Neuropsychologia 77 (2015) 242–252
Contents lists available at ScienceDirect
Neuropsychologia
journal homepage: www.elsevier.com/locate/neuropsychologia
Event-related potentials in performance monitoring are influenced by
the endogenous opioid system
Daniela M. Pfabigan a,n,1, Jürgen Pripfl a,1, Sara L. Kroll a,b, Uta Sailer c, Claus Lamm a,n
a
Social, Cognitive and Affective Neuroscience Unit, Department of Basic Psychological Research and Research Methods, Faculty of Psychology, University of
Vienna, Liebiggasse 5, A-1010 Vienna, Austria
b
Department of Psychiatry, Psychotherapy and Psychosomatics, University of Zurich, Lenggstrasse 31, CH-8032 Zurich, Switzerland
c
Department of Psychology, Faculty of Social Sciences, University of Gothenburg, Haraldsgatan 1, SE-40530 Gothenburg, Sweden
art ic l e i nf o
a b s t r a c t
Article history:
Received 6 March 2015
Received in revised form
26 July 2015
Accepted 29 August 2015
Available online 1 September 2015
Recent research suggests that not only the dopamine neurotransmitter system but also the endogenous
opioid system is involved in performance monitoring and the generation of prediction error signals.
Heightened performance monitoring is also associated with psychopathology such as internalizing disorders. Therefore, the current study investigated the potential link between the functional opioid peptide
prodynorphin (PDYN) 68 bp VNTR genetic polymorphism and neuronal correlates of performance
monitoring.
To this end, 47 healthy participants genotyped for this polymorphism, related to high-, intermediate-,
and low-expression levels of PDYN, performed a choice-reaction task while their electroencephalogram
(EEG) was recorded. On the behavioural level, no differences between the three PDYN groups could be
observed. EEG data, however, showed significant differences. High PDYN expression individuals showed
heightened neural error processing indicated by higher ERN amplitudes, compared to intermediate and
low expression individuals. Later stages of error processing, indexed by late Pe amplitudes, and stimulusdriven conflict processing, indexed by N2 amplitudes, were not affected by PDYN genotype.
The current results corroborate the notion of an indirect effect of endogenous opioids on performance
monitoring, probably mediated by the mesencephalic dopamine system. Overall, enhanced ERN amplitudes suggest a hyper-active performance monitoring system in high PDYN expression individuals, and
this might also be an indicator of a higher risk for internalizing disorders.
& 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords:
Prodynorphin (PDYN)
ERN
Late Pe
Reward prediction error signals
Mesencephalic dopamine system
Opioid system
1. Introduction
To successfully navigate in our social world, we have to be able
to adapt our behaviour to a (sometimes persistently) changing
environment. Recent research assumes that this kind of cognitive
and behavioural adaptation (i.e., performance monitoring) involves learning processes that are driven by so-called (reward)
prediction error signals-RPEs (Friston, 2010). RPEs indicate the
discrepancy between expected and received outcomes and account for internal expectation updates of stimulus-outcome relations (Rescorla and Wagner, 1972). Mesencephalic dopamine
transmission is considered as their neuronal correlate (Schultz,
2007; Schultz, et al., 1997). It has been posited by Holroyd and
Coles (2002) that phasic dips in mesencephalic dopamine transmission signal the need for adaptation after error commission to
n
Corresponding authors. Fax: þ 43 1 4277 47193.
E-mail addresses: [email protected] (D.M. Pfabigan),
[email protected] (C. Lamm).
1
DMP and JP contributed equally to this manuscript.
the anterior cingulate cortex (ACC). Thereby, ACC neuron activity is
disinhibited to allow modification of task performance which
leads to a sharp negative deflection on the scalp (the so-called
Error-Related Negativity component; Falkenstein, et al., 1991;
Gehring, et al., 1993) – thus reflecting RPE signals. However, recent
animal research on fear conditioning suggests the involvement of
another neurotransmitter system, which is the opioid system, in
generating RPEs (Cole and McNally, 2007; Matzel, et al., 1988;
McNally and Cole, 2006). These studies showed that fear blocking
was prevented after administration of an opioid antagonist in
mice. Moreover, Pecina, et al. (2014) suggested that opioid-mediated placebo responses can also be seen as prediction error signals.
In these studies manipulating the opioid system, the expected
outcome (i.e., the predicted pain signal) did not match the actual
outcome (i.e., the actually perceived nociceptive stimulus) – which
describes exactly the discrepancy between expected and received
outcomes as reflected by prediction error signals. Although previous research assumed that endogenous opioids are primarily
involved in nociception and analgesia, but also in hedonic control
and reward processing (Le Merrer, et al., 2009), the mentioned
http://dx.doi.org/10.1016/j.neuropsychologia.2015.08.028
0028-3932/& 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
D.M. Pfabigan et al. / Neuropsychologia 77 (2015) 242–252
studies support the assumption that opioids are also involved in
the generation of RPE, possibly also during performance
monitoring.
The aim of the current study was to investigate RPEs and their
relation to genetic variation affecting the function of the human
endogenous opioid system. We measured event-related potentials
(ERPs) in individuals with a naturally occurring polymorphism of
the opioid peptide prodynorphin (PDYN). PDYN is the precursor of
the endogenous opioid peptide dynorphin which is involved in
locomotor activity, stress response, food consumption, sexual and
anxiety-related behaviour, and drug intake (Bodnar, 2011; Bruijnzeel, 2009). Dynorphin acts as moderately selective agonist for the
kappa (κ) opioid receptor (Chavkin et al., 1982). Importantly, dynorphin-like peptides and κ-opioid receptors are expressed and
localized, among others, in mesolimbic-mesocortical systems (i.e.,
ventral tegmental area, nucleus accumbens, and prefrontal areas
(Shippenberg, 2009)) and are assumed to exert tonic inhibitory
control over striatal dopamine release (Bruijnzeel, 2009; Kreek
et al., 2002; Lutz and Kieffer, 2013; Margolis et al., 2006; Steiner
and Gerfen, 1998). In particular, dynorphin peptides lower basal,
but also drug-induced dopamine levels in these systems (Kreek,
et al., 2005). Dopamine neurons of the ventral tegmental area (the
starting point of the mesolimbic dopamine system) receive input
from neurons innervated by dynorphin and also express κ-opioid
receptors. Activation of these κ-opioid receptors via dynorphin
decreases dopamine release in the respective areas (Knoll and
Carlezon, 2010; Margolis et al., 2006; Wee and Koob, 2010).
The dynorphin and κ-opioid receptor system is widely distributed within the central nervous system. It constitutes a sort of
neuro-modulatory master system within the brain together with
other opioid receptor types and ligands (Chavkin et al., 1982;
Corbett et al., 1982). Its main role is to function as a negative
feedback inhibition within the circuits it modulates (for a review
see Steiner and Gerfen, 1998). The dynorphin and κ-opioid receptor system is setting the threshold for the initiation of the
negative feedback inhibition loop and is thereby able to fine-tune
the excitability of these neuronal circuits. Whereas the dampening
of dopaminergic effects by increased dynorphin function may act
as a compensatory mechanism at the cellular level (Steiner and
Gerfen, 1998), increased negative feedback will also affect activity
in the pathways it is contained in, and consequently alter behaviour. For example, the dynorphin and κ-opioid receptor system is
also involved in regulating the excitability of the reward system
(Bruijnzeel, 2009). Therefore, imbalance of the negative dynorphin-dopamine feedback loop in the reward system might be associated with clinical pathologies characterized by either hyper- or
hyposensitivity to rewards, such as substance abuse or affective
disorders (Knoll and Carlezon, 2010; Shippenberg, 2009; Tejeda
et al., 2012).
Based on previous research on gene expression (Zimprich,
et al., 2000), we targeted participants with a functional polymorphism in the promotor region of the PDYN gene (68 bp VNTR).
This 68 bp repeat has been the subject of several functional analyses studies showing that one- and two-repeat haplotypes have
lower inducibility in vitro as well as a lower PDYN expression level
in vivo than three- and four-repeat haplotypes (Nikoshkov, et al.,
2008). Individuals with high, intermediate, and low PDYN expression rates were distinguished in the current study. We assumed that group-specific dynorphin expression would influence
striatal dopamine release and that the differential modulation of
dopamine release should become evident in amplitude variation
of ERPs related to prediction error signals and performance
monitoring.
