Rumination, Mood and Cognitive Performance

doi:10.4236/psych.2013.43A034

Paper Menu >>

Journal Menu >>

Psychology

2013. Vol.4, No.3A, 224-231

Published Online March 2013 in SciRes (http://www.scirp.org/journal/psych) http://dx.doi.org/10.4236/psych.2013.43A034

224

Rumination, Mood and Cognitive Performance

Jay K. Brinker, Mel Campisi, Laura Gibbs, Rebekah Izzard

Research School of Psychology, The Australian National University, Canberra, Australia

Email: [email protected]

Received October 29th, 2012; revised December 10th, 2012; accepted January 9th, 2013

Rumination, DEFINED AS REPETITIVE, RECURRENT AND UNCONTROLLABLE THINKING, has

been implicated in cognitive impairments, particularly dysexecutive function, in people with depressed

mood, what is now critical is an examination of the link between rumination and cognitive impairment

independent of mood. Rumination’s direct relationship to basic cognitive functions may help explain the

stability of rumination, and how it predicts future episodes of depressed mood. Two experiments exam-

ined how trait and experimentally induced rumination, and negative mood relate to attention and inhibi-

tion at varying levels of cognitive load in non-depressed participants. Study 1 found trait rumination sig-

nificantly predicted errors of attention when cognitive load was low or high, but not at a medium level.

Trait rumination also interacted with frequency of task unrelated thoughts to predict performance. Nega-

tive mood significantly predicted errors of inhibition also when the cognitive load was low or high. Study

2 expanded on the first by including two additional measures of executive function. Results were repli-

cated for the first study and showed a ruminative thought style predicted inhibition, reaction time and set

shifting in the additional measures.

Keywords: Rumination; Mood; Executive Functions; Cognitive Load; Off-Task Thoughts

Introduction

There are many ways to conceptualise rumination; a response

to depressed mood (Nolen-Hoeksema, Parker, & Larson, 1994;

Verhaeghen, Joorman, & Khan, 2005), an emotion regulation

strategy (Nolen-Hoeksema, Wisco, & Lyubomirsky, 2008), a

form of self-focused attention (Mor & Winquist, 2002), as a

process or style of thinking (Brinker & Dozois, 2009) and so on.

These conceptions are not necessarily contradictory, but simply

different ways of considering and examining a complex phe-

nomenon. In its most general sense, rumination is a style of

thinking characterised by repetitive, recurrent, intrusive and un-

controllable thoughts (Brinker & Dozois, 2009). The relation-

ship between rumination and cognitive impairment has been

demonstrated, but what is not known is 1) the relationship be-

tween rumination and cognition independent of depressed

mood; and 2) examination of the causal relationship between

rumination and impairment. These may be addressed by meas-

uring rumination as an individual difference in a non-depressed

sample using a measure of the process of rumination, and by

experimentally manipulating rumination to observe changes in

cognitive performance.

Within the different conceptions of rumination, it is generally

agreed that rumination is a stable phenomenon (Brinker & Do-

zois, 2009; Nolen-Hoeksema, Wisco, & Lybomirsky, 2008)

with serious clinical implications due to the consistent rela-

tionship with depressed mood. Rumination is related to greater

depressed mood, longer depressive episodes and the occurrence

of future depressive episodes (Brinker & Dozois, 2009; Nolen-

Hoeksema, Wisco, & Lybomirsky, 2008). The link to future

episodes of depression raises two important points. First, even

when negative mood is ameliorated, the process of thinking in a

ruminative style may continue, and the vulnerability posed by

rumination may continue. Second, the stability of ruminative

thinking suggests the process may be related to individual dif-

ferences in basic cognitive function—in particular executive

functioning.

Executive functions are a group of cognitive abilities that

enable goal directed behaviour including attentional control,

inhibitory systems and cognitive flexibility. Regardless of con-

tent, ruminative thoughts reflect an unintentional shift in atten-

tion away from the task at hand (Martin & Tesser, 1996), and

the failure of cognitive inhibitory systems to limit attention to

task-relevant information (Carson, Peterson, & Higgins, 2003;

Hertel, 1997; Joormann, 2004). In other words people who ru-

minate are unable to inhibit the ruminative thoughts and main-

tain attention elsewhere.

Research has demonstrated a relationship between rumina-

tion and attentional inflexibility (Davis & Nolen-Hoeksema,

2000; Lo, Lau, Cheung, & Allen, 2012). Watkins and Brown

(2002) suggest that rumination reduces cognitive flexibility by

overloading cognitive resources, primarily in the form of defi-

cits in switching and inhibition (Miyake et al., 2000). Whit-

mer and Banich (2007) suggest that switching attention from

one conceptual set to another is difficult for individuals who

ruminate and show rumination is linked to difficulties with

inhibiting previously relevant task sets in a non-depressed sam-

ple. Davis and Nolen-Hoeksema (2000) also found evidence for

attentional inflexibility, with ruminators making more per-

severation errors on the Wisconsin Card Sorting Test and tak-

ing more time on a measure of psychomotor speed compared to

non-ruminators.

