Emotion Down-Regulation Diminishes Cognitive Control A Neurophysiological Investigation
Emotion Down-Regulation Diminishes Cognitive Control A Neurophysiological Investigation
Emotion Down-Regulation Diminishes Cognitive Control A Neurophysiological Investigation
investigation
Nicholas M. Hobson
University of Toronto
Department of Psychology
1265 Military Trail
Toronto, Ontario M1C 1A4, Canada
e-mail: nick.hobson@utoronto.ca
telephone: 416-208-4868
fax: 416-287-7642
Emotion Regulation and the ERN 2
Abstract
cognitive process, devoid of emotion. In contrast to this dominant view, a growing body of
clinical and experimental research indicates that cognitive control and its neural substrates, in
particular the error-related negativity (ERN), are moderated by affective and motivational
factors, reflecting the aversive experience of response conflict and errors. To add to this growing
line of research, here we use the classic emotion regulation paradigm – a manipulation that
promotes the cognitive reappraisal of emotion during task performance – to test the extent to
which affective variation in the ERN is subject to emotion reappraisal, and also to explore how
design, forty-one university students completed three identical rounds of a go/no-go task while
normally, without engaging any reappraisal strategy (control). Results showed attenuated ERN
mediation analysis revealed that the association between reappraisal style and attenuated ERN
was mediated by changes in reported emotion ratings. An indirect effects model also revealed
which, in turn, predicted poorer task performance. Taken together, these results suggest that the
ERN appears to have a strong affective component that is associated with indices of cognitive
The efficient pursuit of our everyday goals depends critically upon our capacity to detect and
outcomes across a variety of social, personality and affective domains, including academic
attainment (Hirsh & Inzlicht, 2010; Johns, Inzlicht, & Schmader, 2008), the restriction of racially
prejudiced behaviors (Payne, Shimizu, & Jacoby, 2005; Amodio et al., 2008), and the regulation
of negative emotions (Compton et al., 2008). While it has been shown that performance
monitoring is associated with a number of positive outcomes, the precise nature of these control
functions is currently disputed. In recent years, considerable debate has centered on the extent to
which affective processes drive evaluative and executive aspects of cognitive control (Botvinick,
2007; Inzlicht & Al-Khindi, 2012; Legault & Inzlicht, 2013). In the current study, we
demonstrate that conflict monitoring functions are not devoid of emotion, but that they also
possess inherent affective qualities. Through the use of the explicit emotion-regulation paradigm,
we offer evidence that affective experience is a component of conflict detection and performance
monitoring functions.
Performance Monitoring
Over the past two decades, considerable research has investigated the neural substrates of
is an event-related potential (ERP) called the error-related negativity (ERN or Ne; Falkenstein,
Hohnsbien, & Hoormann, 1990; Gehring, Goss, Coles, Meyer, & Donchin, 1993). The ERN
commission. Converging evidence from ERP source localization techniques (Dehaene, Posner,
& Tucker, 1994; Gehring, Himle, & Nisenson, 2000; van Veen & Carter, 2002), functional
neuroimaging (Botvinick, Nystrom, Fissell, Carter & Cohen, 1999; van Veen & Carter, 2002),
and intracerebral EEG recordings (Pourtois et al., 2010) propose the Anterior Cingulate Cortex
(ACC) as the most likely neural generator of this early neurophysiological response to errors. As
the ERN is present in a variety of cognitive tasks (Riesel, Weinberg, Endrass, Meyer & Hajcak,
2013), across multiple stimulus presentation and response modalities (Endrass, Reuter, &
Kathmann, 2007; Falkenstein, Hoormann, Christ, & Hohnshbein, 2000; Holroyd, Dien, & Coles,
1998), the component is widely assumed to reflect the operation of a generic, multi-modal,
Cognitive neuroscience approaches have traditionally led the theoretical framing of the
ERN. Importantly, these cognitive accounts view the ERN as a correlate of executive functions
responsible for the strategic regulation of cognition and performance. Botvinick, Braver, Barch,
Carter, and Cohen (2001) postulated that the ERN reflects conflict monitoring functions that
reside within the ACC. Under this framework, high response conflict occurs after errors due to
the transient co-activation of opposing response channels representing the committed error and
the task-appropriate, correct response (Gehring & Fencsik, 2001; Yeung, Botvinick, & Cohen,
2004). This conflict monitoring hypothesis further suggests that executive aspects of cognitive
control are up-regulated as a function of the conflict strength computed by the ACC. An
alternative model (Holroyd & Coles, 2002) proposes that the ERN reflects reinforcement
learning processes, driven by functional interactions between the ACC and the mesencephalic
dopamine system. According to this view, the ERN reflects a discrepancy between a desired or
expected outcome (i.e., a correct response) and the actual outcome (i.e., an error response).
Emotion Regulation and the ERN 5
Consequently, the ERN reflects an early neurocognitive indicator that ongoing events are
evaluated as ‘worse’ than expected (Holroyd & Coles, 2002; Stahl, 2010). In turn, these
reinforcement learning signals train the ACC to select the appropriate ‘motor controller’ to guide
future performance (Holroyd & Coles, 2002). Importantly, while the conflict monitoring and
reinforcement learning models provide divergent accounts of the precise computational basis of
the ERN (see Gehring, Liu, Orr & Carp, 2012 for a review), these cognitive models mutually
assume that the ERN reflects performance monitoring processes while failing to consider the
In addition to preceding increased cognitive control, errors are aversive events (Hajcak &
Foti, 2008), which are associated with subjective experiences of distress (Spunt, Lieberman,
Cohen & Eisenberger, 2012). The aversive quality of errors is supported by multiple
conductance (Hajcak, McDonald & Simons, 2003; O’Connell et al., 2007), potentiated startle
response (e.g., Hajcak & Foti, 2008; Riesel, Weinberg, Moran, & Hajcak, 2013), and contraction
of the corregator supercilii (frowning) muscle ( indstr m, attson- rn, ol ar & Olsson,
2013). Furthermore, intracranial local field potential recordings in humans reveal error-related
activity in the ACC, but also in deeper limbic structures commonly associated with a ective
processing, such as the amygdala ( r dil et al., 2002; Pourtois et al., 2010), suggesting that
error-monitoring involves the integration of both affective and cognitive information processing
(Pourtois et al., 2010). What’s more, both contemporary and historic perspectives o ACC
function have emphasised the sensitivity of this neural structure to both cognitive control and
negative affect/arousal (Ballantine, Cassidy, Flanagan, & Marino, 1967; Corkin & Hebben,
Emotion Regulation and the ERN 6
1981; Rainville, Duncan, Price, Carrier, & Bushnell, 1997; Shackman et al., 2011). A recent
neuroimaging meta-analysis indicates that largely overlapping portions of the ACC track the
seemingly heterogeneous processes related to negative affect, cognitive control, and pain
(Shackman et al., 2011). These findings, along with others that advocate for the integration of
emotion and cognition (e.g., Zelazo & Cunningham, 2007), provide further evidence that
affective properties play an important role in cognitive processes like performance monitoring
and behavioral control. Building off of the theoretical integration of emotion and cognition, these
studies indicate that error commission produces a psychophysiological profile consistent with a
negative affective event (see Gross, 1998; indstr m et al., 2013; Pourtois et al., 2010; Spunt et
al., 2012).
