Content uploaded by John W. Schwieter
Author content
All content in this area was uploaded by John W. Schwieter on Aug 30, 2022
Content may be subject to copyright.
RESEARCH ARTICLE
The COMT gene modulates the relationship between bilingual
adaptation in executive function and decision-making: an EEG study
Dongxue Liu
1,2
•John W. Schwieter
3
•Wenxin Liu
1,2
•Li Mu
1,2
•Huanhuan Liu
1,2
Received: 13 May 2022 / Revised: 11 July 2022 / Accepted: 27 July 2022
The Author(s), under exclusive licence to Springer Nature B.V. 2022
Abstract
Bilingual adaptive control mechanisms appear to be linked to congenital genetic factors such as dopamine (DA) genes.
However, it is unclear as to whether acquired cognitive exercise can vanquish innate influences that allow bilingual
executive advantages to be shown in other cognitive areas. In the present study, we examine the relationship between gene-
dependent executive control and decision-making by targeting the enzyme catecholamine-O-methyltransferase (COMT)
and employing electroencephalography (EEG). Chinese-English bilinguals (N = 101) participated in a language switching
task and the Iowa Gambling Task (IGT). The findings showed that COMT Val158Met polymorphism played a complex
role in decision-making and bilingual executive control processing: Bilinguals with Valine (Val) homozygotes had poorer
performance in the IGT, while Methionine (Met) carriers had larger switch costs in the language switching task. Second,
the cross-task relationships varied among bilinguals with different COMT genotypes: Bilinguals with Met allele genotypes
showed larger switch costs and better performance on the IGT. These findings suggest that bilinguals who carry Met allele
are equipped with more efficient adaptive mechanisms of executive functions that are generalized to other cognitive
domains.
Keywords Bilingualism Electroencephalogram Decision-making Executive functions Adaptation
Introduction
A growing body of research has found a relationship
between decision-making and executive functions both
among individuals with healthy and impaired cognitive
control abilities (Bechara and Martin 2004; Brand et al.
2005,2006; Delazer et al. 2007). The findings suggest that
executive function impairment or increased cognitive load
negatively impacts decisions made (Jameson et al. 2004;
Norman and Shallice 1986). As is the case with all humans,
bilinguals make decisions in their daily lives, but they are
unique in that their communication and thus, their
decisions, may involve interacting in and often switching
between more than one language. This linguistic ‘‘mental
exercise’’ may strengthen certain executive functions that
are observable both in behavioral performance and func-
tional brain activity patterns (Gunnerud et al. 2020; Kwon
et al. 2021). Furthermore, previous studies have found that
executive functions may be modulated by genetic factors
that play a role in the availability of dopamine (DA) in the
neural structures underlying cognitive control (Hernandez
et al. 2015; Vaughn et al. 2016).
Individual differences in executive functions can be
attributable to genetic factors (Friedman et al. 2008). The
enzyme catecholamine-O-methyltransferase (COMT), for
instance, is critical in the metabolic degradation of DA,
which has attracted considerable attention for being an
important gene involved in human cognitive function
(Sannino et al. 2015; Smolka et al. 2005). The COMT gene
on chromosome 22q11 has a functional polymorphism, in
which a G/A substitution at codon 158 causes an alteration
from valine (Val) to methionine (Met), commonly called
Val158Met polymorphism (Chen et al. 2004). The Val
allele is related to enhanced enzymatic activity, and thus
&Huanhuan Liu
abcde69503@126.com
1
Research Center of Brain and Cognitive Neuroscience,
Liaoning Normal University, Dalian 116029, China
2
Key Laboratory of Brain and Cognitive Neuroscience,
Dalian 116029, Liaoning Province, China
3
Language Acquisition, Multilingualism, and Cognition
Laboratory / Bilingualism Matters, Wilfrid Laurier
University, Waterloo, Canada
123
Cognitive Neurodynamics
https://doi.org/10.1007/s11571-022-09867-2(0123456789().,-volV)(0123456789().,-volV)
results in faster inactivation of extracellular dopamine in
the brain, especially the prefrontal cortex (PFC) (Chen
et al. 2004;Tsaietal.2003). Therefore, the COMT
Val158Met polymorphism influences the efficiency of
prefrontally-guided cognitive function, particularly execu-
tive functions, language abilities, working memory, and
attentional control (Egan et al. 2001; Barnett et al.
2007,2008). Moreover, due to differences in DA avail-
ability, Met carriers show better performance on tasks that
include stable representations, while Val carriers perform
better in those requiring flexible performance (Mier et al.
2010; Nolan et al. 2004). Met carriers implicate additional
cognitive control resources when a task requires cognitive
flexibility (Colzato et al. 2010; Witte and Flo
¨el 2012).
Furthermore, studies have revealed that COMT Val158Met
genotypes may affect risky decisions. For instance, indi-
viduals who are Methionine (Met) homozygous carriers
with low DA activity exhibit more risky choices and poorer
performance (Colzato et al. 2010; van den Bos et al. 2009)
and show higher susceptibility to losses than Val allele
carriers (Gao et al. 2016).
Nonetheless, it is unclear as to whether the apparently
contradictory effect of COMT Val158Met polymorphism
affects the relationship between decision-making and
executive functions in bilinguals (Manes et al. 2010;
Toplak et al. 2010). In the present study, to fill this gap, we
administered the Iowa Gambling Task (IGT), a typical
emotion-based decision task, whose link with executive
functions has been debated, in addition to a language
switching task (LST), a well-established picture-naming
paradigm used to measure bilingual executive control (see
Methods for details). The IGT is based on a long
exploratory learning process aimed to anticipate long-term
risks in decision making and is used to assess risky deci-
sion-making behavior in laboratories (Bechara et al. 1994;
Marco-Pallares et al. 2008). In the task, participants choose
a card from four decks (choice evaluation stage; two
advantageous decks that have more conservative outcomes
and two disadvantageous decks with more risky options).
Feedback is immediately provided after each selection so
that participants learn the characteristic of each deck
(feedback evaluation stage). Accordingly, participants must
utilize feedback to learn the deck characteristics and form
an efficient strategy to facilitate making the optimal deci-
sion, which is inseparable from the demands of updating
and set-shifting abilities (Di Rosa et al. 2017; Shadlen and
Kiani 2013).
In the IGT, there are three electrophysiological effects
(i.e., event-related potential (ERP) components) that are
related to choice evaluation: N2
IGT
reflects the detection of
conflicting alternatives, i.e., a premotor inhibition process
(Mennes et al. 2008); P3
deck
reflects stimuli encoding or
motivational significance of the eliciting stimuli (Polich
2007); Late Positive Potential (LPP) is associated with
long-term learning (Panitz et al. 2015) and may influence
response conflict when determining choices (Kelly and
O’Connell 2013; Murphy et al. 2015; O’Connell et al.
2012). In the feedback evaluation stage, there also three
components of interest: feedback-related negativity (FRN),
a negative wave reflecting earlier feedback evaluation (Di
Rosa et al. 2017); P3
feedback
is related to a late, top-down,
controlled feedback evaluation process associated with
allocation of attentional resources and updating of working
memory; and LPP
feedback
reflects the processing of stimuli
with motivational significance or the downstream cate-
gorical processing of output from an evaluation system
(Duncan-Johnson and Donchin 1977).
In the LST, bilinguals are required to name pictures and
switch back and forth between their two languages. The
task elicits a series of executive functions, such as conflict
monitoring, set-shifting, inhibitory control, and working
memory updating, as needed to regularly engage in two
languages (Abutalebi et al. 2008). Switch costs—calcu-
lated as the difference between switch trials and repeat
trials—have been measured through ERPs and reaction
times (RTs) and can serve as an index of language control,
which is, in fact, a subdomain of general cognitive control
(Blanco-Elorrieta and Pylkka
¨nen 2016; Liu et al. 2016).
