Personality and Individual Differences 55 (2013) 898–903

Contents lists available at ScienceDirect

Personality and Individual Differences

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

Intrusive thoughts mediate the association between neuroticismand cognitive function

0191-8869/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.paid.2013.07.019

⇑ Corresponding author. Address: Center for Healthy Aging, The PennsylvaniaState University, 402 BBH Building, University Park, PA 16802, USA. Tel.: +1 814 8639980.

E-mail address: mjs56@psu.edu (M.J. Sliwinski).

Elizabeth Munoz a, Martin J. Sliwinski a,⇑, Joshua M. Smyth a, David M. Almeida a, Heather A. King b

a Pennsylvania State University, University Park, PA, USAb Center for Health Services Research in Primary Care, Health Services Research and Development Service, Department of Veterans Affairs Medical Center, Durham, NC, USA

a r t i c l e i n f o a b s t r a c t

Article history:Received 30 March 2013Received in revised form 3 July 2013Accepted 14 July 2013Available online 6 August 2013

Keywords:NeuroticismCognitionIntrusive thoughtsRepetitive thinkingRuminationWorryNegative affect

Although research has established a negative association between trait neuroticism and cognition, little isknown about the mechanisms that underlie this relationship. We examined the tendency to experienceintrusive thoughts and negative affect as potential mediators of the relationship between neuroticismand cognitive performance. We hypothesized that the tendency to experience intrusive thoughts reflectsineffective attentional control and would account for the relationship between neuroticism and cognitiveperformance over and above the mediating effect of negative affect. Three hundred seventeen adults(Mage = 49.43) completed a series of attention-demanding cognitive tasks as well as self-report measuresof intrusive thoughts, negative affect, and neuroticism. Intrusive thoughts mediated the associationbetween trait neuroticism and cognitive performance beyond negative affect. These findings are consis-tent with the hypothesis that the tendency to experience intrusive thoughts is a mechanism throughwhich trait neuroticism influences cognitive performance.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Neuroticism is a dimension of personality characterized byemotional distress (Larsen & Ketelaar, 1991), lability (Eid & Diener,1999), and reactivity (Bolger & Zuckerman, 1995). Those with highlevels of neuroticism are at increased risk for poor physical andpsychological health (Mroczek & Spiro, 2007; Lahey, 2009). Highlevels of neuroticism are also associated with facets of cognitivehealth, including inefficient cognitive performance (Robinson & Ta-mir, 2005), cognitive decline (Wilson et al., 2005), and increasedrisk of Alzheimer’s disease (Duchek, Balota, Storandt, & Larsen,2007; Wilson, Arnold, Schneider, Li, & Bennett, 2007). However,the psychological mechanisms through which neuroticism influ-ences cognitive function remain largely uninvestigated. Onehypothesis is that high neuroticism individuals exhibit less effi-cient cognitive processing due to elevated ‘‘mental noise’’ causedby mental preoccupations with task-irrelevant intrusive thoughts(IT) and distress (Robinson & Tamir, 2005). The current study pro-vides an explicit test of this hypothesis by examining the media-tional effects of individual differences in IT and emotionaldistress on the neuroticism–cognition relationship.

1.1. Neuroticism and cognitive performance

Studies have generally found negative associations between neu-roticism and a broad range of cognitive functions in both cross-sec-tional and prospective longitudinal designs. Cross-sectional studiesshow negative associations between neuroticism and attention-demanding cognitive tasks that comprise fluid intelligence, suchas episodic memory, numeric and abstract reasoning, and tasks ofperceptual speed (Jorm et al., 1993; Moutafi, Furnham, & Paltiel,2005). In contrast, neuroticism and performance on crystallizedtasks are not robustly associated (Costa, Fozard, McCrae, & Bosse,1976). Prospective studies have shown longitudinal associations be-tween neuroticism and different indices of cognitive decline andimpairment. Individuals high in neuroticism are at increased riskfor developing mild cognitive impairment as well as Alzheimer’sdisease and other dementias (Wilson et al., 2003). Neuroticism isalso a risk factor for cognitive decline in the absence of dementia—those who score high in neuroticism decline an average of thirty per-cent faster than those low in neuroticism (Wilson et al., 2005). Alto-gether, cross-sectional and longitudinal evidence indicates thateffects are most pronounced for attention-demanding tasks, suchas episodic and working memory. These findings raise the questionof what specific processes drive the association between neuroti-cism and cognitive performance in attention-demanding tasks.

