COGNITION AND EMOTION, 2017

VOL. 31, NO. 5, 1023–1030

http://dx.doi.org/10.1080/02699931.2016.1179174

BRIEF ARTICLE

Understanding the role of mind wandering in stress-related working

memory impairments
Jonathan B. Banksa and Adriel Boalsb
a
College of Psychology, Nova Southeastern University, Fort Lauderdale, FL, USA; bDepartment of Psychology, University of North
Texas, Denton, TX, USAD

ABSTRACT ARTICLE HISTORY

Mind wandering has been identified as a possible cause for stress-related working Received 15 July 2015
memory (WM) task impairments following laboratory stressors. The current study Revised 2 March 2016
attempted to induce mind wandering regarding negative, positive, or neutral Accepted 10 April 2016
events using an expressive writing task and examined the impact on WM task
KEYWORDS
performance. We examined the role of mind wandering in understanding the Working memory; mind
impact of life stress on WM. Additionally, we explored the role of thought wandering; stress; thought
suppression on the relationship between mind wandering and WM. One hundred suppression
and fifty participants completed WM measures before (Time 1) and after (Time 2)
the writing manipulation. The writing manipulation did not alter mind wandering or
WM task performance. Time 1 WM predicted mind wandering during the Time 2
WM task, which subsequently predicted poorer Time 2 WM task performance. The
impact of daily life stress on WM was mediated by mind wandering. Trait levels of
thought suppression moderated the impact of mind wandering on WM. Specifically,
higher levels of suppression resulted in stronger negative impact of mind
wandering on WM task performance. Findings are discussed in terms of the impact
of mind wandering on WM task performance.

Understanding factors that influence state fluctuations
in working memory (WM) performance is critical given
the stable nature of WM and its role in higher order
cognitive functions (Ilkowska & Engle, 2010). Psycho-
logical stress from laboratory manipulations
(Schoofs, Preuss, & Wolf, 2008) and life event stress
(Klein & Boals, 2001; Sliwinski, Smyth, Hofer, &
Stawski, 2006) result in decrements in WM perform-
ance. These decrements may be a result of mind wan-
dering (Curci, Lanciano, Soleti, & Rimé, 2013; Klein &
Boals, 2001). Mind wandering refers to thoughts
about personal concerns or goals unrelated to the
current task (Smallwood & Schooler, 2006). These
task-unrelated thoughts (TUTs) have been shown to
impair performance on lab-based cognitive tasks
(McVay & Kane, 2009) and self-rated performance on
everyday tasks (McVay, Kane, & Kwapil, 2009).
WM has been proposed to be a critical factor for
explaining how and why mind wandering occurs. The
control failures × personal concerns model (McVay &
Kane, 2010) suggests mind wandering reflects an inter-
action between personal concerns triggered by the
environment and failure of WM to control attention.
Mind wandering can be triggered by priming partici-
pants with sufficient personal concerns such that WM
fails to prevent mind wandering from occurring
(McVay & Kane, 2013). Importantly, this model suggests
the extent to which mind wandering occurs is con-
trolled by WM. Smallwood and Schooler’s (2006) execu-
tive resource model suggests WM explains how mind
wandering is supported after it occurs, such that WM
resources enable mind wandering. In this view higher
WM individuals may engage in a task and mind
wander concurrently given sufficient available
resources. However, lower WM capacity individuals
lack sufficient resources to simultaneously support
mind wandering and successful task performance.
Mind wandering may be facilitated by WM in a

