Neural Correlates Related to Anxiety in Attentional Inhibition Control: an Erp Study
We investigated the effect of anxiety on attentional inhibitory control using event-related potentials (ERPs) and measuring the response time. Nineteen participants performed a multisource interference task; three-digit numbers and emotional faces were used as visual stimuli. The P100 and P300 ERP...
Saved in:
| Date: | 2014 |
|---|---|
| Main Authors: | , , |
| Format: | Article |
| Language: | English |
| Published: |
Інститут фізіології ім. О.О. Богомольця НАН України
2014
|
| Series: | Нейрофизиология |
| Online Access: | http://dspace.nbuv.gov.ua/handle/123456789/148375 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Journal Title: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Cite this: | Neural Correlates Related to Anxiety in Attentional Inhibition Control: an Erp Study / S. H. Seo, Ch. K. Lee, S. K. Yoo // Нейрофизиология. — 2014. — Т. 46, № 5. — С. 482-491. — Бібліогр.: 37 назв. — англ. |
Institution
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-148375 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-1483752019-02-19T01:29:46Z Neural Correlates Related to Anxiety in Attentional Inhibition Control: an Erp Study Seo, S.H. Lee, Ch.K. Yoo, S.K. We investigated the effect of anxiety on attentional inhibitory control using event-related potentials (ERPs) and measuring the response time. Nineteen participants performed a multisource interference task; three-digit numbers and emotional faces were used as visual stimuli. The P100 and P300 ERP-components and response time values were used to evaluate whether ERP parameters and behavioral responses are associated with increased anxiety. Higher anxiety was associated with a longer latency and reduced amplitude of the P300 component at F3, whereas higher anxiety was associated with a shorter latency and higher amplitude of P300 at F4. The longer P300 latency at F3 was especially related to the response time to the target number with negative-expression faces as distracters. Ми досліджували вплив рівня тривожності на гальмівний контроль уваги з використанням реєстрації пов’язаних із подією потенціалів (ППП) та вимірювання часу реакції. 19 тестованих виконували мультистимульний тест з інтерференцією; як візуальні стимули застосовували зображення тризначних чисел та облич з емоціональними виразами. Параметри Р100 та Р300 компонентів ППП та час реакції брали до уваги для того, щоб визначити можливість асоційованості параметрів ППП та поведінкових відповідей з рівнем тривожності. Підвищена тривожність була асоційована з довшими латентними періодами та зниженою амплітудою компонента Р300 у відведенні F3, тоді як такий рівень тривожності асоціювався з меншими величинами латентного періоду цієї ж самої хвилі та її вищою амплітудою у відведенні F4. Більший латентний період Р300 у відведенні F3 особливо чітко корелював із часом відповіді на цільове число на тлі облич із негативними виразами як відволікаючих факторів. 2014 Article Neural Correlates Related to Anxiety in Attentional Inhibition Control: an Erp Study / S. H. Seo, Ch. K. Lee, S. K. Yoo // Нейрофизиология. — 2014. — Т. 46, № 5. — С. 482-491. — Бібліогр.: 37 назв. — англ. 0028-2561 http://dspace.nbuv.gov.ua/handle/123456789/148375 159.952+612.821.2+612.014 en Нейрофизиология Інститут фізіології ім. О.О. Богомольця НАН України |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| language |
English |
| description |
We investigated the effect of anxiety on attentional inhibitory control using event-related
potentials (ERPs) and measuring the response time. Nineteen participants performed a multisource interference task; three-digit numbers and emotional faces were used as visual stimuli.