As measures of performance monitoring, three ERPs were assessed: Error-Related Negativity (ERN; Falkenstein et al., 1991;
Gehring et al., 1993), Error Positivity (Pe; Falkenstein et al., 1991,
243
2000), and conflict N2. The ERN component is a negative deflection peaking within the first 100 ms after erroneous responses at
fronto-central electrodes and is assumed to be generated in the
anterior midcingulate cortex (aMCC), i.e. a brain area important for
conflict monitoring and behavioural regulation (Debener et al.,
2005; Hoffmann and Falkenstein, 2010; Vogt, 2005), see also Bauer
et al. (2003) and Pfabigan et al. (2013). The Pe component is a
positive deflection indicating conscious error processing (Nieuwenhuis et al., 2001) or affective responses after errors (Falkenstein et al., 2000). The Pe is sometimes differentiated in an early
and a late component since two distinct peaks can be temporally
discriminated (Tops et al., 2013; Van Veen and Carter, 2002) – as in
the current study. The early Pe is often assumed to reflect rebounding ERN activity (Falkenstein et al., 1995). Thus, investigating
later Pe aspects might be more informative. Indeed, it is in particular the late Pe component (around 300–600 ms after error
commission) which shows enhanced amplitudes for aware compared to unaware errors (Endrass, et al., 2007). The late Pe component has been hypothesized to originate from inferior frontal
gyrus and anterior insula (Ullsperger, et al., 2010). The current
study focused on the late Pe component. The conflict N2 component is a negative-going stimulus-locked ERP indicating visual
template mismatch. It peaks within 200–300 ms after stimulus
onset over fronto-central electrodes, and the aMCC is also assumed
to be its neuronal generator (Folstein and Van Petten, 2008;
Nieuwenhuis et al., 2003; Yeung et al., 2004). Whereas ERN and Pe
amplitudes are assumed to be modulated by phasic mesencephalic
dopamine release reflecting RPEs (Holroyd and Coles, 2002), the
conflict N2 is believed to be unaffected by changes in mesencephalic dopamine, and has been linked to noradrenergic functions
(Warren and Holroyd, 2012). The link between ERN amplitude
variation and dopamine is further corroborated by psychopharmacological studies. Administration of dopamine agonists
increased ERN amplitudes (Barnes et al., 2014; De Bruijn et al.,
2004), whereas administration of dopamine antagonists attenuated ERN amplitudes in comparison to placebo (De Bruijn, et al.,
2006; Zirnheld, et al., 2004). Moreover, investigations of several
functional polymorphisms in dopamine-related genes showed
variation in ERN amplitudes (Agam et al., 2014; Biehl et al., 2011;
Kramer et al., 2007), as well as interactions between the functional
polymorphisms and dopamine antagonists (Mueller et al., 2011,
2014). Variation of Pe amplitudes after administration of dopamine agonists/antagonists or in relation to functional dopamine
polymorphisms is not consistently reported (but see Althaus et al.,
2010).
The present study aimed to investigate the influence of the
PDYN 68 bp VNTR polymorphism on RPEs reflected in ERN amplitude variation. In particular, we expected amplitude differences
between high and low PDYN expression participants. Since high
levels of PDYN exert tonic inhibitory control over dopamine release, we expected decreased ERN amplitudes in high PDYN expression individuals compared to low PDYN expression ones since
the tonic inhibitory control might act in a similar way as the administration of dopamine antagonists, shown to affect ERN amplitudes in previous studies (De Bruijn et al., 2006; Zirnheld et al.,
2004). We had no directional hypotheses concerning the late Pe
component as previous results were not consistent, or the intermediate PDYN expression group and ERP amplitude variation, but
included it for exploratory reasons. The dopamine-independent
N2 component served as control condition to explore potential
influences of the PDYN polymorphism on aMCC activity unrelated
to the link between opioid and dopamine systems (e.g., Bruijnzeel,
2009). Therefore, we only expected effects of stimulus congruency
on N2 amplitudes. Furthermore, we explored behavioural indices
of performance monitoring (reaction times, error rates, post-error
slowing) and personality characteristics associated with reward
244
D.M. Pfabigan et al. / Neuropsychologia 77 (2015) 242–252
sensitivity, impulsivity, and risk for substance abuse in the three
PDYN groups.
2. Material and methods
2.1. Participants
Initially, 286 healthy Caucasian volunteers were genotyped on
the prodynorphin (PDYN) 68 bp variable nucleotide tandem repeat
(VNTR) polymorphism. This functional polymorphism is located in
the PDYN promotor region with one to four repeats of a 68 bp
segment with one binding site per repeat for the transcription
factor AP-1 (c-Fos/c-Jun) (Zimprich et al., 2000). Alleles with three
or four repetitions of the 68 bp VNTR (denoted by H) are associated with higher levels of mRNA and consequently higher levels
of PDYN peptides (i.e., equitable to enhanced dopamine inhibition), compared to alleles with one or two repetitions (denoted by
L) (Nikoshkov et al., 2008; Zimprich et al., 2000).
After genotyping, all volunteers were assigned to one of three
groups with high (HH), intermediate (HL or LH, in the following
denoted by LH), or low (LL) PDYN expression. Genotypes were
distributed as follows in the screening sample: HH (n ¼142), LH
(n ¼113), LL (n ¼31). The allelic distribution of LL, LH, and HH
PDYN polymorphism was in Hardy–Weinberg equilibrium, χ2 (2)¼
1.39, p ¼0.500, demonstrating that the screened population was
genetically homogeneous for the PDYN genotype distribution.
Twenty participants per group (matched for age, sex, education;
tobacco, alcohol, coffee, and energy drink consumption) were invited to the EEG experiment using a double-blind procedure.
Several participants had to be excluded from further analyses
because of technical problems (n ¼3), too high or too low error
rates (n ¼8; error rates higher than 30% or less than six errors in
total (Olvet and Hajcak, 2009)), and outliers exceeding the interquartile range of the ERP amplitudes in question (n ¼2). The final
sample consisted of 16 participants in the HH group (nine women,
mean age 2475.34 years), 16 participants in the LH group (eight
women, mean age 23 73.29 years), and 15 participants in the LL
group (nine women, mean age 257 7.65 years). All participants
were right-handed (Oldfield, 1971), had normal or corrected-tonormal vision, and reported no past or recent neurological or
psychiatric disorders. All gave written informed consent prior to
the experiment, which was conducted in accordance with the
Declaration of Helsinki (7th revision, 2013, http://www.wma.net/
en/20activities/10ethics/10helsinki/) and local guidelines of the
University of Vienna. It was further approved by the ethics board
of the Medical University of Vienna.
Participants were administered several questionnaires prior to
the experimental session to assess psychological constructs related
to reward and error processing. The BIS/BAS Scale (Carver and
White, 1994) was administered to assess sensitivity of behavioural
inhibition and approach systems, i.e., reward and punishment
sensitivity. The Barratt Impulsiveness Scale (BIS-11; Patton, et al.,
1995) was administered to assess impulsive personality traits. The
Substance Use Risk Profile Scale (SURPS; Woicik et al., 2009) was
administered to assess individual risk for substance abuse on four
dimensions (hopelessness, anxiety sensitivity, impulsivity, sensation seeking).
2.2. Genetic analyses
Using a self-collection kit for collection and storage, saliva
samples were collected to determine DNA sequences (Oragene
DNA, DNA Genotek, Ottawa, Canada). DNA was extracted using a
commercial kit (Qiagen, Hilden, Germany). PDYN genotyping was
performed according to established procedures at the DNA
laboratory of the Department of Neurology at the Medical University in Vienna. In detail, purified DNA was diluted into a PCR
reaction mix consisting of 20 mM Tris–HCl (pH: 8.8), 50 mK KCl,
1.5 m MgCl2, deoxynucleotide triphosphates each at 0.4 mM,
10pmol of each primer, and 0.6 U of Taq polymerase in a total
volume of 30 ml. Amplification conditions were 30 s at 94 °C, 45 s
at 62 °C, and 45 s at 72 °C for 30 cycles using the following primers
(which are flanking the entire promotor region): upstream (P1),
5′-AGC AAT CAG AGG TTG AAG TTG GCA GC; downstream (P2), 5′GCA CCA GGC GGT TAG GTA GAG TTG TC. The resulting amplification products were resolved on a 2.5% agarose gel stained with
ethidium bromide.
2.3. Task
The stimuli were presented using Cogent 2000 v1.32, developed by the Cogent 2000 team at FIL and ICN and Cogent Graphics
developed by John Romaya at the LON at the Wellcome Department of Imaging Neuroscience (on a Pentium IV, 3.00 GHz). Participants were seated about 70 cm in front of a 19 in. cathode ray
tube monitor (Philips 201 P; 75 Hz refresh rate) in a shielded
chamber. Participants performed an arrowhead version of the
Eriksen Flanker task (Eriksen and Eriksen, 1974). Five-arrow
strings were presented centrally on the screen. Half of the trials
comprised of congruently aligned arrays (o o o o o ;
4 4 4 4 4), the other half of incongruently aligned arrays
(o o 4 o o; 4 4 o 4 4). Participants’ task was to indicate the
left- or right-hand orientation of the middle arrow via a button
press with the right hand on a keyboard as quickly and accurately
as possible. When the middle arrow pointed to the left, the left
arrow key had to be pressed with the index finger. When the
middle arrow pointed to the right, the right arrow key had to be
pressed with the middle finger.
The experiment started with 20 training trials to familiarize
participants with the task. Each trial started with a white fixation
cross on a black screen presented for 2000 ms. Next, the four
flanking arrows of the five-arrow string were presented for 100 ms
in white colour on black background. Subsequently, the middle
arrow was blended into the string for another 35 ms. This sequential presentation order was introduced by Kopp et al., (1996)
to enhance interfering effects and thereby increase error rates.
Immediately afterwards, the screen turned black for 600 ms and
participants responded via button press to indicate the orientation
of the middle arrow. Button presses were followed by an intertrial-interval presenting a fixation cross for the respective
2000 ms. The same number of congruent (160) and incongruent
(160) flanker arrays was presented pseudo-randomly mixed. Participants were given a short rest after 160 trials. The task took
about 15 min to be completed. All participants received a fixed
amount of monetary remuneration at the end of the experiment.