We examine the N2
L
effect in the language schema stage,
which reflects conflict detection and monitoring, height-
ened demands on cognitive control, and the late positive
component (LPC), which occurs in the lexical response
stage and indexes language control during lexical access
(Elke and Wiebe 2017; Liu et al. 2020; Verhoef et al.
2009).
Previous studies have been narrowly focused by exam-
ining how a single aspect of executive functions (i.e.,
inhibition) interacts with decision-making in behavioral-
based experiments (Ouerchefani et al. 2019; Toplak et al.
2010). There has not yet been a study conducted which
explores the effects of COMT Val158Met polymorphism
on decision-making and executive functions. Here, we
divided the IGT into choice evaluation and feedback pro-
cessing, and the LST into the language schema and lexical
response stages to capture the modulatory effect of COMT
Val158Met polymorphism on the relationship between
decision-making and executive control in bilinguals. We
conduct analyses at behavior level and neural coding level
separately, including (1) analyses of covariance (ANCO-
VAs) with sex as a covariate (Sannino et al. 2015; Mamiya
et al. 2016) to compare the effects of COMT genotypes on
performance during two tasks described below; (2) time-
resolved correlations to characterize the effect of COMT
polymorphism on subtle fluctuations in brain-behavior
relationships within each task; and (3) linear regression
analyses with genotype as a categorical variable to examine
Cognitive Neurodynamics
123
whether the gene-dependent executive function advantages
of bilinguals correlated with decision-making. Based on
previous studies, we hypothesize that in the LST, Met
carriers will have larger switch costs, particularly in the
language schema stage. We anticipate that this effect will
be correlated with behavioral switch costs. We also expect
that in the IGT, Val/Val carriers will show increased neural
activity and poorer decision-making performance. Overall,
the present study provides a robust examination of the
effects of genetic factors on the relationship between
decision-making and executive functions in bilinguals.
Methods
In the present study, we examine processing patterns from
behavioral performance and neural activity using elec-
troencephalography (EEG). We investigate whether and
how gene-dependent (COMT Val158Met polymorphism)
executive functions of bilinguals can be adaptively exten-
ded to decision-making performance. We administered two
tasks: the Iowa Gambling Task (IGT) (Bechera et al. 1994),
a typical decision task based on a long exploratory learning
process that involves choice and feedback evaluation; and a
LST, a picture-naming experiment that involves language
schema selection and lexical retrieval (Abutalebi et al.
2008). The order of the two tasks was counterbalanced
across participants. Both tasks were administered using
E-Prime 2.0 software. Visual stimuli were presented on the
center of a 17-inch computer screen with 1024 9768
resolution.
Participants
One hundred and one Chinese-English bilinguals (75
females, 26 males, mean age: 22.5, range: 19–30) with
normal or corrected-to-normal vision participated in the
study. These individuals were native speakers of Chinese
and reported started to learn English at an average age of
8.98 years (SD = 2.15 years, range: 3–13 years). All par-
ticipants were right-handed, and reported no psychological,
cognitive, and motor impairments. The study was approved
by the Ethics Committee of Research Center of Brain and
Cognitive Neuroscience at Liaoning Normal University
and all participants provided their informed consent before
beginning the experiment.
Participants completed an English proficiency test, the
paper-and-pen version of the Oxford Quick Placement test
(OPT) (Geranpayeh 2003), and self-rated their language
abilities on a six-point scale in which ‘‘1’’ indicated no
knowledge and ‘‘6’’ indicated perfect knowledge (Liu et al.
2021). Paired-sample t-tests revealed that the participants’
L1 was significantly stronger than L2 in listening
(t= 18.11, p\.001), speaking (t= 20.66, p\.001),
reading (t= 17.74, p\.001), and writing (t= 12.34,
p\.001). These self-ratings and OPT scores were similar
to intermediate Chinese-English bilinguals in prior
research (Liu et al. 2016,2021), demonstrating unequal
proficiency between their two languages.
DNA extraction and COMT genotyping
The genomic DNA was extracted from peripheral blood
leukocytes of all participants, which were collected with
anticoagulant ethylene diamine tetraacetic acid tubes. The
COMT Val158Met polymorphism was genotyped using
MassARRAY Flight Mass Spectrometry. Single nucleotide
polymorphism (SNP) was sorted according to dbSNP
database. Three primers were designed for COMT
Val158Met genotying, including two polymerase chain
reaction (PCR) primers (1st: ACGTTGGATGACCCAGC
GGATGGTGGATTT, 2nd: ACGTTGGATGTTTTCCAG
GTCTGACAACGG), and one unextended primer (UEP)
(aagtCATGCACACCTTGTCCTTCA). The gene fragment
containing this SNP was amplified by PCR. Following this,
the dNTP generated was phosphorylated by alkaline
phosphatase reaction to form ddNTP. The UEP was used
for single base extension reaction in the ddNTP system to
form a single base extension product complementary to the
SNP genotype to be detected, and the results were sub-
jected to resin purification and chip sampling. Finally, the
samples were analyzed by MALDI-TOF mass spectrome-
ter. Based on the principle that the flight time of ions
generated by radiation ionization of matrix molecules in a
vacuum environment is directly proportional to mass,
genotyping was obtained.
The COMT genotyping found that 51 of the participants
were homozygous for Val/Val, 44 were heterozygous for
Val/Met, and 6 were homozygous for Met/Met. This dis-
tribution was consistent with Hard-Weinberg expectations
(v
2
= .77, df =2,p= .38). The Met carriers were grouped
together (Met/Met and Val/Met) for analyses (Albaugh
et al. 2010; Colzato et al. 2010). All analyses that examined
the COMT genotype effect were conducted among a group
of Met carriers (N = 50) and a group of Val/Val carriers
(N = 51). An independent-sample t-test indicated that there
was no difference in L1 or L2 proficiency between the Met
and Val/Val carriers (see Table S1).
Experimental tasks
Iowa gambling task
The IGT is widely used to assess risky decision-making
behavior over a course of time in laboratory settings
(Bechara et al. 1994). During the task, participants must
Cognitive Neurodynamics
123
develop a long-term profitable monetary scenario in a sit-
uation of uncertainty: a conflict between the chance of
encountering an immediate large reward (¥100) in two
disadvantageous decks of cards (Decks A and B: -¥250 per
10 cards) and the chance of encountering an immediate
small reward (?¥50) in two advantageous decks (Decks C
and D: ?¥250 per 10 cards). Each participant began the
IGT with ¥2,000 and performed 100 trials, each of which
followed the same sequence of choice evaluation and
feedback evaluation stages.
As shown in Fig. 1, in the task, participants first saw a
fixation cross for 500 ms on the center of the computer
screen. Four decks of cards then appeared (choice evaluation
stage): unbeknownst to the participants, two of the decks
were advantageous (ADs) in that they had more conservative
outcomes; and two were disadvantageous (DDs) in that they
contained more risky outcomes. Feedback in the L1 imme-
diately appeared for 1200 ms after each selection so that
participants learned the characteristics of the decks. The
feedback displayed the economic outcome of their decision
(i.e., win or loss in yuan) and their accumulated winnings/
losses. The participants were instructed to earn as much
money as possible and to freely switch between deck choices
at any time and as often as desired. The dependent measure
was the total net score (the number of selections of advan-
tageous decks minus the number of selections of disadvan-
tageous decks) and the advantageous deck selection ratio
(the number of selections of advantageous decks divided by
the total number of selections) for each block of 20 cards and
for the overall task.