The tendency to experience recurrent and intrusive thoughtsrepresents a possible psychological mechanism underlying the

E. Munoz et al. / Personality and Individual Differences 55 (2013) 898–903 899

relationship between neuroticism and cognitive performance.Intrusive thoughts (IT) reflect a range of related concepts such asworry and rumination, and occur with greater frequency in highneuroticism individuals (Muris, Roelofs, Rassin, Franken, & Mayer,2005; Nezlek, 2005; Suls & Martin, 2005). IT may contribute to theneuroticism–cognition relationship by depleting attentional re-sources that are required to effectively perform attention-demand-ing cognitive tasks. Preoccupations with task-irrelevant intrusivethoughts can cause ‘‘mental noise’’ in high neuroticism individuals,resulting in less efficient and more variable cognitive processing(Robinson & Tamir, 2005; Robinson, Wilkowski, & Meier, 2006).Support for this hypothesis comes from analyses showing that highneuroticism individuals are more variable on response time tasks,possibly due to more frequent attentional lapses. Related work byEysenck and colleagues suggests that performance-related IT inter-feres with online processing and results in impaired performance(Eysenck & Calvo, 1992; Eysenck, Derakshan, Santos, & Calvo,2007).

Individual differences in the tendency to experience IT are asso-ciated with lower working memory in college students (Klein &Boals, 2001) and older adults (Stawski, Sliwinski, & Smyth, 2006).A recent study also found that rumination was associated with im-paired mental set shifting (Altamirano, Miyake, & Whitmer, 2010).Stawski et al. (2006) proposed that the experience of IT acts as adual-task load that depletes attentional resources resulting in im-paired performance in attention-demanding tasks. Consistent withthis reasoning, studies have found that individuals who tend toexperience more IT have lower working memory capacity (Kaneet al., 2007) and are more likely to make more mistakes when theyexperience IT (McVay, Kane, & Kwapil, 2009).

An explicit test of whether IT mediates the relationship be-tween neuroticism and cognition has yet to be conducted. Testingthe role of IT as a mediator requires a simultaneous evaluation ofnegative affect (NA) against IT because elevated NA is both a corefeature of neuroticism and related to greater frequency and inten-sity of IT (Moberly & Watkins, 2008; Muris et al., 2005; Nolen-Hoeksema, Wisco, & Lyubomirsky, 2008; Robinson et al., 2006).Studies have also linked NA with slower information processingand with declines in performance on attention-demanding tasks(Backman, Hill, & Fursell, 1996; Dotson, Resnick, & Zonderman,2008). Thus, we examine if IT associated with neuroticism medi-ates the neuroticism–cognition relationship beyond the effects ofNA.

1.2. Hypotheses

First, we hypothesized that individual differences in IT wouldmediate the association between neuroticism and cognitive func-tion. Because of the robust empirical association between IT andNA, we examined this hypothesis in two steps. First, we testedfor mediation by examining both IT and NA in separate modelsand second, by examining them simultaneously in the same model.Our second hypothesis was that individual differences in IT wouldmediate the effect of neuroticism on cognition beyond the effectsof trait NA.

In addition to testing these hypotheses, we also examined ageas a moderator of the neuroticism-intrusive thought-cognitionrelationship. Because previous research suggests that increasingage in adulthood is associated with diminished inhibitory controlthat could increase susceptibility to the effects of intrusivethoughts (Hasher, Zacks, & May, 1999), we expected stronger asso-ciations among these variables in older compared to youngeradults. Specifically, we examined two predictions: (a) That theassociation between neuroticism and IT would increase with age,and (b) That the effect of IT on cognitive performance would bestronger with advancing age in adulthood.

2. Method

2.1. Participants

Three hundred seventeen adults were recruited using advertise-ments in local newspapers, flyers in community centers, and otherpublic venues. Participants were given an introduction to the studyand informed consent was obtained as approved by the SyracuseUniversity Institutional Review Board. Participants were compen-sated $75. Recruitment was stratified by age to obtain a uniformage distribution. Average age of the current sample was 49(SD = 17.23, range = 19–83), and 49.69% were female. The averageyears of education were 13 years (SD = 2.71); 52.1% of participantswere white, 37.6% black, 1.5% Hispanic, and 8.7% other.

