THE ROLE OF HAND OF ERROR AND STIMULUS ORIENTATION IN THE
RELATIONSHP BETWEEN WORRY AND ERROR-RELATED BRAIN ACTIVITY:
IMPLICATIONS FOR THEORY AND MEASUREMENT
By
Yanli Lin

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Psychology – Master of Arts
2015

ABSTRACT
THE ROLE OF HAND OF ERROR AND STIMULUS ORIENTATION IN THE
RELATIONSHP BETWEEN WORRY AND ERROR-RELATED BRAIN ACTIVITY:
IMPLICATIONS FOR THEORY AND MEASUREMENT
By
Yanli Lin
Anxious apprehension/worry is associated with exaggerated error monitoring; however, the
precise mechanisms underlying this relationship remain unclear. The current study tested the
hypothesis that the error monitoring-worry relationship involves left-lateralized linguistic brain
activity by examining the relationship between worry and error monitoring, indexed by the errorrelated negativity (ERN), as a function of hand of error (Experiment 1) and stimulus orientation
(Experiment 2). Results revealed that worry was exclusively related to the ERN on right-handed
errors committed by the linguistically dominant left hemisphere. Moreover, the right-hand ERNworry relationship emerged only when stimuli were presented horizontally (known to activate
verbal processes) but not vertically. Together, these findings suggest that the worry-ERN
relationship involves left hemisphere verbal processing, elucidating a potential mechanism to
explain error monitoring abnormalities in anxiety. Implications for theory and measurement are
discussed.

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................................... iv
LIST OF FIGURES .........................................................................................................................v
INTRODUCTION ...........................................................................................................................1
EXPERIMENT 1 METHOD ...........................................................................................................7
Participants ...........................................................................................................................7
Task ......................................................................................................................................7
Physiological Recording and Data Reduction .....................................................................9
Overview of Analyses ........................................................................................................10
EXPERIMENT 1 RESULTS .........................................................................................................11
Performance Measures .......................................................................................................11
ERN....................................................................................................................................12
EXPERIMENT 1 DISCUSSION ...................................................................................................13
EXPERIMENT 2 METHOD .........................................................................................................15
Participants .........................................................................................................................15
Physiological Recording and Data Reduction ...................................................................16
Overview of Analyses ........................................................................................................16
EXPERIMENT 2 RESULTS .........................................................................................................18
Performance Measures .......................................................................................................18
ERN....................................................................................................................................20
EXPERIMENT 2 DISCUSSION ...................................................................................................23
GENERAL DISCUSSION ............................................................................................................24
Summary of Findings .........................................................................................................24
Theoretical Significance ....................................................................................................25
Practical Significance.........................................................................................................29
Conclusion .........................................................................................................................30
APPENDICES ...............................................................................................................................31
APPENDIX A: TABLES .................................................................................................. 32
APPENDIX B: FIGURES .................................................................................................35
REFERENCES ..............................................................................................................................40

iii

LIST OF TABLES

Table 1. Summary of Experiment 1 Behavioral & ERP Measures ........................................................33
Table 2. Summary of Experiment 2 Behavioral & ERP Measures ............................................................. 34

iv

LIST OF FIGURES

Figure 1. Experiment 1 Response Locked ERPs ........................................................................................ 36
Figure 2. Experiment 1 Scatter Plot....................................................................................................37
Figure 3. Experiment 2 Response Locked ERPs..................................................................................38
Figure 4. Experiment 2 Scatter Plot....................................................................................................39

v

INTRODUCTION

Anxious apprehension or worry is a distinct dimension of anxiety that is characterized by
incessant concern for the future and verbal ruminations about negative outcomes (Barlow, 1991,
2002; Nitschke, Heller, Imig, McDonald, & Miller 2001; Nitschke, Heller, Palmieri, & Miller,
1999). Neurologically, worry is associated with increased activity in a left-lateralized frontal
network of brain regions thought to mediate language generation and cognitive control processes
(Engels et al., 2007; Engels et al., 2010, Silton et al., 2011; Spielberg et al., 2013). This pattern
of neural activation suggests that the effects of worry on everyday goal-directed behavior may be
characterized by the interaction between excessive sub-vocal verbalization and the compensatory
exertion of cognitive control in response to worry-related interference (Eysenck, Derakshan,
Santos, & Calvo, 2007; Eysenck & Derakshan, 2011; Moser, Moran, Schroder, Donnellan, &
Yeung, 2013; Warren et al., 2013).
One neurophysiological marker of cognitive control processes that may index this
interplay is the error-related negativity (ERN), a negative deflection in the human event-related
brain potential (ERP) following the commission of an error (see Gehring, Liu, Orr, & Carp, 2012
for a review). Although still debated, the ERN is thought to reflect processes related to cognitive
control such as early error correction (Yeung & Summerfield, 2012), conflict monitoring
(Yeung, Botvinick, & Cohen, 2004; Yeung & Cohen, 2006), or reinforcement learning (Holroyd
& Coles, 2002). The ERN is thought to be generated in medial prefrontal regions, including the
anterior cingulate cortex (ACC) and the supplementary motor area (SMA; see Gehring et al.,
2012 for a review).
Studies indicate that individuals with elevated symptoms of worry—including chronic
worriers as well as those with generalized anxiety disorder (GAD) and obsessive-compulsive
1

disorder (OCD)—exhibit increased ERN amplitudes on forced-choice tasks (Gehring, Himle, &
Nisenson, 2000; Johannes et al., 2001; Weinberg, Klein, & Hajcak, 2012; Weinberg, Olvet, &
Hajcak, 2010). It has been posited that the ERN may be exclusively related to worry because
GAD and OCD are phenomenologically characterized by verbal ruminations. Indeed, Hajcak,
McDonald, & Simons (2003) and Moser, Moran, & Jendrusina (2012) demonstrated that the
ERN is modulated by worry but not anxious arousal—the physiological dimension of anxiety—
indicating that worry may be uniquely related to exaggerated error-related brain mechanisms (see
Moser et al., 2013 for a meta-analysis and review). Moreover, Zambrano-Vazquez and Allen (in
press) provided further evidence for this specific relationship by showing that the exaggerated
ERN present in OCD is primarily due to elevated worry.
Despite the growing evidence pointing to the specific relationship between worry and the
ERN, the functional significance of this relationship is not well understood. However, Moser et
al. (2013) recently proposed the compensatory error monitoring hypothesis (CEMH), suggesting
that the enlarged ERN reflects compensatory post-error processing in response to worry-induced
reductions in active goal maintenance. The theory is predicated on the notion that worry reduces
the available capacity of the central executive system (Eysenck & Calvo, 1992; Eysenck et al.,
2007; Eysenck & Derakshan, 2011). With attention diverted to off-task processes (i.e., worries),
anxious individuals must compensate by putting forth more cognitive effort to maintain a
standard level of performance. The enlarged ERN in anxiety is therefore considered to be a
reflection of compensatory effort that subserves performance maintenance. This explanation
accounts for the observation that although anxious people exhibit increased ERNs relative to
non-anxious people, there is typically no difference in performance accuracy on simple forced-

2

choice tasks such as the flanker and Stroop tasks (Hajcak et al., 2003; Vocat, Pourtois, &
Viulleumier, 2008; Schroder & Moser, 2014).
Consistent with the CEMH, higher levels of worry have been found to be associated with
increased dACC activity (Silton et al., 2011) and reduced rACC activity (Bishop, Duncan, Brett,
& Lawrence, 2004; Engels et al., 2007). This contrasting pattern of ACC activation suggests that
worries may impede conflict resolution and proactive control processes, thus leading to the
compensatory recruitment of cognitive control through reactive control processes (Braver, 2012).
Given that a number of source localization studies have localized the ERN to medial prefrontal
regions (Dehaene, Posner, & Tucker, 1994; Ullsperger & von Cramon, 2001; van Veen & Carter,
2002), the ERN-anxious apprehension relationship likely involves overlapping control
mechanisms that drive the aforementioned cognitive and behavioral disruptions associated with
anxiety. Consequently, the present study aimed to advance the understanding of such
mechanisms by exploring a potential nuance of the ERN-worry relationship: the role of verbal
processing.
Studies show that worry is predominantly a verbal process, and that it is the verbal form
of worry, rather than imagery-based worry, which depletes working memory and enhances
attention to threatening information (Rapee, 1993; Hayes, Hirsch, & Mathews, 2008; Leigh &
Hirsch, 2011; Williams, Mathews, & Hirsch, 2014). Demonstrating a clear relationship between
verbal processes and worry, Engels et al. (2007) found that worry was associated with greater
activity in the left inferior frontal gyrus—an area involved in sub-vocal articulatory rehearsal and
maintenance of verbal information (Awh et al., 1996; Zatorre, Meyer, Gjedde, & Evans, 1996;
Fletcher & Henson, 2001). Furthermore, recent research has highlighted the impact of verbal
rumination on cognitive functioning; verbal worry delays processing of negative emotional