2.4. Data acquisition and analyses
Reaction times were defined as the interval from middle arrow
onset until button press. In line with recently proposed procedures, trials with reaction times faster than 200 ms were discarded
from all analyses (Hajcak et al., 2005; Wiswede et al., 2009a,b).
Individual mean reaction times were calculated participant- and
condition-wise. Analysis of the mean reaction times was performed using two two-way repeated measures ANOVAs with the
between-subject factor group (HH vs. LH vs. LL) and either the
within-subject factor response type (correct vs. error) or stimulus
type (congruent vs. incongruent). Furthermore, error percentage
was calculated per participant and analysed via a two-way repeated-measures ANOVA with the factors group and stimulus type.
Additionally, to assess post-error slowing (PES; Rabbitt, 1966),
D.M. Pfabigan et al. / Neuropsychologia 77 (2015) 242–252
reaction times of correct trials were extracted participant-wise
before and after erroneous trials, i.e., applying the so-called
PESrobust method (Dutilh et al., 2012). This procedure was chosen
to avoid problems of the traditional PES calculation in which postcorrect and post-error trials are not evenly distributed over the
time series of the experiment which could cause spurious effects
(Dutilh et al., 2012). These mean reaction times were analysed via
another two-way repeated-measures ANOVA with the factors
group and sequence (pre-error trials vs. post-error trials). Questionnaire data were compared by one-way ANOVAs with group as
between-subject factor for each subscale.
EEG data were recorded from 61 Ag/AgCl ring electrodes with a
DC amplifier (NeuroPrax, neuroConn GmbH, Ilmenau, Germany).
Online EEG recordings were referenced to an electrode on the
forehead. Additional electrodes were placed above and below the
left eye and on the outer canthi to assess eye-movements. Two
pre-experimental eye-movement calibration tasks were performed for subsequent artefact correction. Electrode impedances
were kept below 2 kΩ using a skin scratching procedure (Picton
and Hillyard, 1972). Signals were sampled at 500 Hz for digital
storage. Offline EEG data analyses were performed using EEGLAB
6.03b, implemented in Matlab 7.5.0 (The MathWorks, Inc., Natick,
MA). EEG data were re-referenced to averaged linked mastoids. A
high-pass filter (cut-off frequency 0.1 Hz) and a low-pass filter
(cut-off frequency 30 Hz, roll-off 6 dB/octave) were applied subsequently. Independent component analysis (ICA; Bell and Sejnowski, 1995; Lee et al., 1999) was performed on the concatenated
continuous pre-experimental calibration tasks and all experimental trials. Independent components reflecting eye-movement
activity were discarded separately for each participant. Epochs
were either time-locked to participants’ responses to assess ERN
and late Pe amplitudes, starting 400 ms prior to the button press
and lasting for 1000 ms (baseline: 200 to 100 ms prior to the
button press), or time-locked to the onset of the flanking arrows to
asses conflict-N2 amplitudes, staring 200 ms prior to the onset of
the flanking arrows (baseline: 100 to 0 ms prior to stimulus
onset) and lasting for 1000 ms. A semi-automatic artefact correction was applied to all epoched data. Artefact-afflicted trials
meeting the criteria of voltage values exceeding 775 mV or voltage drifts larger than 50 mV were labelled automatically. These
trials were eventually rejected in case visual inspection also indicated artefact affliction. Subsequently, artefact-free trials were
averaged separately for each participant for the response-locked
conditions: (1) trials including correct responses after congruent
and incongruent stimulus arrays – correct-response, (2) trials including incorrect responses after congruent and incongruent stimulus arrays – error–response. On average, 32.28 errors
(SD ¼21.76), with a minimum of seven errors per participant (Olvet and Hajcak, 2009), were subjected to analyses. For stimuluslocked data, the following artefact-free trials were averaged:
(3) trials including congruent stimulus arrays with subsequent
correct responses, and (4) trials including incongruent stimulus
arrays with subsequent correct responses. ERN, late Pe, and N2
amplitudes were assessed at midline electrode sites Fz, Cz, and Pz,
which is consistent with previous literature (Gehring et al., 1993;
Pfabigan et al., 2013; Wiswede et al., 2009a,b). ERN amplitudes
were assessed as the difference between the most negative peak
within -100 to 100 ms in relation to the response and the preceding positive peak (De Bruijn et al., 2006; Kramer et al., 2007).
Moreover, to gain a measure of neural activity specific to errors for
the response-locked data, ΔERN amplitudes were calculated by
subtracting averaged correct from averaged error trials per participant to remove processes common to both error and correct
trials (Luck, 2005; Meyer et al., 2012). These difference waves were
baseline-corrected in the time interval -100 to 0 ms to allow better
comparisons of the three groups since the positive peak preceding
245
the ΔERN peak was now included in the baseline interval. Subsequently, peaks of ΔERN (in the interval 100 to 100 ms) and the
preceding positive peak were assessed individually at electrodes
Fz, Cz, and Pz; and then subtracted from each other. Late Pe amplitudes were assessed as the most positive peak 200–600 ms
after the response. N2 amplitudes were assessed as the difference
between the most negative peak within 300–500 ms after the
onset of the flanking arrows and the preceding positive peak.
ERPs were analysed separately for all participants with threeway repeated measures ANOVAs with the within-subject factors
electrode site (Fz, Cz, Pz) and type (error vs. correct for ERN and late
Pe components; congruent vs. incongruent for the N2) and the
between-subject factor group (HH vs. LH vs. LL). Additionally for
ΔERN, a two-way ANOVA with the between-subject factor group
and the within-subject factor electrode site was calculated to assess
error specific activity in the three groups. Furthermore, a Pearson
correlation was calculated between the PDYN allele number per
participant and ΔERN values to corroborate potential group differences also on a dimensional level. Spearman correlations were
calculated for HH and LL participants to investigate the association
between the ERP components per condition (at Cz for ERN and Pe
components; at Fz for the N2 component) and the sub-scales of
the BIS/BAS, BIS-11, and SURPS (which included mostly not normally-distributed data).
If not stated otherwise, significant interaction effects were explored with Tukey's HSD post-hoc tests. Group effects were analysed with a priori planned linear contrasts. If indicated by Mauchley's test of sphericity, degrees of freedom were adapted applying the Greenhouse–Geisser (GG) correction. Partial etasquared is reported for significant results to demonstrate effect
sizes of the current ANOVA model (Kirk, 1996). All statistical
analyses were performed using PASW Statistics 18.0 (IBM SPSS
Statistics, Somer, NY) and Statistica 6.0 (StatSoft Inc., Tulsa OK).
The alpha level was set at pr 0.05. For the correlation analyses
between the questionnaire data and the ERPs, we corrected the
significance threshold because of multiple testing using the Bonferroni correction (pcorr r 0.004).
3. Results
3.1. Behavioural results
Mean reaction times were faster for incorrect compared to
correct responses (F(1,44) ¼474.32, p o0.001, ηp2 ¼ 0.92). No group
(F(2,44) ¼1.92, p¼ 0.159) or interaction effects (F(2,44) ¼0.90,
p¼ 0.413) were observed. Mean reaction times were also faster for
congruent compared to incongruent stimulus arrays (F(1,44) ¼
724.62, po 0.001, ηp2 ¼0.94). No group (F(2,44) ¼2.17, p¼ 0.126) or
interaction effects (F(2,44) ¼0.44, p ¼0.649) were observed. Error
percentage was higher for incongruent than congruent stimulus
arrays (F(1,44) ¼25.42, po 0.001, ηp2 ¼0.36). No group (F(2,44) ¼
0.76, p ¼0.474) or interaction effects (F(2,44) ¼1.06, p¼ 0.356)
were observed. A significant post-error slowing effect was observed (F(1,44) ¼79.41, p o0.001, ηp2 ¼ 0.64) – correct responses
were given slower in trials following error occurrence as compared
to trials preceding them. Again, no group (F(2,44) ¼ 1.44, p ¼0.249)
or interaction effects (F(2,44) ¼0.13, p ¼0.876) were observed.
The questionnaire analyses revealed no significant group differences for the BIS/BAS Scales (all p-values 40.317), the BIS-11
Scales (all p-values 40.189), or for the SURPS (all p-values
40.063). Table 1 provides behavioural and questionnaire results
per group.
246
D.M. Pfabigan et al. / Neuropsychologia 77 (2015) 242–252
Table 1
Means and standard deviation (SD) of reaction times (in ms), error percentage, and
post-error slowing (in ms) of the respective conditions and mean scores and SD of
the questionnaire data (BIS/BAS scales, BIS-11, and SURPS) are depicted separately
for high, intermediate, and low PDYN expression participants.
HH PDYN group
LH PDYN group
LL PDYN group
M
M
M
SD
SD
Table 2
Means and standard deviation (SD) of peak-to-peak ERP values for ERN, late Pe, and
N2 components (in μV) for the separate conditions and of peak-to-peak ERP values
for ΔERN (in μV), separately for high, intermediate, and low PDYN expression
participants at electrodes Fz, Cz, and Pz.