2.2. Measures

Cognitive tasks were administered in a fixed order over two in-lab sessions that were approximately one week apart. The follow-ing tasks were administered in the first session: trail making test,logical memory I, existence choice reaction time (CRT) task, digitspan, verbal set switching, reading span, verbal paired associates,number comparison, and the auditory verbal learning test (AVLT).The rest of the tasks were administered during the second sessionin the following order: category fluency, parity CRT, magnitudeCRT, word span, number set switching, counting span, Raven’s pro-gressive matrices, letter fluency, orientation CRT, location CRT, spa-tial span, spatial set switching, operation span, and the Shipleyvocabulary test. A number of health measures were also collectedduring these in-lab sessions and participants provided two cortisolsamples across five consecutive days following the first session—these measures were not used in the current analysis. Between ses-sions, participants completed questionnaires assessing personality,intrusive thoughts, affect, and life experiences.

2.2.1. Fluid intelligenceFluid intelligence was measured using Raven’s progressive

matrices (Raven, 1958). This task measured the extent to whichindividuals could think adaptively. Participants were presentedwith a series of incomplete abstract figures and they chose oneof several abstract figures that best completed the figure. Thedependent measure was the number of abstract figures completedcorrectly out of 30.

2.2.2. Episodic memoryEpisodic memory was measured by three tests that assessed

participants’ ability to recall previously learned information. Inthe logical memory I test, the experimenter read two stories tothe participants and they were then asked to recall as much ofthe story as possible (Wechsler, 1997). In the verbal paired associ-ates (Wechsler, 1997), the examiner read eight pairs of unrelatedwords to participants at a rate of one pair every three seconds.The researcher then gave the participants the first word in the pairand they were asked to produce the second. This was repeated fourtimes, with the same pairs presented in a different fixed randomorder each time. In the AVLT (Schmidt, 1996), participants were gi-ven one minute to study a list of 15 unrelated words. At the end ofthe minute, participants were given one minute to recall as manywords as possible. Raw scores from each of the standard tasks werethe dependent variables.

2.2.3. Primary memoryParticipants completed two tests of short term memory: digit

span and word span (Conway, Cowan, Bunting, Therriault, &Minkoff, 2002). Participants saw a series of digits or words one at

900 E. Munoz et al. / Personality and Individual Differences 55 (2013) 898–903

a time for one second each. After being presented between 5 and 9digits, the participant was asked to recall all of the digits they couldremember in the order they saw them. The word span followed asimilar procedure except that the total number of words in eachseries ranged from 4 to 8 words. Participants performed two trialsat each list length for both tasks. The dependent measures were thetotal number of items recalled in the correct position.

2.2.4. Working memoryParticipants completed three tests of working memory. In the

counting span, participants were presented with a series of dis-plays that included dark blue circles (targets) and dark bluesquares and light blue circles (distractors). After seeing 2 to 6 dis-plays, three question marks ‘‘???’’ prompted the participant to re-call their total count of dark blue circles from the series they hadjust seen. In the operation span, participants verified a series ofequations aloud while trying to remember the words presentedon the screen. They were then prompted to recall all the wordsfrom that series upon its completion (series length ranged from 2to 5). In the reading span, participants verified sentences aloudwhile trying to remember simultaneously presented unrelated let-ters and were prompted to recall all the letters at the end of eachseries (series length ranged from 2 to 5). The dependent measurefor these tasks was the total number of items recalled in the correctposition (Conway et al., 2002; Daneman & Carpenter, 1980).

2.2.5. Processing speedParticipants completed six CRT tasks. In the existence task, par-

ticipants classified a given word as either living or non-living (Goff-aux, Phillips, Sinai, & Pushkar, 2006). In the size task, participantsclassified a given word as either bigger or smaller than a basketball(Mayr & Kliegl, 2000). In the magnitude task participants indicatedwhether a three-digit number was higher or lower than 500. In theparity task, participants indicated whether a three-digit numberwas odd or even. In the orientation task, participants indicatedwhether an arrow presented on the screen was pointing left orright. In the location task, participants indicated whether an arrowpresented on the screen was above or below the midpoint of thescreen. For these measures, the dependent variable was averagereaction time (RT) for correct trials with greater RT scores indicat-ing slower information processing.

2.2.6. Positive and negative affectIn the Positive and Negative Affect Schedule-Expanded Form

(Watson & Clark, 1994), participants indicated the extent to whicha series of adjectives described how they felt in general. Ratingswere made from not at all (1) to extremely (5). Only negative affectitems were used: irritable, afraid, upset, guilty, nervous, hostile,

Table 1Pearson correlation coefficients and descriptive statistics.