As mentioned, Watkins and Brown (2002) argue perform-

ance impairment may result from limited cognitive resources,

and rumination may further consume resources that would oth-

erwise be directed towards the primary task. Increasing the

cognitive demands of the current external task has been found

to lead to a reduction in the frequency of off-task thoughts (e.g.

J. K. BRINKER ET AL.

Teasdale et al., 1995). Antrobus (1968) observed that increas-

ing the speed of stimulus presentation in a signal-detection task

increased the cognitive demands for processing task related

information, resulting in an improvement in task performance.

This was attributed to more cognitive resources needing to be

directed towards the task, leaving fewer cognitive resources

available for the production of off-task thoughts. These results

suggest that attention-demanding tasks may disrupt rumination

(Teasdale et al., 1995). Lyubomirsky and colleagues (2003)

tested the effect of dysphoric rumination on participant’s per-

formance on three academic tasks (reading a passage, watching

a videotaped lecture and proofreading written text) and found

that decrements in task performance tended to be greater for

tasks with mild to moderate difficulty. Lyubomirsky and col-

leagues (2003) also found that dysphoric participants who ru-

minated were slower at the tasks than the other comparison

groups even if their accuracy was equivalent.

Typically, rumination is tested in participants experiencing

dysphoric mood (natural or induced) with much research sug-

gesting it is the combination of negative mood and rumination

that produces effects (Lyubomirsky, Kasri, & Zehm, 2003).

However, there is reason to suppose that a ruminative style of

thinking may be related to impaired executive functions re-

gardless of mood. The previous research examining intrusive

thoughts, inhibition and executive function in relation to cogni-

tive performance illustrates differences in non-depressed sam-

ples without inducing negative mood (Carson, Peterson, & Hig-

gins, 2003, Teasdale et al., 1995, Whitmer & Banich, 2007).

Rumination has been linked to executive function (Davis &

Nolen-Hoeksema, 2000) and Williams, Suchy and Rau (2009)

argue that there are individual differences in executive function

both in conjunction with and independent of mood. Given the

firm basis of research in rumination and depression combined,

it is now time to examine the influence of rumination in a

non-depressed sample without inducing negative mood. It is

hypothesised that rumination, even in the absence of depressed

mood, will be related to impairment in executive function.

Study 1

Several variables are involved when considering the above

findings; attention, inhibition, cognitive load, task unrelated

thoughts and both trait and induced rumination. This project

will examine these variables together by employing a continu-

ous performance task to directly manipulate cognitive load and

measure attention and inhibition. Further, a baseline measure of

trait rumination and experimentally induced rumination will be

included. It is hypothesised that participants in the rumination

condition will show poorer performance than those in a distrac-

tion condition, on attention and inhibition when cognitive load

is low. It is also predicted that participants high on trait rumina-

tion will perform more poorly than those with low trait rumina-

tion. Finally, greater frequency of off-task thoughts will be

related to poorer cognitive performance.

Method

Participants

One-hundred first year psychology students (31 male and 69

female) from the Australian National University participated in

the study ranging in age from 17 - 57 years (M = 19.83, SD =

4.77). Participants WERE RECRUITED VIA POSTERS

PLACED AROUND THE DEPARTMENT AND ON THE

FIRST YEAR COURSE WEBSITE. THEY received one hour

of course credit for their participation. THERE WAS NO SET

INCLUSION OR EXCLUSION CRITERIA AND PARTICI-

PANTS WERE NOT RECRUITED BASED ON PREVIOUS

MENTAL HEALTH HISTORY.

Materials

Positive and Negative Affect Schedule (PANAS-X; Watson

& Clark, 1994): THIS MEASURE WAS INCLUDED TO

CONTROL FOR BASELINE MOOD. The PANAS consists of

20 mood adjectives: 10 positive (PANAS-P) and 10 negative

(PANAS-N). Respondents are asked to rate on a 5-point scale

the degree to which they have experienced each of these moods

in the past 24 hours. The scale ranges from 1 (not at all) to 5

(very much). These scales were used as a baseline measure of

the degree of positive and negative affect participants were

experiencing prior to completing the study. The Chronbach’s

alphas for this study were .87 for the PANAS-P and .85 for the

PANAS-N.

Ruminative Thought Styles Questionnaire (RTS; Brinker &

Dozois, 2009). The RTS is a measure of a general ruminative

thought style characterised by repetitive, recurrent, intrusive

and uncontrollable thinking. It consists of 20 statements that

respondents rate as self descriptive or not on a Likert-type scale

from 1—not at all like me to 7—very much like me. The total

scores range from 20 to 140 with higher scores indicating

greater ruminative thinking. The RTS shows good internal re-

liability (α = .87 to .92), good test-retest reliability (r = .80) and

good convergent and divergent validity (Brinker & Dozois,

2009). Cronbach’s alpha for the current study is .91.