3

stimuli and impairs the efficiency of inhibitory functioning (Spielberg et al., 2013; Warren et al.,
2013). Given that (a) the ERN-anxiety relationship is primarily driven by worry, and (b) worry
typically involves sub-vocal rehearsal, we propose that verbal processing may be an integral
component of the ERN-worry relationship. Elucidating a link between verbal activity and the
ERN-worry relationship would both provide support for and add specificity to a critical aspect of
the CEMH—that the enlarged ERN observed in anxious individuals reflects the recruitment of
reactive control in response to the depletion of on-task processing by task-irrelevant verbal
processes. Such a link would suggest that the ERN-anxiety relationship is essentially a function
of an automatized compensatory response to the impediments of off-task processing on active
goal maintenance. In this study, we aimed to test this hypothesis through two experimental
paradigms.
In the first experiment, we sought to determine whether the association between worry
and the ERN differs between errors committed by the right and left hands. Because worry
involves a strong linguistic component that is predominantly localized to the left hemisphere, we
hypothesized that the ERN-worry relationship would be strongest for errors committed with the
right hand which is controlled by the left hemisphere. This approach is supported by substantial
research into the relationship between lateralized neural functionality and hand motor activity.
Lateralized dual task interference studies have shown that right and left-hand motor responses
reliably activate the left and right hemisphere, respectively, providing support for hands as a
differentiating index of hemispheric activity (Kinsbourne & Cook, 1971; Chang & Hammond,
1987; Ashton & Mcfarland, 1991). Furthermore, performance monitoring ERPs have been
shown to be sensitive to hemispheric manipulations such as selective stimulus presentation to the

4

left or right visual field (Buchsbaum & Fedio, 1970; Mo, Xu, Kay, & Tan, 2011; Simon-Thomas
& Knight, 2005).
Of particular relevance, transcranial magnetic stimulation (TMS) studies have shown that
worry enhances motor preparation by increasing corticospinal excitability (Oathes, Bruce, &
Nitschke, 2008), and that verbal processing alters the excitability of the left but not the right
motor cortex (Meister et al., 2003; Olivieri et al., 2004; Buccino et al., 2005; Pullvermuller,
Shtyrov, & Ilmoniemi, 2005a). Moreover, a recent study by Hochman, Orr, and Gehring (2014)
linked the ERN with motor processing, demonstrating that the ERN may be modulated by the
potency between error and correct motor representations. Consequently, increased left motor
cortex excitability may correspond to an enhanced potency differential for right-handed error
responses and the generation of a bigger ERN. Taken together, these studies suggest that worryrelated verbal processes may influence lateral motor execution—potentially increasing the
likelihood of executing a right-hand motor response via enhanced excitability of the left motor
cortex. Based on this theoretical foundation, we predicted that (a) the ERN elicited by righthanded mistakes, controlled by the linguistically dominant left hemisphere, would be more
closely coupled with worry, and (b) that participants with higher levels of worry would commit
more right-handed errors.
The second experiment served to compliment the first by attempting to verify that the
ERN-worry relationship is associated with verbal processing and not the neurophysiology of the
dominant response hand. We aimed to accomplish this by manipulating the orientation of the
flanker stimuli (horizontal vs. vertical presentation). The rationale for this design comes from
studies examining the effects of stimulus orientation on lateralized linguistic processing (Jordan
& Patching, 2003; Lindell, Nichols, Castles, 2003; Lindell, 2006) and problem solving (Trbovich

5

& LeFevre, 2003; Beilock, Rydell, & McConnell, 2007; Decaro, Rotar, Kendra, & Beilock,
2010). For instance, Lindell (2006) reviewed several studies showing the absence of a left
hemisphere advantage for word recognition when words are presented vertically, suggesting a
differential effect of stimulus orientation on the recruitment of lateralized linguistic processing.
Specifically, horizontal stimuli appear to elicit more automized word recognition relative to
vertical stimuli. The link between orientation and linguistic processing is further reinforced by
studies that find a lateralized advantage for vertical stimuli from languages in which vertical
presentation is culturally normative (Hellige & Yamauchi, 1999; Nakagawa & Sukigara, 2000).
Moreover, Trbovich and LeFevre (2003) discovered that the orientation of stimuli impacts the
type of resources used to solve problems. Solving horizontally presented arithmetic problems
was mediated by verbal processes whereas solving vertically presented problems relied on spatial
processing. Borrowing Trbovich and LeFevres' paradigm, Decaro et al. (2010) found that verbal
worry compromised the ability to solve horizontal problems but not vertical problems thereby
highlighting a direct link between worry and stimulus orientation.
In summary, the first experiment sought to determine whether the ERN-worry
relationship differed by hand of error, and the second experiment aimed to examine whether this
relationship was dissociable by stimuli orientation. We reasoned that if verbal processing is leftlateralized, then right-handed ERNs should be more strongly correlated with worry relative to the
left-hand. Furthermore, if processing of horizontal stimuli is more linguistically mediated than
vertical stimuli, we predicted that the ERN-worry relationship would be exclusive to right-hand
horizontal trials but not vertical trials. Taken together, such findings would provide compelling
evidence that the ERN-worry relationship is influenced by verbal processing.

6

EXPERIMENT 1
METHOD
Participants
Fifty-one female undergraduates participated in the study for course credit. We selected
only females because the ERN is most strongly related to worry in females (Moran, Taylor, &
Moser, 2012) and because women are twice as likely to suffer from an anxiety-related disorder
(Kessler et al., 2005; Kessler, Petukhova, Sampson, Zaslavsky, & Wittchen, 2012). The sample
consisted of 41 right-handed participants and 2 left-handed participants; 8 participants failed to
report handedness. Handedness was assessed using a dichotomized single-item self-report
measure (i.e., asking participants to indicate whether they were right or left handed). Nine
participants were excluded from the analyses because of a failure to follow instructions regarding
the stimulus-response mapping (see below) that resulted in an error rate exceeding 50% during
one of the experimental blocks. The final sample consisted of 42 participants (33 right-handed; 2
left-handed; 7 unreported). Participants ranged in age from 18 to 21 years (M = 18.74, SD =
1.07). No participants discontinued their involvement once beginning the experiment.
Task
Participants completed a letter version of the Eriksen Flankers task (Eriksen & Eriksen,
1974). Participants were seated approximately 60 cm in front of a computer monitor and
instructed to respond to the center letter of a five-letter string in which the target was either
congruent (e.g., MMMMM or NNNNN) or incongruent (e.g., MMNMM or NNMNN) with the
surrounding (i.e., flanking) letters. Characters were displayed in a standard white font on a black
background and subtended 1.3˚ of visual angle vertically and 9.2˚ horizontally. The task was

7

administered on a Pentium R Dual Core computer using E-Prime software (Psychology Software
Tools, Inc), which facilitated stimuli presentation and response measurement.
During each trial, flanking letters were presented 35 ms prior to target letter onset after
which all five letters remained on screen an additional 100 ms (accumulating to 135 ms total trial
time). Participants were given 1000 ms to respond before the next intertrial interval began. A
fixation cross (+) was presented during the intertrial interval, which varied from 1200 ms to 1700
ms. The experimental session included 480 trials grouped into 12 blocks of 40 trials. Participants
were instructed to respond as quickly and as accurately as possible using either their left or right
index finger to select the left ("A”) and right ("L") keyboard buttons, respectively, which
corresponded to the target identity. Letters making up the trial stimuli differed across block pairs:
Blocks 1 and 2, "M" and "N"; Blocks 3 and 4 "F" and "E"; Blocks 5 and 6, "O" and "Q"; Blocks
7 and 8, "T" and "I"; Blocks 9 and 10, "V" and "U"; Blocks 11 and 12, "P" and "R". Prior to each
block, instructions regarding the specific target-button assignment was presented. Performance
feedback was not provided.
Following the flanker task, participants completed a battery of self-report questionnaires
including the Penn State Worry Questionnaire (PSWQ; Meyer, Miller, Metzger, & Borkovec,
1990) and the Anxious Arousal (AA) subscale of the Mood and Anxiety Symptom Questionnaire
(MASQ; Watson & Clark, 1991). Borrowing from Nitschke et al. (2001), the PSWQ was used as
a measure of anxious apprehension and the MASQ-AA was used as a measure of anxious
arousal. The MASQ-AA was primarily included to test the specificity of the ERN relationship
with worry (e.g., Moser et al., 2012).