HH PDYN
group
373.34
428.94
409.21
322.98
30.37
38.50
32.29
22.14
345.08
403.79
383.07
307.32
37.19
42.25
35.77
23.99
362.64
423.15
399.71
323.60
31.47
45.07
33.72
50.47
ERN
Fz
Cz
Pz
Error percentage
congruent
incongruent
total
total number of
errors
Post-error
slowing
pre-error
post-error
BIS/BAS scales
BIS Total
BAS Total
BAS-Drive
BAS-Fun Seeking
BAS-Reward
Responsiveness
BIS-11
BIS-11 Total
Attentional
Impulsiveness
Motor
Impulsiveness
Non-planning
Impulsiveness
SURPS
Hopelessness
Anxiety Sensitivity
Impulsivity
Sensation Seeking
1.72
18.40
20.12
27.31
2.38
12.55
13.72
19.37
1.25
24.49
25.74
38.25
1.16
16.81
16.99
25.68
LL PDYN
group
SD
M
Reaction times
congruent
incongruent
correct
error
LH PDYN
group
2.54
19.21
21.75
31.20
3.48
10.93
12.94
19.42
late Pe
Fz
Cz
Pz
N2
382.04
409.06
34.74
38.58
362.30
386.41
36.34
39.58
375.67
399.54
38.14
34.13
2.70
3.09
3.13
2.84
3.25
0.64
0.39
0.45
0.60
0.49
2.90
3.10
3.10
2.85
3.31
0.46
0.37
0.38
0.57
0.51
2.90
3.10
3.10
2.85
3.31
0.46
0.37
0.38
0.57
0.51
59.56
15.50
9.03
2.63
64.20
15.87
7.88
2.83
62.73
17.13
12.72
3.60
22.94
3.91
24.00
3.57
23.00
4.94
21.13
4.79
24.33
3.75
22.60
5.67
12.25
10.31
9.50
17.25
2.96
3.38
2.19
3.55
11.53
12.00
11.20
14.87
2.95
3.14
2.04
4.64
12.80
13.07
10.27
15.40
3.49
3.01
3.31
3.40
3.2. ERP results
Means and standard deviations for all ERP measures are provided in Table 2. Fig. 1 depicts amplitude courses of the three
groups for response-locked and stimulus-locked ERPs. For the ERN
ANOVA, significant main effects were observed for electrode site (F
(2,88) ¼ 19.42, p(GG) o0.001, ηp2 ¼0.31) and type (F(1,44) ¼ 102.29,
p o0.001, ηp2 ¼0.70). Moreover, significant interaction effects
were found for electrode site type (F(2,88) ¼ 18.11, p o0.001,
ηp2 ¼0.29) and group type (F(2,44) ¼3.34, p ¼0.045, ηp2 ¼0.13).
The main effect for group (F(2,44) ¼ 2.09, p ¼0.136) and the remaining interactions (all p-values 40.557) were not significant.
Tukey post-hoc tests showed that ERN amplitudes were comparable at electrode sites Fz and Cz after errors (p¼ 0.999) and at
electrode sites Fz, Cz, and Pz after correct responses (all p-values
40.340). In general, errors led to more negative ERN amplitudes
than positive responses (all p-values o 0.001). Concerning the
group type interaction, correct responses yielded comparable
ERN amplitudes in all three group (all p-values 4 0.999). Erroneous responses yielded enhanced ERN amplitudes compared to
Fz
Cz
Pz
SD
M
SD
M
SD
Correct
Error
Correct
Error
Correct
Error
4.13
14.30
3.56
14.02
3.65
9.89
3.39
7.93
3.36
9.32
3.43
5.63
3.58
9.26
3.26
10.25
3.36
6.45
2.66
2.85
3.26
3.70
2.68
3.04
4.15
10.54
3.61
10.09
2.27
6.45
2.36
4.62
1.98
5.94
2.37
3.04
Correct
Error
Correct
Error
Correct
Error
3.16
7.66
1.77
8.86
5.74
5.51
6.23
7.55
7.22
5.66
6.24
6.05
0.55
5.98
5.62
6.07
8.38
1.51
5.87
5.46
5.45
6.10
4.13
3.71
1.25
6.35
3.20
7.46
8.03
3.73
4.18
7.06
5.06
8.33
6.94
6.50
4.57
8.25
2.74
6.53
2.12
4.71
3.30
4.84
3.04
5.09
2.76
3.29
3.59
6.89
3.07
6.19
3.58
3.93
3.06
2.61
4.33
2.89
4.39
3.14
3.55
6.60
2.01
5.22
2.43
4.81
1.82
3.57
1.45
3.07
1.74
3.49
12.41
12.41
7.87
7.36
8.28
5.19
7.04
7.82
4.43
3.57
3.35
1.80
7.46
8.25
4.68
4.49
5.85
3.36
Congruent
Incongruent
Congruent
Incongruent
Congruent
Incongruent
ΔERN
Fz
Cz
Pz
correct ones in the three groups (all p-values o 0.001). Although
ERN amplitudes after errors were most pronounced in HH participants, the comparison with LL (p ¼0.288) and LH participants
(p ¼0.166) did not reach significance. LL and LH participants did
not differ regarding their ERN amplitudes after errors (p ¼0.999).
When assessing the ERN component as the difference between
incorrect and correct trials to extract error-specific processes (i.e.,
ΔERN), the ANOVA model resulted in significant main effects for
electrode site (F(2,88) ¼325.77, p o0.001, ηp2 ¼0.37) and group (F
(2,44) ¼4.47, p ¼0.017, ηp2 ¼0.17). ΔERN amplitudes were more
pronounced at Fz and Cz compared to Pz (both p-values o 0.001).
Importantly, HH participants had more pronounced ΔERN amplitudes than LL (p¼ 0.019) and LH (p¼ 0.010) participants. Amplitudes of LL and LH participants did not differ from each other
(p ¼0.828) – see Fig. 2. No interaction effect was found (F(4,88) ¼
0.57, p ¼0.683).
The late Pe ANOVA model showed significant main effects for
electrode site (F(2,88) ¼51.12, p(GG) o 0.001, ηp2 ¼ 0.54) and type (F
(1,44) ¼122.18, p o0.001, ηp2 ¼ 0.74), but no significant effect for
group (F(2,44) ¼1.63, p ¼0.208). The electrode site type interaction was significant (F(2,88) ¼38.74, p o0.001, ηp2 ¼0.47), the
others did not reach significance (all p-values 40.541). Tukey
post-hoc tests showed that again error commission yielded largest
Pe amplitudes at all electrode sites (all p-values o0.001). Errors
lead to comparable late Pe amplitudes at electrodes Fz and Cz
(p ¼0.709), which were in turn more positive than all other conditions (all p-values o0.001).
The N2 ANOVA model showed significant main effects for
electrode site (F(2,88) ¼ 22.51, p(GG) o0.001, ηp2 ¼0.34) and type (F
(1,44) ¼34.47, p o0.001, ηp2 ¼0.44), and a significant interaction of
electrode site type (F(2,88) ¼7.41, p ¼0.001, ηp2 ¼0.14). The factor
group (F(2,44) ¼0.32, p ¼0.730) and the remaining interaction effects were not significant (all p-values 40.112). Post-hoc tests of
the significant interaction effect showed that N2 amplitudes were
D.M. Pfabigan et al. / Neuropsychologia 77 (2015) 242–252
247
Fig. 1. Grand average waveforms separately for high, intermediate and low PDYN expression participants of response-locked correct and error trials at Cz (upper panel) and
of stimulus-locked congruent and incongruent trials at Fz (lower panel). Time point zero indicates participants’ button press (upper panel) and the onset of the four flanking
arrow stimuli (lower panel). Negative is drawn upwards per convention.
most pronounced at Fz during incongruent stimulus arrays (all pvalues o0.004). At all electrode sites, incongruent stimulus arrays
elicited larger N2 amplitudes than congruent ones (all p-values
o0.001).
3.3. Correlation analyses
Supporting the observed group differences for ΔERN amplitudes, a significant positive correlation was found between averaged ΔERN values at Fz and Cz and individual allele numbers
(r ¼ 0.340, p ¼0.019) – higher ΔERN amplitudes were associated
with higher allele numbers.
We did not find significant correlations between the ERPs and
the BIS-11 scales (all pcorr-values 4 0.020) or the BIS/BAS scales
(all pcorr-values 40.006) in HH or LL participants. For the SURPs,
we observed a significant correlation between N2 amplitudes for
incongruent stimulus arrays and the Hopelessness scale in HH
participants (rs ¼ 0.745, pcorr ¼0.001), no other correlation
reached the significance level (all pcorr-values 4 0.006).
4. Discussion
This is the first study investigating whether or not genetic
variation related to endogenous opioids is implicated in the
modulation of electrophysiological correlates of reward prediction
error signals (RPEs) and performance monitoring in humans. Although all groups showed comparable behavioural performance,
high PDYN expression individuals showed enhanced error-related
neural activity (ΔERN amplitudes) in comparison to intermediate
and low PDYN expression ones – which was observable in significant group differences, but also when using a dimensional
approach to assess the relation between the PDYN polymorphism
and RPEs. In contrast, no group differences were observed for late
Pe and N2 amplitudes.
4.1. PDYN group differences
The enhanced ΔERN amplitudes in individuals with high PDYN
expression in comparison to those with intermediate and low
PDYN expression rates corroborate the assumption of an indirect
248
D.M. Pfabigan et al. / Neuropsychologia 77 (2015) 242–252
Fig. 2. Grand average waveforms separately for high, intermediate and low PDYN expression participants of ΔERN amplitudes at Cz (left panel). Negative is drawn upwards
per convention; time point zero indicates participants’ button press. The right-hand panel depicts scalp topographies of mean ΔERN activation in the time window 0–50 ms
after button press, separately for the three groups.
effect of PDYN on electrophysiological correlates of performance
monitoring and RPEs via its relation to the dopaminergic system –
in particular during error commission. Nevertheless, the observed
effects were not in the expected direction since individuals with
the high PDYN expression polymorphism showed enhanced ΔERN
amplitudes and not diminished ones, as hypothesized.