The P100 and P300 ERP-components and response time values were used to evaluate whether
ERP parameters and behavioral responses are associated with increased anxiety. Higher
anxiety was associated with a longer latency and reduced amplitude of the P300 component
at F3, whereas higher anxiety was associated with a shorter latency and higher amplitude of
P300 at F4. The longer P300 latency at F3 was especially related to the response time to the
target number with negative-expression faces as distracters. |
| format |
Article |
| author |
Seo, S.H. Lee, Ch.K. Yoo, S.K. |
| spellingShingle |
Seo, S.H. Lee, Ch.K. Yoo, S.K. Neural Correlates Related to Anxiety in Attentional Inhibition Control: an Erp Study Нейрофизиология |
| author_facet |
Seo, S.H. Lee, Ch.K. Yoo, S.K. |
| author_sort |
Seo, S.H. |
| title |
Neural Correlates Related to Anxiety in Attentional Inhibition Control: an Erp Study |
| title_short |
Neural Correlates Related to Anxiety in Attentional Inhibition Control: an Erp Study |
| title_full |
Neural Correlates Related to Anxiety in Attentional Inhibition Control: an Erp Study |
| title_fullStr |
Neural Correlates Related to Anxiety in Attentional Inhibition Control: an Erp Study |
| title_full_unstemmed |
Neural Correlates Related to Anxiety in Attentional Inhibition Control: an Erp Study |
| title_sort |
neural correlates related to anxiety in attentional inhibition control: an erp study |
| publisher |
Інститут фізіології ім. О.О. Богомольця НАН України |
| publishDate |
2014 |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/148375 |
| citation_txt |
Neural Correlates Related to Anxiety in Attentional Inhibition Control: an Erp Study / S. H. Seo, Ch. K. Lee, S. K. Yoo // Нейрофизиология. — 2014. — Т. 46, № 5. — С. 482-491. — Бібліогр.: 37 назв. — англ. |
| series |
Нейрофизиология |
| work_keys_str_mv |
AT seosh neuralcorrelatesrelatedtoanxietyinattentionalinhibitioncontrolanerpstudy AT leechk neuralcorrelatesrelatedtoanxietyinattentionalinhibitioncontrolanerpstudy AT yoosk neuralcorrelatesrelatedtoanxietyinattentionalinhibitioncontrolanerpstudy |
| first_indexed |
2025-07-12T19:15:14Z |
| last_indexed |
2025-07-12T19:15:14Z |
| _version_ |
1837469764772954112 |
| fulltext |
NEUROPHYSIOLOGY / НЕЙРОФИЗИОЛОГИЯ.—2014.—T. 46, № 5482
UDC 159.952+612.821.2+612.014
S. H. SEO,1 Ch. K. LEE,2 and S. K. YOO3
NEURAL CORRELATES RELATED TO ANXIETY IN ATTENTIONAL
INHIBITION CONTROL: AN ERP STUDY
Received September 12, 2013
We investigated the effect of anxiety on attentional inhibitory control using event-related
potentials (ERPs) and measuring the response time. Nineteen participants performed a multi-
source interference task; three-digit numbers and emotional faces were used as visual stimuli.
The P100 and P300 ERP-components and response time values were used to evaluate whether
ERP parameters and behavioral responses are associated with increased anxiety. Higher
anxiety was associated with a longer latency and reduced amplitude of the P300 component
at F3, whereas higher anxiety was associated with a shorter latency and higher amplitude of
P300 at F4. The longer P300 latency at F3 was especially related to the response time to the
target number with negative-expression faces as distracters.
Keywords: anxiety, attentional inhibitory control, event-related potentials (ERPs),
multi-source interference task.
1Department of Computer Engineering, Kyungnam University, Changwon,
Republic of Korea.
2Center for Bionics, Korea Institute of Science and Technology, Seoul,
Republic of Korea.
3Department of Medical Engineering, College of Medicine, Yonsei University,
Seoul, Republic of Korea.
Correspondence should be addressed to S. K. Yoo
(e-mail: shseotwin@kyungnam.ac.kr, chungki@kist.re.kr, sunkyoo@yuhs.ac)
INTRODUCTION
Existing evidence indicates that high anxiety is
often associated with impaired performance in
various cognitive tasks [1, 2]. A variety of models
have been proposed to account for this deficit [3].
An attentional control theory [1] posits that high
anxiety adversely affects central executive functions
such as inhibition and attention shifting. The
above-mentioned functions play a key role in self-
regulation; deficits in the executive functioning are
identified as a fundamental component in a variety of
developmental psychopathologies, including attention
deficit hyperactivity disorder, conduct disorder, and
substance abuse [4, 5]. Inhibition is an important
regulatory function that uses attentional control
to suppress attentional resources directed to task-
irrelevant stimuli.