8

Psychophysiological Recording and Data Reduction
Participants were fitted with a 64 channel stretch-lycra cap. Continuous
electroencephalographic activity was recorded using the ActiveTwo BioSemi system (BioSemi,
Amsterdam, The Netherlands). Recordings were taken from 64 Ag-AgCl electrodes placed in
accordance with the 10/20 system. Two additional electrodes were placed on the left and right
mastoids. Electrooculogram (EOG) activity generated by eye movements and blinks was
recorded at FP1 and at three electrodes placed inferior to the left pupil and on the left and right
outer canthi approximately 1 cm from the pupil. During data acquisition, the Common Mode
Sense active electrode and Driven Right Leg passive electrode formed the ground, as per
BioSemi’s design specifications. All signals were digitized at 512 Hz using ActiView software
(BioSemi).
Offline analyses were conducted using BrainVision Analyzer 2 (BrainProducts, Gilching,
Germany). Scalp electrode recordings were referenced to the numeric mean of the mastoids and
band-pass filtered with cutoffs of 0.1 and 30 Hz (12 dB/oct rolloff). Ocular artifacts were
corrected using the regression method developed by Gratton, Coles, and Donchin (1983).
Physiological artifacts were detected using a computer-based algorithm such that trials in which
the following criteria were met were rejected: a voltage step exceeding 50 μV between
contiguous sampling points, a voltage difference of more than 200 μV within a trial, or a
maximum voltage difference less than 0.5 μV within a trial. Trials were removed from ERP and
behavioral analyses if the RT fell outside of a 200-800 ms time window. The response-locked
data were segmented into individual epochs beginning 200 ms before response onset and
continuing for 800 ms following the response. To quantify response-locked ERPs, the average
activity in the 200-ms window preceding response onset was subtracted from each data point

9

subsequent to the response. The ERN was then quantified as the average activity occurring
between 0 and 100 ms post-response at the fronto-central recording site FCz—where the ERN
was maximal.
Overview of Analyses
Behavioral and ERP measures were statistically analyzed using SPSS software (Version
21.0). MASQ-AA scores were highly kurtotic (z = 2.2) and therefore transformed in order to
normalize the distribution (z = .69). Behavioral data were submitted to repeated measures
analyses of variance (ANOVAs). Number of errors was submitted to a one-way hand of error
(right vs. left) ANOVA. Overall RTs were submitted to a 2 (hand: right vs. left) X 2 (accuracy:
correct vs. error) ANOVA. Post-error adjustments were examined by submitting RT and
accuracy following error and correct trials to separate 2 (hand: right vs. left) X 2 (response type:
post-error vs. post-correct) ANOVAs. When significant interactions emerged, follow-up t-tests
were conducted to aid in the interpretation of results. Degrees of freedom varied slightly among
the F-tests because of performance variability (e.g., a participant making no left handed errors is
excluded from analyses involving hand of error). For ERP analyses, the ERN was submitted to a
2 (hand: right vs. left) X 2 (accuracy: correct vs. error) ANOVA. Partial eta squared η 2p is
reported as an estimate of effect size in ANOVA models where .05 represents a small effect, .1 a
medium effect, and .2 a large effect (Cohen, 1973). Separate correlational analyses examining
associations between ERN and PSWQ and MASQ-AA scores were conducted to test the main
hypotheses of the current investigation. Specifically, statistical contrasts between dependent
correlations were calculated using Hotelling’s t-test (Hotelling, 1940). Correlation coefficients
(i.e., rs) ranging from .1-.29 are considered small effects, rs ranging from .30-.49 medium
effects, and rs greater than .50 large effects (Cohen, 1988).

10

EXPERIMENT 1 RESULTS
Descriptive statistics for behavioral and ERP measures for Experiment 1 are presented in
Table 1.
Performance Measures
Overall flanker task accuracy was high (M percent correct = 90.73%, SD = 4.31%).
Participants made an average of 42.81 errors (SD = 19.78), with more right-handed errors (M =
23.21, SD = 11.70, range = 3–49) than left-handed errors (M = 19.60, SD = 9.50, range = 3–39,
F(1, 41) = 8.75, p = .005, η 2p = .18). Consistent with our prediction, correlational analysis
showed that as PSWQ scores increased, participants made more right-handed errors (r = .33, p =
.03) but not more left-handed errors (r = .19, p = .23). MASQ-AA scores were not significantly
correlated with accuracy in either hand (rs < |.12|, ps > .41).
The analysis of RTs revealed a main effect of accuracy, confirming faster RTs on error
trials (M = 372.89, SD = 42.14) than on correct trials (M = 438.28, SD = 34.94, F(1, 41) =
163.70, p < .001, η 2p = .80), consistent with a speed-accuracy trade-off.. A main effect of hand
revealed faster overall RTs with the right hand (M = 425.50, SD = 34.34) than the left hand (M =
438.01, SD = 37.67, F(1, 41) = 8.40, p = .006, η 2p = .17). There was not a significant Hand X
Accuracy interaction (F(1,41) < 1). Neither anxiety measure was significantly correlated with
RTs (rs < |.24|, ps > .12).
Analysis of post-error RT replicated the post-error slowing (PES) effect such that correct
responses were slower on trials after errors (M = 465.66, SD = 58.63) than after corrects (M =
435.84, SD = 34.13, F(1, 41) = 22.27, p < .001, η 2p = .35). The Hand X Response Type
interaction did not approach significance, however (F(1, 41) < 1), indicating that PES did not
differ by hand. PES, calculated as the difference between post-error correct RTs minus post11

correct correct RTs, across both hands were not correlated with any anxiety measure (rs < |.19|,
ps > .22).
Post-error accuracy analysis revealed that participants were numerically, but not
statistically, more accurate after correct responses (M percent correct = 91.23%, SD = 4.19%)
than after errors (M percent correct = 87.57%, SD = 13.37%, F(1, 41) = 3.04, p = .09, η 2p = .07).
There was no main effect of hand on post-response accuracy, however (F(1, 41) = 2.33, p = .13,
η 2p = .05), nor was there a significant Hand X Response Type interaction (F(1, 41) < 1). Postaccuracy difference, calculated as the difference between post-error accuracy minus post-correct
accuracy, across both hands were not correlated with the anxiety measures (rs < |.26|, ps > .09).
ERN
As expected, the main effect of Accuracy was significant (F(1,41) = 14.63, p < .001, η 2p =
.26 ), indicating larger negativity on error trials compared to correct trials (see Fig. 1). However,
there was no main effect of hand (F(1, 41) < 1) nor was there a significant Hand X Accuracy
interaction (F(1, 41) < 1).
Critical to the main aims of the current investigation, correlational analysis showed that,
as hypothesized, higher PSWQ scores were only associated with greater ERN on right-handed
errors (r = -.33, p = .04) but not left-handed errors or correct responses in either hand (rs < |.04|,
ps > .78; see Fig. 2). Hotelling’s t-test revealed that these correlations differed significantly
(t(39) = -2.25, p = .03, d = .72). Consistent with our previous work (Moser et al., 2012),
MASQ-AA scores were not significantly correlated with ERNs or CRNs in either hand (rs <
|.19|, ps > .24) indicating the ERN was uniquely related to worry.

12

EXPERIMENT 1 DISCUSSION
Findings from Experiment 1 were consistent with our hypothesis that the ERN-worry
relationship would be strongest for errors made with the right hand. Thus, these results provide
support for our prediction that the ERN-worry relationship is associated with lateralized verbal
activity. Also as predicted, the PSWQ-ERN relationship differed by hand of error—PSWQ
scores were positively correlated with the number of right-handed errors but not left-handed
errors. An in-depth discussion of these results will be reserved for the general discussion.
In Experiment 2, we sought to further explore the role of verbal processing in the ERNworry relationship. First, we modified the orientation of the flanker stimuli to examine if the
ERN elicited by the right hand would differ based on vertical or horizontal presentation. We
hypothesized that since horizontal stimuli have been shown to recruit verbal processes, then the
ERN-worry relationship should be stronger on horizontal trials; importantly, obtaining
differential ERN-worry correlations based on orientation would provide compelling evidence to
rule out the possibility that the ERN-worry relationship is driven solely by the neurophysiology
of the right hand. Second, since right-handed individuals comprised the majority of the
participants in Experiment 1, we recruited an equal number of left and right-handed participants
to rule out the likelihood that the results of Experiment 1 were merely a function of
handedness—specifically the possibility that worry simply enhances the error potency and
subsequent response conflict in the dominant hand. Lastly, we replaced the letters comprising the
flanker stimulus with arrows to reduce the verbal nature of the stimuli, thereby minimizing a
potential confound that could interfere with the primary manipulation of interest (i.e., the effect
of orientation).

13

Overall, the purpose of Experiment 2 was to strengthen our conceptualization by
subjecting alternative interpretations of the results of Experiment 1 to falsification. Specifically,
our predictions were that: (a) the ERN-worry correlation would emerge on horizontal trials but
not vertical trials, and (b) on horizontal trials, the ERN-worry relationship would only be present
for right-handed errors but not left-handed errors—replicating the results of Experiment 1 with
an equal number of right and left-handed participants.

