doi:10.1093/scan/nsr007
SCAN (2012) 7, 289^295
The error-related negativity relates to sadness
following mood induction among individuals
with high neuroticism
Doreen M. Olvet1,2 and Greg Hajcak3
1
The Zucker Hillside Hospital, Psychiatry Research, North Shore – Long Island Jewish Health System, Glen Oaks, NY, 11004, 2The Feinstein
Institute for Medical Research, North Shore – Long Island Jewish Health System, Manhasset, NY, 11030 and 3Department of Psychology,
Stony Brook University, Stony Brook, NY, 11794, USA
The error-related negativity (ERN) is an event-related potential (ERP) that indexes error monitoring. Research suggests that the
ERN is increased in internalizing disorders, such as depression and anxiety. Although studies indicate that the ERN is insensitive
to state-related fluctuations in anxiety, few studies have carefully examined the effect of state-related changes in sadness on the
ERN. In the current study, we sought to determine whether the ERN would be altered by a sad mood induction using a
between-subjects design. Additionally, we explored if this relationship would be moderated by individual differences in neuroticisma personality trait related to both anxiety and depression. Forty-seven undergraduate participants were randomly assigned
to either a sad or neutral mood induction prior to performing an arrow version of the flanker task. Participants reported greater
sadness following the sad than neutral mood induction; there were no significant group differences on behavioral or ERP measures. Across the entire sample, however, participants with a larger increase in sad mood from baseline to post-induction had a
larger (i.e. more negative) ERN. Furthermore, this effect was larger among individuals reporting higher neuroticism. These data
indicate that neuroticism moderates the relationship between the ERN and changes in sad mood.
Keywords: event-related potential; error-related negativity; emotion; affect; sad mood induction
INTRODUCTION
The error-related negativity (ERN) is a negative waveform in
the event-related potential (ERP) that occurs 50 ms after
an erroneous response (Falkenstein et al., 1991; Gehring
et al., 1993). In general, the ERN is thought to represent
error monitoring, and contemporary models focus on functions that include reinforcement learning (Holroyd and
Coles, 2002) and conflict monitoring (Yeung et al., 2004).
Studies also indicate that the ERN is influenced by motivational and affective factors. For example, the ERN is
increased when errors are more monetarily valuable or personally significant (Gehring et al., 1993; Hajcak et al., 2005;
Kim et al., 2005). The ERN is also larger among individuals
with anxiety disorders (Gehring et al., 2000; Johannes et al.,
2001; Ruchsow et al., 2005; Ladouceur et al., 2006; Endrass
et al., 2008; Hajcak et al., 2008) and individuals endorsing
high levels of negative affect and related traits (Luu et al.,
2000; Hajcak et al., 2003, 2004; Boksem et al., 2006). In
contrast to the impact of trait anxiety, the ERN appears insensitive to changes in state levels of anxiety. For example,
one study exposed individuals who were spider phobic to a
tarantula while performing a flanker task; the ERN did not
change as a result of this rather intense provocation of
Received 1 August 2010; Accepted 1 February 2011
Advance Access publication 7 March 2011
Correspondence should be addressed to Doreen M. Olvet, The RAP Program, 75-59 263rd Street, Glen
Oaks, NY 11004, USA. E-mail: [email protected]
anxiety (Moser et al., 2005). Other studies have demonstrated that the ERN does not change after successful treatment for anxiety disorders (Ladouceur et al., 2007; Hajcak
et al., 2008).
The relationship between the ERN and depression is less
straightforward than the relationship between the ERN and
anxiety. Some studies have demonstrated an increased ERN
among individuals with major depressive disorder (MDD;
Chiu and Deldin, 2007; Holmes and Pizzagalli, 2008,
2010). However, other recent studies report that severe depression may result in a reduced ERN (Schrijvers et al., 2008,
2009; Olvet et al., 2010). In light of these findings, it has been
suggested that the relationship between the ERN and MDD
is non-linear (Schrijvers et al., 2008; Olvet et al., 2010) such
that mild to moderate depression is related to an increased
ERN, whereas more severe depression is related to a reduced
ERN (Olvet et al., 2010).
Two studies have reported changes in the ERN in the context of viewing emotional pictures used to induce short-term
changes in emotion (Larson et al., 2006; Wiswede et al.,
2009). Larson and colleagues found an increased ERN on
trials that included pleasant pictures (Larson et al., 2006),
whereas Wiswede and colleagues reported an increased ERN
on trials that followed the presentation of unpleasant pictures, but only in the first half of the task (Wiswede et al.,
2009). More recently, van Wouwe and colleagues (2010)
presented either a positive (i.e. scenes from The Lion King
ß The Author (2011). Published by Oxford University Press. For Permissions, please email: [email protected]
290
SCAN (2012)
or The Little Mermaid) or neutral (i.e. street scenes) film clip
to participants prior to performing a continuous performance task and found that the ERN was increased in subjects
in the positive condition. Although these studies suggest that
the presentation of pleasant and unpleasant pictures can
modulate the ERN, quantitative measures of mood were
not obtained. Furthermore, the contrary results and variability in methodology in these studies make the results difficult
to integrate.