Rumination/Distraction Induction (Lyubomirsky & Nolen-

Hoeksema, 1993): This task IS THE EXPERIMENTAL MA-

NIPULATION AND was designed to induce either a rumina-

tive or distractive style of thought in participants, and has been

widely used in rumination research (e.g. Rimes & Watkins,

2005; Watkins & Moulds, 2005; Lyubomirsky, Kasri, & Zehm,

2003). The rumination condition asks participants to think

about a series of 45 self-focused statements, (e.g. “think about

what your feelings might mean”), and is designed to focus par-

ticipants on the meaning, causes and consequences of their

current feelings (Nolen-Hoeksema, Wisco, & Lyubomirsky,

2008). The distraction condition required participants to focus

on 45 statements designed to shift participants’ focus away

from themselves, (e.g. “think about a boat sailing on the

ocean”). Each induction lasted for eight minutes. While these

are typically used with individuals currently experiencing de-

pressed mood they may be of use with a healthy sample by

inducing self-focused attention, a component of the ruminative

response style.

Continuous Performance Task (CPT): THIS TASK MEAS-

URES ATTENTION AND INHIBITION—THE DEPEND-

ENT VARIABLE. Participants completed a computer task

involving attending to a series of letters presented on the screen

and responding by pressing the space bar whenever an “X” was

presented. Letters were presented in nine blocks of 20 letters,

five of which were “X”s in each block. All letters were dis-

played for 250 ms each. The inter-stimulus intervals were 1, 2

and 4 seconds (fast, medium and slow speeds, respectively)

creating 3 levels of cognitive load—high, medium and low.

Participants viewed three blocks for each of the speeds. Par-

ticipants were also given a ‘practice round’ before commencing

J. K. BRINKER ET AL.

the task. The task took approximately eight minutes to complete,

and produced two outcome measures for each of the speeds:

number of correct hits (attention) and number of false hits

(failure to inhibit). Because the key comparison is across ex-

perimental groups, not across speed blocks, the order of pres-

entation speed was held constant for all participants.

Off-Task Thoughts: Following the CPT, participants reported

the type and frequency of off-task thoughts they experienced

AS A POSSIBLE MEDIATING VARIABLE. Participants

reported frequency of task related and unrelated thoughts, and

frequency of repeating thoughts using Likert scales ranging

from 1 (not at all) to 7 (frequently).

Mood Rating Scale: A visual analogue scale (VAS) was in-

cluded as a brief, repeatable mood measure to assess changes in

participant’s mood throughout the study. Based on Lyubomir-

sky & Nolen-Hoeksema’s (1995) measure, participants were

asked to indicate the degree of positive or negative affect they

were currently experiencing by moving a marker up or down a

100 mm line representing a continuum from 0 (extremely nega-

tive affect) to 100 (extremely positive affect). Higher scores on

these reflect lower negative mood.

Procedure

Participants completed the study in a computer lab in groups

of 5 - 10 and were randomly assigned to either the rumination

or distraction condition. All sections of the study were pre-

sented and completed by participants on computer, and partici-

pants simply clicked on the “Next” button to move through the

protocol. Participants first completed the PANAS-X and the

first visual analogue scale mood measure (Mood 1) to assess

their baseline mood, and the RTS to assess their trait ruminative

thinking style. THESE VARIABLES ACT AS CONTROL

WHEN EXAMINING THE EFFECT OF THE EXPERIMEN-

TAL MANIPULATION. They were then presented with their

assigned induction procedure. Participants were asked to rate

their current mood using a second visual analogue scale (Mood

2) and then complete the Continuous Performance Task. Par-

ticipants reported the off-task thoughts they experienced during

the CPT and completed a final VAS (Mood 3).

Results and Discussion

T-tests revealed no significant between group differences for

age (t = –1.46, p = .15) or mood as measured by the PANAS

(positive mood t = –.70, p = .49; negative mood t = 1.62, p

= .11) and VAS1 prior to the induction (t = –.87, p = .34).

However, despite participants being randomly assigned to con-

ditions, there was a significant difference between experimental

groups on baseline ruminative thought styles (t = 2.23, p < .05)

with the rumination group reporting higher levels of trait rumi-

native thought than the distraction group (M = 91.06 and 83.24

respectively). Because of this, analyses examining the RTS will

be interpreted with caution.

Correlations were conducted to examine the relationships

between baseline measures prior to the inductions (Table 1).

All correlations were significant and in the expected direction

(Papageorgiou & Siegle, 2003) with GREATER rumination

being significantly related to greater negative mood as meas-

ured by the PANAS-N and the visual analogue scales.

Group comparisons on post induction differences in mood

show the inductions produced a difference in mood between the

groups that was near to significance (t = 1.946, p = .06). Be-

cause of the imbalance in RTS scores across the experimental

groups, with the rumination group having higher RTS scores,

this difference in post-induction mood may be an artifact of

assignment bias due to the high correlation between ruminative

thought styles and negative mood. Hierarchical linear regres-

sion examined the prediction of post induction mood, with

baseline mood entered in the first step, and induction condition

and trait rumination entered into the second step. Only baseline

mood (β = –.223, p = .03) and ruminative thought style (β =

–.507, p = .000) were significant predictors.

Linear regression analyses predicted performance on the six

CPT variables (i.e. correct and false hits for each of the 3 levels

of cognitive load). Baseline mood was entered into the first step,

with RTS score and induction condition entered in the second.

