especially vulnerable to attentional lapsescaused by self-referential mentation or anxiousvigilance during shock threat. This assumptionwas not supported in the present study. Despiteobjective (startle) and subjective (retrospectiverating) measures of heightened anxiety in thethreat condition, shock threat did not increasethe overall rate of TUTs; it increased anxiouscognition (threatTUTs) at the expense of non-anxious cognition (nonthreatTUTs), but not atthe expense of TRTs. One possibility is that therewas a tendency for increased TUTs during theshock threat but that this tendency was neutralisedby increased attention to the task, suggesting thatsubjects were prioritising the task over the threat.Our previous results with an n-back task duringshock threat suggest that only cognitively demand-ing tasks strongly inhibit or limit subjective andphysiological responses to shock threat (Vytal,Cornwell, Arkin, & Grillon, 2012). Hence, onewould expect that in a low-demanding cognitivetask, shock threat would lead to increasesin TUTs.

This study had strengths and limitations.Among the strengths, we manipulated anxiety ina within-subject design and relied on well-estab-lished methods of fear induction and measurement(Grillon & Baas, 2003) and thought assessment(Gruberger et al., 2011; Teasdale et al., 1995).Regarding the latter method, the correlationsbetween TRTs and RT variability, as well asthreatTUTs and subjective anxiety and fear-potentiated startle provide some validity to our

approach. Among the limitations, the probemethodology has its drawbacks. Probing forthoughts might interfere with the same processwe attempted to study and might change thenature of the SART task by making it less tediousor by interfering with the task itself (Giambra,1995). Yet, alternatives to the probe thoughtmethodology, such as retrospective questionnaires,have their distinct weaknesses (Ericsson & Simon,1980). The introduction of thought probes did notseem to have disrupted our threat experiment sincewe replicated our previous finding (Robinson et al.,2013a). To minimise interference from the probethought methodology, we restricted the questionsto a minimum. As a result, we did not assessthoughts that could have helped to interpret thedata further. For example, we did not assessworries unrelated to the shock threat, i.e., worriesabout performance (e.g., I am doing poorly) orunrelated to the experiment (e.g., I am going tothe dentist tonight). However, two findings sug-gest that we successfully assessed anxious thoughtsrelated to the shock threat: first, the threatTUTswere more frequent in the threat than the safecondition, and second the threatTUTs werecorrelated with subjective and objective (fear-potentiated startle) measures of anxiety. It ispossible that worries about the experiment wereincluded in TRTs (and worries unrelated to theexperiment were classified as threatTUTs). Thismight have weakened our ability to find correla-tions between TRTs and performance.

To summarise, we replicated our previousfinding that shock threat improves performanceon SART. We showed that this effect appliedessentially to subjects with low attention control.These subjects’ performance was impaired dur-ing nogo trials in the safe condition (comparedto the high attention control subjects), but nogoperformance was restored in the threat condi-tion, perhaps as a result of non-specific increasein cortical arousal and/or improved inhibitorycontrol of prepotent responses. We also foundthat, while the levels of threat-related TUTsincreased in the threat condition, they did so atthe expense of non–threat-related TUTs, notTRTs. This may partly explain the reason why

GRILLON ET AL.

10 COGNITION AND EMOTION, 2015

Dow

nloa

ded

by [

NIH

Lib

rary

] at

14:

20 2

9 A

pril

2015

worrisome thoughts did not interfere withperformance.

Disclosure statement

No potential conflict of interest was reported by theauthors.

Funding

This work was supported by the Intramural ResearchProgram of the National Institutes of Mental Health[grant number MH002798] (Protocol 01-M-0185).

REFERENCES

Bar-Haim, Y., Lamy, D., Pergamin, L., Bakermans-Kranenburg, M. J., & van IJzendoorn, M. H. (2007).Threat-related attentional bias in anxious and nonan-xious individuals: A meta-analytic study. PsychologicalBulletin, 133(1), 1–24. doi:10.1037/0033-2909.133.1.1

Baumeister, R. F. (1991). Escaping the self: Alcoholism,

spirituality, masochism, and other flights from the

burden of selfhood. New York: Basic Books.Bellgrove, M. A., Hester, R., & Garavan, H. (2004).