In the current study, we focused specifically on sadness as
a discrete emotion linked to depression, and sought to determine if the ERN would be altered by a sad mood induction using a between-subjects design. Based on the literature
documenting enhanced ERNs in mild-to-moderate depression (Chiu and Deldin, 2007; Holmes and Pizzagalli, 2008),
we hypothesized that the induction of sad mood would lead
to an increased (i.e. more negative) ERN. Further, we predicted that those subjects who became sadder would demonstrate the largest increase in ERN magnitude. We also
explored whether this effect would be moderated by individual differences in neuroticism. Neuroticism reflects the tendency to experience negative emotions, and is a risk factor
for both anxiety and depression (Hettema et al., 2006). Thus,
individuals high in neuroticism are at increased risk for
internalizing psychopathology, are more emotionally reactive, and might be characterized by a larger increase in the
ERN following a sad mood manipulation.
METHODS
Participants
Forty-seven undergraduates (27 males, 20 females) participated in the current study. Participants completed four tasks
in a fixed order during the course of this study, including the
flanker task and a gambling task. Data from the gambling
task has been published elsewhere (Foti and Hajcak, 2010).
For the flanker task, data from one participant was lost due
to experimenter error and one participant who made less
than six errors was excluded (Olvet and Hajcak, 2009); therefore, the final sample consisted of 45 participants (20 females). Reaction time (RT) data from one subject was
missing due to experimenter error. Twenty-two participants
[11 females; age: mean ¼ 18.41; standard deviation (s.d.) ¼
1.05] were randomly assigned to the neutral mood induction, and 23 participants (9 females; age: mean ¼ 18.74;
s.d. ¼ 1.29) were randomly assigned to the sad mood induction. The two groups were comparable with respect to age
[t(43) ¼ 0.94, P ¼ 0.35], gender [2(1) ¼ 0.54, P ¼ 0.46]
and ethnicity [2(4) ¼ 4.29, P ¼ 0.37]. Informed consent
was obtained from participants prior to the experiment.
This research was formally approved by the Stony Brook
University Institutional Review Board. No participants discontinued their participation in the experiment once procedures had begun, and all participants received course
credit for their participation.
D. M. Olvet and G. Hajcak
Mood induction
The sad and neutral mood induction paradigms were based
on the guidelines provided by Rottenberg et al. (2007) for
using film clips to elicit discrete emotions. Each mood induction consisted of 2–5-min film clips and a song that was
played in the background throughout the entire experiment.
In the neutral mood induction, the film clips used were from
Alaska’s Wild Denali, and the song used was Robert Ronne’s
Meditation No. 19. In the sad mood induction, the film clips
used were from The Champ and My Girl, and the song used
was Gabriel Faure’s Piano Quintet No. 1 in D Minor
(Op. 89). The order of the film clips was counterbalanced
across subjects. Upon completing the computer tasks, a
pleasant mood was induced in all participants using an
amusing film clip of funny cats.
To assess current mood throughout the experiment, the
valence scale of the Self-Assessment Manikin was used (Lang,
1980). Participants were asked to rate their current mood
ranging from one (happy) to nine (sad). This measure was
administered at five points throughout the experiment:
before and after each of the two film clips, and again at
the conclusion of the experiment. For this study, the ratings
of interest were those taken at baseline and immediately
following the first film clip (i.e. immediately before the
flanker task). At the end of the experiment, individuals
filled out the Big Five Inventory (BFI; John and Srivastava,
1999) to assess the personality dimension of neuroticism.
Task and materials
The present task was administered on a Pentium D class
computer, using Presentation software (Neurobehavioral
Systems Inc., Albany, CA, USA) to control the presentation
and timing of all stimuli. The task was an arrow version of
the flanker task (Eriksen and Eriksen, 1974). On each trial,
five horizontally aligned arrowheads were presented. Half of
all trials were compatible (‘< < < < <’ or ‘> > > > >’) and
half were incompatible (‘< < > < <’ or ‘> > < > >’) and the
order of compatible and incompatible trials was random. All
stimuli were presented for 200 ms followed by an interval
that varied randomly from 2300 to 2800 ms.