14

EXPERIMENT 2
METHOD
Participants
Seventy-four female undergraduates participated in the study for course credit. Per our
recruitment strategy, the sample consisted of 37 right-handed participants and 37 left-handed
participants. Similar to Experiment 1, handedness was assessed using a dichotomized single-item
self-report measure (i.e., asking participants to indicate whether they were right or left handed).
Three participants were excluded from analyses because of response-mapping errors that resulted
in an error rate exceeding 50% across all trials. Two participants were excluded because of
excessive movement during the task. The final study sample consisted of 69 participants (33
right-handed; 36 left-handed). Participants ranged in age from 18 to 37 (M = 19.49, SD = 2.49),
not differing significantly from the sample in Experiment 1 (t(98) = -1.70, p = .09). No
participants discontinued their involvement once beginning the experiment.
Participants completed an arrow version of the Eriksen Flankers task (Eriksen & Eriksen,
1974). In contrast to Experiment 1, arrows were selected over letters to isolate the effect of
spatial orientation. Participants were instructed to respond to the center arrow of a five-arrow
array in which the target was either congruent (e.g., >>>>> or <<<<<) or incongruent (e.g.,
>><>> or <<><<) with the surrounding flanker arrows. Presentation of the arrows differed in
orientation. On horizontal trials, equidistant arrows were presented horizontally at the center of
the screen, whereas on verticals trials, the arrows appeared vertically with arrows pointing left or
right. Arrows were displayed in a standard white font on a black background and subtended 1.3˚
of visual angle vertically and 9.2˚ horizontally for horizontal trials, and subtended approximately

15

9.2˚ of visual angle vertically and 1.3˚ horizontally for vertical trials. The task was administered
on a Pentium R Dual Core computer using E-Prime software (Psychology Software Tools, Inc).
During each trial, all five arrows were presented simultaneously for 100 ms. Participants
had 900 ms to respond to the target before the response was deemed missing and excluded from
data analysis. A fixation cross (+) was presented during the intertrial interval, which varied from
800 ms to 1200 ms. The experimental session included 800 trials, consisting of 10 blocks of 40
horizontal trials and 10 blocks of 40 vertical trials. Orientation was randomized across blocks of
the task and incongruent and congruent trials were also randomized within each block. Using the
keyboard, participants were instructed to respond by pressing the “A” (for a target "<") or “L”
(for a target ">") keyboard button as quickly and as accurately as possible. Performance
feedback was not provided.
Following the flankers task, participants completed a series of questionnaires including
the PSWQ and MASQ-AA as in Experiment 1.
Psychophysiological Recording and Data Reduction
The recording and data reduction procedures were identical to those described in
Experiment 1.
Overview of Analyses
As in Experiment 1, behavioral and ERP measures were statistically analyzed using
rANOVAs. In contrast to Experiment 1, however, MASQ-AA scores did not exhibit skewness
and kurtosis values greater than 2 and were not log transformed. Overall number of errors were
submitted to a 2 (hand: right vs. left) X 2 (orientation: vertical vs. horizontal) repeated measures
ANOVA. Overall RTs were submitted to a 2 (hand: right vs. left) X 2 (accuracy: correct vs.
error) X 2 (orientation: vertical vs. horizontal) ANOVA. Post-error adjustments were examined

16

by submitting RT and accuracy following error and correct trials to separate 2 (hand: right vs.
left) X 2 (response type: post-error vs. post-correct) X 2 (orientation: vertical vs. horizontal)
ANOVAs. When significant interactions emerged, follow-up t-tests were conducted to aid in the
interpretation of results. Partial eta squared η 2p is reported as an estimate of effect size in
ANOVA models. For ERP analyses, the ERN was submitted to a 2 (hand: right vs. left) X 2
(accuracy: correct vs. error) 2 (orientation: vertical vs. horizontal) ANOVA with handedness as a
between-subjects factor. Critical to the main aims of the current investigation, separate
correlational analyses examining associations between ERN and PSWQ and MASQ-AA scores
were conducted. Statistical contrasts between dependent correlations were calculated using
Hotelling’s t-test (Hotelling, 1940). Finally, follow-up analyses were conducted by combining
horizontal trials from both experiments to parse the effect of handedness and experiment.

17

EXPERIMENT 2 RESULTS
Descriptive statistics for behavioral and ERP measures for Experiment 2 are presented in
Table 2.
Performance Measures
Overall accuracy was high (M percent correct = 90.38%, SD = 6.27%). The ANOVA
revealed a main effect of orientation as participants made more errors on vertical trials (M =
39.65, SD = 27.34, range = 4–139) than on horizontal trials (M = 34.81, SD = 23.52, range = 5–
104, F(1, 68) = 9.67, p = .003, η 2p = .13). The main effect was qualified by a significant Hand X
Orientation interaction (F(1,68) = 21.49, p < .001, η 2p = .24). Replicating the results of
Experiment 1, follow-up analysis found that on horizontal trials, there were more right-handed
errors (M = 19.10, SD = 13.13, range = 1–56) than left-handed errors (M = 15.71, SD = 12.79,
range = 2–58, t(68) = 2.58, p = .01). There was no significant difference in errors by hand on
vertical trials (right hand: M = 18.74, SD = 14.54, left hand: M = 20.91, SD = 14.64, t(68) = 1.77,
p = .08). There was no main effect of hand (F(1,68) = .30, p = .59, η 2p < .01). Neither anxiety
measure was correlated with error rates (rs < |.18|, ps > .16).
A main effect of accuracy emerged on RT such that RTs on error trials (M = 360.74, SD
= 35.43) were significantly faster than on correct trials (M = 430.07, SD = 42.44, F(1, 68) =
583.27, p < .001, η 2p = .90). The ANOVA also revealed a main effect of orientation, such that
RTs on vertical trials (M = 404.31, SD = 38.82) were slower than horizontal trials (M = 385.72,
SD = 38.48, F(1, 68) = 53.02, p < .001, η 2p = .44). The main effect of orientation was qualified by
a significant Hand X Orientation interaction (F(1, 68) = 6.63, p = .01, η 2p = .09). Significant hand
differences emerged only when comparing across different orientation (e.g., left horizontal RT

18

vs. right vertical RT) due to slower overall RTs on vertical trials. There was no main effect of
hand (F(1,68) < 1). No other significant interactions emerged (Fs < 1.37, ps > .25, η 2p s < .02).
Interestingly, correlational analyses showed that as PSWQ scores increased, right-handed
vertical error RTs increased (r = .28, p = .02). Correlations between PSWQ and other RTs were
non-significant (rs < |.19|, ps > .16). MASQ-AA scores were not correlated with RTs (rs < |.24|,
ps > .14).
A main effect of response type confirmed the PES effect; correct responses were slower
on trials after errors (M = 444.96, SD = 54.62) than after corrects (M = 426.3, SD = 42.02, F(1,
67) = 38.49, p < .001, η 2p = .37). There was also a main effect of orientation such that correct
responses were slower on vertical trials (M = 443.72, SD = 46.07) than on horizontal trials (M =
427.02, SD = 50.62, F(1, 67) = 35.30, p < .001, η 2p = .35), mimicking the overall slowing on
vertical trials. There was no main effect of hand (F(1, 67) < 1). There was neither an interaction
between trial type and hand nor between trial type and orientation (Fs(1, 67) < 1). PES,
calculated as the difference between post-error correct RTs minus post-correct correct RTs, was
not correlated with any anxiety measure (rs < |.19|, ps > .11).
Overall, participants were not more accurate after errors (M percent correct = 90.90%, SD
= 9.71%) than after correct responses (M percent correct = 90.46%, SD = 6.15%, F(1, 67) < 1).
There were no main effects of Orientation, Hand or Response Type (Fs < 3.70, ps > .06, η 2p s <
.05). No significant interactions emerged for post-response accuracy analyses (Fs < 3.48, ps >
.07, η 2p s < .05). There were no significant correlations between the anxiety measures and postresponse accuracy (rs < |.21|, ps > .09).

19

ERN
A 2(Accuracy) X 2(Orientation) X 2(Hand) repeated measures ANOVA was conducted
with handedness as a between-subjects factor. As expected, there was a main effect of Accuracy
(F(1,65) = 83.84, p < .001, η 2p = .56) such that the ERN was larger following errors than correct
trials. There was also a main effect of Orientation (F(1,65) = 4.98, p = .03, η 2p = .07) such that
activity following vertical trials was more negative (M = -.43 SD = 5.19) than horizontal trials
(M = .19, SD = 4.28, t(67) = 2.31, p = .02; see Fig. 3). The main effects were qualified by an
Accuracy X Orientation interaction (F(1,65) = 6.61, p = .01, η 2p = .09). The ERN was
significantly larger on vertical error trials (M = -3.65, SD = 6.20) compared to horizontal error
trials (M = -2.32, SD = 5.11, t(68) = 2.78, p = .01). There was no difference in CRN between
vertical correct trials (M = 2.68, SD = 5.40) and horizontal correct trials (M = 2.59, SD = 4.94,
t(68) = .48, p = .64). An unpredicted 4-way Hand X Accuracy X Orientation X Handed
interaction also emerged (F(1,65) = 4.84, p = .03, η 2p = .07). We parsed this interaction by
splitting the data by handedness. There was not a significant Hand X Accuracy X Orientation
interaction in the right-handed participants (F(1,31) = 1.78, p = .19, η 2p = .05), but a marginally
significant interaction emerged in the left-handed participants (F(1,34) = 3.32 p = .08, η 2p = .09).
Consequently, we examined whether a differential interaction among Accuracy, Hand, and
Orientation would emerge for left-handed participants. No significant interactions emerged (Fs <
1.45, ps > .24, η 2p s < .04) and we concluded that the 4-way interaction revealed no clear effects
with respect to handedness on ERN modulation. No other significant differences emerged (Fs <
2.47, ps > .12, η 2p s < .04).