CONTACT Jonathan B. Banks jonathan.banks@nova.edu

© 2016 Informa UK Limited, trading as Taylor & Francis Group
1024 J. B. BANKS AND A. BOALS
context-dependent manner such that higher WM
capacity individuals are more likely to engage in
mind wandering than lower WM individuals on tasks
that place a low demand on WM resources, such as
low-perceptual load visual search tasks, due to excess
WM resources (Levinson, Smallwood, & Davidson,
2012). Lower WM capacity individuals may lack suffi-
cient resources, even on low-demand tasks, to simul-
taneously mind wandering and complete the task.
Although models differ in predictions of when
mind wandering will impair task performance, all
agree impairments are likely to occur on tasks
placing large demands on attentional resources,
including WM tasks. Randall, Oswald, and Beier
(2014) demonstrated that higher WM capacity was
associated with fewer TUTs, but occurrences of TUTs
impair ongoing task performance, with greater impair-
ments on more complex tasks. Multiple studies have
demonstrated impairments in WM due to mind wan-
dering induced by negative stressful life events
(Klein & Boals, 2001), state anxiety (Banks, Tartar, &
Tamayo, 2015), and negative affect (Curci et al.,
2013). On resource demanding tasks, including the
WM tasks in these studies, the occurrence of mind
wandering was sufficient to reduce performance.
When someone experiences mind wandering, they
may attempt to suppress such thoughts to remain
focused on the relevant task. However, attempts to
suppress unwanted thoughts consume WM resources,
leading to further impairments in relevant task per-
formance (Klein & Bratton, 2007; Wegner, 1994).
Therefore, suppression or attempts to suppress TUTs
may be partially responsible for mind wandering-
related impairments, rather than mind wandering
itself. Hence examining the impact of thought sup-
pression may further our understanding of the conse-
quences of mind wandering. If performance
impairments are due to suppression, rather than
mind wandering, it would suggest that mind wander-
ing is not supported by WM resources, as Smallwood
and Schooler (2006) propose.
The first aim of the current study was to manipulate
rates of TUTs and WM task performance via a writing
task requiring participants to write about either a
future negative, future positive, or a neutral life
event. The writing task was designed to induce TUTs
by priming personal goals. WM performance and
TUTs were measured during an initial assessment
(Time 1) and one week later, following the writing
task (Time 2). Our first hypothesis is that participants
who write about a future negative, but not future
positive or neutral events, will experience increased
TUTs and subsequent decreases in WM task perform-
ance. We hypothesised that the positive writing task
would not result in these changes due to the greater
ability of negative information to capture attention
(Pratto & John, 1991).
The second aim of the study was to examine the
role of WM capacity in controlling mind wandering.
Consistent with the control failures × personal concerns
model (McVay & Kane, 2010), our second hypothesis
was that TUT rates would be negatively related to
WM task performance and Time 1 WM would predict
rates of TUTs at Time 2. Although Time 1 WM perform-
ance will be influenced by mind wandering (Unsworth
& Robison, 2016), this initial measurement, prior to the
writing task, can be used as a predictor for subsequent
TUT percentage and WM performance. Our third
hypothesis was that trait levels of suppression
attempts would moderate the impact of TUTs on
WM. Specifically, higher levels of suppression attempts
would result in poorer WM performance.
A third aim of the current study was to examine the
relationship between self-report measures of stress
and WM. Participants completed self-report measures
of experienced daily stressors, life stress, and intrusive
and avoidant thinking. Our fourth hypothesis was that
TUTs would mediate the negative relationship
between self-report measures of stress and intrusive
and avoidant thoughts, during the seven days prior
to Time 2, and WM.
Methods
Participants
One hundred and fifty undergraduates (84 females; M
age = 21.28 years) from the University of North Texas
participated for partial course credit. Participants
were randomly assigned to the negative event (n =
50), positive event (n = 50), or control condition (n =
50). Twenty-one participants did not return to com-
plete Time 2 (85% retention rate). No differences
were found on Time 1 measures between participants
who did and did not return for Time 2. Three partici-
pants failed to fully complete the questionnaire
measures during Time 2, resulting in a final Time 2
sample size of N = 126 (negative event condition n =
45, positive event condition n = 41, and control con-
dition n = 40). Sample size was determined using
power analyses estimating sample size for the
writing manipulation and effect of stress on WM,