that the component is modulated by both state and trait emotional factors. First, larger ERNs
Gehring, Himle, & Nisenson, 2000) and Generalized Anxiety Disorder (GAD, Weinberg, Olvet
& Hajcak, 2010; Weinberg, Klein & Hajcak, 2012), postulating that errors are particularly
threatening for such cohorts (Hajcak, 2012). Similarly, enhanced ERNs have also been observed
among non-patient groups, such as those high in anxious anticipation (Moser, Moran, &
Jendrusina, 2012); trait-anxiety (Aarts & Pourtois, 2010); trait negative affect (Luu, Collins, &
Tucker, 2000; Hajcak, McDonald, & Simons, 2003, 2004; Yasuda, Sato, Miyawaki, Kumano, &
Kuboki, 2004; Santesso, Bogdan, Birk, Goetz, Holmes, & Pizzagalli, 2012); neuroticism (Pailing
& Segalowitz, 2004; Eisenberger et al., 2005; Olvet & Hajcak, 2012); and obsessive-compulsive
personality symptoms (Hajcak & Simons, 2002), indicating that error-related threat sensitivity is
also a feature of sub-clinical samples. Second, state increases in the ERN have been observed in
Emotion Regulation and the ERN 7
task contexts where errors are punished (Potts, 2011; Riesel, Weinberg, Endrass, Kathmann, &
Yeung & Simons, 2005); when performance contexts include derogatory external eedbac
(Wiswede, nte & sseler, 200 a); or when negatively-valenced pictures are presented
between trials of a lan er tas (Wiswede, nte & sseler, 200 b). Not all studies, however,
find these effects. Larson et al. (2006), for example, found an increase in ERN amplitude when
pleasant pictures, but not negative ones, were superimposed between task trials; Clayson,
Clawson, & Larson (2012) found that, despite changes in emotion ratings, manipulating state
affect had little influence on ERN amplitude; and Larson et al. (2013) found that valence and
arousal did not differentiate ERN amplitude, although difference wave ERN (ERN minus CRN)
was related to arousal but not valence. In light of these findings, several authors have proposed
that the ERN tracks the affective or motivational significance of errors and is modulated by both
contextual and dispositional factors (Amodio, Master, Yee, & Taylor, 2008; Gehring &
Willoughby, 2002; Hajcak & Foti, 2008; Luu, Collins, & Tucker, 2000; Riesel et al., 2012).
Important for current concerns, however, the evidence of the affective and motivational variation
in ERN amplitude is still currently mixed and little is known about how exactly these emotion
properties relate to instrumental behaviors of cognitive control and performance monitoring (e.g.,
Thus, several questions remain, and accounting for the specific role of affective processes
conflict detection process, or if affect itself plays a key role in the initiation of control (c.f.,
Yeung et al., 2004). An instrumental view o ‘on-tas ’ emotion has recently been iterated by
Emotion Regulation and the ERN 8
several authors. The affect alarm model proposes that the affective sting experienced during
conflict and errors act as a distress signal, warning that instrumental control is needed
(Bartholow et al., 2005; Inzlicht & Al-Khindi, 2012). Similarly, Botvinick (2007) proposed that
the ACC might conduct a cost-benefit analysis of on-going information processing, with conflict
registering as one potential cost, which is then met with the up-regulation of cognitive control.
What is also less known is the distinction between incidental and integral affect and the
differentiated effects they may have on the neurophysiological and behavioral correlates of
performance monitoring. The majority of studies looking at the link between emotion, ERN, and
undergo a negative or positive mood induction while viewing valenced images (e.g., Larson,
Gray, Clayson, Jones, & irwan, 201 ; Wiswede, nte & sseler, 200 b). However, the
effects of integral-based affect are less known. The negative emotion states that are integrally
related to the task itself naturally arise when dealing with conflict detection and error
commission, and, as research has shown, can often map onto feelings of anxiety (Inzlicht & Al-
Khindi, 2012) and frustration (Spunt et al., 2012). Given this important distinction, it stands to
reason that integral, on-task affect – separate from incidental, discrete emotion states – has a
unique effect on performance monitoring processes. Thus, the aim of the current study is to test
whether ERN amplitude and behavioral control can be altered when manipulating the appraisal
Emotion Regulation
Pioneering theoretical advances from the psychoanalytical (Breuer & Freud, 1957; Freud,
1946) and stress/coping traditions (Lazarus, 1966) led to the empirical and formal investigation
of emotion regulation (Gross, 1998). A broad process model of emotion regulation distinguishes
focused strategies occur before the emotional response, and most notably take the shape of
Importantly, the two responses have different affective trajectories and selective
consequences of emotional experience, and as such are generally viewed as the more adaptive
form of emotion regulation (Gross, 1998; Gross, 2002; Gross & Thomson, 2007; Ochsner &
Gross, 2005).
In light o cognitive reappraisal’s capacity to alter the emotional experience; and given the
association between emotion and the ERN, we wondered if cognitive reappraisal – the deliberate
up- and down-regulation o one’s emotions – can differentially influence ERN amplitude. If this
shows to be the case, it would provide further evidence for the influence of affective processes
on the ERN, and performance monitoring more generally. The current study borrows from a
recent neuroimaging study by Ichikawa et al. (2011) in which they found that emotion
brain regions, which, in turn, predicted subsequent errors during a cognitive control task. To our
knowledge, however, this is the only study to date which has looked the effect of reappraisal of
Emotion Regulation and the ERN 10
integral negative affect on cognitive control. The question remains of whether the specific
specifically, are similarly influenced by the appraisal of task-related emotion processing. Thus,
the aim of the present study was to test the following related hypotheses: i) the ERN can be
emotions will lead to a dampened ERN amplitude while up-regulating one’s emotions will lead
to a heightened ERN amplitude, and ii) the ERN modulations impacted by reappraisal strategy
will subsequently impact behavioral correlates of cognitive control; that is, a dampened ERN
amplitude will be related to poorer cognitive control (i.e., more errors) while a heightened ERN
certain affective qualities and that the ability to monitor effectively can be altered by the
manipulated task-related negative affect through emotion reappraisal regulation strategies during
a cognitive control task. The goal for the current study, therefore, is to formally investigate the
link between integral negative affect and the neural and behavioral markers of performance
monitoring.