The functional neuroanatomical correlates ofresponse variability: Evidence from a responseinhibition task. Neuropsychologia, 42, 1910–1916.doi:10.1016/j.neuropsychologia.2004.05.007

Birk, J.L., Dennis, T. A., Shin, L. M., & Urry, H. L.(2011). Threat facilitates subsequent executive con-trol during anxious mood. Emotion, 11, 1291–1304.doi:10.1037/a0026152

Bishop, S. J. (2009). Trait anxiety and impoverishedprefrontal control of attention. Nature Neuroscience,12(1), 92–98. doi:10.1038/nn.2242

Chiles, W. D. (1958). Effects of shock-induced stresson verbal performance. Journal of Experimental Psy-chology, 56(2), 159–165. doi:10.1037/h0042915

Corbetta, M., Patel, G., & Shulman, G. L. (2008). Thereorienting system of the human brain: From envir-onment to theory of mind. Neuron, 58, 306–324.doi:10.1016/j.neuron.2008.04.017

Cornwell, B. R., Baas, J. M. P., Johnson, L., Holroyd,T., Carver, F. W., Lissek, S., & Grillon, C. (2007).Neural responses to auditory stimulus deviance underthreat of electric shock revealed by spatially-filteredmagnetoencephalography. Neuroimage, 37, 282–289.doi:10.1016/j.neuroimage.2007.04.055

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

Derryberry, D., & Rothbart, M. K. (1997). Reactive

and effortful processes in the organization of tem-

perament. Development and Psychopathology, 9, 633–652. doi:10.1017/S0954579497001375

Easterbrook, J. A. (1959). The effect of emotion on cue

utilization and the organization of behavior. Psycholo-gical Review, 66, 183–201. doi:10.1037/h0047707

Ericsson, K. A., & Simon, H. A. (1980). Verbal reports asdata. Psychological Review, 87, 215–251. doi:10.1037/0033-295X.87.3.215

Evans, D. E., & Rothbart, M. K. (2007). Developing amodel for adult temperament. Journal of Research in

Personality, 41, 868–888. doi:10.1016/j.jrp.2006.11.002Eysenck, M. W., Derakshan, N., Santos, R., & Calvo,

M. G. (2007). Anxiety and cognitive performance:Attentional control theory. Emotion, 7, 336–353.doi:10.1037/1528-3542.7.2.336

First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J.

B. W. (2002). Structured Clinical Interview for DSM-

IV-TR Axis I Disorders, Research Version, Patient

Edition. (SCID-I/P). New York: Biometrics

Research, New York State Psychiatric Institute,

November 2002.Forster, S., Nunez Elizalde, A. O., Castle, E., &

Bishop, S. J. (2015). Unraveling the anxious mind:

anxiety, worry, and frontal engagement in sustained

attention versus off-task processing. Cerebral Cortex,25, 609–618. doi:10.1093/cercor/bht248

Giambra, L. M. (1995). A laboratory method for invest-igating influences on switching attention to task-unrelated imagery and thought. Consciousness and

Cognition, 4(1), 1–21. doi:10.1006/ccog.1995.1001Grahn, J. A., & Manly, T. (2012). Common neural

recruitment across diverse sustained attention tasks.PloS One, 7, e49556. doi:10.1371/journal.pone.0049556.t002

Gray, J. A. (1987). The psychology of fear and stress.Cambridge: Cambridge University Press.

Grillon, C. (2008). Models and mechanisms of anxiety:

Evidence from startle studies. Psychopharmacology,

199, 421–437. doi:10.1007/s00213-007-1019-1Grillon, C., & Baas, J. M. (2003). A review of the

modulation of the startle reflex by affective states andits application in psychiatry. Clinical Neurophysiology,114, 1557–1579. doi:10.1016/S1388-2457(03)00202-5

Gruberger, M., Ben-Simon, E., Levkovitz, Y., Zangen,A., & Hendler, T. (2011). Towards a neuroscience of

ANXIETY AND SUSTAINED ATTENTION

COGNITION AND EMOTION, 2015 11

Dow

nloa

ded

by [

NIH

Lib

rary

] at

14:

20 2

9 A

pril

2015

mind-wandering. Frontiers in Human Neuroscience, 5,56. doi:10.3389/fnhum.2011.00056

Hansen, A. L., Johnsen, B., & Thayer, J. F. (2009).

Relationship between heart rate variability and cog-

nitive function during threat of shock. Anxiety, Stress,

and Coping, 22(1), 77–89. doi:10.1080/10615800802272251

Healy, B. (2010). The effect of attentional control andheart-period variability on negative affect and traitanxiety. The Journal of General Psychology, 137(2),140–150. doi:10.1080/00221301003645079

Helton, W. S., Kern, R. P., & Walker, D. R. (2009).Conscious thought and the Sustained Attention toResponse Task. Consciousness and Cognition, 18,600–607. doi:10.1016/j.concog.2009.06.002

Herrmann, M. J., Jacob, C., Unterecker, S., & Fallgatter,A. J. (2003). Reduced response-inhibition in obsess-ive–compulsive disorder measured with topographicevoked potential mapping. Psychiatry Research, 120,265–271. doi:10.1016/S0165-1781(03)00188-4

Horowitz, M. J., & Becker, S. S. (1971). Cognitiveresponse to stress and experimental demand. Journalof Abnormal Psychology, 78(1), 86–92. doi:10.1037/h0031386