with significance level set at p = .05 and desired
power = .95. Smallest reported effects for the writing
manipulation on WM and TUTs and stress on WM
(Curci et al., 2013; partial η 2 = .04; Klein & Boals,
2001; r’s= −.36 to −.46, respectively) were used to esti-
mate required sample size. Required sample size
ranged from N = 94, for the effect of stress, to N = 96,
for the writing manipulation. Sample size was
chosen to exceed estimated requirements. A sample
size of n = 40, per group, provides a power = .94 for
effects as small as η 2 = .03.
Measures
Working memory. Participants completed the Auto-
mated Operation Span Task (AOSPAN) and the Auto-
mated Reading Span Task (RSPAN; Unsworth, Heitz,
Schrock, & Engle, 2005). Participants maintained a
series of letters in memory while completing a second-
ary task. The AOSPAN task requires participants to
verify the correctness of the solution to a simple math-
ematical equation ((3 × 2) + 1 = 6). The RSPAN task
requires participants to verify the meaningfulness of
sentences (“The ship sailed across the dishwasher”).
In both tasks, a capital letter (out of 12 possible
letters) appears for 250 ms, 200 ms after the reading
or operation verification response. After a set of 3–7
verification-letter pairs, a grid containing all 12 poss-
ible letters was presented. Participants recalled all
letters from the set in the order presented. Each set
length (3–7 items) was presented 3 times for a total
of 15 sets per task. WM tasks were scored by
summing the total number of items recalled in the
correct serial position (Conway et al., 2005).
Mind wandering. Mind wandering was measured
during the WM tasks, using two measures, probe-
caught TUTs and self-reported percentage of TUTs.
Thought probes (McVay & Kane, 2009) measured
what participants were thinking prior to the probe
(probe-caught TUTs). Participants were asked, “What
were you just thinking about” and selected one of
five options: (a) Task-related thought, exclusively, (b)
Task performance/evaluative thoughts, (c) Task-unre-
lated, neutral content, (d) Task-unrelated, positive
content, (e) Task-unrelated, negative content. During
Time 2, a sixth response option was added to allow
individuals to select an option about the writing
manipulation: (f) Task-unrelated, about the event I
wrote about. Thought probes were inserted into WM
tasks following recall of the letters at the end of
each set. Thought probes were classified as being
COGNITION AND EMOTION 1025
on-task (option a), task-related interference (option
b), or off-task (options c–f). Percentage of probe-
caught TUTs was calculated by summing the
number of off-task response options selected and
dividing by the number of probes.
Following thought probe response, participants
indicated what percentage of their thoughts was off-
task during the previous set (between 0% and
100%). Percentage of self-report TUTs was calculated
by averaging the self-reported off-task thought per-
centages. Fifteen probes and self-report off-task
thought percentage questions were inserted into
each WM task, resulting in a total of 30 probes and
self-report TUT percentages during the 2 tasks.
Daily Inventory of Stressful Events (Almeida,
Wethington, & Kessler, 2002). The Daily Inventory of
Stressful Events (DISE) is a seven-item scale measuring
the number of daily stressors experienced. Partici-
pants indicated if an event had occurred in the last
24 hours. The DISE is scored by summing the
number of items endorsed, creating a possible range
from 0 to 7.
Life Experiences Scale (Sarason, Johnson, & Siegel,
1978). The Life Experiences Scale (LES) is a 47-item
checklist measuring the amount of life stress partici-
pants have experienced. Participants indicate if the
event occurred in the last six months. If it occurred,
they rated the direction (negative or positive) and
degree of the impact on a scale from 0 (no impact)
to 3 (extremely positive/negative). Scores were com-
puted by summing impact ratings of negatively
rated events.
Impact of Events Scale (Horowitz, Wilner, & Alvarez,
1979). The Impact of Events Scale (IES) is a 15-item
questionnaire measuring the frequency of avoidant
and intrusive thoughts experienced during the last
seven days about a nominated event. Participants
respond to statements (e.g. “I thought about it when
I didn’t mean to”) on a scale from 1 (not at all) to 5
(extremely). Items were summed to produce an
overall intrusion and avoidance score.
White Bear Suppression Inventory (Wegner &
Zanakos, 1994). The White Bear Suppression Inventory
(WBSI) is a 15-item questionnaire measuring the
degree to which individuals are likely to engage
in thought suppression or experience intrusive
thoughts. Participants respond to items (e.g. “I