Methods
participated for course credit. We report how we determined our sample size, all data exclusions,
all manipulations, and all measures in the study. We decided, a priori, to terminate data
collection at the end of the term provided that we had upwards of 40 participants at that point (a
sample size that is not unlike previous studies with similar methodological and repeated
Emotion Regulation and the ERN 11
measures designs; e.g., Krompinger, Moser, & Simons, 2008). Seven participants were excluded
from all analyses due to computer/hardware malfunction (n = 5), too few errors (n = 1), or high
EEG artifact rates (>35% artifacts; n=1). This left us with a sample of 41 participants (29
females, 12 males; mean age = 19 years, SD = 1.64 years). Participants were told that the
purpose of the study was to investigate the role of emotion and personality on cognitive
performance.
regulation, and a control. The order of condition was counterbalanced across participants.
participants were given the ollowing instructions: “For the next part of the task, we ask that you
adopt a detached attitude as you complete the task. Think about the task in a cold, emotion-free,
analytical way. View the task from a third-person perspective. Try to remove and disengage
yourself from the task as much as possible.” In the up-regulation condition, the instructions were
as follows: “For the next part of the task, we ask that you adopt an involved attitude as you
complete the task. Immerse yourself in the task; really feel all the emotions going through you as
you complete the task. Think of the task as being personally very important to you, as being vital
for your self-identity.” Last, in the control condition, participants were asked to complete the task
as they normally would, without any further instructions. The instructions were a specific form
of reappraisal and differed from traditional positive/negative reappraisal that is often used in the
(see Gross, 1998). Furthermore, the context in which the reappraisal instructions were given
Emotion Regulation and the ERN 12
differs from previous studies where, for instance, participants are given specific instructions to
reappraise their emotions in direct response to a particular event like positive or negative
valenced images (Hajcak & Nieuwenhuis, 2006) or errors (Ichikawa et al., 2011). Contrasted
with these studies, the current reappraisal instructions did not explicitly mention the target of
reappraisal (i.e., commission of errors). Rather, participants were asked to reappraise any
emotions throughout the duration of the task. By using these non-specific reappraisal
instructions, our aim was to have participants down- and up-regulate their affective states as they
naturally arose during their performance, so that, in effect, any efforts of reappraisal were aimed
task. Participants were instructed to press a button i they saw a “go” stimulus (i.e., the letter )
and to re rain rom pushing the button i they saw a “no-go” stimulus (i.e., the letter W). On each
trial, a fixation cross was presented in the middle of the screen for 300-700 ms, followed by
either a “go” or “no-go” stimulus or 100 ms. Participants were given a maximum time of 500
ms to respond on each trial. Within each condition, participants completed four blocks, each
consisting o 40 “go” trials and 10 “no-go” trials. Trials were presented randomly within blocks.
For each condition, participants’ average reaction time or go and no-go (i.e., errors of
commission) responses were measured; the number of errors of commission and the number of
emotional experiences across the different types of reappraisal strategies, participants were asked
after each condition to rate on a 7-point Likert scale how involved they were, ranging from
“detached” to “involved”, and also how emotional they were, ranging rom “ ull o emotion” to
Emotion Regulation and the ERN 13
“emotionless.” The second rating chec s were scaled in the reverse in order to insure that
participants were paying close attention; they were reverse scored during analyses. A higher
score therefore indicated being more involved and emotional. The two checks had a low internal
scale.
Neurophysiological Recording
Continuous EEG was recorded during the go/no-go task using a stretch Lycra cap
embedded with 32 tin electrodes (Electro-Cap International, Eaton, OH). Recordings used
average ear and a forehead channels as reference and ground, respectively. The continuous EEG
was digitized using a sample rate of 512 Hz, and electrode impedances were maintained below 5
Ω during recording. O line, EEG was analyzed with Brain Vision Analyzer 2.0 (Brain
Products GmbH, Munich, Germany). EEG data was corrected for vertical electro-oculogram
artifacts (Gratton, Coles, & Donchin, 1983) and digitally filtered offline between 0.1 and 30 Hz
(FFT implemented, 24 dB, zero phase-shift Butterworth filter). The signal was corrected using a
200 ms baseline which commenced 200 ms before the response. An automatic procedure was
employed to detect and reject artifacts. The criteria applied were a voltage step of more than 25
µV between sample points, a voltage difference of 150 µV within 150 ms intervals, voltages
above 85 µV and below -85 µV, and a maximum voltage difference of less than 0.50 µV within
100 ms intervals. These intervals were rejected from individual channels in each trial. An epoch
was defined as 200 ms before and 800 ms after the response. Data for these epochs were
averaged within participants independently for correct and incorrect trials, and then grand-
averaged within the respective emotion regulation conditions. Error and correct-related brain
activity was defined as the mean amplitude between 0 ms and 100 ms post-response at the
Emotion Regulation and the ERN 14
frontocentral electrode, FCz. We opted to use a mean amplitude measure for the ERPs as such
measures provide more reliable ERP measurements than peak amplitudes (Luck, 2005). ERN
calculations were based on no fewer than five artifact-free error trials (Olvet & Hajcak, 2009).
In light of recent evidence pointing to the weak internal consistency of the Go/Nogo task in ERN
calculations (Meyer, Bress, & Proudfit, in press), it is important to note that the average total
number of error trials that were used in the ERN averaging were well above five (M = 16.6; SD
= 6.7).
Results
Analyses revealed that self-report levels of involvement and emotional feelings during
task performance reflected the changing reappraisal strategies (See Table 1). A 3-way repeated
measures ANOVA with rated level of task involvement as the dependent variable revealed a
significant main effect, F (2,40) = 47.04, p < .001, p2 = .547, such that participants’ involvement
ratings differed for each reappraisal condition. Post-hoc pairwise comparisons revealed that
compared to the control round (M = 4.56, SD = 1.81), participants’ involvement ratings were
significantly less during the down-regulation round (M = 2.78, SD = 1.68), F (1,40) = 29.09, p <
.001, p2 = .427; and significantly more during the up-regulation round (M = 5.85, SD = 1.46), F
Similarly, a 3-way repeated measures ANOVA with rated emotional feelings as the
dependent variable revealed a significant main effect, F (2,40) = 23.96, p < .001, p2 = .381, such
that participants’ emotional ratings di ered or each reappraisal condition. Post-hoc pairwise
comparisons revealed that compared to the control round (M = 3.60, SD = 1.6 ), participants’
Emotion Regulation and the ERN 15
emotion ratings were significantly less during the down-regulation round (M = 2.70, SD = 1.28),
F (1,40) = 10.06, p = .003, p2 = .205 ; and significantly more during the up-regulation round
(M = 4.80, SD = 1.58), F (1,40) = 47.91, p < .001, p2 = .551. Together, this suggests that
participants’ sel -reported emotion and level of involvement reflected our manipulation
instructions to down-regulate, up-regulate, or perform the task normally (i.e., control) across the
experiment.
The behavioral data revealed that reappraisal strategies had no direct effects on
performance (see Table 1). The average percent correct for Go trials was 85.1% in the control
condition, 85.1% in the down-regulation condition, and 87.3% in the up-regulation condition.