Procedure
Following a brief description of the experiment, electroencephalograph sensors were attached. Participants viewed
the film clip (with pre- and post-mood ratings) and then
performed the flanker task followed by another task, an additional film clip, and two other tasks. In the flanker task,
participants were instructed to press the right mouse
button if the center arrow was facing to the right and to
press the left mouse button if the center arrow was facing
to the left. Participants performed a practice block containing 30 trials and were instructed to be both as accurate and
fast as possible. The actual task consisted of 10 blocks of 30
trials (300 trials total) with each block initiated by the participant. Feedback was given at the end of each block to
ERN relates to sadness
maintain performance between 75% and 90% accuracy. The
task lasted for 10 min. Participants filled out the self-report
questionnaire at the end of the experiment.
Psychophysiological recording, data reduction and
analysis
The continuous EEG activity was recorded using an
ActiveTwo head cap and the ActiveTwo BioSemi system
(BioSemi, Amsterdam, The Netherlands). The EEG was
pre-amplified at the electrode with a gain of one; EEG data
was digitized at 64-bit resolution with a sampling rate of
512 Hz using a low-pass fifth-order sinc filter with a
half-power cutoff of 102.4 Hz using ActiView Software
(BioSemi). Recordings were taken from 64 scalp electrodes
based on the 10/20 system, as well as two electrodes placed
on the left and right mastoids. The electro-oculogram (EOG)
generated from blinks and eye movements were recorded
from four facial electrodes: two 1 cm above and below
the participant’s right eye (vertical EOG bipolar recording),
one 1 cm to the left of the left eye and one 1 cm to the
right of the right eye (horizontal EOG bipolar recording). As
per BioSemi’s design, each electrode was measured online
with respect to a common mode sense electrode that
formed a monopolar channel.
Off-line analysis was performed using Brain Vision
Analyzer software (Brain Products, Gilohing, Germany).
EEG data were re-referenced to the numeric mean of
the mastoids and band-pass filtered with cutoffs of 0.1 and
30 Hz. The EEG was segmented for each trial, beginning
400 ms before each response and continuing for 1000 ms.
The EEG was corrected for blinks and eye movements
using the method developed by Gratton et al. (1983).
Specific intervals for individual channels were rejected in
each trial using a semi-automated procedure, with physiological artifacts identified by the following criteria: a voltage
step of >50.0 mV between sample points, a voltage difference
of 300.0 mV within a trial, and a maximum voltage difference
of <0.50 mV within 100 ms intervals; all segments were also
visually inspected for additional artifacts.
Response-locked average ERPs were computed for correct
and error trials. The ERN and correct response negativity
(CRN) were quantified on error and correct trials, respectively, as a peak measure (i.e. the most negative peak in a
100 to 100 ms window relative to response onset) at electrode site FCz, where the ERN was maximal.1 The error
positivity (Pe) and correct positivity (Pc) were evaluated on
error and correct trials, respectively, as the average activity
from 200 to 400 ms at Pz following response onset. A 200 ms
window from 400 to 200 ms prior to response onset
served as the baseline. Behavioral measures included the
number of error trials for each subject, average reaction
time (RT) on error and correct trials, and post-error slowing
1
An area measure of the ERN and the CRN was also scored on error and correct trials, respectively, from
response onset to 100 ms (i.e. 0–100 ms) at FCz.
SCAN (2012)
291
(PES; RT on correct trials after errors minus RT on correct
trials after correct responses).
Statistical analysis
All data was checked for normality prior to analysis. The
only measure that resulted in skewness and kurtosis scores
that deviated from normality was the Pe (both on error and
correct trials), which was due to two subjects with scores that
were outliers (i.e. 5 s.d. from the mean in both cases).
Removing these data-points resulted in normality statistics
that were below 2 with no change in the overall pattern of
results.
In all cases, behavioral and ERP data were statistically
evaluated using SPSS General Linear Model software
(Version 16.0; SPSS Inc., Chicago, IL, USA). For state
mood scores, a 2 (group: sad vs neutral mood induction) 2
(time: before vs after induction) mixed model analysis of
variance (ANOVA) was used to detect differences between
the groups. For the behavioral and ERP measures, a 2
(group: sad vs neutral mood induction) 2 (trial type: correct vs error trial) mixed model ANOVA was used to detect
differences between the groups. Independent sample t-tests
were performed for follow-up post-hoc comparisons. Group
differences on post-error slowing and neuroticism scores
were evaluated with a one-way ANOVA. The Pearson correlation coefficient (r) was used to examine the relationship
between ERPs, neuroticism scores and mood ratings
across groups. Fisher’s z-test was used to examine differences
between correlation coefficients.