Language switching task
The LST, in which simple line drawings are named aloud
as quickly and accurately as possible in the L1 or L2
according to a color cue, has been used as a measure of
language control (Verhoef et al. 2009). Throughout the
task, bilinguals must orchestrate a series of executive
functions, such as conflict monitoring, flexibility,
inhibition, and working memory, as needed to regularly
engage and inhibit the two languages (Abutalebi et al.
2008; Blanco-Elorrieta and Pylkka
¨nen 2016).
The stimuli of 24 black-and-white line drawings
(15 cm 915 cm) were taken from a standardized picture
inventory (Snodgrass and Vanderwart 1980; Zhang, and
Yang 2003). The Chinese names of the drawings were two-
characters in length and their English equivalents were one-
or two-syllable words containing three to six letters. Before
the experiment, we asked a separate group of age-matched
participants (N = 35) from the same population, but who did
not participate in the formal experiment, to judge their
familiarity with the Chinese and English names of the pic-
tures. There were no significant differences between their
familiarity in the two languages (L1: M= 4.85 ±.08, L2:
M= 4.84 ±.12, t(34) = .69, p= .760).
As shown in Fig. 2, in the task, participants first saw a
red or blue box color cue for 250 ms on the center of the
screen followed by a blank for 500 ms. A picture then
appeared until the participant named it aloud in the cued
language or until 2000 ms passed. Finally, a blank screen
lasted for 1000 ms before the next trial started. Participants
completed 12 practice trials before the formal experiment.
There were 5 blocks of 98 trials each in the formal
experiment. Each block was designed as a dual-language
context and contained 2 practice trials followed by 48
repeat trials and 48 switch trials randomly distributed
throughout. The color-language associations were coun-
terbalanced across participants. We classified trials as
either switch or repeat trials. In switch trials, the response
language of the target picture was different from that of the
previous trial (L1–L2 or L2–L1). In repeat trials, the
response language was the same as the preceding trial (L1–
L1 or L2–L2). The trial sequences were randomized
throughout the task so that participants could not anticipate
upcoming trial types. We ensured that the trial type could
only be the same for a maximum of three trials.
RTs were recorded by a PSTSR-BOX connected to a
microphone. To ensure that EEG signals were spared from
Fig. 1 Experimental design and
time course of the IGT. After
choosing a deck, feedback
appeared with an outcome
according to the characteristics
of deck. Words in green font
represent money earned and
words in red font show money
lost. The accumulated amount
of money appeared in black
font. (Color figure online)
Cognitive Neurodynamics
123
oral artifacts, we instructed participants to name items
quietly (Liu et al. 2020). However, there were some
naming responses that were too low to be recorded by the
microphone and as such, we excluded these trials from the
data analyses (see Table S5). The order of the two tasks,
which were administered at Liaoning Normal University,
was counterbalanced across participants.
Behavioral data analyses
Iowa gambling task
In the IGT, behavioral performance of interest is the net
score of the total number of cards selected from the
advantageous decks (i.e., Decks C and D) minus the total
number of cards selected from disadvantageous decks A
and B (i.e., [C ?D] -[A ?B]), with a higher score
indicating superior performance. Moreover, in each block,
we also calculated the advantageous decks selection ratio
by dividing the number of choices from C and D by the
number of trials (i.e., [C ?D]/[A ?B?C?D]). Typi-
cally, IGT performance gradually improves as subjects
learn the rules of the deck (Bechera and Martin 2004;
Ouerchefani et al. 2019). Considering that the effect of
COMT Val158Met has been reported to be sex-dependent
(Mamiya et al. 2016; Sannino et al. 2015), we performed a
two-way ANCOVA, with sex as a covariate and subjects as
a random variate, to test the main and interaction effects of
the COMT genotype (Met carriers vs. Val homozygotes)
and Block (1, 2, 3, 4, 5) selection ratio.
Language switching task
The behavioral indicators of interest in the LST are switch
costs—calculated as the difference between switch trials
and repeat trials. We excluded RTs that were M±3SD for
each participant to eliminate the potential impact brought
by outlying values. To satisfy the assumption of normal
distribution, we then performed the log transform followed
by a two-way ANCOVA on accuracy and naming latencies
with sex as a covariate and subjects as the random variate.
The ANCOVA examined the effects of genotypes and trial
types (switch vs. repeat trials).
ERP analyses
During both tasks, we selected five electrodes located in
central scalp region (FCz, Fz, Cz, CPz, Pz) for analyses
(Stauder et al. 2006; Yu and Zhou 2006).
Iowa gambling task
In the IGT, there are three electrophysiological effects of
interest that occur during choice evaluation: the N2
IGT
component is associated with the detection of conflicting
alternatives and reflects a premotor inhibition process; the
P3
deck
component occurs during stimuli-encoding; the Late
Positive Potential (LPP) is associated with long-term learn-
ing and is said to influence response conflict when deter-
mining choices (Kelly and O’Connell 2013; O’Connell et al.
2012; Murphy et al. 2015). During the feedback evaluation
stage, there are also three components of interest: feedback-
related negativity (FRN) reflects earlier feedback evaluation;
the P3
feedback
component demonstrates the late, top-down,
controlled feedback evaluation process associated with the
allocation of attentional resources and working memory
updating; and the LPP
feedback
represents the processing of
stimuli with motivational significance or downstream cate-
gorical processing of output from the evaluation system
(Duncan-Johnson and Donchin 1977; Polich 2007; Polich
and Kok 1995).
Language switching task
In the LST, the ERP component of interest that occurs
during language schema selection is the N2
L
effect which
reflects conflict detection, monitoring, and demands on
cognitive control. During the lexical retrieval stage is the
late positive component (LPC) which indexes language
control involved or required during lexical access (Abu-
talebi et al. 2008; Verhoef et al. 2009; Blanco-Elorrieta
et al. 2016).
Fig. 2 Experimental design and
time course of the LST. Blue
squares imply that the upcoming
picture is to be named in the L2,
red squares mean that the
picture is to be named in the L1.
The color-language association
was counterbalanced across
participants. (Color
figure online)
Cognitive Neurodynamics
123
We extracted the amplitude of the ERP components
across trials and conditions for each subject and conducted
two-way ANCOVAs on the ERP components (N2
IGT
,
P3
deck
, LPP
deck
) during the choice evaluation stage, with
sex as a covariate and subjects as a random variate, to
investigate the main and interactive effects of genotype and
advantageous (AD) and disadvantageous (DD) deck choi-
ces. During the feedback evaluation stage, two-way
ANCOVAs were performed on the ERP components (FRN,
P3
feedback
, LPP
feedback
) during the feedback evaluation
stage, with sex as a covariate and subjects as a random
variate to estimate genotype and feedback (win and loss)
effects on feedback processing. The ANCOVAs examined
the genotype and trial type (switch and repeat) effects on
language control (N2
L
, LPC).
EEG acquisition and preprocessing
EEG data were recorded from 64 Ag/AgCl electrodes
placed according to the extended 10–20 positioning sys-
tem. The signal was recorded from eemagine (ANT Neuro)
at a 1 kHz sampling rate and referenced online to the CPz
electrode. Impedances were kept below 5 kX. Offline
processing was down-sampled to 500 Hz and referenced
offline to the average of M1 and M2. ERP activity was
filtered online within a bandpass of 0.1 and 100 Hz and
refiltered offline with a highpass filter of 0.1 Hz and a
lowpass filter of 30 Hz. Epochs with excessive eye
movements or other movement artifacts (e.g., muscular
artifacts, channel noise) were identified by visual inspec-
tion and manually removed. Ocular artifact reduction was
performed through Independent Component Analysis
(ICA) rejection by using EEGLAB (Makeig et al. 1996).