ERN amplitude variation is mostly seen as a reinforcement
learning signal (Holroyd and Coles, 2002) or as a signal of motivational error significance (Gehring and Willoughby, 2002; Inzlicht and Al-Khindi, 2012; Yeung et al., 2005). The influential
reinforcement-learning-theory (RL-theory; Holroyd and Coles,
2002) relates ERN amplitude variation to changes in phasic dopamine transmission in the mesencephalon. Events worse than
expected are assumed to yield phasic mesencephalic dopamine
decrease which is associated with enhanced ERN amplitudes (and
concurrently reflecting a negative RPE); whereas events better
than expected are assumed to yield phasic mesencephalic dopamine increase which is associated with diminished ERN amplitudes (reflecting a positive RPE). Recent theories on prediction
errors suggest that ERN or Feedback-Related Negativity (FRN)2
amplitude variation rather reflects an absolute reward prediction
error signal without a numerical sign (Alexander and Brown, 2011;
Chase et al., 2011; Hauser et al., 2014; Talmi et al., 2013) than
outcomes better or worse than expected. These unsigned prediction error signals are assumed to represent the surprise which is
accompanying unexpected outcomes (Hayden et al., 2011). Thus,
the current results of error-specific ΔERN enhancement point towards the assumption that high PDYN expression individuals
showed enhanced RPE signals in comparison to low and
2
The Feedback-Related Negativity component (FRN; Miltner et al., 1997) is an
indicator of external performance monitoring, whereas the ERN component reflects
internal performance monitoring processes.
intermediate PDYN expression ones. Regarding performance
monitoring ERPs, the effects of different PDYN polymorphisms do
not seem to be comparable to the effects of psychopharmacological studies directly manipulating dopamine neurotransmission.
Assuming that PDYN expression levels modulate mesencephalic
dopamine levels differently, the theory of tonic vs. phasic dopamine effects (Grace, 1991, 1993; Moore et al., 1999) might help to
explain the results. The theory claims that two processes are involved in regulating the dynamics of dopamine transmission in
limbic and striatal areas. On the one hand, there is transient phasic
dopamine release with high amplitudes which is mediated by
burst firing of dopamine neurons. On the other hand, there is
constant low-level tonic dopamine release which is mediated by
baseline firing of dopamine neurons and corticostriatal glutamatergic afferents. The phasic dopamine release is supposed to
transmit prediction error signals. Importantly, it was proposed
that tonic dopamine release regulates the amplitude of the phasic
dopamine bursts by stimulating autoreceptors located on the dopaminergic neurons. Thus, tonic dopamine release determines the
responsivity of the dopamine system per se and serves to suppress
phasic dopamine release via the activation of these autoreceptors
(see Floresco et al., 2003; Grace, 1991). Consequently, low tonic
levels of dopamine – caused by high levels of dynorphin, as suggested in high PDYN expression individuals (Zimprich et al., 2000)
– might, for a given prediction error signal, result in a relatively
higher dopamine release and thereby could explain enhanced
ΔERN amplitudes in high PDYN expression individuals compared
to intermediate and low PDYN expression ones.
However, other authors argued that phasic dopamine dips
should be more pronounced in case of high tonic dopamine levels
because there is a larger range available for the phasic dip (Kenemans and Kahkonen, 2011). More precisely, when tonic dopamine levels are high, even small phasic changes will result in
D.M. Pfabigan et al. / Neuropsychologia 77 (2015) 242–252
relatively large ERN differences. This assumption is in line with
ERN amplitude enhancement observed after administration of
dopamine agonists (Barnes et al., 2014; De Bruijn et al., 2004), but
cannot be reconciled with the current results. Future studies are
therefore needed which clarify the interaction of the variants of
the functional PDYN 68 bp VNTR polymorphism, resulting dynorphin levels, and dopamine transmission in mesencephalic structures, and how this affects ERN amplitude variation.
Further linking RPE, opioid and dopamine systems, recent
studies using electrophysiological, functional, and structural imaging methods reported neuronal generators for ERN/FRN components in dopamine-innervated striatal areas (Becker et al., 2014;
Carlson et al., 2015; Carlson et al., 2011; Foti et al., 2011). In particular Becker et al. (2014) suggested that the ventral striatum
exerts a modulating contribution to the FRN scalp topography. Also
dynorphin peptides and κ-opioid receptors are accumulated in
ventral striatal areas (Fallon and Leslie, 1986; Svingos et al., 1989,
2001) – thereby suggesting a shared contribution to scalp-recorded ERPs reflecting RPEs.
Importantly, no group differences were observed for late Pe
amplitudes, i.e., more elaborate stages after error commission.
Therefore, we assume that conscious error awareness (Nieuwenhuis et al., 2001) or affective error responses (Falkenstein
et al., 2000) were comparable in the three PDYN groups. Thus,
despite initial hyperactivity of the performance monitoring system
in high PDYN expression individuals, they were nevertheless able
to regulate later processing stages to be comparable with the activation patterns of the two other groups.
4.2. Implications of ERN enhancement
There is an ongoing debate as to whether enhanced ERN/ΔERN
amplitudes functionally indicate a positive or a negative adaptation of the performance monitoring system – either heightened
awareness of cognitive conflict or an overactive performance
monitoring system (Larson et al., 2014). Increased ERN amplitudes
have been reported to be an individual trait marker, but have also
been observed after experimental state manipulations. For example, ERN amplitudes were enhanced in individuals scoring high
on negative affect scales compared to low-scoring individuals
(Hajcak et al., 2004; Luu et al., 2000). Concerning state manipulations, ERN enhancement was observed after the presentation of
derogatory feedback (Wiswede et al., 2009b), after the induction
of self-relevant failure (Unger et al., 2012), after the induction of
feelings of helplessness (Pfabigan et al., 2013), and after punishment threat (Riesel et al., 2012). In this context, ERN amplitude
enhancement has been mostly interpreted as an indicator of motivational significance (Gehring and Willoughby, 2002; Inzlicht
and Al-Khindi, 2012; Yeung et al., 2005). Errors might be more
salient events for negative affect-prone individuals or after the
induction of negative affective states. Thus, one could speculate
that high PDYN expression participants – with high levels of PDYN
and consequently low levels of dopamine as continuous/chronic
state-belong to the group of this negative affect-prone individuals
in which ERN amplitude enhancement could indicate heightened
risk for internalizing disorders (Olvet and Hajcak, 2008), which are
characterized by anxious, fearful, and depressive symptoms and
behavioural tendencies (Krueger, 1999; Vollebergh et al., 2001).
Recent research suggests indirectly that the dynorphin and κopioid receptor system is implicated in different aspects of psychopathology (Tejeda et al., 2012). For example, McLaughlin et al.,
(2005) summarized that κ-opioid receptor agonists induce “prodepressive” effects while κ-opioid receptor antagonists induce
“anti-depressant” effects. Furthermore, dynorphins are also implicated in the maintenance of substance abuse disorders (Margolis et al., 2008; Wee and Koob, 2010). Relating prediction error
249
signals and these disorders, studies investigating performance
monitoring in substance-use disorder individuals observed decreased ERN amplitudes in comparison to healthy controls (Franken et al., 2007; Morie et al., 2014; Sokhadze et al., 2008); even in
high-risk groups (Euser et al., 2013) – thereby indicating a hypoactive performance monitoring system in these individuals. Thus,
the dynorphin and κ-opioid receptor system might be indirectly
implicated in either hyper- or hypo-reactivity to errors. Research
assessing the relation between different PDYN polymorphism
variants, performance monitoring, and clinical symptoms might be
a promising research avenue in the future.
Moreover, our interpretation of enhanced motivational significance of erroneous events in high PDYN expression individuals
is in line with recent findings suggesting that they also display
enhanced sensitivity for upcoming rewards (Votinov, et al., 2014).
Importantly, in the current study, all participants received the
same financial bonus after task completion to avoid confounding
effects of external incentives on performance monitoring ERPs
(Van den Berg et al., 2012). Performance-based external incentives
might have induced additional motivation in particular in individuals more sensitive to expected rewards (Votinov et al., 2014).
Thus, both the current data and the study by Votinov et al. (2014)
point towards a hyper-active performance monitoring system in
high PDYN expression individuals.
4.3. Conflict N2 results
As expected, N2 amplitude variation differentiated between
congruent and incongruent stimulus arrays (Nieuwenhuis et al.,
2003), thereby further validating the current paradigm. Both ERN
and conflict N2 amplitudes are assumed to be generated within
the aMCC (Debener et al., 2005; Gruendler et al., 2011; Hoffmann
and Falkenstein, 2010; Van Veen and Carter, 2002; Yeung et al.,
2004) and Yeung et al. (2004) even proposed that both reflect similar cognitive processes related to conflict monitoring (either
response- or stimulus-driven). However, in contrast to ERN amplitudes, conflict N2 amplitudes were not affected by PDYN genotype. Differences in PDYN availability might only affect particular
stages of performance monitoring such as the automatic evaluation of the error information (reflected in ΔERN amplitude differences), and not the initial perceptual processing of stimulusdriven conflict as reflected in N2 amplitudes.