Measure 1 2 3 4 5 6

1. Neuroticism –2. Intrusive thoughts .55* –3. Episodic memory �.11* �.22* –4. Primary memory �.16* �.28* .46* –5. Working memory �.17* �.29* .57* .66* –6. Processing speed .02 .17* �.44* �.54* �.51*

7. Fluid IQ �.18* �.24* .49* .48* .56* �8. Negative affect .61* .46* �.10� �.15* �.17*

9. Education �.20* �.31* .41* .37* .41* �10. Gender (female = 1; male = 0) a �.00 .08 .11* �.09 �.0411. Age �.22* �.20* �.22* �.25* �.22*

a Correlations with gender reflect Spearman correlations.* p < .05.� p < 10.

jittery, ashamed, scared, and distressed. One total score was obtainedfrom these items with higher scores indicating greater negativeaffect.

2.2.7. NeuroticismParticipants rated whether 10 statements accurately described

how they perceived themselves in general on a scale of 1 (veryinaccurate) to 5 (very accurate). Items were taken from the Inter-national Personality Item Pool (Goldberg et al., 2006) and includedstatements such as ‘‘am not easily bothered by things.’’ A compos-ite score was calculated where higher scores indicated greaterneuroticism.

2.2.8. Intrusive thoughtsWe assessed the experience of intrusive thoughts using the

White Bear Suppression Inventory (Wegner & Zanakos, 1994). Thisscale consisted of 15 items designed to assess the general experi-ence of intrusive thoughts and what the individual does to controlthese thoughts (e.g., ‘‘I wish I could stop thinking about certainthings’’ and ‘‘I often do things to distract myself from mythoughts’’). Responses were made on a 5-point scale from stronglydisagree to strongly agree and higher scores indicated a greatertendency to experience intrusive thoughts.

3. Results

3.1. Descriptive statistics and correlations

Descriptive statistics and correlations were conducted usingSAS (SAS Institute, 2008). Apart from fluid intelligence, multipleindicators were used to assess all cognitive ability constructs.Based on exploratory and confirmatory factor analysis, we con-structed z-score composites for processing speed, primary, epi-sodic, and working memory. For all subsequent analyses, these z-score composite measures were used as the dependent measures.As shown in Table 1, all of the measures were within acceptableranges for assumptions of normality for the regression analyses.Overall, the relationships among the variables were in the ex-pected direction. Neuroticism was associated with lower cognitiveperformance in most tasks (rs range from �.25 to �.11, p < .05) butwas not associated with processing speed (r = .02, ns). Neuroticismwas also positively associated with IT (r = .55, p < .05) and IT wasassociated with lower performance on fluid intelligence, episodic,primary, and working memory (rs range from �.29 to �.22) as wellas with slower response times on the processing speed task(r = .17, p < .05).

7 8 9 10 M SD Skew Kurtosis

25.83 17.23 0.32 �0.1146.44 12.95 �0.28 �0.48

0.03 0.81 0.06 0.270.01 0.85 0.06 0.270.01 0.89 0.14 �0.39

– �0.02 0.86 1.66 4.37.49* – 20.53 5.99 �0.85 0.22.04 �.14* – 17.28 6.87 1.18 1.30.23* .47* �.18* – 13.34 2.71 0.22 0.38.08 �.04 �.08 .03 – 0.50 0.50.29* �.20* �.21* .12* .03 49.43 17.23 0.46 �1.02

E. Munoz et al. / Personality and Individual Differences 55 (2013) 898–903 901

3.2. Mediation

Using Mplus software (Muthen & Muthen, 1998), we tested sta-tistical mediation using bootstrap methodology to evaluate the sig-nificance of the indirect effect (Edwards & Lambert, 2007).Preliminary analyses showed that both NA and IT mediated the ef-fect of neuroticism on the cognitive variables when examined sep-arately. However, our primary interest was in the simultaneousmediation effects of IT and NA (see Fig. 1). Results in Table 2 showthat the total effect of neuroticism was significant for fluid intelli-gence, primary, and working memory (p < .05), marginally signifi-cant for episodic memory (p = .05) and not significant forprocessing speed. We tested for mediation effects between neurot-icism and processing speed because mediation can still exist underthis circumstance (MacKinnon & Fairchild, 2009). The direct effectof neuroticism was not significant for any of the cognitive mea-sures after IT and NA were introduced to the model (Bs range from�.045 to �.001, ns). The indirect effect of neuroticism through ITwas statistically significant in all cases, (Bs range from �.054 to�.011, p < .05 and B = .010, p < .05 for speed) providing supportfor our hypothesis that IT mediates the neuroticism–cognitionrelationship. IT accounted for between 65% and 90% of the total ef-fect of neuroticism on primary, episodic, and working memory andfor 52% of the effect of neuroticism on fluid intelligence. Addition-ally, the indirect effect through NA was not significant for any ofthe cognitive measures (Bs range from �.005 to .003, ns). AlthoughNA covaried with both neuroticism and with all the cognitive tasks,it does not account for a significant amount of variance when IT isalso included as an indirect path. This result indicates that themediating effect of IT on the neuroticism–cognition relationshipis independent of individual differences in NA.