20

Correlational analyses were supportive of our predictions. Analyzing horizontal trials by
hand revealed a differential relationship between the ERN and PSWQ. Consistent with our
hypothesis and the results of Experiment 1, PSWQ scores were only correlated with the ERN on
horizontals trials following right-handed errors (r = -.35, p < .01) but not left-handed errors (r = .15, p = .24; see Fig. 4). No significant relationships emerged for errors committed on vertical
trials (rs < |.13|, ps > .24). Indeed, Hotelling’s t-test revealed a moderate difference between right
and left-handed ERN-worry correlations on horizontal trials (t(66) = -1.80, p = .08, d = .44), but
not on vertical trials (t(66) = .06, p = .95, d = .01). To examine the potential effects of
handedness, we conducted separate correlational analyses in right and left-handed participants.
Although the horizontal right-hand ERN-PSWQ relationship was slightly attenuated in lefthanded participants (r = -.28, p = .09) relative to right-handed participants (r = -.42, p = .02),
Fisher’s r-to-z transformation revealed that this difference was negligible (z = .63, p = .53).
These null results suggest that handedness plays a minimal role in modulating the right-hand
PSWQ-ERN relationship. MASQ-AA scores were not significantly correlated with any ERN (rs
< |.14|, ps > .26). Neither PSWQ nor MASQ-AA scores were correlated with the CRN (rs < |.12|,
ps > .34).
To the extent that an alpha of less than .05 represents a meaningful difference (Cumming,
2013), the absence of a statistically significant contrast between horizontal right-hand and lefthand ERN-PSWQ correlations was likely due to a lack of statistical power. Consequently, we
combined the horizontal trial data from Experiment 1 and 2 (N = 110) and submitted the ERN to
a single two-level within subjects factor model (Hand of Error: Right vs. Left) with PSWQ
scores entered simultaneously as a continuous predictor and experiment (Experiment: 1 vs. 2) as
a between-subjects factor. As expected, the main effect of Hand of Error (F(1, 107) = 7.07, p <

21

.01, η 2p = .06) was qualified by a significant Hand of Error X PSWQ interaction (F(1, 107) =
6.81, p = .01, η 2p = .06). Higher PSWQ scores were associated with bigger right-handed ERNs (r
= -.32, p < .01) but not left-handed ERNs (r = -.05, p = .59; t(107) = -2.83, p < .01, d = .55).
Critically, no other interactions emerged, including the Hand of Error X PSWQ X Experiment
interaction (Fs < .13, ps > .71)—suggesting that Experiment did not moderate the Hand of Error
X PSWQ interaction.
To likewise examine the effect of handedness across studies, we repeated the analysis
with handedness as a between-subjects factor. Again, the main effect of Hand of Error (F(1,100)
= 8.77, p < .01, η 2p = .08) was qualified by a significant Hand of Error X PSWQ interaction
(F(1,100) = 7.00, p = .01, η 2p = .07). Importantly, no other interactions emerged, including the
Hand of Error X PSWQ X Handedness interaction (Fs < 1.76, ps > .19)—supporting the
aforementioned correlational analysis that the right-hand ERN-PSWQ correlation was similar
across left and right-handed participants. Handedness was not a significant between-subjects
factor (F(1, 100) = .60, p = .69, η 2p < .01). Taken together, these results suggest that a
differential ERN-PSWQ relationship by hand of error held across both studies irrespective of
experiment or handedness.

22

EXPERIMENT 2 DISCUSSION
In line with our hypothesis, the ERN was significantly correlated with PSWQ scores on
horizontal trials and not vertical trials. Moreover, we replicated the results of Experiment 1 after
accounting for handedness, showing that the ERN-PSWQ relationship was exclusive to righthanded horizontal trials. A combined analysis of the horizontal trials showed that that the righthand ERN-PSWQ relationship held across both experiments and was not moderated by
handedness. Together, these results provide initial evidence that verbal mechanisms play a role
in the ERN-worry relationship.

23

GENERAL DISCUSSION
Summary of Findings
The overarching goal of the study was to examine the role of verbal processing in the
ERN-worry relationship through two experiments involving different tasks and participants.
Based on evidence indicating that worry is associated with left-frontal verbal brain activity, we
predicted that the ERN-worry relationship would be largest on error trials committed with the
right hand, which is controlled by the left hemisphere—demonstrating the significance of left
hemispheric verbal processes in the anxiety-error monitoring relationship.
Across both experiments, the ERN-worry relationship differed by hand of error—that is,
ERN and PSWQ scores were only significantly correlated on right-handed horizontal error trials.
That the ERN-worry relationship was exclusive to right-handed errors on horizontal trials is
consistent with our prediction that the ERN-worry relationship involves left hemispheric verbal
processing. Experiment 2 demonstrated that (1) the ERN elicited by right-handed mistakes on
vertical trials was not significantly correlated with worry scores, and (2) the right-hand ERNworry relationship was not moderated by handedness. Critically, these findings are inconsistent
with alternative hypotheses that the ERN-worry relationship is driven by the neurophysiology of
the right hand or exclusively a function of the dominant response.
The finding that the ERN-worry correlation differed based on stimulus orientation is
entirely consistent with research showing that verbal processes are recruited during presentation
of horizontal stimuli but not vertical stimuli (Lindell et al., 2002; Jordan & Patching, 2003;
Trbovich & LeFevre, 2003; Beilock et al., 2007; Decaro et al., 2010). Furthermore, the slight
attenuation of the right-hand ERN-worry correlation in left-handed participants supports the
involvement of left-hemispheric verbal processing given that a higher proportion of left-handed

24

people (27% in strong left-handers relative to 4% in strong right-handers) exhibit righthemisphere language dominance (Knecht et al., 2000). Replicating Moser and colleagues (2012),
anxious arousal—as indexed by MASQ-AA scores—was unrelated to the ERN. Taken together,
these results support the hypothesis that the anxiety-error monitoring relationship may be
specific to verbal worry. In addition to advancing the specificity of the anxiety-error monitoring
relationship, the present findings add a unique vantage point to a growing body of literature
showing that left hemispheric verbal processing dissociates anxious apprehension from anxious
arousal (Heller, Nitschke, Etienne, & Miller 1997; Nitschke et al., 1999; Hofmann et al., 2005;
Engels et al., 2007).
Theoretical Significance
The functional model proposed by Hochman et al. (2014) offers a framework in which to
elucidate the mechanistic underpinnings of our findings. Based on results from two novel ERN
paradigms, Hochman and colleagues suggested that the ERN indexes processes dedicated to
aborting an error. This conclusion follows from the parallel task set model (PTS, Seymour and
Schumacher, 2009) that proposes that conflict occurs when a request to prepare a motor response
differs from a request that is already in process. Specifically, this type of conflict occurs between
prepotent unplanned errors and preplanned correct responses. The more prepotent the error
representation, the more difficult it is to suppress the error response, and consequently, the more
likely it is that an error will be committed. Upon error commission, the preplanned correct
response becomes corrective and its automatized execution must wait until the error
representation is suppressed. Consistent with the PTS model, Hochman et al. (2014) found that
the wider the disparity between the prepotency of the error representation and the nonpotency of
the correct representation, the larger the ERN. For example, they found that the ERN was larger