always try to put problems out of my mind”) on a
scale from 1 (strongly disagree) to 5 (strongly agree).
Items are summed so that higher scores indicate
greater suppression.
1026 J. B. BANKS AND A. BOALS
Procedure
During Time 1 participants completed the AOSPAN
and RSPAN in a counterbalanced order, with thought
probes and self-reported off-task thought percentage
inserted into each task. Participants then completed
the DISE and WBSI. Time 2 occurred 2–10 days follow-
ing Time 1. Control condition participants wrote for 2
min about the prior days activities in a non-emotional
manner.
Negative event condition participants nominated
their worst fear, such as a fear of spiders, snakes, or
heights. Positive event condition participants nomi-
nated a possible future event with the best possible
outcome. Participants in the negative and positive
event conditions then imagined an event in which
they would have to deal with their nominated event.
Participants engaged in an expressive writing task
for 2 min in which they imagined how they might
feel experiencing the event and described the experi-
ence in detail.
Participants then completed the AOSPAN and
RSPAN in the same order as Time 1. Thought probes
and self-reported off-task thought percentages were
inserted into both WM tasks. Finally, participants com-
pleted the LES, IES, DISE, and a demographics
questionnaire.
Results
All measures were examined for outliers. Any values
beyond three standard deviations were replaced
with a three standard deviation value. This resulted
in replacement of six values on the AOSPAN at Time
1, six values on the AOSPAN at Time 2, two values
on the RSPAN at Time 1, and three values on the
RSPAN at Time 2. A composite WM score was
created by averaging z-scores of the AOSPAN and
RSPAN scores following outlier transformations. A
composite mind wandering score, TUT composite,
was created by averaging z-scores of probe-caught
TUTs and self-report TUT percentage.
Effects of writing task
To test the first hypothesis that writing about a
future negative event, but not positive or neutral
events, will increase TUTs and decrease WM perform-
ance, two mixed-model ANOVAs were conducted.
No effects of condition, time, or time by condition
interactions were found for the WM composite score,
all p’s > .05. A significant effect of time was found on
the TUT composite, F (1, 123) = 24.66, p < .0001,
partial η 2 = 0.16, such that TUTs increased from Time
1 (M = 0.00, SD = 0.84) to Time 2 (M = 0.37, SD = 0.98),
but no effect of condition or time by condition inter-
action, p’s > .05. Based on the lack of significant differ-
ences between conditions on TUTs or WM, all three
conditions were collapsed for subsequent analyses.
WM, mind wandering, and suppression
To test the second hypothesis that TUTs would be
negatively related to WM performance and WM per-
formance would predict TUTs on a subsequent task,
we conducted a correlational analysis. Consistent
with our hypothesis, WM was negatively related to
TUTs at Time 1, r (146) = −.55, p < .0001, and Time 2,
r (125) = −.42, p < .0001. Not surprisingly, WM at
Time 1 was strongly related to WM at Time 2, r (125)
= .77, p < .0001. To determine if the relationship
between WM at Time 1 and Time 2 was mediated by
TUTs during Time 2, a mediation analysis was con-
ducted using the process macro for SAS (Hayes,
2013). The method provides a bootstrapped 95% con-
fidence interval (CI) around the indirect effect. The
indirect effect is interpreted as significant when the
CI does not contain zero, indicating the mediation
effect is different from zero (Preacher & Hayes,
2008). As given in Table 1, a significant indirect
effect was observed, such that Time 2 TUTs mediated
the relationship between WM at Time 1 and Time
2. Due to the possibility that the relationship
between WM at Time 1 and Time 2 was due to the
influence of TUTs on performance at both time
points, TUTs at Time 1 were included as a covariate
in the model. The mediation model remained signifi-
cant, indirect effect = 0.04, bootstrapped SE = 0.02,
bootstrapped 95% CI [0.002, 0.11].
To test the third hypothesis that the tendency to
suppress thoughts should moderate the impact of
TUTs on WM, a regression analysis was conducted.
Time 1 TUTs, WBSI score, and the interaction
between WBSI and TUTs were included as predictors
of Time 1 WM. The overall model was significant,
R 2 = .34, F(3, 141) = 24.45, p < .0001, with the inter-
action between WBSI and TUTs serving as the only sig-
nificant predictor, B = −0.014, t = −2.03, p < .05. Simple
slopes analyses examined the relationship between
Time 1 TUTs and Time 1 WM at three levels of the
WBSI – the mean and ±1 standard deviation from
the mean. Analyses revealed a significant negative
 COGNITION AND EMOTION 1027