A moderated multiple regression (MMR) analysis (Aiken
and West, 1991) was used to test whether neuroticism moderated the relationship between the change in mood and the
ERN minus the CRN (ERN) in the entire sample. First,
continuous independent variables (i.e. change in mood
score and neuroticism) were centered by subtracting the
mean from each of the individual datapoints. Second, an
interaction term was created by multiplying the centered
values of change in mood score by neuroticism scores.
Finally, for the MMR, mood condition (i.e. sad vs neutral
condition), change in mood scores and neuroticism were
entered in step 1, and mood condition, change in mood
score, neuroticism and the interaction term were entered in
step 2.
RESULTS
Self-report ratings
Mean neuroticism scores, and mood scores before and after
the presentation of the film clips are presented in Table 1. The
neutral and sad mood groups did not differ on neuroticism
scores [F(1,43) <1]. Participants reported being sadder following both films [F(1,43) ¼ 22.41, P < 0.001], and those assigned to the sad film had higher sadness ratings overall
[F(1,43) ¼ 27.78, P < 0.001]; importantly, there was an interaction between the two factors [F(1,43) ¼ 18.18, P < 0.001]
such that change in mood scores (i.e. post-induction mood
292
SCAN (2012)
D. M. Olvet and G. Hajcak
Table 1 Trait neuroticism, mood ratings, performance and ERP data means
(and s.d.’s)
Sad
condition
(N ¼ 23)
Trait neuroticism scores
Neuroticism
Mood ratings
Pre-induction
Post-induction
Performance
Error reaction time (ms)
Correct reaction time (ms)
Post-error slowing (ms)
No. of errors
ERPs
ERN Peak (mV)
CRN Peak (mV)
ERN Peak (mV)
Pe Area (mV)
Pc Area (mV)
Neutral
condition
(N ¼ 22)
23.53 (5.97)
22.36 (6.99)
3.87 (1.36)
5.61 (1.34)
3.14 (0.99)
3.23 (0.97)
329
392
19
32
337
398
14
36
3.90
7.41
13.35
16.24
3.83
(34)
(32)
(28)
(13)
(9.24)
(8.30)
(7.61)
(8.97)
(7.28)
1.12
9.07
11.55
19.24
4.45
(36)
(37)
(34)
(15)
(6.84)
(6.42)
(4.73)
(14.38)
(5.60)
scores minus baseline mood scores) were larger among individuals who watched the sad film [t(43) ¼ 4.32, P < 0.001).
Correlations between neuroticism and change in mood scores
did not reach significance in the entire sample (r ¼ 0.11,
P ¼ 0.49), nor in the neutral (r ¼ 0.07, P ¼ 0.77) and sad
mood groups separately (r ¼ 0.39, P ¼ 0.07).
Behavioral performance
Accuracy, RT and PES data are presented in Table 1.
Participants were faster on error than correct trials
[F(1,42) ¼ 184.31, P < 0.001], however, the groups did not
differ in RT [F(1,42) <1], nor was there a significant interaction between trial type and group [F(1,42) <1].
Participants in the neutral and sad mood groups did not
differ on PES [F(1,42 <1] or number of errors
[t(43) ¼ 1.03, P ¼ 0.31]. Correlations between neuroticism
and performance measures did not reach significance in
the entire sample (error reaction time: r ¼ 0.06, P ¼ 0.72;
correct reaction time: r ¼ 0.08, P ¼ 0.62; PES: r ¼ 0.15,
P ¼ 0.32; number of errors: r ¼ 0.01, P ¼ 0.94), nor in the
neutral (error reaction time: r ¼ 0.13, P ¼ 0.58; correct reaction time: r ¼ 0.12, P ¼ 0.59; PES: r ¼ 0.27, P ¼ 0.23; number
of errors: r ¼ 0.06, P ¼ 0.79) and sad mood groups separately
(error reaction time: r ¼ 0.00, P ¼ 0.99; correct reaction time:
r ¼ 0.32, P ¼ 0.14; PES: r ¼ 0.03, P ¼ 0.91; number of
errors: r ¼ 0.06, P ¼ 0.79).
ERPs
Grand average response-locked ERPs at FCz and scalp topographies of the ERN are presented in Figure 1, and the average ERP values are presented in Table 1. The ERN was
significantly more negative than the CRN [F(1,43) ¼ 129.19,
P < 0.001], and individuals in the neutral and sad mood
condition did not differ from one another on ERN/CRN
amplitudes [F(1,43) ¼ 1.09, P ¼ 0.30]; the trial type group
interaction was not significant [F(1,43) < 1).2 The Pe was
significantly more positive than the Pc [F(1,43) ¼ 109.93,
P < 0.001), and neither the main effect of group [F(1,43)
<1] nor the trial type group interaction [F(1,43) <1] was
significant.