IT

N C

NA

aIT

aNA bNA

bIT

c’

Fig. 1. Illustration of a multiple mediation design in which neuroticism (N) has aneffect on cognition (C) through intrusive thought (IT) or negative affect (NA) asindicated by the products of aIT by bIT or aNA by bNA.

Table 2Individual path estimates and multiple mediation estimates for cognitive variables.

Dependent variable aIT bIT aNA bNA

Primary memory .818 (.074) * �.013 (.004) * .514 (.044) * �.007Working memory .818 (.074) * �.013 (.004) * .514 (.044) * �.009Episodic memory .818 (.074) * �.011 (.004) * .514 (.044) * �.001Fluid intelligence .818 (.074) * �.067 (.028) * .514 (.044) * �.007Processing Speed .818 (.074) * .012 (.004) * .514 (.044) * .005

Note: The small values necessitate reporting results to three decimal points. ResultsIT = intrusive thoughts, NA = negative affect.* p < .05.� p < .10.

3.3. Age moderation

We tested for age moderation of the effect between neuroticismon IT and of the effect of IT and cognition (Edwards & Lambert,2007). Results showed that the effect of neuroticism on IT didnot vary across age (B = .000, ns) and that age did not significantlymoderate the effect of IT on any of the composite cognitive vari-ables (Bs = .000, ns). Contrary to our expectations, the effect of neu-roticism on IT and the effect of IT on cognition was age invariant.

4. Discussion

Our results supported our two main hypotheses. First, we foundthat individual differences in intrusive thoughts (IT) accounted forthe relationships between trait neuroticism and performance onattention-demanding cognitive tasks. Consistent with other empir-ical work (i.e., Eysenck & Calvo, 1992; Eysenck et al., 2007), theseresults suggest that a cognitive tendency to experience IT indicatesineffective attentional control that results in lower cognitive per-formance. Second, the disposition to experience IT mediated theneuroticism–cognition relationship independent of trait negativeaffect (NA). These results demonstrate that individual differencesin IT are more important than trait NA in accounting for the rela-tionship between neuroticism and cognitive performance. Analysisof age moderation did not support our expectation that the effectsof IT on cognition would increase with advancing age in adulthood.

4.1. Neuroticism, intrusive thoughts, and cognitive performance

We consider three possible explanations for our finding that ITmediated the relationship between neuroticism and cognitive per-formance. First, individuals with high levels of trait neuroticismmay be more susceptible to experiencing IT under situations thattax their attentional control (e.g., Muris et al., 2005; Robinson & Ta-mir, 2005). Among these individuals, the experience of IT during acognitive task may directly interfere with performance. This expla-nation is consistent with ‘‘mental noise’’ (Robinson & Tamir, 2005)and ‘‘processing efficiency’’ (Eysenck & Calvo, 1992; Eysenck et al.,2007) accounts which postulate that ITs disrupt online cognitiveprocessing. A second possible explanation is that neuroticism-re-lated IT correlates with cognitive performance because it is a mar-ker of individual differences in attentional resources. Thus,individuals with high trait neuroticism perform worse on atten-tion-demanding tasks not because they are more likely to experi-ence disruptive IT while performing a task, but because theyhave less effective attentional resources required to successfullyperform these tasks (e.g., Kane et al., 2007). That is, increased fre-quency of IT and poorer working memory do not bear a direct rela-tionship with each other, but rather both result from a diminishedcapacity of attentional control.