Hu, N., He, S., & Xu, B. (2012). Different efficienciesof attentional orienting in different wanderingminds. Consciousness and Cognition, 21(1), 139–148.doi:10.1016/j.concog.2011.12.007

Johnson, K. A., Barry, E., Bellgrove, M. A., Cox, M.,Kelly, S. Dáibhis, A., … Gill, M. (2008). Dissoci-ation in response to methylphenidate on responsevariability in a group of medication naïve childrenwith ADHD. Neuropsychologia, 46, 1532–1541.doi:10.1016/j.neuropsychologia.2008.01.002

Judah, M. R., Grant, D. M., Mills, A. C., & Lechner,W. V. (2013). Factor structure and validation of theAttentional Control Scale. Cognition and Emotion,28, 433–451. doi:10.1080/02699931.2013.835254

Klinger, E. (1971). Structure and functions of fantasy.New York, NY: Wiley-Interscience.

Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1992). Amotivational analysis of emotion: Reflex-cortex connec-tions. Psychological Science, 3(1), 44–49. doi:10.1111/j.1467-9280.1992.tb00255.x

Langner, R., & Eickhoff, S. (2013). Sustaining atten-tion to simple tasks: A meta-analytic review of theneural mechanisms of vigilant attention. PsychologicalBulletin, 139, 870–900. doi:10.1037/a0030694

Manly, T., Robertson, I. H., Galloway, M., & Hawkins,K. (1999). The absent mind: Further investigations of

sustained attention to response. Neuropsychologia, 37,661–670. doi:10.1016/S0028-3932(98)00127-4

McVay, J. C., & Kane, M. J. (2009). Conducting the trainof thought: Working memory capacity, goal neglect,and mind wandering in an executive-control task.Journal of Experimental Psychology. Learning, Memory,

and Cognition, 35, 196–204. doi:10.1037/a0014104McVay, J. C., & Kane, M. J. (2010). Does mind

wandering reflect executive function or executivefailure? Comment on Smallwood and Schooler(2006) and Watkins (2008). Psychological Bulletin,136, 188–197. doi:10.1037/a0018298

McVay, J. C., & Kane, M. J. (2012a). Drifting fromslow to "D’oh!": Working memory capacity andmind wandering predict extreme reaction times andexecutive control errors. Journal of Experimental

Psychology. Learning, Memory, and Cognition, 38,525–549. doi:10.1037/a0025896

McVay, J. C., & Kane, M. J. (2012b). Why doesworking memory capacity predict variation in readingcomprehension? On the influence of mind wander-ing and executive attention. Journal of ExperimentalPsychology: General, 141, 302–320. doi:10.1037/a0025250

Mechias, M.-L., Etkin, A., & Kalisch, R. (2010). Ameta-analysis of instructed fear studies: Implicationsfor conscious appraisal of threat. Neuroimage, 49,1760–1768. doi:10.1016/j.neuroimage.2009.09.040

Nigg, J. T. (2006). Temperament and developmentalpsychopathology. Journal of Child Psychology and

Psychiatry, 47, 395–422. doi:10.1111/j.1469-7610.2006.01612.x

Pardo, J. V., Fox, P. T., & Raichle, M. E. (1991).Localization of a human system for sustained atten-tion by positron emission tomography. Nature, 349(6304), 61–64. doi:10.1038/349061a0

Peebles, D., & Bothell, D., eds. (2004). Modelling

performance in the Sustained Attention to Response

Task. Pittsburgh, PA: Carnegie Mellon University/University of Pittsburgh.

Pessoa, L. (2009). How do emotion and motivationdirect executive control? Trends in Cognitive Sciences,13, 160–166. doi:10.1016/j.tics.2009.01.006

Quay, H. C. (1997). Inhibition and attention deficithyperactivity disorder. Journal of Abnormal Child Psy-

chology, 25(1), 7–13. doi:10.1023/A:1025799122529Righi, S., Mecacci, L., & Viggiano, M. (2009). Anxiety,

cognitive self-evaluation and performance: ERPcorrelates. Journal of Anxiety Disorders, 23, 1132–1138. doi:10.1016/j.janxdis.2009.07.018

GRILLON ET AL.

12 COGNITION AND EMOTION, 2015

Dow

nloa

ded

by [

NIH

Lib

rary

] at

14:

20 2

9 A

pril

2015

Robertson, I. H., Manly, T., Andrade, J., Baddeley, B.T., & Yiend, J. (1997). “Oops!”: Performancecorrelates of everyday attentional failures in trau-matic brain injured and normal subjects. Neuropsy-chologia, 35, 747–758. doi:10.1016/S0028-3932(97)00015-8

Robinson, O. J., Krimsky, M., & Grillon, C. (2013a).The impact of induced anxiety on response inhibi-tion. Frontiers in Human Neuroscience, 7, 69.