The average percent correct for NoGo trials was 57.5% for the control, 58.4% for the down-
regulation, and 59.5% for the up-regulation. A 3 (reappraisal strategy: down-regulation vs. up-
regulation vs. control) X 2 (response: error vs. correct) repeated measures analysis of variance
(ANOVA) with reaction time (RT) as the dependent variable revealed a significant main effect
of response, F (2, 40) = 263.87, p < .001, p2 = .867, such that RTs on error trials, regardless of
reappraisal condition, was significantly faster than RTs on correct trials. A 3 (reappraisal
strategy: down-regulation vs. up-regulation vs. control) X 2 (error type: omission vs.
commission) repeated measures ANOVA with error rate as the dependent variable revealed a
significant main effect of error type, F(1,40) = 83.12, p < .001, p2 = .670, such that participants,
again regardless of reappraisal strategy, committed significantly more commission than omission
The ERN
A 2 (response type: error vs. correct) X 3 (reappraisal condition: down-regulation vs. up-
regulation vs. control) repeated measures ANOVA revealed a significant main effect of response
type, F(1, 40) = 103.11, p < 0.001, p2 = .72, indicating that there is an increased negative going
ERP in response to errors (M = -4.89 V, SD = 5.20) versus correct responses (M = 2.76 V, SD
= 5.20; see Figure 1). Importantly, the main effect of response type was subsumed under a
significant interaction with reappraisal strategy, F (1, 40) = 5.034, p = 0.03, p2 = .112.
Analyses of simple main effects for correct response type revealed that CRN amplitude
in the up-regulation condition (M = 3.37, SD = 2.54) was significantly larger (i.e., more positive)
than the CRN amplitude in both the control condition (M = 2.80 V, SD = 2.61), F (1, 40) =
4.73, p = 0.04, p2 = .106, and the down-regulation condition (M = 2.14 V, SD = 2.47), F (1,
40) = 12.77, p < .01, p2 = .242. The difference between down-regulation and control was
marginally significant, F (1, 40) = 3.27, p = .08, p2 = .076. Analyses of simple main effects for
error response type revealed that ERN amplitude in the down-regulation reappraisal condition
(M = -3.22 V, SD = 4.61) was significantly smaller (i.e., less negative) than the ERN amplitude
in the control condition (M = -4.67 V, SD = 5.38), F (1, 40) = 4.213, p = 0.04, p2 = .095.
However, there was not a difference between the up-regulation ERN (M = -4.15 V, SD = 5.54)
and the control ERN, F (1, 40) = 0.517, p = 0.476, p2 = 0.013; nor was the difference between
the down- and up-regulation ERN significant, F (1, 40) = 1.917, p = 0.174, p2 = .046.
Analyses using the difference wave approach were also conducted in order to avoid
issues of interpretability of raw ERP components (Luck, 2005). A repeated measures ANOVA
with the difference wave scores (Error Amplitude minus Correct Amplitude, ΔE N) as the
Emotion Regulation and the ERN 17
dependent variable revealed a significant main effect of strategy, F (2, 40) = 5.034, p = 0.02, p2
= .130, suggesting a difference in ΔE N across reappraisal strategies (see panel D in Figure 1).
Mirroring our findings from the traditional ERP analyses, pairwise comparisons revealed that the
negative than that observed in both the control condition (M = -7.54 V, SD = 5.45), F (1,40) =
5.96, p = .019, p2 = .130; and up-regulation condition (M = 7.53 V, SD = 5.46), F (1,40) =
8.34, p = .006, p2 = .173. However, no significant difference was found between the control and
These results provide partial support for our hypotheses: We found that in response to
errors, the down-regulation strategy led to a dampened ERN; however, the up-regulation strategy
did not lead to an increased ERN amplitude. This suggests that emotion regulation reappraisal
strategies affect ERN amplitude, but only when participants selectively down-regulated their
and involvement ratings, we used a multicategorical mediation model (Preacher & Hayes, 2008;
in press). We used bootstrap analysis with 5,000 samples to obtain parameter estimates for the
specific indirect effects. Table 2 presents the 95% bias-corrected confidence intervals for the
indirect effects of emotion ratings and involvement ratings on the relationship between condition
and ΔE N. A confidence interval that does not contain zero indicates a statistically significant
indirect effect, and, consequently, mediation (Preacher & Hayes, 2008). The confidence intervals
for specific indirect effects indicate that emotion ratings mediated the relationship between both
Emotion Regulation and the ERN 18
down-regulation and up-regulation reappraisal and ΔE N; involvement ratings did not act as a
significant mediator in either case (see Figure 2). Furthermore, the effects were not significant
when ERN and CRN were included in the model. When included alone, neither error- nor
correct-related performance monitoring processing was predictive of the effects; only difference-
wave activity, ΔE N, was affected by mediation. This aligns well with ERP research practices
which argue that difference-waves can be helpful in isolating and drawing inferences from
waveform components as they have lower signal-to-noise ratio than those of the original ERP
waveforms (Luck, 2005). Taken together then, the emotion and involvement ratings, though
similar in how they were affected by condition manipulation, seem to be tapping into different
constructs. Indeed, the mediation analyses show that emotion and involvement ratings
this shows that emotion or affect in particular – rather than involvement or engagement – is what
affect cognitive performance, we conducted multiple indirect effects tests. We tested the models
Hayes, 2004). The two main models that we tested differed only in terms of their initial predictor
variable; the first including the categorical variable of reappraisal condition (i.e., down-
regulation, up-regulation, and control) and the second, as a more direct test, including the
highlight the distinction between the statistical terms mediated effect and indirect effect
Emotion Regulation and the ERN 19
(Holmbeck, 1997). In a mediation effect, the assumption is that the total effect X on Y be
significant initially; there is no such requirement in the testing of an indirect effect. For the
present analyses therefore, an indirect effects test was justified despite the absence of an initial
total effect of reappraisal condition/emotion ratings on error rates (Preacher & Hayes, 2004;
Hayes, 2009). For contrasting views on the requirement that the total effect be significant, see
Collins, Graham, and Flaherty (1998) and Shrout and Bolger (2002).
For the first model (reappraisal condition as our predictor variable) we used an indirect
effects test for a multicategorical independent variable (in this case, the reappraisal strategy:
down-regulation, up-regulation, and control; Preacher & Hayes, 2008; 2013). Similar to the
mediation analysis above, the significance of the relative indirect effects was tested using
bootstrap analysis with 5,000 samples to obtain parameter estimates. A confidence interval that
does not include zero indicates a statistically significant indirect effect (Preacher & Hayes, 2008;
see Table 3). We fit the data to our model separately for the intervening variables, ERN, CRN,
and ΔE N. The results show non-significant indirect effects for condition on task performance
through ERN (Down-regulation: [-0.24, 1.86]; Up-regulation: [-0.94, 1.41]), as well as through
CRN (Down-regulation: [-1.09, 0.09]; Up-regulation: [-0.12, 1.05]). The third test of our model,
however, revealed a significant indirect effect when including ΔE N in the model: The down-
regulation condition, specifically, was positively related to ΔE N, which, in turn, was positively
related to the number of errors during task performance (0.05, 2.09). The indirect effect for up-
regulation was not significant (-1.10, 0.91). Figure 3 illustrates the results of our analyses.