Table 1. Mediation analysis results.

Independent Mediator Dependent Effect of IV Effect of M to Total effect Direct effect Indirect 95% CI for Kappa-
variable (IV) (M) variable (DV) on M (a) DV (b) (c) (c′ ) effect indirect effect squared
WM T1 TUTs T2 WM T2 −0.43**** −0.14* 0.83**** 0.77**** 0.06 0.2, 0.14 0.09
DISE T1 TUTs T1 WM T1 0.08* −0.64**** −0.01 0.04 −0.05 −0.11, −0.01 0.11
DISE T2 TUTs T2 WM T2 0.03 −0.43**** 0.06 0.08+
−0.02 −0.06, 0.03 0.03
IES TUTs T2 WM T2 0.02** −0.44**** −0.004 0.005 −0.01 −0.02, −0.002 0.12
LES TUT T2 WM T2 0.01 −0.41**** −0.01 −0.01 −0.005 −0.02, 0.002 0.05
Note: Values represent unstandardised path coefficients; +p < .07, *p < .05, **p < .01, ***p < .001, ****p < .0001, Time 1 analysis N = 145, Time 2
analysis N= 126, bootstrap value = 5000 samples for all analyses. Bold values represent the models with significant indirect effects.

relationship between TUTs and WM at lower levels of
WBSI, B = −0.49, t = −7.74, p < .0001, mean levels of
WBSI, B = −0.62, t = −8.27, p < .0001, and higher
levels of WBSI, B = −0.76, t = −7.59, p < .0001, such
that the relationship between TUTs and WM increased
as WBSI increased.
Self-report stress, mind wandering, and WM
stress on WM. Additionally, we examined the role of
thought suppression in altering the impact of mind
wandering on WM performance. The first aim of the
study was to manipulate psychological stress using a
writing task. The writing task did not alter WM task
performance or mind wandering. TUTs increased
across time points in all conditions, possibly due to a
decrease in task novelty (Randall et al., 2014).