4.4. Behavioural and questionnaire results
The behavioural results are in line with previous results, replicating findings of faster reaction times for error responses, lower
error rates for congruent trials (Eriksen and Eriksen, 1974), and
post-error slowing effects (Rabbitt, 1966) – thus generally validating
the administered experimental paradigm. However, the three PDYN
groups did not differ in any behavioural measure. This is in line
with Weinberg et al., (2012) who summarize that ERN amplitude
enhancement is rarely accompanied by behavioural effects.
Individuals might be able to compensate behaviourally for the
observed neuronal alterations when performing simple stimulus–
response tasks as in the current study (Miller, 1996). Importantly,
the comparable error rates in all groups strengthen the current
results since ERN amplitude variation is also influenced by error
frequency (Hajcak et al., 2003; Yeung et al., 2004). Thus, the observed group effects are not related to group-wise error frequency.
Interestingly, when confronting high PDYN expression individuals
with more demanding cognitive control tasks such as a reversal
learning paradigm, enhanced error rates in high compared to low
PDYN expression individuals were observed. The more complex
performance monitoring and behavioural adaptation processes
seem to be less flexible in high compared to low PDYN expression
250
D.M. Pfabigan et al. / Neuropsychologia 77 (2015) 242–252
individuals. This assumption is also reflected in fewer cortical activation patterns and decreased functional connectivity in brain
areas associated with cognitive control (Votinov et al., 2015).
Since the current groups were matched for age, sex, and education, and since the applied questionnaires did not reveal reliable
group differences, the current ERP group differences can also not
be explained by these variables.
Limitations of the current study pertain to the small sample
size investigated. Although the effect sizes of the ERN/ΔERN ANOVA results can be considered as medium to high (according to
effect size classifications based on partial eta-squared; Kirk, 1996),
future studies should nevertheless strive for larger group size
when investigating the effects of functional polymorphisms on
ERP correlates. Moreover, the current results only demonstrate
indirect evidence for effects of the endogenous opioid system on
performance monitoring. Our hypotheses were based on findings
by Zimprich et al. (2000) and Nikoshkov et al. (2008). Contrary to
these authors, recent evidence claims that alleles with fewer repeat copies of the 68 bp VNTR polymorphism are associated with
higher transcriptional activation and not those with more repeat
copies (Rouault et al., 2011). Further pharmaco-genetic studies are
therefore necessary to investigate the link between functional
polymorphisms and PDYN expression rates.
5. Conclusion
The current results indicate an impact of genetic variation in
the endogenous opioid system on specific electrophysiological
correlates of performance monitoring and reward prediction error
signals. In particular, we observed a hyper-active performance
monitoring system probably based on enhanced prediction error
signals in individuals with the high PDYN expression polymorphism (Zimprich et al., 2000) compared to those with intermediate
and low PDYN expression. It was the automatic response evaluation stage, but not earlier perceptual conflict or later conscious
error processing stages, which differentiated high expression
PDYN participants from the other participants. This hyper-reactivity to committed errors despite successful behavioural compensation might be a characteristic of the probable association
between PDYN, dynorphin and dopamine levels, and internalizing
disorders.
Conflict of interest
All authors declare that this research project was conducted in
the absence of any commercial or financial relationship that could
be construed as a potential conflict of interest.
Acknowledgements
Parts of this study were presented at the 53rd Annual Meeting
of the Society for Psychophysiological Research, held in Firenze,
Italy. We thank Alexander Zimprich and Fritz Zimprich for the
genetic analyses and Mikhail Votinov and Christoph Huber-Huber
for technical assistance. This study was supported by the Austrian
Science Fund (FWF): P22813-B09. Participants' financial remuneration was funded by a scholarship of the University of
Vienna awarded to SLK.
References
Agam, Y., Vangel, M., Roffman, J.L., Gallagher, P.J., Chaponis, J., Haddad, S., Goff, D.C.,
Greenberg, D.S., Wilhelm, S., Smoller, J.W., Manoach, D.S., 2014. Dissociable
genetic contributions to error processing: a multimodal neuroimaging study.
PLoS ONE 9, e101784.
Alexander, W.H., Brown, J.W., 2011. Medial prefrontal cortex as an action-outcome
predictor. Nat. Neurosci. 14, 1338–1344.
Althaus, M., Groen, Y., Wijers, A.A., Minderaa, R.B., Kema, I.P., Dijck, J.D., Hoekstra, P.
J., 2010. Variants of the SLC6A3 (DAT1) polymorphism affect performance
monitoring-related cortical evoked potentials that are associated with ADHD.
Biol. Psychol 85, 19–32.
Barnes, J.J., O’Connell, R.G., Nandam, L.S., Dean, A.J., Bellgrove, M.A., 2014. Monoaminergic modulation of behavioural and electrophysiological indices of error
processing. Psychopharmacology 231, 379–392.
Bauer, H., Pripfl, J., Lamm, C., Prainsack, C., Taylor, N., 2003. Functional neuroanatomy of learned helplessness. Neuroimage 20, 927–939.
Becker, M.P., Nitsch, A.M., Miltner, W.H., Straube, T., 2014. A single-trial estimation
of the feedback-related negativity and its relation to BOLD responses in a timeestimation task. J. Neurosci. 34, 3005–3012.
Bell, A.J., Sejnowski, T.J., 1995. An information-maximization approach to blind
separation and blind deconvolution. Neural Comput. 7, 1129–1159.
Biehl, S.C., Dresler, T., Reif, A., Scheuerpflug, P., Deckert, J., Herrmann, M.J., 2011.
Dopamine transporter (DAT1) and dopamine receptor D4 (DRD4) genotypes
differentially impact on electrophysiological correlates of error processing. PLoS
One 6, e28396.
Bodnar, R.J., 2011. Endogenous opiates and behavior: 2010. Peptides 32, 2522–2552.
Bruijnzeel, A.W., 2009. kappa-Opioid receptor signaling and brain reward function.
Brain Res. Rev. 62, 127–146.
Carlson, J.M., Foti, D., Harmon-Jones, E., Proudfit, G.H., 2015. Midbrain volume
predicts fMRI and ERP measures of reward reactivity. Brain Struct. Funct. 220,
1861–1866.
Carlson, J.M., Foti, D., Mujica-Parodi, L.R., Harmon-Jones, E., Hajcak, G., 2011. Ventral
striatal and medial prefrontal BOLD activation is correlated with reward-related
electrocortical activity: a combined ERP and fMRI study. Neuroimage 57,
1608–1616.
Carver, C.S., White, T.L., 1994. Behavioral inhibition, behavioral activation, and affective responses to impending reward and punishment: the BIS/BAS Scales. J.
Personal. Soc. Psychol. 67, 319–333.
Chase, H.W., Swainson, R., Durham, L., Benham, L., Cools, R., 2011. Feedback-related
negativity codes prediction error but not behavioral adjustment during probabilistic reversal learning. J. Cogn. Neurosci. 23, 936–946.
Chavkin, C., James, I.F., Goldstein, A., 1982. Dynorphin is a specific endogenous ligand of the kappa opioid receptor. Science 215, 413–415.
Cole, S., McNally, G.P., 2007. Temporal-difference prediction errors and Pavlovian
fear conditioning: role of NMDA and opioid receptors. Behav. Neurosci. 121,
1043–1052.
Corbett, A.D., Paterson, S.J., McKnight, A.T., Magnan, J., Kosterlitz, H.W., 1982. Dynorphin and dynorphin are ligands for the kappa-subtype of opiate receptor.
Nature 299, 79–81.
De Bruijn, E.R., Sabbe, B.G., Hulstijn, W., Ruigt, G.S., Verkes, R.J., 2006. Effects of
antipsychotic and antidepressant drugs on action monitoring in healthy volunteers. Brain Res. 1105, 122–129.
De Bruijn, E.R.A., Hulstijn, W., Verkes, R.J., Ruigt, G.S.F., Sabbe, B.G.C., 2004. Druginduced stimulation and suppression of action monitoring in healthy volunteers. Psychopharmacology 177, 151–160.
Debener, S., Ullsperger, M., Siegel, M., Fiehler, K., Von Cramon, D.Y., Engel, A.K.,
2005. Trial-by-trial coupling of concurrent electroencephalogram and functional magnetic resonance imaging identifies the dynamics of performance
monitoring. J. Neurosci. 25, 11730–11737.
Dutilh, G., van Ravenzwaaij, D., Nieuwenhuis, S., van der Maas, H.L.J., Forstmann, B.
U., Wagenmakers, E.-J., 2012. How to measure post-error slowing: a confound
and a simple solution. J. Math. Psychol. 56, 208–216.
Endrass, T., Reuter, B., Kathmann, N., 2007. ERP correlates of conscious error recognition: aware and unaware errors in an antisaccade task. Eur. J. Neurosci. 26,
1714–1720.
Eriksen, B.A., Eriksen, C.W., 1974. Effects of noise letters upon the identification of a
target letter in a nonsearch task. Percept. Psychophys. 16, 143–149.
Euser, A.S., Evans, B.E., Greaves-Lord, K., Huizink, A.C., Franken, I.H., 2013. Diminished error-related brain activity as a promising endophenotype for substanceuse disorders: evidence from high-risk offspring. Addict. Biol. 18, 970–984.