The mean number of ICs rejected was 2.51 ±1.53 per
participant (Val/Val genotype: 2.65 ±1.74; Met carriers:
2.38 ±1.28). The continuous recordings were cut into
epochs ranging from -200 to 800 ms relative to the onset
of each trial. For the LST, the mean number of ICs that
were rejected per participant was 1.45 ±1.88 (Val/Val
genotype: 1.65 ±2.46; Met carriers: 1.24 ±1.00), with
epochs cue-locked between -200 and 1600 ms. Baseline
correction was performed in reference to pre-stimulus
activity (-200 to 0 ms). Signals exceeding ±80 lVin
any given epoch were automatically discarded. The mean
(and SD) number of rejected and accepted epochs per
condition across genotypes are shown in Table S5. All
preprocessing was performed by EEGLAB (Brunner et al.
2013; Delorme and Makeig 2004).
Results
COMT effect on behavioral performance
in decision-making
To examine the genotype effect on decision-making, we
conducted a two-way ANCOVA with the main factor block
[block 1–5] and genotype on AD selection ratios (see
Table S2). Although other studies have found that perfor-
mance in the IGT gradually improves as subjects master
the rules of the deck (Bechara and Martin 2004; Ouerch-
efani et al. 2019), in the present study, there was no sig-
nificant main effect of block, F(4,98) = 1.72, p= .286,
genotype, F(1,98) = 1.14, p= .286, or interaction.
COMT effect on behavioral performance
in language control
The accuracy analyses from the ANCOVA on sequence
and genotype demonstrated a significant main effect of
sequence, F(1,98) = 134.70, p\.001, reflecting that
responses to switch trials were less accurate
(M= .95 ±.22) than to repeat trials (M= .97 ±.17) (see
Table S3). Switch costs in RTs showed an effect of
sequence, F(1,98) = 192.07, p\.001, such that there were
slower responses to switch trials (M= 851 ±245 ms)
compared to repeat trials (M= 820 ±232 ms). There was
also an effect of COMT genotype, F(1,98) = 232.89,
p\.001, with faster responses for Met carriers
(M= 818 ±230 ms) compared to individuals with the
Val/Val genotype (M= 852 ±246 ms).
COMT effect on the time course of decision-
making
To examine how the decision-making process is altered by
genetic variations of COMT genotype, we conducted a
two-way ANCOVA using deck choice and genotype as
main factors (Fig. 3and Table S4). In the choice evaluation
stage, the results showed significant genotype effects on
N2
IGT
,F(1,98) = 28.32, p\.001, P3
deck
,F(1,98) = 3.99,
p= .046, and LPP
deck
,F(1,98) = 7.82, p= .005 (Fig. 3A).
Particularly, the N2
IGT
effect among Met carriers
(M=-1.21 ±9.87 lV) was stronger than for Val/Val
(M=-.13 ±9.68 lV). P3
deck
amplitudes were more
robust for individuals with the Val/Val genotype
(M= 4.02 ±10.80 lV) than Met carriers
(M= 3.59 ±11.01 lV). The LPP
deck
yielded a marginally
Cognitive Neurodynamics
123
Fig. 3 Waveforms and mean amplitudes of ERP components of Val/
Val and Met carriers depicting brain potentials and scalp topography
in the IGT. In (A,B), dashed boxes represent early (200–300 ms),
middle (320–420 ms), and late (420–800 ms) time windows during
chioce evaluation. In (C,D), dashed boxes represent early
(230–330 ms), middle (350–450 ms), and late (450–800 ms) time
windows during feedback evaluation. ADs Advantageous decks, DDs
Disadvantageous decks. *p\.05; **p\.01; *** p\.001
Cognitive Neurodynamics
123
significant deck 9genotype interaction, F(1,98) = 3.20,
p= .074. Follow-up t-tests for each deck group showed
that ADs evoked greater LPP
deck
effects for Val/Val
compared to Met carriers (t= 9.65, p= .002), However, no
such LPP
deck
enhancement was observed for DDs (t= .57,
p= .451) (Fig. 3B).
To investigate whether the stronger coding of decks
observed among Val/Val genotypes was beneficial or
adverse to behavioral performance, we performed time-
resolved brain-behavior correlations focused on the entire
EEG signal (Fig. 4). We averaged the amplitude for each
card group at each time point across the trials for both
genotypes and correlated this value with their behavioral
score. The neural coding showed different relationships
with changes in performance between the two genotypes.
Specifically, in the evaluation of ADs, a significant corre-
lation between the magnitude of encoding and scores was
detected for individuals with the Val/Val genotype at the
cluster window of 278–332 ms (Fig. 4A: window tes-
ted = 100–1000 ms; mean rho = -.31, FDR p= .047),
but not for Met carriers (mean rho = -.05, p[.691).
Furthermore, Val/Val carriers showed a significant corre-
lation for AD encoding and scores in the cluster window of
582–760 ms (mean rho = -.29, FDR p= .047). Likewise,
Val/Val carriers had a marginally significant correlation
between DD encoding and scores at the time clusters of
228–350 ms (Fig. 4B: window tested = 200–1000 ms;
mean rho = -.31, FDR p= .053) and 580–800 ms (mean
rho = -.27, FDR p= .053). To compare relationships
between neural processing and behavioral performance, we
Fig. 4 Time-resolved correlations and time-averaged correlations
between neural coding and performance across COMT genotypes.
(A) Spearman’s rho values and scatter plot between amplitudes of
advantageous deck (Ad) evaluation and scores in IGT. (B) Spearman’s
rho values and scatter plot between amplitudes of disadvantageous
desk (DD) evaluation and scores in IGT. (C) Spearman’s rho values
and scatter plot between the neural switch costs and behavioral switch
costs in LST. Dark lines represent the significant time windows after
FDR correction. *p \.05
Cognitive Neurodynamics
123
also performed time-averaged correlations on the time
windows. We found that greater magnitude of deck eval-
uation among Val/Val carriers was significantly correlated
with poorer performance (Fig. 4).
ANCOVAs of feedback and genotype were performed
on feedback processing (Table. S4). In early feedback
evaluation, no significant differences or interactions were
detected for the FRN effect (Fs\2.38, ps[.123). For
P3
feedback
, there was a significant effect of feedback,
F(1,98) = 19.95, p\.001, and genotype, F(1,98) = 8.07,
p= .005, but no interaction emerged, F(1,98) = 2.12,
p= .145. Specifically, loss trials revealed a greater
enhancement on P3
feedback
than win trials (M= 7.71 ±
9.80 lV, and M= 6.73 ±9.52 lV, respectively). Met
carriers evoked a greater P3
feedback
effect than Val/Val
carrier (M= 7.22 ±9.72 lV, and M= 6.75 ±9.47 lV,
respectively). Similarly, for LPP
feedback
, there were signif-
icant effects for feedback, F(1,98) = 109.25, p\.001,
genotype, F(1,98) = 14.76, p\.001, and a significant
interaction between feedback 9genotype, F(1,98) = 8.82,
p= .003. Follow-up tests on LPP
feedback
magnitudes
evoked by losses (compared to wins) showed sensitivity to
COMT genotype, F(1,98) = 14.08, p\.001: LPP
feedback
magnitudes were significantly stronger for Met carriers
(M= 13.27 ±10.22 lV) compared to Val/Val carriers
(M= 11.72 ±10.44 lV). Such an LPP
feedback
enhance-
ment was also found during feedback processing despite
genotype: Loss trials revealed a more robust LPP effect
than win trials (ps\.001) (Fig. 3D). There were no sig-
nificant associations between feedback-related neural cod-
ing and behavioral scores in each genotype (window tested:
0–1000 ms; Win: mean rho values \.07, ps[.415; Loss:
mean rho values \.06, ps[.595).