Across the entire sample, there was a significant correlation between the change in mood scores (i.e. post-induction
mood scores minus baseline mood scores) and the ERN
minus the CRN (i.e. ERN; r ¼ 0.40, P < 0.01), such that
a larger change in mood score (i.e. a greater increase in
sadness) related to a larger ERN (see Figure 2).3 The correlation between the ERN and change in mood score was
significant in the sad mood condition (r ¼ 0.42, P < 0.05)
and there was a trend toward a similar relationship in the
neutral mood condition (r ¼ 0.30, P ¼ 0.17) as well. Using
a Fisher’s z-test, the difference between the correlation coefficients was non-significant (z ¼ 0.41, P ¼ 0.68). Thus, across
both groups, those participants who became sadder following
the mood induction were characterized by larger (i.e. more
negative) error-related brain activity. This finding was driven
by the relationship between the post-induction mood scores
and the ERN (r ¼ 0.31, P < 0.05), not the pre-induction
mood scores and the ERN (r ¼ 0.07, P ¼ 0.64). There was
also a trend towards a significant relationship between the
ERN and the change in mood scores (r ¼ 0.28, P ¼ 0.07).
Neither the pre- nor post-induction mood scores correlated
with the CRN (pre-induction: r ¼ 0.14, P ¼ 0.37; postinduction: r ¼ 0.13, P ¼ 0.40). Correlations between neuroticism and ERP measures did not reach significance in the
entire sample (ERN: r ¼ 0.15, P ¼ 0.32; CRN: r ¼ 0.13,
P ¼ 0.41; ERN: r ¼ 0.11, P ¼ 0.49; Pe: r ¼ 0.11, P ¼ 0.47;
Pc: r ¼ 0.09, P ¼ 0.58), nor in the neutral (ERN: r ¼ 0.18,
P ¼ 0.43; CRN: r ¼ 0.12, P ¼ 0.60; ERN: r ¼ 0.21, P ¼ 0.35;
Pe: r ¼ 0.17, P ¼ 0.45; Pc: r ¼ 0.23, P ¼ 0.30) or sad mood
groups separately (ERN: r ¼ 0.18, P ¼ 0.42; CRN: r ¼ 0.17,
P ¼ 0.45; ERN: r ¼ 0.08, P ¼ 0.73; Pe: r ¼ 0.03, P ¼ 0.89;
Pc: r ¼ 0.05, P ¼ 0.83).
A moderated multiple regression (MMR) analysis was
used (Aiken and West, 1991) to test whether neuroticism
moderated the relationship between the change in mood
and the ERN in the entire sample. Data from the MMR
analysis is presented in Table 2. The MMR analysis revealed
that neuroticism scores significantly moderated the relationship between change in mood scores and the ERN
[Figure 3; F(4,40) ¼ 3.48, P < 0.05). Among individuals
2
Using an area measure of the ERN and the CRN, the ANOVA results are consistent with the peak data. The
ERN is significantly more negative than the CRN [F(1,43) ¼ 126.07, P < 0.001], with no effect of group
[F(1,43) <1], nor a trial type group interaction [F(1,43) <1].
3
Using an area measure, correlations were significant between the change in mood score and the ERN
(r ¼ 0.31, P < 0.05), but only reached a trend significance between post-induction mood scores and the
ERN (r ¼ 0.23, P ¼ 0.13). For the MMR analysis, there was a similar trend that neuroticism scores
moderated the relationship between change in mood scores and the ERN [F(4,40) ¼ 1.84, P ¼ 0.14].
ERN relates to sadness
SCAN (2012)
293
Fig. 1 Response-locked ERPs at FCz for error and correct trials for the sad (top, left) and neutral (top, right) conditions. Response onset occurred at 0 ms and negative is plotted
up. Scalp topography of error-related brain activity from 0 to 100 ms post-response for the sad (bottom, left) and neutral (bottom, right) conditions.
Table 2 Moderated multiple regression analysis statistics predicting ERN
Predictor variables
Step 1
Mood condition
Change in mood
Neuroticism
Step 2
Mood condition
Change in mood
Neuroticism
Change in mood Neuroticism
B
Standard
error B
1.19
1.86
0.48
2.18
0.72
0.14
0.10
0.45*
0.05
0.80
2.17
0.02
0.25
2.10
0.71
0.14
0.12
0.06
0.52**
0.02
0.32*
R2 ¼ 0.17 for step 1; R2 ¼ 0.09 for step 2 (P’s 0.05).