Similar to the earlier mediation analysis, these findings suggest that the composite ERP
difference wave form is a stronger model predictor and, thus, a more suitable metric to use
(Luck, 2005). The results from these analyses show that down-regulation predicted a dampened
Emotion Regulation and the ERN 20
test of cognitive control, in other words, predicted worse cognitive control, albeit indirectly and
In the second model we collapsed across condition assignment to test the indirect effect
multilevel structural equation modeling (Preacher, Zyphur, Zhang, 2010). We tested separate
models using emotion/involvement ratings and ERN, CRN, and ΔE N as our X and M
predictors, respectively. Contrary to our hypotheses, all of the models tested were not significant.
These findings are expected, however, given the null effects of the up-regulation condition in our
original repeated measures analyses. Taken together, the indirect effects tests reveal that emotion
task but does so indirectly through dampening amplitude of ΔE N. Although we find no such
indirect effect for emotion/involvement ratings – when collapsing across reappraisal condition –
the significant condition effect is partial evidence in support of the hypothesis that integral
negative affect, as manipulated by reappraisal during task performance, is related to both the
Discussion
The current study is one of the first to demonstrate that antecedent-focused emotion
cognitive control. By asking participants to engage in emotional reappraisal strategies, our study
investigated how top-down cognitive processes of reappraisal color online emotional responding
Emotion Regulation and the ERN 21
during a cognitive control task. This manipulation allowed us to focus in on interactions between
monitoring processes differentiated less between error and correct trials (reduced ERN and
(but not subjective involvement) mediated the relationship between reappraisal condition and
ΔE N. Importantly, these indings provide novel support or emerging a ective accounts of the
ERN (Inzlicht & Al-Khindi, 2012). Most importantly, we also found evidence supporting a link
between affect, monitoring, and behavioral control (no-go error rate); that is, reappraisal
regulate emotional involvement during task performance would selectively impact upon early (<
100 ms) error-related brain activity. We suggest that by deliberately approaching the go/no-go
reduced affective reactivity. And, as erroneous actions are rapidly evaluated as negative events
(Aarts et al., 2012; Aarts et al., 2013; Hajcak & Foti, 2008; indstr m et al., 2013; Pourtois et
al., 2010), this diminished emotional responding likely dampened neural monitoring mechanisms
to the transient distress associated with erroneous actions. Specifically, the data from our study
suggests that the ERN is not only a neural indicator of cognitive processing; but also has
properties of emotion that can be modulated by the same regulatory strategies known to
influence negative emotional experience in other contexts (e.g., Gross, 2002). Interestingly, this
finding complements recent reports that manipulations which reduce negative arousal during
Emotion Regulation and the ERN 22
performance, such as acute alcohol administration (Bartholow, Henry, Lust, Saults, & Wood,
2012) or the misattribution of arousal (Inzlicht & Al-Khindi, 2012), also attenuate the ERN.
neuroscience models of control (e.g., Botvinick et al., 2001; Holroyd & Coles, 2002) are unable
to fully explain the observed associations between emotion regulation, brain, and behavior.
Specifically, our findings stress the role of emotional experience in behavioral regulation: when
participants down-regulated their emotions – removing the full range of possible experienced
emotions – they reported feeling less emotion reactivity and performance monitoring was less
efficient (reduced ERN), which, in turn, was associated with poorer task performance. This
pattern of results is best accounted for by recent proposals that the aversive experience of
response-conflict or errors alerts individuals to challenges, and, in turn, this distress energizes
cognitive control efforts to avoid future negative outcomes (Botvinick, 2007; Inzlicht & Legault,
2012; Schmeichel & Inzlicht, 2013). Consequently, when the emotional pang of error
commission is reduced during emotion down-regulation, the saliency o this “a ective alarm”
signal (Inzlicht & Legault, in press) is diminished, making the individual less likely to engage
corrective control processes (Inzlicht & Al-Khindi, 2012; Bartholow et al., 2012). While we
stress the importance of affect for control, we are also mindful of the inherent difficulty in
“cognitive” or “a ective” processes (e.g., Et in et al., 2011; ray, 2004; Shac man et al., 2011).
We do not wish to create a alse dichotomy between a purely “a ective” or “cognitive” theory o
the ERN and executive control. Instead, we hope that the current study adds to the growing line
of evidence to suggest that emotional and cognitive processes are highly integrated, with
Emotion Regulation and the ERN 23
affective experience playing a central role in cognitive control and the strategic regulation of
In further relation to the behavioral correlates of control, and consistent with previous
research (e.g., Inzlicht & Gutsell, 2007), our findings suggest that the ERN predicts particular
performance outcomes (i.e., overall accuracy). Although widely hypothesized (e.g., Botvinick et
al., 2001; Holroyd & Coles, 2002; Yeung et al., 2004), such a direct relationship between
increased ERN amplitude and improved cognitive performance is not always found (Weinberg et
al., 2012). In light of these mixed results, Weinberg et al. (2012) recently hypothesised that
improved task performance may constitute but one potential adaptive consequence of
performance monitoring. Alternatively, the uncertain threat associated with error commission
might also trigger the mobilization of defensive responses, particularly for groups that
experience errors as being particularly concerning (see also Hajcak, 2012; Proudfit, Inzlicht, &
Mennin, 2013). Consequently, the coupling between ERN amplitude and task performance is
potentially moderated by a number state or trait factors. Given that the relationship between the
ERN and control has been shown to be moderated by variables such as intrinsic motivation
(Bartholow et al., 2012; Legault, Al-Khindi, & Inzlicht, 2012) and the correct attribution of
negative affect (Inzlicht & Al-Khindi, 2012), it may be interesting to see whether differences in
trait emotion regulation – that is, people’s natural disposition to employ one strategy over
another – will also act as one such moderator (e.g., Drabant, McRae, Manuck, Hariri, & Gross,
2009).