COMT effect on the time course of bilingual
executive control
We next examined whether the Val158Met polymorphism in
the COMT gene affects bilingual executive control. Two-
way ANCOVAs were performed on N2 and LPC (see Fig. 5
and Table S4). During the language schema selection stage,
there was a significant effect of sequence, F(1,98) = 4.26,
p= .039, and genotype, F(1,98) = 113.92, p\.001, such
that switch trials (M=-3.91 ±9.92 lV) evoked stronger
N2 effects than repeat trials (M=-3.71 ±9.77 lV).
Likewise, N2 effects were stronger for Met carriers
(M=-4.36 ±10.06 lV) compared to Val/Val carriers
(M=-3.30 ±9.62 lV). There was also a significant
sequence 9genotype interaction effect, F(1,98) = 4.50,
p= .034. Follow-up tests on N2 amplitudes demonstrated a
robust genotype effect in that N2s evoked by Met carriers (vs
Val/Val) was observed in both repeat and switch trials,
Fs[36.96, ps\.001. Moreover, for Met carriers, the N2
effect was more robust in switch trials than repeat trials,
F(1,98) = 8.87, p= .003. This was not the case for Val/Val
carriers, F(1,98) = .01, p= .926. Contrarily, in the lexical
response stage, there was an effect of sequence,
F(1,98) = 11.36, p= .001, and genotype, F(1,98) = 134.89,
p\.001: The LPC was stronger in repeat trials
(M= 1.82 ±13.67 lV) than in switch trials
(M= 1.35 ±13.80 lV), and was stronger in Val/Val
(M= 2.36 ±13.62 lV) compared to Met carriers
(M= .75 ±13.81 lV) during lexical retrieval (Fig. 5A, B).
We also conducted time-resolved correlations between
neural switch costs and behavioral switch costs to explore
the gene-dependent executive function abilities of bilin-
guals. The results demonstrated strikingly different patterns
that were modulated by COMT polymorphism (refer back
to Fig. 4C). A significantly positive association occurred in
Met carriers at the time window of 314–620 ms (window
tested: 270–750 ms; mean rho = .23, FDR p= .049), but
not for Val/Val carriers (mean rho = .02, p= .633). For the
latter individuals, however, there was a marginally signif-
icant correlation between larger neural switch costs and
smaller behavioral switch costs at the time cluster of
1686–1744 ms (window tested: 1200–1900 ms; mean
rho = -.35, FDR p= .066). The time-averaged correla-
tions for 314–620 ms indicated that for Met carriers, uti-
lizing more inhibitory control during language schema
selection was associated with larger switch costs during
picture naming.
Influence of COMT on the relationship
between decision-making and executive
functions
To further investigate whether the effect of COMT influ-
ences the relationship between decision-making and exec-
utive functions among bilinguals, we performed cross-task
linear regression analyses and time-resolved correlations
on the relevant behavioral indicators (net scores in the IGT;
behavioral switch costs in the LST) and brain indicators
(neural encoding of decision-making in the IGT; neural
switch costs in the LST). The results showed that the
COMT genotype significantly affected the relationship
between the two indicators of behavioral performance: the
net scores from the IGT and the switch costs from the LST
(Fig. 6A) (R
2
= .073, p= .006). Specifically, the co-vari-
ant relationship observed in Met carriers was such that
larger switch costs predicted better performance in the IGT
(r= .326, p= .021). For Val/Val carriers, this relationship
was not significant (r= .206, p= .146).
The same linear regression models were conducted on
ERP components to assess whether the executive functions
involved in the two tasks were modulated by the genetic
variation of COMT (Fig. 6B). The results showed
Cognitive Neurodynamics
123
differences among the genotypes: During language schema
selection, Met carriers showed a significant positive rela-
tionship between neural switch costs of N2
L
in the LST and
mean N2
IGT
amplitudes in the IGT (ADs: r= .374, FDR
p= .011; DDs: r= .358, FDR p= .011), while the Val
homozygote subjects marginally displayed the reverse
pattern, ps = .116 (Fig. 6C, D). In the lexical retrieval
stage, the Met carriers showed positive correlations
between switch costs of LPC in the LST and mean N2
IGT
and P3
deck
amplitudes in the IGT (Fig. 6E–H), whereas no
correlation was found for the Val/Val carriers. Consistent
with the behavioral findings, the neural activity during
language switching for bilingual Met carriers has a positive
correlation with the neural encoding involved in deck
evaluation.
Discussion
In the present study, we examined how variations in the
COMT gene affect the relationship between executive
functions and decision-making in bilinguals. Although it
has been claimed that bilinguals have advantages over
monolinguals in certain executive functions (Gunnerud
et al. 2020; Kwon et al. 2021), previous studies have
reported inconsistencies when investigating the relation-
ship between decision-making and executive functions
(Ouerchefani et al. 2019; Toplak et al. 2010). Our results
show that COMT Val158Met genotypes modulate the
relationship. The main findings include:
•The COMT gene affects decision-making and executive
function. Specifically, during the IGT, Val/Val carriers
elicited stronger neural activity in the middle and late
stages of choice evaluation. During the LST, Met
carriers showed stronger executive control in language
schema selection.
•The predictive effect of neural coding on behavioral
performance in the tasks revealed sensitivity to the
COMT gene. Participants with the Val/Val genotype
showed a negative brain-behavior relationship in choice
evaluation in the IGT. However, in the LST, Met
carriers’ neural switch costs during language schema
selection showed a positive brain-behavior correlation.
•Gene-dependent executive function of bilinguals inter-
acts with decision-making: COMT Val158Met poly-
morphism modulated cross-task relationships between
behavioral indicators, and between specific executive
functions.
Taken together, these findings suggest that while bilin-
guals appear to be able to adaptively extend executive
functions to decision-making processes, these effects are
sensitive to genetic factors. The findings from the linear
regression analyses along with time-resolved and time-
averaged correlation analyses revealed significant rela-
tionships between decision-making and executive function
Fig. 5 Waveforms (A) and mean amplitudes of ERP components (B) of Met and Val/Val carriers depicting brain potentials and scalp topography
during the LST. Dashed boxes represent early (250–350 ms) and late (1200–1500 ms) time windows. **p\.01; ***p\.001
Cognitive Neurodynamics
123
for behavioral-behavioral and brain-brain relationships, but
were sensitive to COMT gene variation.
COMT Val158Met polymorphism plays different
roles in choice evaluation and feedback
evaluation during decision-making
among bilinguals
In this study, we assessed the COMT effect on decision-
making processes among bilinguals. Although a main
effect of genotype was absent in the behavioral data, the
analyses of neural indicators illustrated modulating roles of
genotypes in two processing stages involved in the IGT. In
the choice evaluation stage, Val/Val carriers showed a
decreased N2 effect and increased P3
deck
and LPP
deck
effects, particularly for ADs. Met carriers did not show
neural differences between ADs and DDs, suggesting that
these Met carriers exhibit similar monitoring control and
allocation of attention resources on choices of decks. These
patterns are inconsistent with previous studies which sug-
gest that the Met/Met genotype results in enhanced cog-
nition as measured by neurophysiological methods
(Bramon et al. 2006). Furthermore, we found negative
correlations widely distributed in middle and late stages of
Fig. 6 Cross-task correlations across COMT genotypes. (A) The
relationships between the IGT score and switch costs from the LST.
Ribbons show 95% CI of the linear regression. (B) The correlation
matrix of ERP components across genotypes. (C–H) Significant
correlation scatter plots for ERP components across genotypes.
*p\.05; **p\.01; ***p\.001
Cognitive Neurodynamics
123
choice evaluation, indicating that the excessive processing
of cards among Val/Val carriers leads to poorer behavioral
scores. Considering that recent evidence shows that LPP in
long-term learning is modulated by COMT Val158Met
genotypes (Swart et al. 2011), we hypothesize that COMT
genotypes also affect the long-term exploitation of advan-
tageous options.