For correct hits, as a measure of attention, RTS score was a

significant predictor at both the high and low cognitive load but

not the medium (Table 2). Mood was a significant predictor of

Table 1.

Baseline correlations.

PANAS-N Mood 1 M SD

RTS .540** –.604** 87.15 17.91

PANAS-N –.444** 20.47 7.05

Mood 1 .673 .208

Note: **p = .01.

Table 2.

Predicting cognitive performance by degree of cognitive load control-

ling for baseline mood.

False Hits (Inhibition) Correct Hits (Attention)

High Load B SE B  B SE B

Step 1

Mood .034.009 .362 –.040 .008–.444**

Step 2

Mood .031.011 .329** –.013 .008–.147

RTS Score .003.004 .085 –.021 .003–.575**

Condition .113.129 .085 –.110 .101–.086

Medium Load

Step 1

Mood .001.001 .036 .000 .007–.002

Step 2

Mood –.001.002 –.038 .004 .009.053

RTS Score .001.001 .115 –.004 .003–.149

Condition –.016.021 –.081 –.165 .105–.161

Low Load

Step 1

Mood .086.012 .600** –.090 .015–.514**

Step 2

Mood .093.014 .649** –.029 .014–.168*

RTS Score –.006.005 –.106 –.044 .006–.644**

Condition –.105.166 –.052 –.011 .171–.005

Note: *p < .05; **p < .01.

226

J. K. BRINKER ET AL.

correct hits, but only when cognitive load was low. For false

hits, as a measure of failed inhibition, only baseline mood was a

significant predictor, but again, only at high or low cognitive

load (Table 2).

In relation to inhibition, it appears that mood alone accounts

for variance in scores at fast and slow speeds. This supports the

previous literature showing that negative mood is related to

cognitive deficits (Austin, 2001). In particular, Koster, De Lis-

snyder, Derakshan and De Raedt (2011) propose the impaired

disengagement hypothesis as a comprehensive explanation of

the neurocognitive basis of these findings. This theory suggests

that individuals encounter some stressor that prompts self-re-

flective/negative thinking.

Some individuals are unable to disengage from this negative

thinking leading to prolonged rumination and subsequently task

impairments and negative mood. This model is circular with

negative mood feeding back into the loop at the point where the

person fails to disengage from the self-reflective/negative

thinking. The current findings linking negative mood to re-

duced ability to inhibit responses, may be one mechanism by

which this happens—these individuals are unable to inhibit the

self-reflective/negative thoughts. Disengagement may be fur-

ther hampered by the person’s inability to maintain attention

elsewhere due to the ruminative style of the self-reflective/

negative thinking.

Scores on the ruminative thought styles scale predicted errors

of attention after controlling for negative mood. It appears that

mood and ruminative processes each relate to a different im-

pairment in abilities needed to disengage from maladaptive

thoughts and engage their attention elsewhere. The common co-

occurrence of rumination and mood may be a particularly pow-

erful vulnerability.

Following the continuous performance task, participants

were asked to report the frequency of off-task thoughts in gen-

eral, off-task thoughts directly related to the induction proce-

dure and frequency of repeating thoughts. The two induction

groups did not significantly differ in the self-reported frequency

of these thoughts. However, the RTS significantly predicted

general off-task thought frequency, even when controlling for

baseline negative mood and induction condition (β = .241, p

= .05). Hierarchical regressions were used to further explore the

predictive ability of ruminative thought styles and off-task

thoughts on task performance.

Baseline mood was entered into the first step, the main ef-

fects of RTS score and frequency of off-task thoughts were

entered into the second step, and the interaction of these two

variables was entered into the third (Table 3). Ruminative

thought style and frequency of off-task thoughts interacted to

significantly predict errors of attention when cognitive load was

high or low. As suggested by previous research, increasing

cognitive demands may improve performance by preventing the

intrusion of off-task thoughts, but there may be a point where

the demands are too great and a decrement in performance is

seen (Lyubomirsky et al., 2003; Teasdale et al., 1995). These

results suggest it is the combination of a ruminative style of

thinking and the experience of off-task thoughts that predicts

impaired performance.

The results of this study are very interesting and speak to the

possible differential effects of a ruminative thought style and

negative mood. The separation of these two variables may be

artificial because of their high co-occurrence, but recognition of

this divide may have important consequences for treatment.

Table 3.

Rumination and off task thoughts predicting attention.

Correct Hits (Attention)

High Load B SE B

Step 1

Baseline Mood –.040 .008 –.444**

Step 2

Baseline Mood .031 .011 .329**

RTS Score .003 .004 .085

Frequency of Off Task Thoughts .113 .129 .085

Step 3

Baseline Mood –.010 .008 –.113

RTS Score –.003 .008 –.081

Frequency Off Task Thoughts .427 .199 .774*

RTS × Frequency Off Task Thoughts –.005 .002 –1.051*

Low Load

Step 1

Mood .086 .012 .600**

Step 2

Mood .093 .014 .649**

RTS Score –.006 .005 –.106

Condition –.105 .166 –.052

Step 3

Baseline Mood –.022 .013 –.126

RTS Score .001 .013 .013

Frequency Off Task Thoughts 1.157 .321 1.085**

RTS × Frequency Off Task Thoughts –.014 .004 –1.454**

Note: *p < .05; **p < .01.