Total Direct (c0) Indirect through:

IT NA

(.008) �.017 (.005) * �.002 (.007) �.011 (.003) * �.004 (.004)(.008) �.017 (.006) * �.002 (.007) �.011 (.004) * �.005 (.004)(.008) �.010 (.005)� �.001 (.007) �.009 (.003) * .000 (.004)(.056) �.103 (.043) * �.045 (.057) �.054 (.024) * �.004 (.029)(.009) .004 (.006) �.008 (.008) .010 (.004) * .003 (.004)

covaried for age, years of education, and gender. Bootstrap sample size = 5000.

902 E. Munoz et al. / Personality and Individual Differences 55 (2013) 898–903

However, neither of these two accounts can explain findingsfrom longitudinal studies that show increased rates of declineand risk for cognitive impairment in high neuroticism individuals.A third explanation which could account for both cross-sectionaland longitudinal associations between neuroticism and cognitionis that trait IT could also be an indicator of chronic stress (Wilsonet al., 2003, 2005, 2007). The tendency to experience repetitive andintrusive thoughts has been proposed to be a mechanism throughwhich stress leads to poor physical health outcomes (Brosschot,Gerin, & Thayer, 2006; Smyth, Zawadzki, & Gerin, 2013). This cog-nitive style is hypothesized to prolong the stress response becauseindividuals who experience greater IT may relive stressful eventsin their mind even after those events have long passed. A pro-longed physiological stress response due to IT may accumulateover time and lead to cognitive impairment (Sapolsky, Krey, &McEwen, 1986; Glynn, Christenfeld, & Gerin, 2002).

An important limitation of this study is that it involves only across-sectional analysis, precluding statements of the temporalprecedence of elevated states of IT and cognitive performance.Additionally, we could not evaluate whether IT is a marker ofcumulative stress that causes cognitive decline. A longitudinalexamination of IT, stress, and cognition that examines the sequen-tial and dynamic associations among these variables across multi-ple time scales is necessary to determine how IT may propagateand prolong the physiological stress response and negatively affectcognitive performance in both the short- and long-term.

4.2. Neuroticism, intrusive thoughts, and age

We predicted that IT should impact performance on attention-demanding tasks to a greater extent among older as compared toyounger adults (Hasher et al., 1999). Our hypothesis was not sup-ported, as age did not moderate the indirect effect through IT. Ourresults are consistent with an alternative view that older adultsmay not have a specific deficit in inhibitory and attentional control(Verhaeghen, 2011). The negative association between age and ITmay also indicate that older adults are better able to regulate neg-ative intrusive thoughts compared to younger adults. Numerousstudies indicate that older adults adopt more successful self-regu-latory strategies, such as attention-shifting strategies that allowthem to minimize negative emotional states and garner more posi-tive emotional experiences (e.g., Carstensen, Isaacowitz, & Charles,1999). This improvement in self-regulation strategies may also ap-ply to thought control in which older adults may be better able tocontrol intrusive thoughts compared to their youngercounterparts.

We cannot rule out other extraneous factors that could haveinfluenced these results. For instance, neuroticism is related to abroad range of physical health outcomes that can predispose indi-viduals to develop more health problems as they age (e.g., Lahey,2009) and increases in levels of neuroticism are associated withgreater mortality risk (Mroczek & Spiro, 2007). Comparisons ofyounger and older adults on neuroticism may thus be biased byselective survivorship and underestimate differential effects ofneuroticism across the adult lifespan.

5. Conclusion

This study demonstrated that the tendency to experience intru-sive thoughts mediates the relationship between neuroticism andcognitive performance. We found that the effect of intrusivethoughts remains after accounting for negative affect suggestingthat this cognitive characteristic of neuroticism is more predictiveof cognitive performance. The current results extend previous

findings by demonstrating that individuals who score high in traitneuroticism may have poorer cognitive performance because oftheir tendency to experience intrusive thoughts.

6. Author notes

Support for this study provided by the National Institute onAging (Grant AG26728).

The views expressed in this article are those of the authors anddo not necessarily reflect the position or policy of the Departmentof Veterans Affairs or the United States government.

The last author was supported by a post-doctoral fellowshipfrom the Department of Veterans Affairs, Office of Academic Affil-iations, Health Services Research and Development Service (TPP21-020).

References

Altamirano, L. J., Miyake, A., & Whitmer, A. J. (2010). When mental inflexibilityfacilitates executive control: Beneficial side effects of ruminative tendencies ongoal maintenance. Psychological Science, 21, 1377–1382.