*P < 0.05; **P < 0.01.
who scored low in neuroticism, the ERN was not related to
change in mood. However, among individuals who scored
high in neuroticism, those reporting a greater increase in
sadness had a larger ERN than those who reported a smaller increase in sadness. The unstandardized simple slope for
high neuroticism (þ1 s.d.) was 3.68 (r ¼ 0.88) and the
unstandardized simple slope for low neuroticism (1 s.d.)
was 0.38 (r ¼ 0.09). There was no significant effect of
condition in the model.
Fig. 2 Scatterplot depicting the relationship between ERN and changes in sad
mood ratings. Participants in the sad mood condition are indicated by open circles
and participants in the neutral mood condition are indicated by closed circles. Larger
numbers indicate a greater increase in sadness.
DISCUSSION
The sad mood induction appeared successful: those individuals who viewed a sad film and listened to sad music reported increased sadness compared to participants who
viewed a neutral film and listened to neutral music.
Although errors elicited an ERN, error-related brain activity
did not differ between subjects in the sad and neutral mood
294
SCAN (2012)
Fig. 3 Scatterplot depicting the relationship between ERN and changes in sad
mood ratings as a function of trait neuroticism scores.
induction conditions. There was, however, a significant correlation between self-reported change in sadness and the
ERN across the entire sample: greater change in sadness
from pre- to post-mood induction was related to increased
error-related brain activity.
In the sad condition, 18 out of 23 (78%) individuals
reported an increase in sadness, and in the neutral condition,
4 out of 22 (18%) individuals reported an increase in
sadness. Thus, not all participants assigned to the sad
mood condition became sadder, whereas some of the participants assigned to the neutral mood induction became
sadder. It is important to note that the relationship between
sadness and ERN was specific to the change in mood
ERN was not related to baseline mood. Thus, increased
error-related brain activity predicted changes in mood. One
possibility is that the ERN may relate to individual differences in the propensity to become sad following exposure to
both neutral and sad stimuli. For example, the neutral film
clip could have had the effect of reducing a more pleasant
mood to a more neutral one, which is possibly equivalent to
an increase in sad mood if you assume this bipolar dimension. The existing literature suggests that both mildly and
moderately depressed individuals are characterized by
increased error-related brain activity (Chiu and Deldin,
2007; Holmes and Pizzagalli, 2008, 2010). The current
study is in line with these data: changes in mood were relatively minor and participants who became sadder (regardless
of condition) had an increased ERN.
It is difficult to directly compare the current results with
other studies that have examined the effect of affective picture presentation (Larson et al., 2006; Wiswede et al., 2009)
and a positive mood induction (van Wouwe et al., 2010) on
the ERN. Contrary results were reported in the two studies
that presented affective pictures to induce short-term affective changes on a trial-to-trial basis: pleasant (Larson et al.,
2006) and unpleasant pictures (Wiswede et al., 2009) were
found to precede a larger ERN. Rather than using rapidly
D. M. Olvet and G. Hajcak
presented IAPS to manipulate emotional state, the current
study used a well-established method [i.e. participant’s
viewed sad or neutral film clips] for inducing sad mood
prior to the start of the task so that any differences in behavior or ERP measures could be attributed to the differences in
mood state. In addition, we measured sadness both before
and after our sad mood induction to verify its success.
Van Wouwe (2010) used a children’s cartoon to induce positive affect; however, qualitative mood ratings were only obtained at the end of the experiment; furthermore, the two
groups in this study differed on accuracy, which is known to
affect the size of the ERN (Yeung, 2004; Yeung et al., 2004).
Although individuals high in the personality trait of neuroticism can be more susceptible to negative mood inductions
(Larsen and Ketelaar, 1989), neuroticism was unrelated to
change in mood in the current study. However, neuroticism
scores significantly moderated the relationship between
self-reported change in sadness and the ERN. In particular,
for individuals who scored low in neuroticism, change in
mood and the ERN were unrelated; on the other hand,
there was a strong relationship between mood and the
ERN among individuals who scored high in neuroticism.
Therefore, the largest ERN was observed among individuals who were both high in neuroticism and experienced
the greatest increase in sad mood. Both neuroticism
(Kendler et al., 1993; Hettema et al., 2004; Ormel et al.,
2004) and an increased ERN (Gehring et al., 2000;
Johannes et al., 2001; Ruchsow et al., 2005; Ladouceur
et al., 2006; Chiu and Deldin, 2007; Endrass et al., 2008;
Hajcak et al., 2008; Holmes and Pizzagalli, 2008, 2010)
have been related to clinical anxiety and depression; the current study provides further evidence for their joint role in
explaining even short-term changes in mood. The current
findings raise the intriguing possibility that error-related
brain activity can be used to further differentiate emotional
reactivity among individuals high in neuroticism.