25

when more errors were committed, presumably because the error potency was much stronger
than potency of the correct response.
With respect to the current findings, the observation that individuals with greater levels of
worry exhibit greater right-hand ERNs suggests the possibility that worry serves to increase the
prepotency of the right-hand motor response. In support of this notion, worry has been shown to
be associated with greater corticospinal motor excitability, providing evidence that worry
enhances motor preparation (Oathes et al., 2008). Moreover, evidence from the laterality
literature suggests that the motor systems of the hemisphere controlling the dominant hand are
more excitable and exhibit a lower motor response threshold relative to the non-dominant
hemisphere—in other words, the dominant hand is more prepotent (see Hammond, 2002 for a
review). Given that the left hemisphere is dominant in approximately 90% of the population
(Annett, 2004), it follows that for a large majority of people, the right hand is more prepotent
than the left. Consequently, if worry enhances motor preparation and the right hand is more
prepotent than the left hand, then our findings suggest that worriers are more “prepared” to
respond with the right hand relative to the left. Therefore, this enhanced prepotency would
engender more right-handed "slips"—a type of error that, by definition, involves a wide errorcorrect potency disparity, and thus, produce a larger ERN. Critically, we find that PSWQ scores
were positively correlated with the number of right-handed errors in Experiment 1 where the
majority of participants were right-handed, indicating that more worried individuals are indeed
more likely to commit errors with the prepotent right hand.
The finding that the right-hand ERN-worry relationship was exclusive to horizontal trials
and not moderated by handedness adds specificity to the hypothesis by showing that the
prepotency of the right-hand may be exacerbated by the verbal nature of worry—that is,

26

recruitment of verbal processes in the linguistically dominant left hemisphere may "prime" the
left motor cortex. This notion receives support from a number of transcranial magnetic
stimulation (TMS) studies showing that processing action words or sentences alters the
excitability of the left but not the right motor cortex and affects reaction times when the motor
response and the verbal stimuli call the same effector (Olivieri et al., 2004; Buccino et al., 2005;
Pulvermuller et al., 2005a). More specific to the present findings, neurophysiological studies
have found increased excitability in the hand motor area of the language dominant hemisphere
during speech production, demonstrating a functional link between hand motor activity and
language processing (Tokimura, Tokimura, Oliviero, Asakura, & Rothwell, 1996; Seyal, Mull,
Bhullar, Ahmad, & Gage, 1999; Meister et al., 2003). Together, these studies reveal a unique
interaction between the neural systems involved in motor action and language processing.
Indeed, we found that across both Experiment 1 and Experiment 2, there were significantly more
right-handed errors than left-handed errors during the more verbally demanding horizontal trials
but no difference on the spatially demanding vertical trials.
Although our findings appear consistent with the aforementioned literature, our
hypothesis remains speculative due to the lack of empirical evidence demonstrating a direct link
between sub-vocal rehearsal and enhanced motor excitability. A future study involving the
manipulation of verbal working memory during the flanker task may offer a more direct way to
evaluate our hypothesis. Indeed, a prior experiment in our lab supports the promise for such a
design—we found that increased verbal working memory load was associated with a larger ERN
and more errors (Moser et al., 2013). The caveat is, however, that all participants responded with
their right hand using a mouse. Consequently, a future experimental manipulation involving
lateralized response options (left vs. right) may be particularly fruitful. Moreover, employing a

27

handedness inventory measuring handedness gradation (i.e., the extent to which one is left or
right-handed) rather than a dichotomized self-report may yield a more fine-grained analysis of
the relationship between worry and motor potency.
Despite the substantial amount of evidence in support of our proposal, it is important to
acknowledge the absence of a relationship between PSWQ and the number of right-handed
errors in Experiment 2. The failure to replicate the effect of Experiment 1 may be attributed to
differences in the complexity of the task design; that is, the enhanced cognitive load associated
with frequent alternations between horizontal and vertical stimuli may have influenced the error
rate. Furthermore, the use of arrows as the target stimuli may have reduced the verbal nature of
the stimuli and thus diminished the "priming" effect of the left motor cortex. Lastly, the higher
number of left-handed participants in Experiment 2 introduced more variability, as evidenced by
the finding of an attenuated right-hand ERN-PSWQ in left-handed participants. As mentioned
before, this attenuation is compatible with our conceptual model insofar as left handers are less
likely to be left-language dominant (Knecht et al., 2000). Additional empirical investigation will
be needed to further validate our claims here. For instance, a worry induction paradigm using
fMRI to differentiate left and right language dominance may help clarify the specific relationship
among verbal worry, hemispheric dominance, and lateral motor execution.
That we were able to link verbal activity with the ERN-worry relationship provides
compelling support for a central aspect of the CEMH—that sub-vocal worries elicit disruptive
off-task processing (e.g., increasing probability of right-handed error slips through enhanced
excitability) that trigger the engagement of reactive control (see Moser et al. 2013). Moreover,
the present findings enrich the specificity of the CEMH, introducing the possibility that the
interplay between verbally-mediated worry and motor activity may constitute a key component

28

of the ERN-worry relationship—that is, the enlarged ERN may reflect the extent to which the
amount of processing required to abort the error (part of the broader error correction response) is
modulated by worry-induced disparities between the potency of the correct and incorrect motor
responses. In other words, worriers exhibit more error corrective processing than their nonworried counterparts in order to maintain equivalent performance standards—a notion that is
quintessential to the core principle of the CEMH model.
Practical Significance
There have been recent proposals to characterize the ERN as an endophenotype for
internalizing disorders (Hajcak, Franklin, Foa, & Simons, 2008; Olvet & Hajcak, 2008; Hajcak,
2012). Specifically, the ERN is hypothesized to mediate the relationship between the genetic
predisposition for affective disorders and their expression. Although some interpretations of
extant research would lend credence to this proposal, our results argue against an unqualified
interpretation of the ERN as a marker of affective disorders. By uncovering key moderators of
the ERN-worry relationship, the current findings continue to illustrate the nuance and complexity
underlying the link between the enhanced ERN and anxiety. Given the specificity of the ERNworry relationship—that is, an enlarged ERN amplitude is only strongly correlated with worry
(Moser et al., 2012) in women (Moran et al., 2012) on right-handed, horizontal error trials (data
from the current study)—the ERN may not be a generalizable psychopathological risk factor.
Rather, the predictive value of the ERN is most likely contingent on the moderators that uphold
the relationship between the ERN and the construct of interest. Our data suggest that the ERN is
most predictive of cognitive anxiety in females under task conditions in which stimuli are
presented horizontally and errors are produced by the right hand—as it happens, a very common

29

way that data have been collected in the literature to date (Hajcak et al., 2003; Moser, Hajcak, &
Simons 2005; Weinberg et al., 2010; Weinberg et al., 2012).
Conclusion
Through two experimental paradigms, the current study revealed the importance of verbal
mechanisms in the relationship between worry and error processing. Uncovering moderators and
mediators driving the impact of verbal worry on performance monitoring is likely to yield
important theoretical and pragmatic insights for improving the understanding of the nature and
impact of anxiety disorders.

30

APPENDICES

31

APPENDIX A:

TABLES

32

Table 1. Summary of Experiment 1 Behavioral & ERP Measures
Measure

Overall
M

SD

Left Hand

Right Hand

M

SD

M

SD

Error RT

372.89 42.14

438.01

37.67

425.50

34.34

Correct RT

438.28 34.94

444.41

37.29

432.41

34.42

Post-error RT

465.66 58.63

472.19

75.75

459.12

59.81

Post-correct RT

435.84 34.13

442.13

36.15

429.55

33.71

Post-error slowing

29.82

40.95

30.07

62.68

29.57

41.08

Accuracy (% Correct)

90.73

4.31

91.38

4.20

90.20

4.20

Number of errors

42.81

19.78

19.60

9.50

23.21

11.70

Post-error accuracy

87.57

13.37

88.09

14.30

87.45

13.57

Post-correct accuracy

91.23

4.19

91.93

4.03

90.63

4.79

Post-error accuracy difference

-3.66

13.57

-3.84

14.72

-3.18

13.63

ERN

-3.58

5.31

-3.72

7.01

-3.43

5.62

CRN

-0.07

4.31

-0.41

4.75

0.27

4.17

33

Table 2. Summary of Experiment 2 Behavioral & ERP Measures
Measure

Overall

Left Hand

Right Hand

M

SD

M

SD

M

SD

Error RT: Horizontal

350.40

38.05

348.84

49.44

349.68

42.76

Error RT: Vertical

369.48

38.47

374.85

58.75

368.54

40.40

Correct RT: Horizontal

421.04

43.20

418.59

44.36

423.77

43.17

Correct RT: Vertical

439.15

43.67

440.78

46.11

437.98

43.05

Accuracy: Horizontal

90.97

6.01

91.78

6.62

90.15

6.68

Accuracy: Vertical

89.79

6.94

89.19

7.42

90.38

7.44

ERN: Horizontal

-2.18

5.01

-2.36

5.92

-2.02

5.68

ERN: Vertical

-3.49

6.09

-3.37

7.01

-3.60

6.39

CRN: Horizontal

2.70

4.90

2.50

4.88

2.90

5.06

CRN: Vertical

2.79

5.37

2.76

5.40

2.81

5.58

34

APPENDIX B:

FIGURES

35

Figure 1. Experiment 1 Response Locked ERPs.

Note: Response-locked ERPs plotted by accuracy and median-split PSWQ scores separated by
hand of error.

36

Figure 2. Experiment 1 Scatter Plot.