However, when advancing to feedback evaluation,
which involves emotional reward/punishment processing,
the COMT gene has different effects on the middle and late
processing windows. Met carriers showed potentiated
electrophysiological effects (P3
feedback
and LPP
feedback
) and
focused more on (potential) losses than wins. Previous
work has shown that Met carriers are more sensitive to
affective stimuli, especially negative emotions in tasks
involving emotion processing and that Val carriers display
some behavioral advantages regarding emotion processing
(Smolka et al. 2005; Swart et al. 2011). Moreover,
Met allele carriers have demonstrated enhanced sensory
encoding or stronger activation in the PFC and limbic
system for unpleasant/negative stimuli (Bilder et al. 2004).
In the present study, we did not detect brain-behavior
relationships between the enhanced neural coding of
feedback and performance in Met carriers (nor in Val/Val
carriers), which should have led to poorer performance due
to their susceptibilities to losses (Gao et al. 2016).
Nonetheless, our results indicate that Met carriers exhibit
stronger cognitive control resources to successfully avoid
losses (Zhou et al. 2021).
Together, Met carriers show enhanced N2
IGT
but
reduced P3
deck
and LPP
deck
components, underscoring their
indistinguishable card preference during the choice pro-
cessing stage. This implies that more cognitive resources
which are invested in early stages can effectively assist in
long-term learning of deck rules. However, for Val/Val
carriers, stronger neural activities must be applied which
consequently hampers performance. Moreover, in the
feedback processing stage, the COMT gene shows distinct
effects such that Met carriers must exhibit stronger neural
activities to prevent reduce the influence of negative
emotions in order to make rational choices. The different
brain-behavior relationships between Met and Val/Val
carriers suggest that COMT genotypes affect the nature of
focusing on information encoding during the course of
decision-making.
COMT Val158Met polymorphism is associated
with enhanced executive control in bilinguals
In the LST, we found that COMT Val158Met polymor-
phism plays a key role in executive control for bilinguals,
as supported by the significant interaction between
sequence and genotype for N2
L
component. Specifically,
Met carriers showed an enhanced N2
L
effect and higher
neural switch costs than individuals with the Val/Val
genotype. Our data illustrate that Met carriers rely on
additional inhibitory control to suppress non-target lan-
guage interference than Val/Val-subjects in both repeat and
switch trials. Nolan et al. (2004) reported that Met carriers
exhibited significant higher accuracy, but less flexibility in
a task requiring rule changes that are dependent on trial-by-
trial feedback. Colzato et al. (2010) found that Met carriers
had larger switch costs in the task-switching paradigm,
reflecting less cognitive flexibility than Val/Val carriers.
However, this overloaded neural effect for Met carriers
was no longer present in the lexical retrieval: Val/Val
carriers demonstrated larger LPC effects than Met carriers,
and repeat trials elicited stronger neural activity than
switch trials. When also considering that Met carriers had
faster RTs, this COMT effect suggests that for Met carriers,
utilizing and maintaining a higher demanding of executive
control during language schema selection facilitated sub-
sequent lexical access. Hence, it is plausible that Met
carriers show heightened neural activity during the lan-
guage schema selection stage, which demands flexible
inhibitory control over the non-target language, but less
cognitive demands are required during lexical access
(Bilder et al. 2004; Colzato et al. 2010; Witte and Flo
¨el
2012). These findings were further confirmed by correla-
tion analyses in which neural switch costs for Met carriers
in the language schema section stage positively predicted
behavioral switch costs, underscoring the relationship
between poorer executive control of Met carriers and the
difficulties they faced in word production.
COMT Val158Met polymorphism modulates
the relationship between bilingual executive
functions and decision-making
Correlations of behavioral indicators
The positive correlation between behavioral switch costs
and IGT scores suggests that better control abilities are
associated with poorer performance in decision-making.
Although this finding appears counter-intuitive at first, it
only appears to be the case for Met carriers, but not for
individuals with the Val/Val genotype. Prior work has
shown that Met carriers have higher DA levels in the
prefrontal cortex which are beneficial for the stability of
cognitive representations, yet are hampering when needing
to flexibly alter representations. Val/Val-subjects, however,
with lower DA levels in the prefrontal cortex, may be more
amendable to such cognitive flexibility (Bilder et al. 2004;
Witte and Flo
¨el 2012). Our finding that bilingual Met
Cognitive Neurodynamics
123
carriers with better executive control abilities showed
poorer performance in decision-making, although they
acquired the inherent advantages from learning the char-
acteristics of the decks in the IGT, are reflective of these
previous patterns.
Correlations of neural indicators
The findings revealed significant positive correlations
among the neural indicators for the Met carriers, but not for
individuals with the Val/Val genotype. Specifically, for
Met carriers, during language schema selection, neural
switch costs (N2
L
) were positively correlated with N2
IGT
effects during evaluation of ADs and DDs. For Val/Val
carriers, these correlations were not significant and in fact,
trended in the opposite direction. As discussed, greater
neural switch costs reflect the reduced cognitive flexibility
associated with intense neural processing on decks. We
speculate that the executive function advantages argued to
be incurred from bilingualism may be sensitive to genetic
factors. It is precisely due to the significant difference in
executive function between COMT genotypes that distinct
patterns arise. Previous studies have demonstrated that DA
availability profoundly affects learning-related improve-
ments in performance and related changes in brain struc-
ture and functions (Vaughn and Hernandez 2018).
Individuals with the Val/Val genotype have shown larger
gains in working memory compared than individuals with
the Met/Met genotype when being trained, though their
baseline working memory performance before training was
worse (Bellander et al. 2015). Moreover, changes in white
matter have been observed among individuals learning an
L2 when those individuals carried the Val/Val or Val/Met
genotypes, but not the Met/Met genotype (Mamiya et al.
2016). Our findings suggest that Met carriers may have
stronger cognitive benefits from bilingualism than indi-
viduals with the Val/Val genotype. This benefit allows Met
carriers to reduce the inherent demands of cognitive flex-
ibility and readily process stable tasks. However, yet cru-
cially, this hypothesis is not supported by the behavioral
results, which is consistent with the notion that COMT
effects are more detectable at the neural rather than
behavioral level. The findings also demonstrated a positive
relationship between behavioral switch costs in the LST
and decision-making performance in the IGT for Met
carriers, but not for Val/Val carriers, suggesting that lan-
guage control was adaptively generalized to domain-gen-
eral control (Abutalebi and Green 2016; Liu et al. 2021;
Green and Abutalebi 2013).
Conclusion
The present study has shown that COMT genotypes mod-
ulate the relationship between language control and deci-
sion-making among bilinguals. Compared to bilinguals
with the Val/Val genotype, Met carriers showed robust
cross-task correlations between decision-making and
executive function with respect to behavioral performance,
ERP components, and behavior-brain relationships. Bilin-
gual Met carriers are equipped with better adaptive
mechanisms of language control, which is embodied in
both behavioral performance and neural activity. Rather
than focusing on a single variable or method, this study has
also illustrated an ecological valid way of examining the
relationship between two high-level cognitive functions
(decision-making and executive functions) through
behavioral and neural indicators. Overall, the findings
demonstrate that the relationship between decision-making
and executive function among bilinguals is sensitive to
genetic factors—particularly COMT Val158Met—and
provide support for the cross-task adaptation of bilinguals’
executive functions to domain-general cognitive abilities.
1
Supplementary Information The online version contains
supplementary material available at https://doi.org/10.1007/s11571-
022-09867-2.
Acknowledgements This research was supported by Grants from
Youth Foundation of Social Science and Humanity, China Ministry of
Education (21C10165001), and Youth Project of Liaoning Provincial
Department of Education (LJKQZ2021089), Liaoning Social Science
Planning Fund of China (L20AYY001), Dalian Science and Tech-
nology Star Fund of China (2020RQ055), and Research Project on
Economic and Social Development of Liaoning Province (2023l-
slqnkt-054).