While depressed mood may be ameliorated by treatment, the

lingering ruminative thinking style may remain as the link to

future episodes of depressed mood (Brinker & Dozois, 2009).

Trait rumination predicted frequency of off-task thoughts and

the interaction of these predicted impaired performance in the

absence of depressed mood. This speaks to the importance of

stable ruminative thinking styles occurring in everyday life and

the impact it may have on cognitive processes. However, the

inequity between groups on RTS scores requires confirmation

with a second sample to confidently draw conclusions.

Study 2

The unexpected group differences in RTS scores in the first

sample need to be addressed to confirm the findings from study

1 and provide a clearer understanding of how the constructs of

ruminative thought style, mood and cognitive performance are

related. This second study replicates the first but with the ex-

clusion of the unrelated thoughts measure and the addition of

two new measures of cognitive flexibility—The Wisconsin Cart

Sorting Task, and the Simon Task. The Wisconsin Card Sorting

Task (WCST) is a measure of perseveration—a presentation of

the inability to shift set. This task was designed to assess brain

pathology and therefore may not be sensitive enough to assess

J. K. BRINKER ET AL.

individual differences in a healthy population. Given this po-

tential limitation, the Simon Task, a computer based measure of

inhibition based on the Stroop task, was included having been

successfully used in analogue studies using non-impaired sam-

ples (Simon, 2010).

Method

Participants

One hundred and twenty eight participants completed the

current study. The majority of these participants were first year

psychology students at the Australian National University par-

ticipating as a course requirement. The remaining participants

included senior level psychology student volunteers and volun-

teer community members. The study included 65% females and

35% males, with an age range of 18 - 62 (M = 22.3, SD = 9.54);

58% were Caucasian.

Materials

In addition to the materials repeated from study one, this

study includes.

Wisconsin Card Sorting Task (WCST; Grant & Berg, 1948).

Grant and Berg (1948) designed this task as a measure of set

shifting ability. Participants match response cards at the top of

the screen to one of four stimulus cards at the bottom of the

screen. The response cards were four different possible colours

(blue, red, green and yellow), numbers (1 through 4) and four

shapes (triangle, star, circle and cross). The participant is re-

quired to match the response card to the stimulus card by one of

three rules (colour, number or shape) but is not told the rule.

They must deduce the correct sorting category based on feed-

back of “correct” or “incorrect”. After four correct matches the

rule changes and the participant must then identify the new

matching rule based on further feedback. Participants who per-

severate would not be able to complete as many shifts as those

who are able to shift quickly. The variable of interest for this

study is total number of shifts.

Simon Task. This task was based on a modified Stroop Task

the Stroop task, the Simon task requires the respondent to in-

hibit a dominant response. The computer screen presents a se-

ries of red or green arrows oriented left, right or towards the top

of the screen. Participants are directed to respond to the colour

of the arrow by pressing the “j” key for green, and the “f” key

for red requiring them to inhibit responding to the direction of

the arrow. The incongruent trials are expected to incur the

greatest level of errors (i.e. pressing the j key for green, even

though the arrow is pointing left; Hajcak, McDonald & Simons,

2003). There were a total of 200 trials with half congruent and

half incongruent. The key variable of interest for this study is

number of correct responses in the incongruent trials and the

reaction time for these responses. Reaction time for correct

incongruent trials was also recorded.

Procedure

As with study 1, participants were tested in small groups and

were randomly assigned to either a rumination condition or a

distraction condition. They first completed the baseline ques-

tionnaires followed by the induction and then the 3 cognitive

tasks—CPT, WCST and the Simon task. Because the between

group differences were the key comparison, the cognitive meas-

ures were presented in the same order for every participant. As

with the first study, visual analogue scales assessed the partici-

pant’s current mood between the tasks.

Results and Discussion

For this sample, the two experimental groups did not signifi-

cantly differ on age (t = –.104) or baseline measures—RTS (t =

–1.103), PANAS-N (t = –.165) or PANAS-P (t = 1.22). They

also did not differ in their mood rating on the first visual ana-

logue scale prior to the rumination or distraction induction.

Correlations were completed between the baseline measures

and were as predicted (Table 4).

An examination of the change in mood from before and after

the induction shows that prior to induction, the two groups did

not significantly differ (t = –.737) nor following the induction (t

= –1.776, p = .08) although this did approach significance with

those in the rumination condition showing greater negative

mood. To examine the prediction of post-induction mood, a

hierarchical linear regression was completed.

Baseline mood was entered into the first step and induction

condition and RTS score were entered into the second. Both

baseline mood (β = –.235, p = .00) and RTS score (β = –.630, p

= .00) were significant predictors while induction condition was

not.

Hierarchical regressions examined the prediction of conti-

nuous performance scores. As with Study 1, baseline mood was

entered into the first step and RTS and condition in the second.

Results replicate those found in the first study with ruminative

thought style predicting errors of attention in the slow and fast

blocks (Table 5), and baseline negative mood predicting errors

of inhibition in the slow and fast blocks (Table 5). As before,

induction condition was not a significant predictor.