Our findings may also have implications for clinical research. As reviewed previously,
clinical participants and healthy controls (Holmes & Pizzagalli, 2008; Olvet & Hajcak, 2008;
Weinberg et al., 2012). Furthermore, while substantial variation in ERN amplitude appears to be
stable and trait-like among groups with internalizing psychopathologies (Olvet & Hajcak, 2008;
Weinberg et al., 2012; Proudfit, Inzlicht, & Mennin, 2013), state-related changes in error-
monitoring have been reported for anxious (Riesel et al., 2012) and neurotic (Olvet & Hajcak,
2012) individuals. Therefore, investigating the relationship between emotion regulation and
performance monitoring in clinical groups would provide an interesting avenue for future
research and clinical application. More specifically, these psychopathologies are associated with
poor emotional response systems, in addition to the inefficient use of adaptive emotion
regulation strategies (Davidson, 2000; Mather et al., 2004), with interventions such as cognitive-
behavioral therapy aiming to encourage the framing of more realistic cognitive appraisals (for a
review see Butler, Chapman, Formen, & Beck, 2006). And so, ongoing research should continue
to explore whether increased ERN amplitudes in anxious pathologies are immutable (i.e.,
(Hajcak, 2012; Weinberg et al., 2012), one might predict that any possible ERN reductions
resulting from reappraisal in such groups could occur without producing detrimental effects for
behavioral performance. Of course, more research is needed in order to understand the exact
It is important to consider the current findings within the broader framework of the
process model of emotion regulation and to address their limitations (c.f., Gross, 1998). First, in
neural or behavioral markers of cognitive control. Given the proposed relationship between
experience did not increase ERN amplitude. One possibility is that there exists an upper limit to
how much negative emotion individuals experience during a non-valenced, cognitive control
task. Thus, failure to find an effect of up-regulation on the ERN may be due to an upward
boundary condition or ceiling effect. In line with our results, other studies have also failed to find
neural effects of up-regulation strategies (Krompinger, Moser, & Simons, 2008; Moser et al.,
2006), despite an increase in people’s sel -reported emotion ratings (Ochsner et al., 2004).
Another possibility could be due to our methodological design and the fact that the up-regulation
reappraisal condition did not specifically instruct participants to increase only negative affect
during task performance. In other words participants may have been equally as likely to amplify
their positive affect in response to effective task performance as they would have been to
increase their negative affect in response poorer task performance (i.e., error commission); thus
reducing the overall likelihood of the up-regulation manipulation influencing ERN amplitude.
Future studies will benefit from using clean experimental manipulations in order to tease apart
Furthermore, as multiple forms of emotion reappraisal have been identified in the literature
(e.g., Gross, 1998; Gross & Thompson, 2007), it is unclear which specific strategies were
employed during up- and down-regulation in the present study. Importantly, while we welcome
future research that aims to more precisely determine the influence of distinct reappraisal
Emotion Regulation and the ERN 26
shed further light on the present findings. Specifically, as our down-regulation instructions
emphasised performing the task with a detached, non-emotional mindset, it is possible that
explanation of our results. In effect, by deliberately removing focus from the task, participants
may have employed less performance monitoring (i.e., ERN) through general attentional
disengagement, rather than through the explicit reappraisal of emotional experience. Countering
these concerns, however, it should first be noted that mental distraction and emotion regulation
are not mutually exclusive concepts. More specifically, distraction – as a unique form of emotion
regulation - has been found to activate similar neural circuitry as reappraisal and to also lead to
decreased reports of negative affect (e.g., Kanske et al., 2011). In further support of the central
role of emotion regulation in our ERN results, the mediation analysis (see Figure 2) indicated
that self-reported affective experience (but not subjective involvement/engagement) mediated the
relationship between reappraisal condition and ΔE N. Critically, these indings suggest that
rather than attentional disengagement more generally. Yet given the possibility of these
alternative explanations, future research will benefit from exploring the different forms of
reappraisal strategies and their differential effects on neuro-affective markers like the ERN. The
current study, for instance, concentrated on antecedent focused reappraisal strategies, and so it
impacts performance monitoring in a similar manner. Such findings would provide additional
Emotion Regulation and the ERN 27
support for our hypothesis that error-related affective experience is modulated by other emotion-
based processes.
Finally, it may be suggested that cognitive demands induced by reappraisal also influenced
the observed data pattern. More specifically, cognitive load was perhaps increased when
between cognitive capacity and the ERN have recently been subjected to increased consideration
(c.f., Moser et al., 2013), it is important to assess the possible influence of attentional load on the
present results. In order for a capacity explanation to adequately account for the current findings,
there would need to be asymmetrical effects between reappraisal styles, given the null effect in
the up-regulation condition. Current theory, however, does not align with this interpretation. The
production-monitoring hypothesis (Kalisch et al., 2005; Kalisch et al., 2006), for example,
suggests that the prefrontal areas that generate and maintain positive working memory contents
during down-regulation are also needed to actively maintain negative working memory contents
during up-regulation. According to this theory then, if attentional load were the likely operating
mechanism, we would expect to see similar attenuations in ERN amplitude in both down-
regulation and up-regulation. In short, the asymmetrical effects of attentional load on ERN
amplitude deviate from previous theory, and therefore, provide a less compelling explanation of
the present findings. Nevertheless, more research is needed in order to explicitly test how
changes in cognitive capacity modulate ERN amplitude and whether such factors are impacted
Conclusions
behavioral correlates of cognitive control, as observed in the current study, demonstrates the
for by the principle cognitive neuroscience models of cognitive control, which do not explicitly
address the role of affect in cognitive performance. Using emotion regulation strategies as a way
to manipulate levels of emotional involvement and intensity, we were able to show that certain
fluctuations in affective experiences can modulate the ERN, which, in turn, predicts changes in
cognitive performance. It is our hope that these findings will contribute to a more complete
account of performance monitoring – an account which appreciates the central role of emotion in
executive functioning.
Emotion Regulation and the ERN 29
References
Amodio, D. M., Master, S. L., Yee, C. M., & Taylor, S. E. (2008). Neurocognitive components
of the behavioral inhibition and activation systems: implications for theories of self-
Aarts, K., & Pourtois, G. (2010). Anxiety not only increases, but also alters error-monitoring
Aarts, K., De Houwer, J., & Pourtois, G. (2012). Evidence for the automatic evaluation of self-
Aarts, K., De Houwer, J., & Pourtois, G. (2013). Erroneous and correct actions have different
Ballantine, H.T., Cassidy, W.L., Flanagan, N.B., & Marino, R. (1967). Stereotaxic anterior
26, 488-495.
Bartholow, B. D., Henry, E. A., Lust, S. A., Saults, J. S., & Wood, P. K. (2012). Alcohol effects
Bartholow, B. D., Pearson, M. A., Dickter, C. L., Fabiani, M., Gratton, G., & Sher, K. H. (2005).
Strategic control and medial frontal negativity: beyond errors and response conflict.
Botvinick, M.M. (2007). Conflict monitoring and decision making: reconciling two perspectives
366.
Botvinick, M. M., Braver, T. S., Carter, C. S., Barch, D. M.,&Cohen, J. D. (2001). Evaluating
the demand for control: anterior cingulate cortex and crosstalk monitoring.
Botvinick, M. M., Nystrom, L. E., Fissell, K., Carter, C. S., & Cohen, J. D. (1999). Conflict
181.
r dil, M., Roman, R. Falkenstein, M., Daniel, P., Jurak, & Rektro, I. (2002). Error processing -
evidence from intracerebral ERP recordings. Experimental Brain Research, 146, 460-
466.
Breuer, J., & Freud, S. (1957). Studies on hysteria (J. Strachey, Trans.). New York: Basic
Bush, G., Luu, P., & Posner, M. I. (2000). Cognitive and emotional influences in anterior
Butler, A.C., Chapman, J.E., Forman, E.M., & Beck, A.T. (2006). The empirical status of
26, 17-31.