To test the fourth hypothesis that self-report measures

of stress and intrusive and avoidant thoughts will have
a negative impact on WM task performance, and this
relationship is mediated by TUTs, a series of mediation
analyses were conducted examining the self-report
stress measures and WM and TUT measures at the cor-
responding time. As seen in Table 1, a significant
mediation was found such that higher scores on the
DISE during Time 1 lead to higher rates of TUTs, which
lead to lower Time 1 WM scores. No significant relation-
ship was found with the Time 2 measures of DISE, WM,
and TUTs. Likewise, TUTs did not mediate the relation-
ship between LES and WM at Time 2. However, TUTs
mediated the relationship between IES and WM, such
that scores on the IES predicted TUTs and TUTs sub-
sequently predicted poorer WM task performance.
Despite the lack of a significant mediation effect for
the LES on WM performance through TUTs, we hypoth-
esised LES scores would predict IES scores, which in
turn would predict TUTs, resulting in poorer WM per-
formance. We predicted two mediators operating in
succession, IES and TUTs. As seen in Figure 1, a signifi-
cant indirect effect was found, indirect effect= −0.004,
bootstrapped SE = 0.002, bootstrapped 95% CI [−0.01,
−0.001]. Higher LES scores predicted higher IES scores,
which predicted higher rates of TUTs, which predicted
lower WM performance.
Discussion
The current study examined the role of mind wander-
ing as a mediator for the impact of psychological
WM, mind wandering, and suppression
The second aim of the current study was to examine
the relationship between mind wandering and WM.
Although the writing manipulation did not alter WM
task performance or TUT percentage, the results
provide support for the control failures × personal con-
cerns model (McVay & Kane, 2010). Time 1 WM pre-
dicted TUTs at Time 2 and TUTs at Time 2 were
negatively related to WM at Time 2, even when con-
trolling for TUTs at Time 1. This supports the notion
that WM is responsible for preventing the occurrence
of mind wandering but when mind wandering does
occur, WM performance is impaired. However, we
must be cautious in our interpretation since our
Time 1 WM measure cannot be seen as a true WM
capacity measure due to concerns about TUTs impact-
ing WM task performance, thus altering WM capacity
estimates (Unsworth & Robison, 2016), and the
inclusion of thought probes in the Time 1 WM
measure.
Further, control over mind wandering may be
context dependent such that it varies based on the
resources required for the ongoing task. Higher WM
capacity individuals report more frequent TUTs
during a low-perceptual load visual search task than
lower WM capacity individuals, but WM did not
predict TUTs during a high-perceptual load visual
search task (Levinson et al., 2012). Consistent with
alternative views (McVay & Kane, 2009), our results
suggest that during tasks requiring greater attentional
demands, higher WM individuals reported fewer TUTs
1028 J. B. BANKS AND A. BOALS
Figure 1. Mediated mediation model.
Note: Values represent unstandardised path coefficients. The unstandardised regression coefficient for the relationship between LES and WM at
Time 2, prior to controlling for the indirect paths (c) is presented in parentheses. c′ represents the direct path after controlling for the indirect
path. T2 = Time 2. *p < .05, **p < .01, ***p < .001, ****p < .0001, N= 126.
that lower WM individuals. However, it is important to
point out that the visual task used by Levinson differs
in nature and difficulty than the task used in the
current study. Thus comparisons between the two
studies have to be considered in light of these meth-
odological differences.
Following the work of Wegner (1994), it was
expected that suppression of TUTs will negatively
impact task performance. Although directed suppres-
sion of thoughts has been linked with a rebound of to-
be-suppressed thoughts (Wegner, 1994) and
decreases in cognitive task performance (Klein &
Bratton, 2007), our findings suggest that trait levels
of thought suppression moderate the impact of
TUTs on WM performance. WM performance impair-
ments are greatest when individuals exert cognitive
control over TUTs, to return to the current task. The
suppression of mind wandering is a resource demand-
ing process (Wegner, 1994), requiring WM resources to
suppress unwanted thoughts. Although trait level of
thought suppression moderated the impact of TUTs
on WM task performance, TUTs remained predictive
of poorer WM performance at low levels of suppres-
sion, suggesting that suppression alone is not respon-
sible for impaired performance. However, we
examined suppression at a trait level rather than in
response to TUTs during the task. Future work manip-
ulating suppression of TUTs may provide a better
understanding of the role of suppression on the con-
sequences of mind wandering.
Stress, WM, and mind wandering
The third aim of the study was to examine mind wan-
dering as a factor in stress-related WM performance
decrements. The impact of daily stressors on WM
was mediated by TUTs during Time 1, but not during
Time 2. This partially supported our hypothesis that
life stressors increase mind wandering, thereby
decreasing WM performance. It is possible our Time
2 findings differed because increased TUTs between
sessions altered the relationship between these con-
structs. Klein and Boals (2001) suggested that the
impact of life stress on WM was due to intrusive
thoughts, as measured in their study with the IES.
Our results did not show a direct relationship
between the IES or LES and WM, or that IES mediated
the relationship between LES and WM performance.
However, as mentioned by Klein and Boals, IES
scores may not be indicative of thoughts occurring
during the WM tasks. Our results demonstrated the
impact of the IES on WM was mediated by TUTs
during the task. These results must be interpreted cau-
tiously due the correlational nature of the analyses
and the order in which the tasks were completed.
The DISE, IES, and LES were completed after the WM
tasks at Time 2. Mediation analyses require the
outcome variable (WM performance) to occur after
all other variables. However, the DISE, IES, and LES
are retrospective in nature, asking participants to
report events that occurred prior to the Time 2
session. Despite being measured after the WM tasks,
participants reported events that occurred prior to
the WM tasks. Thus examining the impact of these
measures on WM performance is reasonable.
These results appear to add support for the role of
mind wandering in cognitive impairments observed
as a result of psychological stress. Given the mixed
results with the DISE the results need to be interpreted
cautiously. Further, the lack of a direct effect of the LES