Data availability The datasets generated and analyzed in this study
are available in the OSF repository: Liu, H. (2022, February 23).
‘‘COMT Modulation in Cross-task Adaptation for Bilinguals’’.
Retrieved from https://osf.io/2bys8.
Declarations
Conflict of interest The authors declare that they have no conflict of
interest.
1
The present study explored the relationship between bilinguals’
executive function and decision-making abilities. Different from
previous research revealing a correlation between a single aspect of
executive functions, we expected to obtain similar results through two
tasks involving multiple executive functions. While many studies
distinguish between language control (e.g., language-switching) and
domain-general cognitive control (e.g., executive functions) (e.g.,
Green & Abutalebi, 2013), the present study aligns with other work
arguing that language switching involves executive functions (i.e.,
inhibiting, updating, shifting) (e.g., Sikora et al., 2016; Liu et al.,
2021). Readers should be aware of these differences and as research is
ongoing in this area, should use some caution when interpreting our
findings.
Cognitive Neurodynamics
123
References
Abutalebi J, Green DW (2016) Neuroimaging of language control in
bilinguals: neural adaptation and reserve. Biling Lang Cogn
19(4):689–698
Abutalebi J, Annoni J, Zimine I, Pegna A, Seghier M, Lee-Jahnke H,
Lazeyras F, Cappa S, Khateb A (2008) Language control and
lexical competition in bilinguals: an event-related fMRI study.
Cereb Cortex 18(7):1496
Albaugh MD, Harder, VS, Althoff RR, Rettew DC, Ehli EA, Lengyel-
Nelson T, ... Hudziak JJ (2010) COMT Val158Met genotype as a
risk factor for problem behaviors in youth. J Am Acad Child
Adolesc Psychiatry 49(8):841–849
Barnett JH, Jones PB, Robbins TW, Mu
¨ller U (2007) Effects of the
catechol-O-methyltransferase Val158Met polymorphism on
executive function: a meta-analysis of the Wisconsin Card Sort
Test in schizophrenia and healthy controls. Mol Psychiatry
12(5):502–509
Barnett JH, Scoriels L, Munafo
`MR (2008) Meta-analysis of the
cognitive effects of the catechol-O-methyltransferase gene
Val158/108Met polymorphism. Biol Psychiat 64(2):137–144
Bechara A, Martin EM (2004) Impaired decision making related to
working memory deficits in individuals with substance addic-
tions. Neuropsychology 18(1):152–162
Bechara A, Anderson S, Damasio A, Damasio H (1994) Insensitivity
to future consequences following damage to the human
prefrontal cortex. Cognition 50:7–15
Bellander M, Ba
¨ckman L, Liu T, Schjeide B-MM, Bertram L,
Schmiedek F, Lo
¨vde
´n M (2015) Lower baseline performance but
greater plasticity of working memory for carriers of the val allele
of the COMT Val158Met polymorphism. Neuropsychology
29(2):247–254
Bilder RM, Volavka J, Lachman HM, Grace AA (2004) The catechol-
O-methyltransferase polymorphism: relations to the tonic–phasic
dopamine hypothesis and neuropsychiatric phenotypes. Neu-
ropsychopharmacology 29(11):1943–1961
Blanco-Elorrieta E, Pylkka
¨nen L (2016) Bilingual language control in
perception versus action: MEG reveals comprehension control
mechanisms in anterior cingulate cortex and domain-general
control of production in dorsolateral prefrontal cortex. J Neurosci
36(2):290–301
Bramon E, Dempster E, Frangou S, McDonald C, Schoenberg P,
MacCabe JH, Murray RM (2006) Is there an association between
the COMT gene and P300 endophenotypes? Eur Psychiatry
21(1):70–73
Brand M, Kalbe E, Labudda K, Fujiwara E, Kessler J, Markowitsch
HJ (2005) Decision-making impairments in patients with
pathological gambling. Psychiatry Res 133:91–99
Brand M, Labudda K, Markowitsch HJ (2006) Neuropsychological
correlates of decision-making in ambiguous and risky situations.
Neural Netw 19:1266–1276
Brunner C, Delorme A, Makeig S (2013) Eeglab: an open source
matlab toolbox for electrophysiological research. Biomed Eng/
Biomedizinische Technik, 58(SI-1-Track-G),
000010151520134182
Chen J, Lipska BK, Halim N, Ma QD, Matsumoto M, Melhem S,
Weinberger DR (2004) Functional analysis of genetic variation
in catechol-O-methyltransferase (COMT): effects on mRNA,
protein, and enzyme activity in postmortem human brain. Am J
Hum Genet 75(5):807–821
Colzato LS, Waszak F, Nieuwenhuis S, Posthuma D, Hommel B
(2010) The flexible mind is associated with the catechol-O-
methyltransferase (COMT) Val158Met polymorphism: evidence
for a role of dopamine in the control of task-switching.
Neuropsychologia 48(9):2764–2768
Delazer M, Sinz H, Zamarian L, Benke T (2007) Decision-making
with explicit and stable rules in mild Alzheimer’s disease.
Neuropsychologia 45:1632–1641
Delorme A, Makeig S (2004) EEGLAB: An open-source toolbox for
analysis of single-trial EEG dynamics including independent
component analysis. J Neurosci Methods 134(1):9–21
Di Rosa E, Mapelli D, Arcara G, Amodio P, Tamburin S, Schiff S
(2017) Aging and risky decision-making: new ERP evidence
from the Iowa Gambling Task. Neurosci Lett 640:93–98
Duncan-Johnson C, Donchin E (1977) On quantifying surprise: the
variation of event-related potentials with subjective probability.
Psychophysiology 14(5):456–467
Egan MF, Goldberg TE, Kolachana BS, Callicott JH, Mazzanti CM,
Straub RE, Weinberger DR (2001) Effect of COMT Val108/158
Met genotype on frontal lobe function and risk for schizophrenia.
Proc Natl Acad Sci 98(12):6917–6922
Elke S, Wiebe S (2017) Proactive control in early and middle
childhood: an ERP study. Dev Cogn Neurosci 26:28–38
Friedman NP, Miyake A, Young SE, DeFries JC, Corley RP, Hewitt
JK (2008) Individual differences in executive functions are
almost entirely genetic in origin. J Exp Psychol Gen 137(2):201
Gao X, Gong P, Liu J, Hu J, Li Y, Yu H, Zhou X (2016) COMT
Val158Met polymorphism influences the susceptibility to fram-
ing in decision-making: OFC-amygdala functional connectivity
as a mediator. Hum Brain Mapp 37(5):1880–1892
Geranpayeh A (2003) A quick review of the English Quick Placement
Test. UCLES Res Notes 12:8–10
Green DW, Abutalebi J (2013) Language control in bilinguals: the
adaptive control hypothesis. J Cogn Psychol 25(5):515–530
Gunnerud HL, Ten Braak D, Reikera
˚s EKL, Donolato E, Melby-
Lerva
˚g M (2020) Is bilingualism related to a cognitive advantage
in children? A systematic review and meta-analysis. Psychol
Bull 146(12):1059
Hernandez AE, Greene MR, Vaughn KA, Francis DJ, Grigorenko EL
(2015) Beyond the bilingual advantage: the potential role of
genes and environment on the development of cognitive control.
J Neurolinguist 35:109–119
Jameson TL, Hinson JM, Whitney P (2004) Components of working
memory and somatic markers in decision making. Psychon Bull
Rev 11:515–520
Kelly S, O’Connell R (2013) Internal and external influences on the
rate of sensory evidence accumulation in the human brain.