Future studies would need to measure error-related brain
activity both before and after a sad mood induction to determine whether there are transient changes in the ERN
based on neuroticism and the experience of sadnessor
whether the ERN is stable across variation in sadness and
instead can be used to predict sadness in response to emotional challenge as a function of neuroticism. In addition,
future studies might obtain separate ratings of negative and
positive affect, so that the impact of each dimension on the
ERN can be examined separately. Finally, this study was
conducted in a non-clinical sample, and therefore we
cannot determine whether a sad mood manipulation in individuals with MDD or their offspring might produce different results. These are promising avenues for future work
on the ERN.
REFERENCES
Aiken, L.S., West, S.G. (1991). Multiple Regression: Testing and Interpreting
Interactions. Newbury Park, CA: Sage.
ERN relates to sadness
Boksem, M.A., Tops, M., Wester, A.E., Meijman, T.F., Lorist, M.M. (2006).
Error-related ERP components and individual differences in punishment
and reward sensitivity. Brain Research, 1101(1), 92–101.
Chiu, P.H., Deldin, P.J. (2007). Neural evidence for enhanced error detection in major depressive disorder. American Journal of Psychiatry, 164(4),
608–16.
Endrass, T., Klawohn, J., Schuster, F., Kathmann, N. (2008). Overactive
performance monitoring in obsessive-compulsive disorder: ERP evidence
from correct and erroneous reactions. Neuropsychologia, 46(7), 1877–87.
Eriksen, B.A., Eriksen, C.W. (1974). Effects of noise letters on the identification of target letters in a non-search task. Perception and Psychophysics,
16, 143–9.
Falkenstein, M., Hohnsbein, J., Hoormann, J., Blanke, L. (1991). Effects of
crossmodal divided attention on late ERP components. II. Error processing in choice reaction tasks. Electroencephalography and Clinical
Neurophysiology, 78(6), 447–55.
Foti, D., Hajcak, G. (2010). State sadness reduces neural sensitivity to nonrewards versus rewards. Neuroreport, 21(2), 143–7.
Gehring, W.J., Goss, B., Coles, M.G.H., Meyer, D.E., Donchin, E. (1993). A
neural system for error detection and compensation. Psychological Science,
4, 385–90.
Gehring, W.J., Himle, J., Nisenson, L.G. (2000). Action-monitoring dysfunction in obsessive-compulsive disorder. Psychological Science, 11(1),
1–6.
Gratton, G., Coles, M.G., Donchin, E. (1983). A new method for off-line
removal of ocular artifact. Electroencephalography and Clinical
Neurophysiology, 55(4), 468–84.
Hajcak, G., Franklin, M.E., Foa, E.B., Simons, R.F. (2008). Increased
error-related brain activity in pediatric obsessive-compulsive disorder
before and after treatment. The American Journal of Psychiatry, 165(1),
116–23.
Hajcak, G., McDonald, N., Simons, R.F. (2003). Anxiety and error-related
brain activity. Biological Psychology, 64(1–2), 77–90.
Hajcak, G., McDonald, N., Simons, R.F. (2004). Error-related psychophysiology and negative affect. Brain and Cognition, 56(2), 189–97.
Hajcak, G., Moser, J.S., Yeung, N., Simons, R.F. (2005). On the ERN and the
significance of errors. Psychophysiology, 42, 151–60.
Hettema, J.M., Neale, M.C., Myers, J.M., Prescott, C.A., Kendler, K.S.
(2006). A population-based twin study of the relationship between neuroticism and internalizing disorders. The American Journal of Psychiatry,
163(5), 857–64.
Hettema, J.M., Prescott, C.A., Kendler, K.S. (2004). Genetic and environmental sources of covariation between generalized anxiety disorder and
neuroticism. The American Journal of Psychiatry, 161(9), 1581–7.
Holmes, A.J., Pizzagalli, D.A. (2010). Effects of task-relevant incentives on
the electrophysiological correlates of error processing in major depressive
disorder. Cogn Affect Behav Neuroscience, 10(1), 119–28.
Holmes, A.J., Pizzagalli, D.A. (2008). Spatiotemporal dynamics of error
processing dysfunctions in major depressive disorder. Archives of
General Psychiatry, 65(2), 179–88.
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(4), 679–709.
Johannes, S., Wieringa, B.M., Nager, W., et al. (2001). Discrepant target
detection and action monitoring in obsessive-compulsive disorder.
Psychiatry Research, 108(2), 101–110.