Robinson, O. J., Vytal, K., Cornwell, B. R., & Grillon,C. (2013b). The impact of anxiety upon cognition:perspectives from human threat of shock studies.Frontiers in Human Neuroscience, 7, 203.

Rothbart, M. K., & Derryberry, D. (1981). Developmentof individual differences in temperament. In M. E.Lamb & A. L. Brown (Eds.), Advances in develop-

mental psychology (pp. 37–86). Hillsdale: LawrenceErlbaum.

Sehlmeyer, C., Konrad, C., Zwitserlood, P., Arolt, V.,Falkenstein, M., & Beste, C. (2010). ERP indicesfor response inhibition are related to anxiety-relatedpersonality traits. Neuropsychologia, 48, 2488–2495.doi:10.1016/j.neuropsychologia.2010.04.022

Singh, I. L., Dwivedi, C. B., & Sinha, M. M. (1979).Effects of anxiety on vigilance. Perceptual and Motor

Skills, 49(1), 142. doi:10.2466/pms.1979.49.1.142Smallwood, J., Davies, J. B., Heim, D., Finnigan, F.,

Sudberry, M., O'Connor, R., & Obonsawin, M.(2004). Subjective experience and the attentionallapse: Task engagement and disengagement duringsustained attention. Consciousness and Cognition, 13,657–690. doi:10.1016/j.concog.2004.06.003

Smallwood, J., Fitzgerald, A., Miles, L., & Phillips, L.(2009). Shifting moods, wandering minds: Negativemoods lead the mind to wander. Emotion, 9, 271–276. doi:10.1037/a0014855

Spence, K. W., Farber, I. E., & McFann, H. H. (1956).The relation of anxiety (drive) level to performancein competitional and non-competitional paired-associates learning. Journal of Experimental Psycho-

logy, 52, 296–305. doi:10.1037/h0043507Spielberger, C. D. (1983). Manual for the state-trait

anxiety inventory. Palo Alto, CA: Consulting Psy-chologist Press.

Stifter, C. A., & Jain, A. (1996). Psychophysiologicalcorrelates of infant temperament: Stability ofbehavior and autonomic patterning from 5 to 18months. Developmental Psychobiology, 29, 379–391.doi:10.1002/(SICI)1098-2302(199605)29:4<379::AID-DEV5>3.0.CO;2-N

Su, J. C., Tran, A. G. T. T., Wirtz, J. G., Langteau, R.A., & Rothman, A. J. (2009). Driving under theinfluence (of stress): Evidence of a regional increase inimpaired driving and traffic fatalities after the Sep-tember 11 terrorist attacks. Psychological Science, 20(1),59–65. doi:10.1111/j.1467-9280.2008.02257.x

Teasdale, J. D., Proctor, L., Dritschel, B. H., Taylor,M. J., Lloyd, C. A., Nimmo Smith, I., & Baddeley,A. D. (1995). Stimulus-independent thoughtdepends on central executive resources. Memory &Cognition, 23, 551–559. doi:10.3758/BF03197257

Thayer, J. F., & Lane, R. D. (2000). A model ofneurovisceral integration in emotion regulation anddysregulation. Journal of Affective Disorders, 61, 201–216. doi:10.1016/S0165-0327(00)00338-4

Van Dillen, L., & Koole, S. (2007). Clearing the mind:A working memory model of distraction fromnegative mood. Emotion, 7, 715–723. doi:10.1037/1528-3542.7.4.715

Vinski, M. T., & Watter, S. (2013). Being a grumponly makes things worse: A transactional account ofacute stress on mind wandering. Frontiers in Psycho-

logy, 4, 730. doi:10.3389/fpsyg.2013.00730Vytal, K., Cornwell, B., Arkin, N., & Grillon, C.

(2012). Describing the interplay between anxietyand cognition: From impaired performance underlow cognitive load to reduced anxiety under highload. Psychophysiology, 49, 842–852. doi:10.1111/j.1469-8986.2012.01358.x

Wiersema, J., & Roeyers, H. (2009). ERP correlates ofeffortful control in children with varying levels ofADHD symptoms. Journal of Abnormal Child Psycho-

logy, 37, 327–336. doi:10.1007/s10802-008-9288-7Wright, L., Lipszyc, J., Dupuis, A., Thayapararajah, S.,

& Schachar, R. (2014). Response inhibition andpsychopathology: A meta-analysis of go/no-go taskperformance. Journal of Abnormal Psychology, 123,429–439. doi:10.1037/a0036295

ANXIETY AND SUSTAINED ATTENTION

COGNITION AND EMOTION, 2015 13

Dow

nloa

ded

by [

NIH

Lib

rary

] at

14:

20 2

9 A

pril

2015