Note: Scatter-plots depicting the amplitude of overall post-response negativity as a function of
PSWQ score and hand of error. There is a negative correlation between post-response negativity
and PSWQ for right-handed errors but not left-handed errors. As post-response negativity scores
are negative and PSWQ scores are positive, the negative correlation indicates that as PSWQ
score increases, so does post-response negativity amplitude and vice versa.1

1

Removal of the participant in the lower right quadrant of the right scatter-plot does not
significantly affect the magnitude of the relationship between worry and right-handed ERN (r = .28, p = .08).
37

Figure 3. Experiment 2 Response Locked ERPs.

Note: Response-locked ERPs plotted by stimulus orientation and median-split PSWQ scores
separated by hand of error at electrode site FCz.

38

Figure 4. Experiment 2 Scatter Plot.

Note: Scatter-plots depicting the amplitude of overall post-response negativity as a function of
PSWQ score, hand of error, and stimulus orientation. Replicating the results of Experiment 1, a
negative correlation between post-response negativity and PSWQ scores was found only on
right-handed, horizontal trials. As post-response negativity scores are negative and PSWQ scores
are positive, the negative correlation indicates that as PSWQ score increases, so does postresponse negativity amplitude and vice versa.

39

REFERENCES

40

REFERENCES

Annett, M. (2004). Hand preference observed in large healthy samples: classification, norms and
interpretations of increased non-right-handedness by the right shift theory. British
Journal of Psychology, 95, 339-353.
Ashton, R., & McFarland, K. (1991). A simple dual-task study of laterality, sex-differences and
handedness. Cortex, 27, 105-109.
Awh, E., Jonides, J., Smith, E. E., Schumacher, E. H., Koeppe, R. A., & Katz, S. (1996).
Dissociation of storage and rehearsal in verbal working memory: Evidence from positron
emission tomography. Psychological Science, 7, 25-31.
Barlow, D. H. (1991). Disorders of emotion. Psychological Inquiry, 2, 58-71.
Barlow, D. H. (2002). Anxiety and Its Disorders, 2nd Edn. New York, NY: Guilford Press.
Beilock, S. L., Rydell, R. J., & McConnell, A. R. (2007). Stereotype threat and working
memory: Mechanisms, alleviation, and spillover. Journal of Experimental Psychology:
General, 136(2), 256-276.
Bishop, S., Duncan, J., Brett, M., & Lawrence, A. D. (2004). Prefrontal cortical function and
anxiety: Controlling attention to threat-related stimuli. Nature Neuroscience, 7, 184-188.
Braver, T. S. (2012). The variable nature of cognitive control: a dual mechanisms framework.
Trends in Cognitive Sciences, 16, 106-113.
Buccino, G., Riggio, L., Melli, G., Binkofski, F., Gallese, V. & Rizzolatti, G. (2005). Listening
to action-related sentences modulates the activity of the motor system: A combined TMS
and behavioral study. Cognitive Brain Research, 24(3), 355-363.
Buchsbaum, M., & Fedio, P. (1970). Hemispheric differences in evoked potentials to verbal and
nonverbal stimuli in the left and right visual fields. Physiology & Behavior, 5(2), 207210.
Chang, P., & Hammond, G. R. (1987). Mutual interactions between speech and finger
movements. Journal of Motor Behavior, 19, 265-268.
Cohen, J. (1973). Eta-squared and partial eta-squared in fixed factor ANOVA designs.
Educational and Psychological Measurement, 33, 107–112.
Cohen, J. (1988). Statistical Power of Analysis for the Behavioral Sciences (2nd ed). Hillsdale,
NJ: Lawrence Erlbaum Associates, Inc.
Cumming, G. (2013). The new statistics: Why and how. Psychological Science, 1-23.
41

DeCaro, M. S., Rotar, K. E., Kendra, M. S., & Beilock, S. L. (2010). Diagnosing and alleviating
the impact of performance pressure on mathematical problem solving. The Quarterly
Journal of Experimental Psychology: Human Experimental Psychology, 63(8), 16191630.
Dehaene, S., Posner, M. I., & Tucker, D. M. (1994). Localization of a neural system for error
detection and compensation. Psychological Science, 5, 303–305.
Engels, A. S., Heller, W., Mohanty, A., Herrington, J. D., Banich, M. T., Webb, A. G., . . .
Miller, G. A. (2007). Specificity of regional brain activity in anxiety types during
emotion processing. Psychophysiology, 44, 352–363.
Engels, A. S., Heller, W., Spielberg, J. M., Warren, S. L., Sutton, B. P., Banich, M. T., & Miller,
G. A. (2010). Co-occurring anxiety influences patterns of brain activity in depression.
Journal of Cognitive Affective, & Behavioral Neuroscience. 10, 141–156.
Eriksen, B. A., & Eriksen, C. W. (1974). Effects of noise letters upon the identification of a
target letter in a nonsearch task. Perception & Psychophysics, 16, 143–149.
Eysenck, M. W., & Calvo, M. G. (1992). Anxiety and performance: The processing efficiency
theory. Cognition and Emotion, 6, 409-434.
Eysenck, M. W., Derakshan, N., Santos, R., & Calvo, M. G. (2007). Anxiety and cognitive
performance: Attentional control theory. Emotion, 7, 336–353.
Eysenck, M.W., & Derakshan, N. (2011). New perspectives in attentional control theory.
Personality and Individual Differences, 50, 955–960.
Fletcher, P. C., & Henson, R. N. A. (2001). Frontal lobes and human memory: Insights from
functional neuroimaging. Brain, 124, 849–881.
Gehring, W. J., Himle, J., & Nisenson, L. G. (2000). Action-monitoring dysfunction in
obsessive–compulsive disorder. Psychological Science, 11, 1–6.
Gehring, W. J., Liu, Y., Orr, J. M., & Carp, J. (2012). "The error-related negativity (ERN/Ne)",
in Oxford Handbook of Event-Related Potential Components, eds S. J. Luck and E.
Kappenman (New York, NY: Oxford University Press), 231-291.
Gratton, G., Coles, M. G. H., & Donchin, E. (1983). A new method for off-line removal of
ocular artifact. Electroencephalography and Clinical Neurophysiology, 55, 468–484.
Hajcak, G. (2012). What we’ve learned from our mistakes: Insights from error-related brain
activity. Current Directions in Psychological Science, 21(2), 101-106.
Hajcak, G., Franklin, M. E., Foa, E. B., & Simons, R. F. (2008). Increased error-related brain
42

activity in pediatric OCD before and after treatment. The American Journal of
Psychiatry, 165, 116-123.
Hajcak, G., McDonald, N., & Simons, R. F. (2003). Anxiety and error-related brain activity.
Biological Psychology, 64, 77-90.
Hammond, G. (2002). Correlates of human handedness in primary motor cortex: A review and
hypothesis. Neuroscience & Biobehavioral Reviews, 26(3), 285-292.
Hayes, S., Hirsch, C. R., & Mathews, A. (2008). Restriction of working memory capacity during
worry. Journal of Abnormal Psychology, 117, 712–717.
Heller, W., Nitschke, J. B., Etienne, M. A., & Miller, G. A. (1997). Patterns of regional brain
activity differentiate types of anxiety. Journal of Abnormal Psychology, 106(3), 376-385.
Hellige, J. B., & Yamauchi, M. (1990). Quantitative and qualitative hemispheric asymmetry for
processing Japanese Kana. Brain and Cognition, 40, 453-463.
Hochman, E. Y., Orr, J. M., & Gehring, W. J. (2014). Toward a more sophisticated response
representation in theories of medial frontal performance monitoring: The effects of motor
similarity and motor asymmetries. Cerebral Cortex, 24(2), 414-425.
Hofmann, S. G., Moscovitch, D. A., Litz, B. T., Kim, H. J., Davis, L. L., & Pizzagalli, D. A.
(2005). The worried mind: Autonomic and prefrontal activation during
worrying. Emotion, 5(4), 464-475.
Holroyd C. B., & Coles M. G. (2002). The neural basis of human error processing:
Reinforcement learning, dopamine, and the error-related negativity. Psychological
Review, 109, 679-709.
Hotelling, H. (1940). The selection of variates for use in prediction, with some comments on the
general problem of nuisance parameters. Annals of Statistics, 11, 271-283.
Jordan, T. R., & Patching, G. R. (2003). Assessing effects of stimulus orientation on perception
of lateralized words and nonwords. Neuropsychologia, 41, 1693-1702.
Johannes, S., Wieringa, B. M., Nager, W., Rada, D., Dengler, R., Emrich, H. M., . . . . Dietrich,
D. E. (2001). Discrepant target detection and action monitoring in obsessive-compulsive
disorder. Psychiatry Research: Neuroimaging, 108, 101-110.
Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., Walters, E. E. (2005).
Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National
Comorbidity Survey Replication. Archives of General Psychiatry, 62, 593-602.
Kessler, R. C., Petukhova, M., Sampson, N. A., Zaslavsky, A. M., & Wittchen, H. U. (2012).
Twelve-month and lifetime prevalence and lifetime morbid risk of anxiety and mood
43

disorders in the US. International Journal of Methods in Psychiatric Research, 21(3),
169-184.
Kinsbourne, M., & Cook, J. (1971). Generalized and lateralized effects of concurrent
verbalization on a unimanual skill. Quarterly Journal of Experimental Psychology, 23,
341-345.
Knecht, S., Dräger, B., Deppe, M., Bobe, L., Lohmann, H., Flöel, A., . . . . Henningsen, H.
(2000). Handedness and hemispheric language dominance in healthy humans. Brain,
123(12), 2512-2518.
Leigh, E., & Hirsch, C. R. (2011). Worry in imagery and verbal form: Effect on residual working
memory capacity. Behaviour Research and Therapy, 49(2), 99-105.
Lindell, A. K., Nicholls, M. E. R., & Castles, A. E. (2003). The effect of orthographic
uniqueness and deviation points on lexical decisions: Evidence from unilateral and
bilateral-redundant presentations. The Quarterly Journal of Experimental Psychology,
56A, 287-307.
Lindell, A. K. (2006). In your right mind: Right hemisphere contributions to language processing
and production. Neuropsychology Review, 16(3), 131-148.