Backman, L., Hill, R. D., & Fursell, Y. (1996). The influence of depressivesymptomatology on episodic memory functioning among clinicallynondepressed older adults. Journal of Abnormal Psychology, 105, 97–105.

Bolger, N., & Zuckerman, A. (1995). A framework for studying personality in thestress process. Journal of Personality and Social Psychology, 69, 890–902.

Brosschot, J. F., Gerin, W., & Thayer, J. F. (2006). The perseverative cognitionhypothesis: A review of worry, prolonged stress-related physiologicalactivation, and health. Journal of Psychosomatic Research, 60, 113–124.

Carstensen, L. L., Isaacowitz, D., & Charles, S. T. (1999). Taking time seriously: Atheory of socioemotional selectivity. American Psychologist, 54, 165–181.

Conway, A., Cowan, N., Bunting, M., Therriault, D., & Minkoff, S. (2002). A latentvariable analysis of working memory capacity, short-term memory capacity,processing speed, and general fluid intelligence. Intelligence, 30, 163–183.

Costa, P. T., Fozard, J. L., McCrae, R. R., & Bosse, R. (1976). Relations of age andpersonality dimensions to cognitive ability factors. Journal of Gerontology, 31,663–669.

Daneman, M., & Carpenter, P. (1980). Individual differences in working memory andreading. Journal of Verbal Learning and Verbal Behavior, 19, 450–466.

Dotson, V. M., Resnick, S. M., & Zonderman, A. B. (2008). Differential association ofconcurrent, baseline, and average depressive symptoms with cognitive declinein older adults. American Journal of Geriatric Psychology, 16, 318–330.

Duchek, J. M., Balota, D. A., Storandt, M., & Larsen, R. (2007). The power ofpersonality in discriminating between healthy aging and early-stage Alzheimer’s disease. Journal of Gerontology: Psychological Sciences, 62B,353–361.

Edwards, J. E., & Lambert, L. S. (2007). Methods for integrating moderation andmediation: A general analytical framework using moderated path analysis.Psychological Methods, 12, 1–22.

Eid, M., & Diener, E. (1999). Intraindividual variability in affect: Reliability, validity,and personality correlates. Journal of Personality and Social Psychology, 76,662–679.

Eysenck, M. W., & Calvo, M. G. (1992). Anxiety and performance: The processingefficiency theory. Cognition and Emotion, 6, 409–434.

Eysenck, M. W., Derakshan, N., Santos, R., & Calvo, M. G. (2007). Anxiety andcognitive performance: Attentional control theory. Emotion, 71, 336–353.

Glynn, L. M., Christenfeld, N., & Gerin, W. (2002). The role of rumination in recoveryfrom reactivity: Cardiovascular consequences of emotional states.Psychosomatic Medicine, 64, 714–726.

Goffaux, P., Phillips, N. A., Sinai, M., & Pushkar, D. (2006). Behavioural andelectrophysiological measures of task switching during single and mixed-taskconditions. Biological Psychology, 72, 278–290.

Goldberg, L., Johnson, J., Eber, H., Hogan, R., Ashton, M., Cloninger, C., et al. (2006).The International personality item pool and the future of public-domainpersonality measures. Journal of Research in Personality, 40, 84–96.

Hasher, L., Zacks, R. T., & May, C. P. (1999). Inhibitory control, circadian arousal, andage. In D. Gopher & A. Koriat (Eds.), Attention and performance XVII(pp. 653–675). Cambridge, MA: MIT Press.

Jorm, A. F., MacKinnon, A. J., Christensen, H., Henderson, S., Scott, R., & Korten, A.(1993). Cognitive function and neuroticism in an elderly community sample.Personality and Individual Differences, 15, 721–723.

Kane, M. J., Brown, L. H., McVay, J. C., Silvia, P. J., Myin-Germeys, I., & Kwapil, T. R.(2007). For whom the mind wanders, and when. Psychological Science, 18,614–621.

Klein, K., & Boals, A. (2001). The relation of life event stress and working memorycapacity. Applied Cognitive Psychology, 15, 565–579.

Lahey, B. B. (2009). Public significance of neuroticism. American Psychologist, 64,241–256.

E. Munoz et al. / Personality and Individual Differences 55 (2013) 898–903 903

Larsen, R. J., & Ketelaar, T. (1991). Personality and susceptibility to positive andnegative emotional states. Journal of Personality and Social Psychology, 61,132–140.

MacKinnon, D. P., & Fairchild, A. J. (2009). Current directions in mediation analysis.Current Directions in Psychological Science, 18, 16–20.