J Neurosci 33:19434–19441
Kwon YH, Yoo K, Nguyen H, Jeong Y, Chun MM (2021) Predicting
multilingual effects on executive function and individual con-
nectomes in children: an ABCD study. Proc Natl Acad Sci
118(49):e2110811118
Liu H, Liang L, Dunlap S, Fan N, Chen B (2016) The effect of
domain-general inhibition-related training on language switch-
ing: an ERP study. Cognition 146:264–276
Liu H, Zhang Y, Blanco-Elorrieta E, He Y, Chen B (2020) The role of
proactive control on subcomponents of language control:
evidence from trilinguals. Cognition 194:104055
Liu H, Li W, Zuo M, Wang F, Guo Z, Schwieter JW (2021) Cross-
task adaptation effects of bilingual language control on cognitive
control: a dual-brain EEG examination of simultaneous produc-
tion and comprehension. Cereb Cortex
Makeig S, Jung TP, Ghahremani D, Sejnowski TJ (1996) Independent
component analysis of simulated ERP data. Institute for Neural
Computation, University of California: Technical Report INC-
9606
Mamiya PC, Richards TL, Coe BP, Eichler EE, Kuhl PK (2016) Brain
white matter structure and COMT gene are linked to second-
language learning in adults. Proc Natl Acad Sci
113(26):7249–7254
Cognitive Neurodynamics
123
Manes F, Torralva T, Roca M, Gleichgerrcht E, Tristan AB, Hodges
JR (2010) Frontotemporal dementia presenting as pathological
gambling. Nat Rev Neurol 6:347–352
Marco-Pallares J, Cucurell D, Cunillera T, Garcı
´a R, Andre
´s-Pueyo
A, Mu
¨nte TF, Rodrı
´guez-Fornells A (2008) Human oscillatory
activity associated to reward processing in a gambling task.
Neuropsychologia 46(1):241–248
Mennes M, Wouters H, Van Den Bergh B, Lagae L, Stiers P (2008)
ERP correlates of complex human decision making in a
gambling paradigm: detection and resolution of conflict. Psy-
chophysiology 45(5):714–720
Mier D, Kirsch P, Meyer-Lindenberg A (2010) Neural substrates of
pleiotropic action of genetic variation in COMT: a meta-
analysis. Mol Psychiatry 15(9):918–927
Murphy PR, Robertson IH, Harty S, O’Connell RG (2015) Neural
evidence accumulation persists after choice to inform metacog-
nitive judgments. Elife 4:e11946
Nolan KA, Bilder RM, Lachman HM, Volavka J (2004) Catechol
O-methyltransferase Val158Met polymorphism in schizophre-
nia: differential effects of Val and Met alleles on cognitive
stability and flexibility. Am J Psychiatry 161(2):359–361
Norman DA, Shallice T (1986) Attention to action. In: Consciousness
and self-regulation. Springer, Boston, pp 1–18
O’Connell R, Dockree P, Kelly S (2012) A supramodal accumulation-
to-bound signal that determines perceptual decisions in humans.
Nat Neurosci 15:1729–1735
Ouerchefani R, Ouerchefani N, Allain P, Ben Rejeb MR, Le Gall D
(2019) Relationships between executive function, working
memory, and decision-making on the Iowa Gambling Task:
evidence from ventromedial patients, dorsolateral patients, and
normal subjects. J Neuropsychol 13(3):432–461
Panitz C, Hermann C, Mueller EM (2015) Conditioned and
extinguished fear modulate functional corticocardiac coupling
in humans. Psychophysiology 52(10):1351–1360
Polich J (2007) Updating P300: an integrative theory of P3a and P3b.
Clin Neurophysiol 118(10):2128–2148
Polich J, Kok A (1995) Cognitive and biological determinants of
P300: an integrative review. Biol Psychol 41(2):103–146
Sannino S, Gozzi A, Cerasa A, Piras F, Scheggia D, Manago
`F,
Papaleo F (2015) COMT genetic reduction produces sexually
divergent effects on cortical anatomy and working memory in
mice and humans. Cereb Cortex 25(9):2529–2541
Shadlen MN, Kiani R (2013) Decision making as a window on
cognition. Neuron 80(3):791–806
Sikora K, Roelofs A, Hermans D, Knoors H (2016) Executive control
in spoken noun-phrase production: contributions of updating,
inhibiting, and shifting. Q J Exp Psychol 69(9):1719–1740
Smolka MN, Schumann G, Wrase J, Gru
¨sser SM, Flor H, Mann K,
Heinz A (2005) Catechol-O-methyltransferase val158met geno-
type affects processing of emotional stimuli in the amygdala and
prefrontal cortex. J Neurosci 25(4):836–842
Snodgrass JG, Vanderwart M (1980) A standardized set of 260
pictures: norms for name agreement, image agreement, famil-
iarity, and visual complexity. J Exp Psychol Hum Learn Mem
6(2):174
Stauder JE, Smeets EE, van Mil SG, Curfs LG (2006) The
development of visual-and auditory processing in Rett syn-
drome: an ERP study. Brain Dev 28(8):487–494
Swart M, Bruggeman R, Larøi F, Alizadeh BZ, Kema I, Kortekaas R,
Aleman A (2011) COMT Val158Met polymorphism, verbalizing
of emotion and activation of affective brain systems. Neuroim-
age 55(1):338–344
Toplak ME, Sorge GB, Benoit A, West RF, Stanovich KE (2010)
Decision-making and cognitive abilities: a review of associations
between Iowa Gambling Task performance, executive functions,
and intelligence. Clin Psychol Rev 30(5):562–581
Tsai SJ, Yu YWY, Chen TJ, Chen JY, Liou YJ, Chen MC, Hong CJ
(2003) Association study of a functional catechol-O-methyl-
transferase-gene polymorphism and cognitive function in healthy
females. Neurosci Lett 338(2):123–126
van den Bos R, Homberg J, Gijsbers E, den Heijer E, Cuppen E
(2009) The effect of COMT Val158 Met genotype on decision-
making and preliminary findings on its interaction with the
5-HTTLPR in healthy females. Neuropharmacology
56(2):493–498
Vaughn KA, Hernandez AE (2018) Becoming a balanced, proficient
bilingual: predictions from age of acquisition & genetic back-
ground. J Neurolinguist 46:69–77
Vaughn KA, Nun
˜ez AIR, Greene MR, Munson BA, Grigorenko EL,
Hernandez AE (2016) Individual differences in the bilingual
brain: the role of language background and DRD2 genotype in
verbal and non-verbal cognitive control. J Neurolinguist
40:112–127
Verhoef K, Roelofs A, Chwilla D (2009) Role of inhibition in
language switching: evidence from event-related brain potentials
in overt picture naming. Cognition 110:84–99
Witte AV, Flo
¨el A (2012) Effects of COMT polymorphisms on brain
function and behavior in health and disease. Brain Res Bull
88(5):418–428
Yu R, Zhou X (2006) Brain responses to outcomes of one’s own and
other’s performance in a gambling task. NeuroReport
17(16):1747–1751
Zhang Q, Yang Y (2003) The determiners of picture naming latency.
Acta Psychol Sin 35(4):447–454
Zhou Y, Yao M, Fang S, Gao X (2021) A dual-process perspective to
explore decision making in internet gaming disorder: an ERP
study of comparison with recreational game users. Comput Hum
Behav 128:107104
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
Springer Nature or its licensor holds exclusive rights to this article
under a publishing agreement with the author(s) or other rightsh-
older(s); author self-archiving of the accepted manuscript version of
this article is solely governed by the terms of such publishing
agreement and applicable law.
Cognitive Neurodynamics
123
A preview of this full-text is provided by Springer Nature.
Content available from Cognitive Neurodynamics
This content is subject to copyright. Terms and conditions apply.