Emotion Regulation and the ERN 31
Clayson, P. E., Clawson, A., & Larson, M. J. (2012). The effects of induced state negative affect
677-688.
Collins, L. M., Graham, J. W., & Flaherty, B. P. (1998). An alternative framework for defining
Compton, R.J., Robinson, M.D., Ode, S., Quandt, L.C., Fineman, S.L., & Carp, J. (2008). Error-
monitoring ability predicts daily stress regulation. Psychological Science, 19, 702-708.
Corkin, S., & Hebben, N. (1981). Subjective estimates of chronic pain before and after
Davidson, R. J. (2000). Affective style, psychopathology, and resilience: brain mechanisms and
Dehaene, S., Posner, M. I., & Tucker, D. M. (1994). Localization of a neural system for error
Drabant, E.M., McRae, K., Manuck, S.B., Hariri, A.R., & Gross, J.J. (2009). Individual
Eisenberger, N. I., Lieberman, M. D., & Satpute, A. B. (2005). Personality from a controlled
Endrass, T., euter, ., & athmann, N. (2007). E P correlates o conscious error recognition:
aware and unaware errors in an antisaccade tas . European Journal of Neuroscience, 26,
1714-1720.
Et in, A., Egner, T., & alisch, . (2011). Emotional processing in anterior cingulate and medial
Falkenstein, M., Hohnsbein, J., Hoorman, J., & Blanke, L. (1990). Effects of errors in choice
reaction tasks on the ERP under focused and divided attention. In C. H. M. Brunia, A. W.
K. Gaillard, & A. Kok (Eds.), Psychophysiological brain research (Vol. 1, pp. 192–195).
Fal enstein, ., Hoormann, J., Christ, S., & Hohnsbein, J. (2000). E P components on reaction
errors and their unctional signi icance: a tutorial. Biological Psychology, 51, 87-107.
Freud, A. (1 46). The ego and the mechanisms of defense. New Yor : International Universities
Press.
Gehring, W. J., & Fencsik, D. (2001). Functions of the medial frontal cortex in the processing of
Gehring, W. J., Goss, B., Coles, M. G. H., Meyer, D. E., & Donchin, E. (1993). A neural system
Gehring, W.J., Liu, Y., Orr, J.M., Carp, J. (2012). The error-related negativity (ERN/Ne). In S.J.
Gratton, G., Coles, M. G., & Donchin, E. (1983). A new method for off-line removal of ocular
Gross (Ed.), Handbook of emotion regulation (pp. 3-24). New York, NY: Guilford Press.
Hajca , . (2012). What we've learned rom mista es: insights rom error-related brain activity.
Hajca , ., & Foti, D. (2008). Errors are aversive: de ensive motivation and the error-related
Hajcak, G., McDonald, N., & Simons, R. F. (2003). To err is autonomic: error-related brain
895–903.
Hajcak, G., McDonald, N., Simons, R. F. (2004). Error-related psychophysiology and negative
Hajcak, G., Moser, J. S., Yeung, N., & Simons, R. F. (2005). On the ERN and the significance of
Hajcak, G., & Nieuwenhuis, S. (2006). Reappraisal modulates the electrocortical response to
Hajcak, G., Moser, J.S., Holroyd, C.B., & Simons, R. F. (2007). It’s worse than you thought: the
Hayes, M. A. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the new
Hirsh, J., & Inzlicht, M. (2010). Error-related negativity predict academic performance.
Hobson, N. M, Inzlicht, M., & Al-Khindi, T. (2013). Emotional reappraisal strategies: the
Holmbeck, G. N. (1997). Toward terminological, conceptual, and statistical clarity in the study
of mediators and moderators: Examples from the child-clinical and pediatric psychology
Holroyd, C. B., & Coles, M. G. H. (2002). The neural basis of human error processing:
Holroyd, C. B., Dien, J., & Coles, M. G. H. (1998). Error-related scalp potentials elicited by
Ichikawa, N., Siegle, G. J., Jones, N. P., Kamishima, K., Thompson, W. K., Gross, J., &
Ohira, H. (2011). Feeling bad about screwing up: emotion regulation and action
Inzlicht, M., & Al-Khindi, T. (2012). ERN and the placebo: a misattribution approach to
Inzlicht, M. & Gutsell, J. N. (2007). Running on empty: neural signals for self-control failure.
Inzlicht, M., & Legault, L. No pain, no gain: how distress underlies effective self- control (and
The Control Within: Motivation and its Regulation. New York: Psychology Press (in
press).
Johns, M., Inzlicht, M., & Schmader, T. (2008). Stereotype threat and executive resource
Kalisch, R., Wiech, K., Herrmann, K., Dolan, R. J. (2006). Neural correlates of self-distraction
from anxiety and a process model of cognitive emotion regulation. Journal of Cognitive
Kanske, P., Heissler, J., Sch n elder, ongers, A., & Wessa, . (2010). How to regulate
emotion? Neural networks for reappraisal and distraction. Cerebral Cortex, 21,
1379-1388.
Krompinger, J.W., Moser, J.S., & Simons, R.F. (2008). Modulations of the electrophysiological
Larson, M. J., Baldwin, S. A., Good, D. A., & Fair, J. E. (2010). Temporary stability of the error-
Larson, M. J., Gray, A. C., Clayson, P. E., Jones, R., & Kirwan, B. C. (2013). What are the
327-339.
Lazarus, R. S. (1966). Psychological stress and the coping process. New York: McGraw Hill.
Legault, L., Al-Khindi, T., & Inzlicht, M. (2012). Preserving integrity in the face of performance
Legault, L., & Inzlicht, M. (2013). Self-determination, self-regulation, and the brain:
indstr m, B.R., attson- rn, I.B., Golkar, A., & Olsson, A. (2013). In your face: risk
Luck, S. (2005). An introduction to the event-related potential technique. Cambridge, MA: MIT
Press.
Luu, P., Collins, P., & Tucker, D. M. (2000). Mood, personality, and self-monitoring: negative
Mather M., Canli T., English T., Whitfield S., Wais P., Ochsner K., Gabrieli J.D.E., &
Carstensen L.L. (2004). Amygdala responses to emotionally valenced stimuli in older and
Meyer, A., Bress, J. N., & Proudfit, G. H. Psychometric properties of the error-related
Moser, J. S., Hajcak, G., Bukay, E., & Simons, R. F. (2006). Intentional modulation of emotional
Moser, J. S., Moran, T. P., & Jendrusina, A. A. (2012). Parsing relationships between dimensions
Moser, J. S., Moran, T. P., Schroder, H. S., Donnellan, M. B., & Yeung, N. (2013). On the
O'Connell, R. G., Dockree, P. M., Bellgrove, M. A., Kelly, S. P., Hester, R., Garavan, H.,
Robertson, I. H., & Foxe, J. J. (2007). The role of cingulate cortex in the detection of
errors with and without awareness: a high-density electrical mapping study. European
Ochsner, K. N., & Gross, J. J. (2005). The cognitive control of emotion. Trends in Cognitive
Sciences, 9, 242-249.