Mayr, U., & Kliegl, R. (2000). Task-set switching and long-term memory retrieval.Journal of Experimental Psychology: Learning, Memory, & Cognition, 26,1124–1140.

McVay, J. C., Kane, M. J., & Kwapil, T. R. (2009). Tracking the train of thought fromthe laboratory into everyday life: An experience-sampling study of mindwandering across controlled and ecological contexts. Psychonomic Bulletin &Review, 16, 857–863.

Moberly, N. J., & Watkins, E. R. (2008). Ruminative self-focus, negative life events,and negative affect. Behaviour Research and Therapy, 46, 1034–1039.

Moutafi, J., Furnham, A., & Paltiel, L. (2005). Can personality predict intelligence?Personality and Individual Differences, 38, 1021–1033.

Mroczek, D. K., & Spiro, A. (2007). Personality change influences mortality in oldermen. Psychological Science, 18, 371–376.

Muris, P., Roelofs, J., Rassin, E., Franken, I., & Mayer, B. (2005). Mediating effects ofrumination and worry on the links between neuroticism, anxiety anddepression. Personality and Individual Differences, 39, 1105–1111.

Muthen, L. K., & Muthen, B. O. (1998). Mplus (Version 6). Los Angeles, CA: Muthen &Muthen.

Nezlek, J. B. (2005). Distinguishing affective and non-affective reactions to dailyevents. Journal of Personality, 73, 1539–1568.

Nolen-Hoeksema, S., Wisco, B. E., & Lyubomirsky, S. (2008). Rethinking rumination.Perspectives on Psychological Science, 3, 400–424.

Raven, J. (1958). Standard progressive matrices. London: H.K. Lewis.Robinson, M. D., & Tamir, M. (2005). Neuroticism as mental noise: A relation

between neuroticism and reaction time standard deviations. Journal ofPersonality and Social Psychology, 89, 107–114.

Robinson, M. D., Wilkowski, B. M., & Meier, B. P. (2006). Unstable in more ways thanone: Reaction time variability and the neuroticism/distress relationship. Journalof Personality, 74, 311–344.

Sapolsky, R. M., Krey, L. C., & McEwen, B. S. (1986). The neuroendocrinology of stressand aging: The glucocorticoid cascade hypothesis. Endocrine Reviews, 7,284–301.

SAS Institute (2008). SAS (Version 8.2). Cary, NC: SAS Institute, Inc.Schmidt, M. (1996). Rey auditory verbal learning test. Los Angeles: Western

Psychological Services.Smyth, J., Zawadzki, M., & Gerin, W. (2013). Stress and disease: A structural and

functional analysis. Social and Personality Psychology Compass, 7, 217–227.Stawski, R. S., Sliwinski, M. J., & Smyth, J. M. (2006). Stress-related cognitive

interference predicts cognitive function in old age. Psychology and Aging, 21,535–544.

Suls, J., & Martin, R. (2005). The daily life of the garden-variety neurotic: Reactivity,stressor exposure, mood spillover, and maladaptive coping. Journal ofPersonality, 73, 1485–1510.

Verhaeghen, P. (2011). Aging and executive control: Reports of a demise greatlyexaggerated. Current Directions in Psychological Science, 20, 174–180.

Watson, D., & Clark, L. (1994). Manual for the positive and negative affect schedule –Expanded form. University of Iowa.

Wechsler, D. (1997). Wechsler Memory Scale – Third edition. Administration andscoring manual. San Antonio, TX: The Psychological Corporation.

Wegner, D. M., & Zanakos, S. (1994). Chronic thought suppression. Journal ofPersonality, 62, 615–640.

Wilson, R. S., Arnold, S. E., Schneider, J. A., Li, Y., & Bennett, D. A. (2007). Chronicdistress, age related neuropathology, and late-life dementia. PsychosomaticMedicine, 69, 47–53.

Wilson, R. S., Bennett, D. A., Mendes de Leon, C. F., Bienias, J. L., Morris, M. C., &Evans, D. A. (2005). Distress proneness and cognitive decline in a population ofolder persons. Psychoneuroendocrinology, 30, 11–17.

Wilson, R. S., Evans, D. A., Bienias, J. L., Mendes de Leon, C. F., Schneider, J. A., &Bennett, D. A. (2003). Proneness to psychological distress is associated with riskof Alzheimer’s disease. Neurology, 61, 1479–1485.