on WM performance suggests experiencing negative
life events does not automatically result in mind wan-
dering about the event during ongoing tasks.
However, when negative life events lead to thoughts
about the event, increases in TUTs occur and impair
task performance.
Limitations
One limitation in the current study was the lack of an
effect of the writing manipulation. This may be due to
the short duration of the writing task, such that partici-
pants may not have thought about the event enough
to induce sufficient personal concerns to trigger TUTs
(McVay & Kane, 2010). Alternatively, thinking about a
future event may be more difficult than thinking
about a prior event. Thus, it may not trigger sufficient
personal concerns to induce TUTs. A second limitation
in the present study is the use of only two measures of
WM. Many studies examining mind wandering and
WM use three tasks to provide a clear estimate of
WM capacity (see McVay & Kane, 2009). Future
research should replicate the current findings with a
larger number of WM measures.
Conclusion
The current findings present evidence for the role of mind
wandering in determining the impact of stress on WM
task performance. The impact of multiple indices of life
stress, including daily stressors and larger life stressors,
appears to decrease WM task performance due to
increases in mind wandering during the task. New to
the current study, suppression moderates the impact of
mind wandering on WM such that higher levels of sup-
pression attempts are related with a stronger impact of
mind wandering on WM. Our hope is that the results of
the current study increase our understanding of the
impact of thought suppression, which is critical to the cre-
ation of a unified model of mind wandering.
Disclosure statement
No potential conflict of interest was reported by the authors.
References
Almeida, D., Wethington, E., & Kessler, R. (2002). The Daily
Inventory of Stressful Events: An interview-based approach
for measuring daily stressors. Assessment, 9, 41–55.
COGNITION AND EMOTION 1029
Banks, J. B., Tartar, J. L., & Tamayo, B. A. (2015). Examining factors
involved in stress related working memory impairments:
Independent or conditional effects? Emotion, 15(6), 827–836.
Conway, A. R. A., Kane, M. J., Bunting, M. F., Hambrick, D. Z., Wilhelm,
O., & Engle, R. W. (2005). Working memory span tasks: A meth-
odological review and user’s guide. Psychonomic Bulletin and
Review, 12, 769–786. doi:10.3758/BF03196772
Curci, A., Lanciano, T., Soleti, E., & Rimé, B. (2013). Negative emotional
experiences arouse rumination and affect working memory
capacity. Emotion, 13(5), 867–880. doi:10.1037/a0032492
Hayes, A. F. (2013). Introduction to mediation, moderation, and
conditional process analysis: A regression-based approach.
New York, NY: Guilford Press.
Horowitz, M. J., Wilner, N. R., & Alvarez, W. (1979). Impact of event
scale. A measure of subjective stress. Psychosomatic Medicine,
41, 209–218.
Ilkowska, M., & Engle, R. W. (2010). Trait and state differences in
working memory capacity. In A. Gruszka, G. Matthews, & B.
Szymura (Eds.), Handbook of individual differences in cognition.
Attention, memory, and executive control (pp. 295–320).
New York, NY: Springer.
Klein, K., & Boals, A. (2001). The relationship of life event stress
and working memory capacity. Applied Cognitive Psychology,