Falkenstein, M., Hohnsbein, J., Hoormann, J., 1995. Event-related potential correlates of errors in reaction tasks. Electroencephalogr. Clin. Neurophysiol. Suppl.
44, 287–296.
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. Electroencephalogr. Clin. Neurophysiol. 78, 447–455.
Falkenstein, M., Hoormann, J., Christ, S., Hohnsbein, J., 2000. ERP components on
reaction errors and their functional significance: a tutorial. Biol. Psychol. 51,
87–107.
Fallon, J.H., Leslie, F.M., 1986. Distribution of dynorphin and enkephalin peptides in
the rat brain. J. Comp. Neurol. 249, 293–336.
Floresco, S.B., West, A.R., Ash, B., Moore, H., Grace, A.A., 2003. Afferent modulation
of dopamine neuron firing differentially regulates tonic and phasic dopamine
transmission. Nat. Neurosci. 6, 968–973.
Folstein, J.R., Van Petten, C., 2008. Influence of cognitive control and mismatch on
the N2 component of the ERP: a review. Psychophysiology 45, 152–170.
Foti, D., Weinberg, A., Dien, J., Hajcak, G., 2011. Event-related potential activity in
the basal ganglia differentiates rewards from nonrewards: temporospatial
principal components analysis and source localization of the feedback
D.M. Pfabigan et al. / Neuropsychologia 77 (2015) 242–252
negativity. Hum. Brain Mapp. 32, 2207–2216.
Franken, I.H., van Strien, J.W., Franzek, E.J., van de Wetering, B.J., 2007. Error-processing deficits in patients with cocaine dependence. Biol. Psychol. 75, 45–51.
Friston, K., 2010. The free-energy principle: a unified brain theory? Nat. Rev. Neurosci. 11, 127–138.
Gehring, W.J., Goss, B., Coles, M.G.H., Meyer, D.E., Donchin, E., 1993. A neural system
for error detection and compensation. Psychol. Sci. 4, 385–390.
Gehring, W.J., Willoughby, A.R., 2002. The medial frontal cortex and the rapid
processing of monetary gains and losses. Science 295, 2279–2282.
Grace, A.A., 1991. Phasic versus tonic dopamine release and the modulation of
dopamine system responsivity: a hypothesis for the etiology of schizophrenia.
Neuroscience 41, 1–24.
Grace, A.A., 1993. Cortical regulation of subcortical dopamine systems and its
possible relevance to schizophrenia. J. Neural Transm. Gen. Sect. 91, 111–134.
Gruendler, T.O., Ullsperger, M., Huster, R.J., 2011. Event-related potential correlates
of performance-monitoring in a lateralized time-estimation task. PLoS One 6,
e25591.
Hajcak, G., McDonald, N., Simons, R.F., 2003. To err is autonomic: error-related brain
potentials, ANS activity, and post-error compensatory behavior. Psychophysiology 40, 895–903.
Hajcak, G., McDonald, N., Simons, R.F., 2004. Error-related psychophysiology and
negative affect. Brain Cogn. 56, 189–197.
Hajcak, G., Moser, J.S., Yeung, N., Simons, R.F., 2005. On the ERN and the significance
of errors. Psychophysiology 42, 151–160.
Hauser, T.U., Iannaccone, R., Stampfli, P., Drechsler, R., Brandeis, D., Walitza, S.,
Brem, S., 2014. The feedback-related negativity (FRN) revisited: new insights
into the localization, meaning and network organization. Neuroimage 84,
159–168.
Hayden, B.Y., Heilbronner, S.R., Pearson, J.M., Platt, M.L., 2011. Surprise signals in
anterior cingulate cortex: neuronal encoding of unsigned reward prediction
errors driving adjustment in behavior. J. Neurosci. 31, 4178–4187.
Hoffmann, S., Falkenstein, M., 2010. Independent component analysis of erroneous
and correct responses suggests online response control. Hum. Brain Mapp. 31,
1305–1315.
Holroyd, C.B., Coles, M.G.H., 2002. The neural basis of human error processing:
reinforcement learning, dopamine, and the error-related negativity. Psychol.
Rev. 109, 679–709.
Inzlicht, M., Al-Khindi, T., 2012. ERN and the placebo: a misattribution approach to
studying the arousal properties of the error-related negativity. J. Exp. Psychol.
Gen. 141, 799–807.
Kenemans, J.L., Kahkonen, S., 2011. How human electrophysiology informs psychopharmacology: from bottom-up driven processing to top-down control.
Neuropsychopharmacology 36, 26–51.
Kirk, R.E., 1996. Practical significance: a concept whose time has come. Educ. Psychol. Meas. 56, 746–759.
Knoll, A.T., Carlezon Jr., W.A., 2010. Dynorphin, stress, and depression. Brain Res.
1314, 56–73.
Kopp, B., Rist, F., Mattler, U., 1996. N200 in the flanker task as a neurobehavioral
tool for investigating executive control. Psychophysiology 33, 282–294.
Kramer, U.M., Cunillera, T., Camara, E., Marco-Pallares, J., Cucurell, D., Nager, W.,
Bauer, P., Schule, R., Schols, L., Rodriguez-Fornells, A., Munte, T.F., 2007. The
impact of catechol-O-methyltransferase and dopamine D4 receptor genotypes
on neurophysiological markers of performance monitoring. J. Neurosci. 27,
14190–14198.
Kreek, M.J., Bart, G., Lilly, C., LaForge, K.S., Nielsen, D.A., 2005. Pharmacogenetics
and human molecular genetics of opiate and cocaine addictions and their
treatments. Pharmacol. Rev. 57, 1–26.
Kreek, M.J., LaForge, K.S., Butelman, E., 2002. Pharmacotherapy of addictions. Nat.
Rev. Drug Discov. 1, 710–726.
Krueger, R.F., 1999. The structure of common mental disorders. Arch. Gen. Psychiatry 56, 921–926.
Larson, M.J., Clayson, P.E., Clawson, A., 2014. Making sense of all the conflict: a
theoretical review and critique of conflict-related ERPs. Int. J. Psychophysiol. 93,
283–297.
Le Merrer, J., Becker, J.A., Befort, K., Kieffer, B.L., 2009. Reward processing by the
opioid system in the brain. Physiol. Rev. 89, 1379–1412.
Lee, T.W., Girolami, M., Sejnowski, T.J., 1999. Independent component analysis using
anextended infomax algorithm for mixed subgaussian and supergaussian
sources. Neural Comput. 11, 417–441.
Luck, S.J., 2005. An Introduction to the Event-related Technique. MIT Press,
Cambridge.
Lutz, P.-E., Kieffer, B.L., 2013. Opioid receptors: distinct roles in mood disorders.
Trends Neurosci 36, 195–206.
Luu, P., Collins, P., Tucker, D.M., 2000. Mood, personality, and self-monitoring:
Negative affect and emotionality in relation to frontal lobe mechanisms of error
monitoring. J. Exp. Psychol.: Gen. 129, 43–60.
Margolis, E.B., Lock, H., Chefer, V.I., Shippenberg, T.S., Hjelmstad, G.O., Fields, H.L.,
2006. Kappa opioids selectively control dopaminergic neurons projecting to the
prefrontal cortex. Proc. Natl. Acad. Sci. U.S.A. 103, 2938–2942.
Margolis, E.B., Mitchell, J.M., Ishikawa, J., Hjelmstad, G.O., Fields, H.L., 2008. Midbrain dopamine neurons: projection target determines action potential duration and dopamine D(2) receptor inhibition. J. Neurosci. 28, 8908–8913.
Matzel, L., Hallam, S., Miller, R., 1988. Contribution of conditioned opioid analgesia
to the shock-induced associative US-preexposure deficit. Anim. Learn. Behav.
16, 486–492.
McLaughlin, J.P., Li, S., Valdez, J., Chavkin, T.A., Chavkin, C., 2005. Social defeat
251
stress-induced behavioral responses are mediated by the endogenous kappa
opioid system. Neuropsychopharmacology 31, 1241–1248.
McNally, G.P., Cole, S., 2006. Opioid receptors in the midbrain periaqueductal gray
regulate prediction errors during pavlovian fear conditioning. Behav. Neurosci.
120, 313–323.
Meyer, A., Klein, D.N., Torpey, D.C., Kujawa, A.J., Hayden, E.P., Sheikh, H.I., Singh, S.
M., Hajcak, G., 2012. Additive effects of the dopamine D2 receptor and dopamine transporter genes on the error-related negativity in young children. Genes
Brain Behav. 11, 695–703.
Miller, G.A., 1996. How we think about cognition, emotion, and biology in psychopathology. Psychophysiology 33, 615–628.
Miltner, W.H.R., Braun, C.H., Coles, M.G.H., 1997. Event-related brain potentials
following incorrect feedback in a time-estimation task: Evidence for a ‘generic’
neural system for error detection. J. Cogn. Neurosci. 9, 788–798.
Moore, H., West, A.R., Grace, A.A., 1999. The regulation of forebrain dopamine
transmission: relevance to the pathophysiology and psychopathology of schizophrenia. Biol. Psychiatry 46, 40–55.
Morie, K.P., De Sanctis, P., Garavan, H., Foxe, J.J., 2014. Executive dysfunction and
reward dysregulation: a high-density electrical mapping study in cocaine
abusers. Neuropharmacology 85, 397–407.