Emotion Regulation and the ERN 39
Ochsner, K. N., Ray, R. D., Cooper, J. C., Robertson, E. R., Chopra, S., Gabrieli, J. D., & Gross,
J.J. (2004). For better or for worse: neural systems supporting the cognitive down- and
Olvet, D. M., & Hajcak, G. (2009). The stability of error-related brain activity with increasing
Olvet, D. M, & Hajcak, G. (2012). The error-related negativity (ERN) relates to sadness
following mood induction among individuals with high neuroticism. Social, Cognitive,
Payne, B. K., Shimizu, Y., & Jacoby, L. L. (2005). Mental control and visual illusions: toward
324-331.
Pourtois, G., Vocat, R., N'Diaye, K., Spinelli, L., Seeck, M., & Vuilleumier, P. (2010). Errors
recordings in the dorsal anterior cingulate gyrus and amygdala combined with fMRI.
Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect effects
in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36,
717-731.
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and
comparing indirect effects in multiple mediator models. Behavior Research Methods, 40,
879–891.
Preacher, K.J., & Hayes, A.F. Statistical mediation analysis with a multicategorical
press).
Preacher, K. J., Zyphur, M. J., & Zhang, Z. (2010). A general multilevel SEM framework for
Proudfit, G. H., Inzlicht, M., & Mennin, D. (2013). Anxiety and error monitoring: the
Rainville, P., Duncan, G. H., Price, D. D., Carrier, B., & Bushnell, M. C. (1997). Pain affect
encoded in human anterior cingulate but not somatosensory cortex. Science, 277, 968–
971.
Riesel A., Weinberg A., Endrass T., Kathmann N., & Hajcak G. (2012). Punishment has a lasting
Riesel, A., Weinberg, A., Endrass, T., Meyer, A., & Hajcak, G. (2013). The ERN is the ERN is
the ERN? Convergent validity of error-related brain activity across different tasks.
Santesso, D. L., Bogdan, R., Birk, J. L., Goetz, E. L., Holmes, A. J., Pizzagalli, A. D. (2012).
Schmeichel, B. J., & Inzlicht, M. (2013). Incidental and integral effects of emotions on self-
Shackman, A. J., Salomons, T. V., Slagter, H. A., Fox, A. S., Winter, J. J., & Davidson, R. J.
(2011). The integration of negative affect, pain and cognitive control in the cingulate
Shrout, P.E., & Bolger, N. (2002). Mediation in experimental and nonexperimental studies: New
Spunt, R. P., Lieberman, M. D., Cohen, J. R., & Eisenberger, N. I. (2012). The phenomenology
Stahl, J. (2010). Error detection and the use of internal and external error indicators: an
77, 43–52.
Emotion Regulation and the ERN 42
van Veen, V., & Carter, C. S. (2002). The timing of action monitoring in rostral and caudal
Weinberg, A., Olvet, D., & Hajcak, G. (2010). Increased error-related brain activity in
Weinberg, A., iesel, A., & Hajca , . (2012). Integrating multiple perspectives on error- related
Wiswede, D., Münte, T. F., & Rüsseler, J. (2009a). Negative affect induced by derogatory verbal
feedback modulates the neural signature of error-detection. Social, Cognitive, & Affective
Neuroscience, 4, 227-237.
Wiswede, D., Münte, T. F., Goschke, T., & Rüsseler, J. (2009b). Modulation of the error-related
Yasuda, A., Atsushi, S., Miyawaki, K., Kumano, H., Kuboki, T. (2004). Error-related negativity
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, 931-
959.
Zelazo, P.D., & Cunningham W. (2007). Executive function: Mechanisms underlying emotion
regulation. In J. Gross (Ed.), Handbook of emotion regulation (pp. 135-158). New York:
Guilford.
Emotion Regulation and the ERN 43
Figure 1. Upper Panels: Response-locked waveform amplitude at FCz following correct and incorrect
responses on the go/no-go task for the (A) down-regulation reappraisal, (B) control no-reappraisal, (C)
up-regulation reappraisal (D), and the difference wave for each. Lower panels: Spline maps depict the
scalp distribution o the ΔE N (mean activity 0-100 ms) for the (A) down-regulation, (B) control, and (C)
up-regulation conditions.
Figure 2. A multicategorical mediation model of emotion ratings as a mediator of the relation between
reappraisal condition (down- and up-regulate) and difference-wave ERN. Unstandardized regression
coefficients (and the associated standard errors) from a bootstrap procedure are provided along the paths.
Darker outlines indicate significant indirect effects and, consequently, mediation (*p < .05; **p < .01).
Figure 3. An indirect effects model showing the indirect effect of condition assignment (reappraisal
strategy) on cognitive control (i.e., number of errors) through difference waveform (ΔERN).
Unstandardized regression coefficients (and the associated standard errors) from a bootstrap procedure are
provided along the paths (*p < .05; **p < .01).
Table 1. Means across rows with different subscripts differ significantly at p < .05 (two tailed).
Figure 1.
Emotion Regulation and the ERN 45
Figure 2.
Emotion atings
b = 1.20** b = 60. ✝
SE = 0. 7 SE = 4.
c ’ = -4.88
Up- egulation SE = 116.
b = -0. 0**
c ’ = 204.10✝ ΔE N
SE = 0. 7 b = 1. 0**
SE = 0. 71 SE = 116.
Down- egulation
b = 52. 2
SE = 1.77
b = -1.77**
SE = 0. 71
Involvement atings
Emotion Regulation and the ERN 46
Figure 3.
ΔE N
b = -16.16 b = 0.00 *
SE = 64.25 SE = 0.001
c ’ = -0.21
1
SE = 0.802
Up- egulation
Errors
b = 20 .78**
SE = 64.25
c ’ = -0.
Down- egulation 2
SE = 0.84
Emotion Regulation and the ERN 47
Table 1.
Means (SD) for Manipulation Checks (Reported Involvement and Emotional Feeling), Cognitive
Performance on the Go/No-Go Task, and Electroencephalography (EEG) Measures
Omission error rate (%) 14.95a (12.10) 12.75a (11.11) 14.99a (12.13)
Commission error rate (%) 41.61a (17.02) 40.53a (17.62) 42.50a (16.67)
Total number of commission errors 16.8a (6.60) 16.2a (7.04) 17a (6.68)
Overall accuracy rate (%) 79.7a (13.05) 81.7a (12.41) 79.5a (13.05)
Reaction time correct 206.20a (41.34) 201.89a (37.35) 203.56a (40.11)
Reaction time error 148.94a (23.49) 147.08a (23.91) 148.14a (31.77)
Table 2.
95% bias-corrected
Indirect effects through emotion ratings SE b confidence interval
95% bias-corrected
Indirect effects through involvement ratings SE b confidence interval
Table 3.
0% bias-corrected
elative indirect e ect through ΔE N SE b
con idence interval