15, 565–579. doi:10.1002/acp.727
Klein, K., & Bratton, K. (2007). The cost of suppressing stressful
memories. Cognition and Emotion, 21, 1496–1512.
Levinson, D. B., Smallwood, J., & Davidson, R. J. (2012). The per-
sistence of thought: Evidence for a role of working memory
in the maintenance of task-unrelated thinking. Psychological
Science, 23(4), 375–380.
McVay, J. C., & Kane, M. J. (2009). Conducting the train of thought:
Working memory capacity, goal neglect, and mind wandering
in an executive-control. Journal of Experimental Psychology, 35,
196–204. doi:10.1037/a0014104
McVay, J. C., & Kane, M. J. (2010). Does mind wandering reflect
executive function or executive failure? comment on
Smallwood and Schooler (2006) and Watkins (2008).
Psychological Bulletin, 136, 188–197. doi:10.1037/a0018298
McVay, J. C., & Kane, M. J. (2013). Dispatching the wandering
mind? Toward a laboratory method for cuing “spontaneous”
off-task thought. Frontiers in Psychology, 4, 570. doi:10.3389/
fpsyg.2013.00570
McVay, J. C., Kane, M. J., & Kwapil, T.R. (2009). Tracking the train of
thought from the laboratory into everyday life: An experience-
sampling study of mind wandering across controlled and eco-
logical contexts. Psychonomic Bulletin & Review, 16, 857–863.
doi:10.3758/PBR.16.5.857
Pratto, F., & John, O. P. (1991). Automatic vigilance: The attention-
grabbing power of negative social information. Journal of
Personality and Social Psychology, 61, 380–391.
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling
strategies for assessing and comparing indirect effects in mul-
tiple mediator models. Behavior Research Methods, 40, 879–
891. doi:10.3758/BRM.40.3.879
Randall, J. G., Oswald, F. L., & Beier, M. E. (2014). Mind-wandering,
cognition, and performance: A theory-driven meta-analysis of
attention regulation. Psychological Bulletin, 140(6), 1411–1431.
doi:10.1037/a0037428
Sarason, I., Johnson, J., & Siegel, J. (1978). Assessing the impact of
life changes: Development of the life experiences survey.
Journal of Consulting and Clinical Psychology, 46, 932–946.
1030 J. B. BANKS AND A. BOALS
Schoofs, D., Preuss, D., & Wolf, O.T. (2008). Psychosocial
stress induces working memory impairments in
an n-back paradigm. Psychoneuroendocrinology, 33(5),
643–653.
Sliwinski, M., Smyth, J., Hofer, S., & Stawski, R. (2006).
Intraindividual coupling of daily stress and cognition.
Psychology and Aging, 21, 545–557. doi:10.1037/0882-7974.
21.3.545
Smallwood, J., & Schooler, J. W. (2006). The restless mind.
Psychological Bulletin, 132(6), 946–958.
Unsworth, N., Heitz, R., Schrock, J., & Engle, R. (2005). An auto-
mated version of the operation span task. Behavior Research
Methods, 37, 498–505. doi:10.3758/BF03192720
Unsworth, N., & Robison, M. K. (2016). The influence of lapses of
attention on working memory capacity. Memory and
Cognition, 44, 188–196.
Wegner, D. M. (1994). Ironic processes of mental control.
Psychological Review, 101(1), 34–52.
Wegner, D. M., & Zanakos, S. (1994). Chronic thought suppres-
sion. Journal of Personality, 62, 615–640.
Copyright of Cognition & Emotion is the property of Taylor & Francis Ltd and its content
may not be copied or emailed to multiple sites or posted to a listserv without the copyright
holder's express written permission. However, users may print, download, or email articles for
individual use.