Meister, I.G., Boroojerdi, B., Foltys, H., Sparing, R., Huber, W. & Topper, R. (2003). Motor
cortex hand area and speech: implications for the development of language.
Neuropsychologia, 41(4), 401-406.
Meyer, T. J., Miller, M. L., Metzger, R. L., & Borkovec, T. D. (1990). Development and
validation of the Penn State Worry Questionnaire. Behaviour Research and Therapy, 28,
487–495.
Mo, L., Xu, G., Kay, P., & Tan, L. (2011). Electrophysiological evidence for the left-lateral
effect of language on preattentive categorical perception of color. Proceedings of the
National Academy of Sciences, 108 (34), 14026-30.
Moran, T. P., Taylor, D., & Moser, J. S. (2012). Sex moderates the relationship between worry
and performance-monitoring brain activity in undergraduates. International Journal of
Psychophysiology, 85(2), 188-194.
Moser, J. S., Hajcak, G., & Simon, R. F. (2005). The effects of fear on performance monitoring
and attentional allocation. Psychophysiology, 42, 261-268.
Moser, J. S., Moran, T. P., & Jendrusina, A. A. (2012). Parsing relationships between dimensions
of anxiety and action monitoring brain potentials in female undergraduates.
Psychophysiology, 49, 3-10.

44

Moser, J. S., Moran, T. P., Schroder, H. S., Donnellan, M. B., & Yeung, N. (2013). On the
relationship between anxiety and error monitoring: A meta-analysis and conceptual
framework. Frontiers in Human Neuroscience, 7, 1-19.
Nakagawa, A., & Sukigara, M. (2000). Visual word form familiarity and attention in lateral
difference during processing of Japanese Kana words. Brain and Language, 74, 223-237.
Nitschke, J. B., Heller, W., Imig, J. C., McDonald, R. P., & Miller, G. A. (2001). Distinguishing
dimensions of anxiety and depression. Cognitive Therapy and Research, 25, 1–22.
Nitschke, J. B., Heller, W., Palmieri, P. A., & Miller, G. A. (1999). Contrasting patterns of brain
activity in anxious apprehension and anxious arousal. Psychophysiology, 36, 628–637.
Oathes, D. J., Bruce, J. M., & Nitschke, J. B. (2008). Worry facilitates corticospinal motor
response to transcranial magnetic stimulation. Depression & Anxiety, 25, 969-976.
Oliveri, M., Finocchiaro, C., Shapiro, K., Gangitano, M., Caramazza, A., & Pascual-Leone, A.
(2004). All talk and no action: A transcranial magnetic stimulation study of motor cortex
activation during action word production. Journal of Cognitive Neuroscience, 16(3), 374381.
Olvet, D. M., & Hajcak, G. (2008). The error-related negativity (ERN) and psychopathology:
Toward an endophenotype. Clinical Psychology Review, 28, 1343-1354.
Pulvermüller, F. (2005). Brain mechanisms linking language and action. Nature Reviews
Neuroscience, 6, 576-582.
Pulvermüller, F., Shtyrov, Y. & Ilmoniemi, R. (2005a). Brain signatures of meaning access in
action word recognition. Journal of Cognitive Neuroscience, 17(6), 1-9.
Rapee, R. M. (1993). The utilisation of working memory by worry. Behaviour Research and
Therapy, 31, 617-620.
Schroder, H.S., & Moser, J.S. (2014). Improving the study of error monitoring with
consideration of behavioral performance measures. Frontiers in Human Neuroscience, 8:
178.
Seyal, M., Mull, B., Bhullar, N., Ahmad, T., & Gage, B. (1999). Anticipation and execution of a
simple reading task enhance corticospinal excitability. Clinical Neurophysiology, 110,
424–429.
Seymour, T. L., & Schumacher, E. H. (2009). Electromyographic evidence for response conflict
in the exclude recognition task. Cognitive, Affective, & Behavioral Neuroscience, 9(1),
71-82.

45

Silton, R. L., Heller, W., Towers, D. N., Engels, A. S., Edgar, J. C., Spielberg, J. M., . . . Miller,
G. A. (2011). Depression and anxiety distinguish frontocingulate cortical activity during
top-down attentional control. Journal of Abnormal Psychology, 120, 272–285.
Simon-Thomas, E. R., Role, K. O., & Knight, R. T. (2005). Behavioral and electrophysiological
evidence of a right hemisphere bias for the influence of negative emotion on higher
cognition. Journal of Cognitive Neuroscience, 17(3), 517-529.
Spielberg, J. M., De Leon, A. A, Bredemeier, K., Heller, W., Engels, A. S., Warren, S. L., . . .
Miller, G. A. (2013). Anxiety type modulates immediate versus delayed engagement of
attention-related brain regions. Brain and Behavior, 3(5), 532–51.
Tokimura, H., Tokimura, Y., Oliviero, A., Asakura, T., & Rothwell, J. C. (1996).
Speech-induced changes in corticospinal excitability. Annals of Neurology, 40, 628-634.
Trbovich, P. L., & LeFevre, J. A. (2003). Phonological and visual working memory in mental
addition. Memory & Cognition, 31(5), 738-745.
Ullsperger, M., & von Cramon, D. (2001). Subprocesses of performance monitoring: A
dissociation of error processing and response competition revealed by event-related fMRI
and ERPs. NeuroImage, 14, 1387-1401.
van Veen, V., & Carter, C.S. (2002). The timing of action-monitoring processes in the anterior
cingulate cortex. Journal of Cognitive Neuroscience, 14, 593–602.
Vocat, R., Pourtois, G., & Vuilleumier, P. (2008). Unavoidable errors: A spatio-temporal
analysis of time-course and neural sources of evoked potentials associated with error
processing in a speeded task. Neuropsychologia, 46, 2545-2555.
Warren, S. L., Crocker, L. D., Spielberg, J. M., Engels, A. S., Banich, T., Sutton, B. P., . . .
Miller, G. A. (2013). Cortical organization of inhibition-related functions and modulation
by psychopathology. Frontiers of Human Neuroscience, 7, 1-13.
Watson, D., & Clark, L. A. (1991). The Mood and Anxiety Symptom Questionnaire.
Unpublished manuscript. University of Iowa, Iowa City, IA.
Weinberg, A., Klein, D. N., & Hajcak, G. (2012). Increased error-related brain activity
distinguishes generalized anxiety disorder with and without comorbid major depressive
disorder. Journal of Abnormal Psychology, 121, 885-896.
Weinberg, A., Olvet, D. M., Hajcak, G. (2010). Increased error-related brain activity in
generalized anxiety disorder. Biological Psychology, 85, 472–480.
Williams, M, O., Mathews, A., & Hirsch, C. R. (2014). Verbal worry facilitates attention to
threat in high-worriers. Journal of Behavior Therapy and Experimental Psychiatry, 45, 814.
46

Yeung, N., Botvinick, M. M., & Cohen, J. D. (2004). The neural basis of error detection:
Conflict monitoring and the error-related negativity. Psychological Review, 111, 939959.
Yeung, N., & Cohen, J. D. (2006). The impact of cognitive deficits on conflict monitoring:
Predictable dissociation between the ERN and N2. Psychological Science, 17, 164-171.
Yeung, N, & Summerfield, C. (2012). Metacognition in human decision making: Confidence and
error monitoring. Philosophical Transactions of the Royal Society B, 367, 1310-21.
Zatorre, R. J., Meyer, E., Gjedde, A., & Evans, A. C. (1996). PET studies of phonetic processing
of speech: Review, replication, and reanalysis. Cerebral Cortex, 6, 21–30.

47