John, O.P., Srivastava, S. (1999). The big five trait taxonomy: history, measurement, and theoretical perceptives. In: Pervin, L.A., John, O.P., editors.
Handbook of Personality: Theory and Research. New York: Guilford Press,
pp. 102–38.
Kendler, K.S., Kessler, R.C., Neale, M.C., Heath, A.C., Eaves, L.J. (1993).
The prediction of major depression in women: toward an integrated
etiologic model. The American Journal of Psychiatry, 150(8), 1139–48.
SCAN (2012)
295
Kim, E.Y., Iwaki, N., Uno, H., Fujita, T. (2005). Error-related negativity in
children: effect of an observer. Developmental Neuropsychology, 28(3),
871–83.
Ladouceur, C.D., Dahl, R.E., Birmaher, B., Axelson, D.A., Ryan, N.D.
(2006). Increased error-related negativity (ERN) in childhood anxiety
disorders: ERP and source localization. Journal of Child Psychology and
Psychiatry, 47(10), 1073–82.
Ladouceur, C.D., Dahl, R.E., Birmaher, B., Axelson, D.A., Ryan, N.D.
(2007). Decreased Pe, but not ERN, amplitude following treatment of
children diagnosed with an anxiety disorder: preliminary results.
Psychophysiology, 44, s99.
Lang, P.J. (1980). Self-assessment Manikin. Gainesville, FL: The Center for
Research in Psychophysiology, University of Florida.
Larsen, R.J., Ketelaar, T. (1989). Extraversion, neuroticism and susceptibility
to positive and negative mood induction procedures. Personality and
Individual Differences, 10(12), 1221–8.
Larson, M.J., Perlstein, W.M., Stigge-Kaufman, D., Kelly, K.G.,
Dotson, V.M. (2006). Affective context-induced modulation of the
error-related negativity. Neuroreport, 17(3), 329–33.
Luu, P., Collins, P., Tucker, D.M. (2000). Mood, personality, and selfmonitoring: negative affect and emotionality in relation to frontal lobe
mechanisms of error monitoring. Journal of Experimental Psychology.
General, 129(1), 43–60.
Moser, J.S., Hajcak, G., Simons, R.F. (2005). The effects of fear on performance monitoring and attentional allocation. Psychophysiology, 42(3),
261–8.
Olvet, D.M., Hajcak, G. (2009). The stability of error-related brain activity
with increasing trials. Psychophysiology, 46(5), 957–61.
Olvet, D.M., Klein, D.N., Hajcak, G. (2010). Depression symptom
severity and error-related brain activity. Psychiatry Research, 179(1),
30–7.
Ormel, J., Oldehinkel, A.J., Vollebergh, W. (2004). Vulnerability before,
during, and after a major depressive episode: a 3-wave population-based
study. Archives of General Psychiatry, 61(10), 990–6.
Rottenberg, J., Ray, R.D., Gross, J.J. (2007). Emotion elicitation using films.
In: Coan, J.J., Allen, J.J.B., editors. Handbook of Emotion Elicitation and
Assessment. New York: Oxford University Press, pp. 9–28.
Ruchsow, M., Gron, G., Reuter, K., Spitzer, M., Hermle, L., Kiefer, M.
(2005). Error-related brain activity in patients with obsessive-compulsive
disorder and in healthy controls. Journal of Psychophysiology, 19(4),
298–304.
Schrijvers, D., de Bruijn, E.R., Maas, Y., De Grave, C., Sabbe, B.G.,
Hulstijn, W. (2008). Action monitoring in major depressive disorder
with psychomotor retardation. Cortex, 44(5), 569–79.
Schrijvers, D., De Bruijn, E.R., Maas, Y.J., Vancoillie, P., Hulstijn, W.,
Sabbe, B.G. (2009). Action monitoring and depressive symptom reduction in major depressive disorder. International Journal of
Psychophysiology, 71(3), 218–24.
van Wouwe, N.C., Band, G.P.H., Ridderinkhof, K.R. (2010). Positive affect
modulates flexibility and evaluative control. Journal of Cognitive
Neuroscience, 23(3), 524–39.
Wiswede, D., Munte, T.F., Goschke, T., Russeler, J. (2009). Modulation of
the error-related negativity by induction of short-term negative affect.
Neuropsychologia, 47(1), 83–90.
Yeung, N. (2004). Relating cognitive and affective theories of the
error-related negativity. In: Ullsperger, M., Falkenstein, M., editors.
Errors, Conflicts, and the Brain. Current Opinions on Performance
Monitoring. Leipzig: Max Planck Institute of Cognitive Neuroscience,
pp. 63–70.
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(4), 931–59.