Mueller, E.M., Burgdorf, C., Chavanon, M.L., Schweiger, D., Hennig, J., Wacker, J.,
Stemmler, G., 2014. The COMT Val158Met polymorphism regulates the effect of
a dopamine antagonist on the feedback-related negativity. Psychophysiology
51, 805–809.
Mueller, E.M., Makeig, S., Stemmler, G., Hennig, J., Wacker, J., 2011. Dopamine effects on human error processing depend on catechol-O-methyltransferase
VAL158MET genotype. J. Neurosci. 31, 15818–15825.
Nieuwenhuis, S., Richard Ridderinkhof, K., Blom, J., Band, G.P.H., Kok, A., 2001. Error-related brain potentials are differentially related to awareness of response
errors: evidence from an antisaccade task. Psychophysiology 38, 752–760.
Nieuwenhuis, S., Yeung, N., van den Wildenberg, W., Ridderinkhof, K.R., 2003.
Electrophysiological correlates of anterior cingulate function in a go/no-go task:
effects of response conflict and trial type frequency. Cogn., Affect. Behav.
Neurosci. 3, 17–26.
Nikoshkov, A., Drakenberg, K., Wang, X., Horvath, M.C., Keller, E., Hurd, Y.L., 2008.
Opioid neuropeptide genotypes in relation to heroin abuse: dopamine tone
contributes to reversed mesolimbic proenkephalin expression. Proc. Natl. Acad.
Sci. U.S.A. 105, 786–791.
Oldfield, R.C., 1971. The assessment and analysis of handedness: the Edinburgh
inventory. Neuropsychologia 9, 97–113.
Olvet, D.M., Hajcak, G., 2008. The error-related negativity (ERN) and psychopathology: Toward an endophenotype. Clin. Psychol. Rev. 28, 1343–1354.
Olvet, D.M., Hajcak, G., 2009. Reliability of error-related brain activity. Brain Res.
1284, 89–99.
Patton, J.H., Stanford, M.S., Barratt, E.S., 1995. Factor structure of the Barratt impulsiveness scale. J. Clinical Psychology 51, 768–774.
Pecina, M., Stohler, C.S., Zubieta, J.K., 2014. Neurobiology of placebo effects: expectations or learning? Soc. Cogn. Affect. Neurosci. 9, 1013–1021.
Pfabigan, D.M., Pintzinger, N.M., Siedek, D.R., Lamm, C., Derntl, B., Sailer, U., 2013.
Feelings of helplessness increase ERN amplitudes in healthyindividuals. Neuropsychologia 51, 613–621.
Picton, T.W., Hillyard, S.A., 1972. Cephalic skin potentials in electroencephalography. Electroencephalogr. Clin. Neurophysiol. 33, 419–424.
Rabbitt, P.M., 1966. Errors and error correction in choice-response tasks. J. Exp.
Psychol. 71, 264–272.
Rescorla, R.A., Wagner, A.R., 1972. A theory of Pavlovian conditioning: Variations in
the effectiveness of reinforcement and nonreinforcement. Class. Cond. II: Curr.
Res. Theory, 64–99.
Riesel, A., Weinberg, A., Endrass, T., Kathmann, N., Hajcak, G., 2012. Punishment has
a lasting impact on error-related brain activity. Psychophysiology 49, 239–247.
Rouault, M., Nielsen, D.A., Ho, A., Kreek, M.J., Yuferov, V., 2011. Cell-specific effects
of variants of the 68-base pair tandem repeat on prodynorphin gene promoter
activity. Addict. Biol. 16, 334–346.
Schultz, W., 2007. Behavioral dopamine signals. Trends Neurosci. 30, 203–210.
Schultz, W., Dayan, P., Montague, P.R., 1997. A neural substrate of prediction and
reward. Science 275, 1593–1599.
Shippenberg, T.S., 2009. The dynorphin/kappa opioid receptor system: a new target
for the treatment of addiction and affective disorders? Neuropsychopharmacology 34, 247.
Sokhadze, E., Stewart, C., Hollifield, M., Tasman, A., 2008. Event-related potential
study of executive dysfunctions in a Speeded Reaction Task in cocaine addiction. J. Neurother. 12, 185–204.
Steiner, H., Gerfen, C.R., 1998. Role of dynorphin and enkephalin in the regulation of
striatal output pathways and behavior. Exp. Brain Res. 123, 60–76.
Svingos, A.L., Chavkin, C., Colago, E.E.O., Pickel, V.M., 2001. Major coexpression of κopioid receptors and the dopamine transporter in nucleus accumbens axonal
profiles. Synapse 42, 185–192.
Svingos, A.L., Colago, E.E., Pickel, V.M., 1999. Cellular sites for dynorphin activation
of kappa-opioid receptors in the rat nucleus accumbens shell. J. Neurosci. 19,
1804–1813.
Talmi, D., Atkinson, R., El-Deredy, W., 2013. The feedback-related negativity signals
salience prediction errors, not reward prediction errors. J. Neurosci. 33,
8264–8269.
Tejeda, H.A., Shippenberg, T.S., Henriksson, R., 2012. The dynorphin/kappa-opioid
receptor system and its role in psychiatric disorders. Cell. Mol. Life Sci. 69,
857–896.
252
D.M. Pfabigan et al. / Neuropsychologia 77 (2015) 242–252
Tops, M., Koole, S.L., Wijers, A.A., 2013. The Pe of perfectionism: concern over
mistakes predicts the amplitude of a late frontal error positivity. J. Psychophysiol. 27, 84–94.
Ullsperger, M., Harsay, H.A., Wessel, J.R., Ridderinkhof, K.R., 2010. Conscious perception of errors and its relation to the anterior insula. Brain Struct. Funct. 214,
629–643.
Unger, K., Kray, J., Mecklinger, A., 2012. Worse than feared? Failure induction
modulates the electrophysiological signature of error monitoring during subsequent learning. Cogn., Affect., Behav. Neurosci. 12, 34–51.
Van den Berg, I., Shaul, L., Van der Veen, F.M., Franken, I.H.A., 2012. The role of
monetary incentives in feedback processing: why we should pay our participants. NeuroReport 23.
Van Veen, V., Carter, C.S., 2002. The timing of action-monitoring processes in the
anterior cingulate cortex. J. Cogn. Neurosci. 14, 593–602.
Vogt, B.A., 2005. Pain and emotion interactions in subregions of the cingulate gyrus.
Nat. Rev. Neurosci. 6, 533–544.
Vollebergh, W.A., Iedema, J., Bijl, R.V., de Graaf, R., Smit, F., Ormel, J., 2001. The
structure and stability of common mental disorders: the NEMESIS study. Arch.
Gen. Psychiatry 58, 597–603.
Votinov, M., Pripfl, J., Windischberger, C., Kalcher, K., Zimprich, A., Zimprich, F.,
Moser, E., Lamm, C., Sailer, U., 2014. A genetic polymorphism of the endogenous
opioid dynorphin modulates monetary reward anticipation in the corticostriatal loop. PLoS One 9, e89954.
Votinov, M., Pripfl, J., Windischberger, C., Moser, E., Sailer, U., Lamm, C., 2015. A
functional polymorphism in the prodynorphin gene affects cognitive flexibility
and brain activation during reversal learning. Front. Behav. Neurosci. 9, 172.
Warren, C.M., Holroyd, C.B., 2012. The Impact of deliberative strategy dissociates
ERP Components related to conflict processing vs. reinforcement learning.
Front. Neurosci. 6, 43.
Wee, S., Koob, G.F., 2010. The role of the dynorphin-kappa opioid system in the
reinforcing effects of drugs of abuse. Psychopharmacology 210, 121–135.
Weinberg, A., Riesel, A., Hajcak, G., 2012. Integrating multiple perspectives on errorrelated brain activity: the ERN as a neurobehavioral trait. Motiv. Emot. 36,
84–100.
Wiswede, D., Münte, T.F., Goschke, T., Rüsseler, J., 2009a. Modulation of the errorrelated negativity by induction of short-term negative affect. Neuropsychologia
47, 83–90.
Wiswede, D., Münte, T.F., Rüsseler, J., 2009b. Negative affect induced by derogatory
verbal feedback modulates the neural signature of error detection. Soc. Cogn.
Affect. Neurosci. 4, 227–237.
Woicik, P.A., Stewart, S.H., Pihl, R.O., Conrod, P.J., 2009. The Substance Use Risk
Profile Scale: a scale measuring traits linked to reinforcement-specific substance use profiles. Addict. Behav. 34, 1042–1055.
Yeung, N., Botvinick, M.M., Cohen, J.D., 2004. The neural basis of error detection:
Conflict monitoring and the error-related negativity. Psychol. Rev. 111,
931–959.
Yeung, N., Holroyd, C.B., Cohen, J.D., 2005. ERP correlates of feedback and reward
processing in the presence and absence of response choice. Cereb. Cortex 15,
535–544.
Zimprich, A., Kraus, J., Woltje, M., Mayer, P., Rauch, E., Hollt, V., 2000. An allelic
variation in the human prodynorphin gene promoter alters stimulus-induced
expression. J. Neurochem. 74, 472–477.
Zirnheld, P.J., Carrol, C.A., Kieffaber, P.D., O’Donnel, B.F., Shekhar, A., Hetrick, W.P.,
2004. Haloperidol impairs learning and error-related negativity in humans. J.
Cognit. Neurosci. 16, 1098–1112.