The same statistical procedures were used for the two new

measures of cognitive flexibility, the WCST and the Simon

Task (Table 6). Ruminative thought styles predicted percentage

correct on incongruent trials in the Simon task, and both rumi-

native thought style and baseline mood predicted reaction time

for these trials. Further, both baseline mood and RTS score sig-

nificantly predicted number of shifts in the WCST.

These findings support those obtained in Study 1 without the

limitation of the random assignment anomaly. A ruminative

thought style is related to impairments of attention and negative

mood with impairments in inhibition in the continuous perfor-

mance task. This suggests that for the continuous performance

measure, mood and ruminative thinking are differentially re-

lated to attention and inhibition. However, this becomes less

clear when we consider the findings for the Simon task and the

WCST.

The Simon task was included as a measure of inhibition,

however unlike the CPT, it is the ruminative thought style that

significantly predicts performance on this task, with both mood

and ruminative thought style predicting reaction time. Further,

Table 4.

Baseline correlations.

PANAS-N Mood 1 M SD

RTS .476** –.749** 89.58 19.62

PANAS-N –.539** 21.39 7.21

Mood 1 .599 .193

Note: **p < .01.

228

J. K. BRINKER ET AL.

Table 5.

Predicting cognitive performance by degree of cognitive load control-

ling for baseline mood.

False Hits (Inhibition) Correct Hits (Attention)

High Cognitive Load B SE B B SE B

Step 1

Baseline Mood .060 .010 .480** –.064 .014–.387**

Step 2

Baseline Mood .056 .011 .451** .001 .010.004

RTS Score .003 .004 .062 –.051 .004–.820**

Condition .008 .145 .004 .041 .126.017

Medium Load

Step 1

Baseline Mood –.018 .008 –.213 –.018 .011–.148

Step 2

Baseline Mood –.012 .008 –.142 –.018 .012–.150

RTS Score –.006 .003 –.175 .001 .004.017

Condition –.177 .109 –.142 .134 .154.078

Low Load

Step 1

Baseline Mood .025 .008 .257** –.069 .015–.387**

Step 2

Baseline Mood .018 .009 .191+ .003 .010.020

RTS Score .005 .004 .137 –.056 .004–.850**

Condition .000 .122 .000 .047 .127.018

Note: *p < .05; **p < .01; +p = .055.

both mood and RTS score predict performance on the WCST.

These tasks may draw upon different, if overlapping abilities in

executive function.

General Discussion

As hypothesised, a ruminative thought style predicted cogni-

tive impairments when cognitive load was low, and the combi-

nation of rumination and off-task thoughts surpassed the main

effect in predictive ability. This same relationship was found

when cognitive load was high. When task demands are low,

cognitive resources remain available for intrusive thoughts

which then detract from performance on the task at hand

(Teasdale et al., 1995) but there comes a point where the task

demands become too great, and the influence of ruminative

thought returns (Lybomirsky, Kasri, & Zehm, 2003). This is of

considerable clinical relevance when treating clients with de-

pressed mood and a tendency to ruminate. Depression can lead

to social withdrawal, and isolation to the point where the indi-

vidual may spend days or even weeks with very little external

stimulation (Watson & Andrews, 2002). In this situation, the

cognitive load would be minimal, leaving room for intrusive

ruminative thoughts to thrive. Watson and Andrews (2002)

suggest that this withdrawal and minimization of external dis-

tractions is an adaptive behaviour to help focus the individual

on finding solutions to their stressors. However, this adaptive-

behaviour may become maladaptive when paired with a dispo

Table 6.

Predicting simon and wisconsin card sort task scores.

Simon Task Incongruent B SE B 

Step 1

Baseline Mood –.143 .176 –.073

Step 2

Baseline Mood .171 .191 .087

RTS Score –.263 .071 –.359**

Induction Condition –3.831 2.453 –.135

Simon Task Reaction Time

Step 1

Baseline Mood .519 .813 .057

Step 2

Baseline Mood 1.895 .892 .209*

RTS Score –1.119 .334 –.332**

Induction Condition –9.933 11.470 –.076

WCST Number of Shifts

Step 1

Baseline Mood –.051 .038 –.117

Step 2

Baseline Mood –.121 .042 –.280**

RTS Score .056 .016 .349**

Induction Condition .230 .540 .037

Note: *p < .05; **p < .01.

sition towards a ruminative style of thinking. At the other end,

numerous stressors, may increase cognitive load to a point

where mood and rumination again reduce the ability to inhibit

intrusive thoughts and to maintain attention elsewhere, such as

on problem solving strategies that might otherwise resolve the

stressors.

Koster, De Lissnyder, Derakshan and De Raedt (2011) pro-

pose the impaired disengagement hypothesis and describe a

circular process of stress, ruminative thinking, negative mood,

further ruminative thinking and so on. If one is predisposed to a

ruminative style of thinking, as a result of or concomitant to an

individual difference in executive function, the experience of

this stressor may indeed trigger this maladaptive cycle, espe-

cially if the stressor consumes considerable cognitive resources,

or compels the individual to withdraw thereby eliminating ex-

ternal demands or distracters. In these two situations, the indi-

vidual would not be able to maintain attention elsewhere such

as identifying strengths and resources or on solution generation.

A key consideration in this process is the apparent differen-

tial influences of mood and ruminative thinking. Intuitively,

rumination would seem to be a combination of inability to in-

hibit ruminative thoughts and the inability to maintain attention

elsewhere. However current findings suggest that negative

mood is the phenomenon related to the inability to inhibit and

this has been supported in previous research (Lau et al., 2007).

Rogers et al. (2004) describe mixed results from numerous

studies examining negative mood and executive dysfunction,

but state that most studies have reported deficits in set-shifting.

Much of the previous research has measured set-shifting using

J. K. BRINKER ET AL.

the Wisconsin Card Sorting Task, which does not easily differ-

entiate between the inhibition of previous set and the maintain-

ing of attention on the new set. Executive functions are a com-

plex set of abilities and reside in a number of brain structures

limiting the ability to examine them in isolation (Rogers et al.,

2004). In reality, attempts to differentiate executive functions

may be as artificial as trying to separate depression and rumina-

tion since the interaction between these functions is inevitable.

However, it does suggest that the amelioration of depressed

mood may not necessarily influence the tendency to ruminate,

and this lingering behaviour, may pose a vulnerability to future

depressive episodes.

Finally, it is important to note that the experimental induc-

tions used in these experiments did not produce differences in

mood or cognitive performance and this makes perfect sense

given the theory on which they were created (Lyubomirsky &

Nolen-Hoeksema, 1995; Nolen-Hoeksema, Parker, & Larson,

1994). In the response style theory, individuals ruminate in re-

sponse to existing depressed mood. Asking participants to

spend time focusing on themselves, how they feel and the pos-

sible causes and consequences of those feelings would not pro-

duce consistent outcomes if those individuals are not all ex-

periencing negative mood. Viewing rumination as a style of

thinking is not in conflict with the response style theory, but

simply a different perspective that examines the process of

rumination. New induction procedures that mimic the process

of rumination, independent of content and mood, are needed to

undertake an experimental examination and to establish causal

direction. Current findings show that a ruminative style of

thinking predicts subsequent performance, and it may be that

rumination impairs executive function. However, it seems more

intuitive that structure leads function, and therefore individual

differences in executive function dispose some individuals to a

ruminative style of thinking. Until it is possible to manipulate

one and to observe changes in the other, the causal direction in

this relationship will remain unknown. This area of research has

many more questions than answers, but in time the relationship

between ruminative thinking, executive function and mood will

be clarified and provide further guidance in treating depressed

mood.

REFERENCES

Antrobus, J. S. (1968). Information theory and stimulus-independent

thought. British Jou rnal of Psychology, 59, 423-430.

doi:10.1111/j.2044-8295.1968.tb01157.x

Austin, M., Mitchell, P., & Goodwin, G. M. (2001). Cognitive deficits

in depression. British Journal of Psychiatry, 178, 200-206.

doi:10.1192/bjp.178.3.200

Brinker, J. K., & Dozois, D. J. A. (2009). Ruminative thought style and

depressed mood. Journal of Clinical Psychology, 65, 1-19.

doi:10.1002/jclp.20542

Carson, S. H., Peterson, J. B., & Higgins, D. M. (2003). Decreased la-

tent inhibition is associated with increased creative achievement in

high functioning individuals. Journal of Personality and Social Psy-

chology, 85, 499-506. doi:10.1037/0022-3514.85.3.499

Davis, R. N., & Nolen-Hoeksema, S. (2000). Cognitive inflexibility among

ruminators and non-ruminators. Cognitive Therapy and Research, 24,

699-711. doi:10.1023/A:1005591412406

Grant, D. A., & Berg, E. A. (1948). A behavioral analysis of reinforce-

ment and ease of shifting to new responses in a Weigel-type card-

sorting problem. Journal of Experimental Psychology, 38, 404-411.

doi:10.1037/h0059831

Hajcak, G., McDonald, N., & Simons, R. F. (2003). To err is autonomic:

Error-related brain potentials, ANS activity, and post-error compen-

satory behavior. Psychophysiology, 40, 895-903.

doi:10.1111/1469-8986.00107

Hertel, P. T. (1997). On the contributions of deficient cognitive control

to memory impairments in depression. Cognition & Emotion, 11,

569-584. doi:10.1080/026999397379890a

Joorman, J., Yoon, K. L., & Zetsche, U. (2007). Cognitive inhibition in

depression. Applied an d Preventative Psychology, 12, 128-139.

doi:10.1016/j.appsy.2007.09.002

Koster, E. H. W., De Lissnyder, E., Derakshan, N., & De Raedt, R.

(2001). Understanding depressive rumination from a cognitive sci-

ence perspective: The impaired disengagement hypothesis. Clinical

Psychology Review, 31, 138-145. doi:10.1016/j.cpr.2010.08.005

Lau, M., Christensen, B. K., Hawley, L. L., Gemar, M. S., & Segal, Z.

V. (2007). Inhibitory deficits for negative information in persons

with major depressive disorder. Psychological Medicine, 37, 1249-

1259. doi:10.1017/S0033291707000530

Lo, B. C. Y., Lau, S., Cheung, S., & Allen, N. B. (2012). The impact of

rumination on internal attention switching. Cognition and Emotion,

26, 209-203. doi:10.1080/02699931.2011.574997

Lyubomirsky, S., Kasri, F., & Zehm, K. (2003). Dysphoric rumination

impairs concentration on academic tasks. Cognitive Therapy and Re-

search, 27, 309-330. doi:10.1023/A:1023918517378

Lyubomirsky, S., & Nolen-Hoeksema, S. (1993). Self-perpetuating pro-

perties of dysphoric rumination. Journal of Personality and Social

Psychology, 65, 339-349. doi:10.1037/0022-3514.65.2.339

Lyubomirsky, S., & Nolen-Hoeksema, S. (1995). Effects of self-focused

rumination on negative thinking and interpersonal problem solving.

Journal of Personality a nd Social Psychology, 69, 176-190.

doi:10.1037/0022-3514.69.1.176

Martin, L. L., & Tesser, A. (1996). Some ruminative thoughts. In R. S.

Wyer Jr. (Ed.), Ruminative thoughts (pp. 1-47). Hillsdale, NJ: Erl-

baum.

Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter,

A., & Wager, T. D. (2000). The unity and diversity of executive

functions and their contribution to complex “frontal-lobe” tasks: A

latent variable analysis. Cognitive Psycholo gy, 41, 49-100.

doi:10.1006/cogp.1999.0734

Mor, N., & Winquist, J. (2002). Self-focused attention and negative

affect: A meta-analysis. Psychological Bulletin, 128, 638-662.

doi:10.1037/0033-2909.128.4.638

Nolen-Hoeksema, S. Parker, L., & Larson, J. (1994). Ruminative cop-

ing with depressed mood following loss. Journal of Personality and

Social Psychology, 67, 92-104. doi:10.1037/0022-3514.67.1.92

Nolen-Hoeksema, S., Wisco, B. E., & Lyubomirsky, S. (2008). Re-

thinking rumination. Perspectives on Psychological Science, 3, 400-

424. doi:10.1111/j.1745-6924.2008.00088.x

Papageorgiou, C., & Siegle, G. J. (2003). Rumination and depression:

Advances in theory and research. Cognitive Therapy and Research,

27, 243-245. doi:10.1023/A:1023918331490

Rimes, K. A., & Watkins, E. (2005). The effects of self-focused rumi-

nation on global negative self-judgements in depression. Behaviour

Research and Therapy, 43, 1673-1681.

doi:10.1016/j.brat.2004.12.002

Rogers, M. A., Kasai, K., Matsuo, K., Fukuda, R., Iwanami, A., Naka-

gome, K., Fukuda, M., & Kato, N. (2004). Executive and prefrontal

dysfuction in unipolar depression: A review of neuropsychological

and imaging evidence. Neuroscience Research, 50, 1-11.

doi:10.1016/j.neures.2004.05.003

Teasdale, J. D., Dritschel, B. H., Taylor, M. J., Proctor, L., Lloyd, C. A.,

Nimmo-Smith, I., & Baddeley, A. D. (1995). Stimulus-independent

thought depends on central executive resources. Memory & Cogni-

tion, 23, 551-559. doi:10.3758/BF03197257

Verhaeghen, P., Joorman, J., & Khan, R. (2005). Why we sing the blues:

The relation between self-reflective rumination, mood, and creativity.

Emotion, 5, 226-232. doi:10.1037/1528-3542.5.2.226

Watkins, E., & Brown, R. G. (2002). Rumination and executive func-

tion in depression: An experimental study. Journal of Neurology,

Neurosurgery and Psychi atr y, 72, 400-402.

doi:10.1136/jnnp.72.3.400

Watkins, E., & Moulds, M., (2005). Distinct modes of ruminative self-

focus: Impact of abstract versus concrete rumination on problem solv-

230

J. K. BRINKER ET AL.

ing in depression. Emotion , 5, 319-328.

doi:10.1037/1528-3542.5.3.319

Watson, D., & Clark, L. A. (1994). Manual for the positive and nega-

tive affect schedule-expanded form. Iowa City, IA: The University of

Iowa. http://www.psychology.uiowa.edu/faculty/clark/panas-x.pdf

Watson, P. J., & Andrews, P. W. (2002). Toward a revised evolutionary

adaptionist analysis of depression: The social navigation hypothesis.

Journal of Affective Disor d ers, 72, 1-14.

doi:10.1016/S0165-0327(01)00459-1

Whitmer, A. J., & Banich, M. T. (2007). Inhibition versus switching

deficits in different forms of rumination. Psychological Science, 18,

546-553. doi:10.1111/j.1467-9280.2007.01936.x

Williams, P. G., Suchy, Y., & Rau, H. K. (2009) Individual differences

in executive functioning: Implications for stress regulation. Annals of

Behavioral Medicine, 37, 126-140. doi:10.1007/s12160-009-9100-0