Most research on the effects of anxiety on inhibitory
control has compared task performances under low
and high distraction conditions [1]. Research on
anxiety and inhibitory control is based on several
assumptions. First, high anxiety should impair the
processing efficiency whether the distracting stimuli
are task-irrelevant ones presented by the experimenter
or induced by worrying thoughts. Generally, the
performance effectiveness can be defined as the quality
of performance, and the processing efficiency can be
defined as the relationship between the performance
effectiveness and spending of resources or efforts.
The processing efficiency is high when performance
effectiveness is high, while the use of resources is low.
Eysenck and Derakshan [6] have suggested that anxiety
typically impairs the processing efficiency to a greater
extent than the performance effectiveness. Second, the
adverse effects of anxiety under distraction conditions
are greater when the task-irrelevant stimuli are threat-
related rather than neutral. In several studies, it was
suggested that anxious individuals have an attentional
bias for threat-related stimuli, and that they have more
difficulties in disengaging attention from such stimuli
[7]. Further, recent research has demonstrated that
anxiety can influence inhibitory control even in the
absence of threat [8, 9].
Assumptions about attentional inhibitory control
have been tested to obtain direct estimates of such
control using various tasks. Derakshan [10] found,
using the antisaccade task [11] that involved the
measurement of eye movements, that high-anxiety vs.
low-anxiety individuals demonstrated significantly
longer antisaccade latencies. Additionally, as was
found in other studies, high-anxiety individuals
NEUROPHYSIOLOGY / НЕЙРОФИЗИОЛОГИЯ.—2014.—T. 46, № 5 483
NEURAL CORRELATES RELATED TO ANXIETY IN ATTENTIONAL INHIBITION CONTROL
had longer antisaccade latencies in response to the
inhibited target. Ansari [12] also found that highly
anxious individuals have lower ERP activity in
the period prior to onset of the inhibited target on
antisaccade trials.
Several studies have also investigated the effects of
anxiety on inhibition of attention toward distracting
stimuli. As was found [13], angry facial expressions
are detected faster in a crowd of happy expressions,
whereas happy face targets were the slowest to be
detected in a crowd of angry faces using a visual search
task [14], suggesting that fear-motivated processing
may occur more rapidly than non-fear processing.
Additionally, Fox [15, 16] found that individuals with
a high-anxiety state took a significantly longer time
to disengage attention from an angry face compared
to those with low anxiety in an emotional spatial-
cuing task. Neuroimaging studies also demonstrated
that highly trait-anxious individuals showed a reduced
prefrontal activity and a slowed target identification
response when the task did not fully occupy the
attentional resources in a letter-search task [9]. This
research suggests that trait anxiety may be linked to
reduced activation of the prefrontal attentional control
mechanisms that inhibit distracter processing, even
when threat-related stimuli are absent. Taken together,
these findings provide evidence for the assumption
that anxiety impairs the processing efficiency and
inhibitory control.
Neurocognitive models of executive attentional
control implicate the lateral prefrontal cortex (PFC),
including the dorsolateral and ventrolateral PFCs
(DLPFC and VLPFC, respectively), as well as the
anterior cingulate cortex (ACC), which are involved
in executive attentional control, particularly in
the allocation of resources during a conflicting or
distracting situations [17-19]. The involvement of the
DLPFC related to attentional control was demonstrated
through a task-manipulating response conflict, where
task-irrelevant stimuli promote a response required by
the current target [20, 21].
There are many studies where the effects of anxiety
on cognitive performance were estimated. However,
the data for research assessing the relationship
between anxiety and inhibitory attentional control
with emotional distracters are scarce. Further, a few
studies examined the relationship between anxiety
and attentional control using direct measures such
as parameters of event-related potentials (ERPs) and
functional magnetic resonance imaging (fMRI). It
is important to understand the neural mechanisms
involved when anxiety impairs the effective inhibitory
control. Previous studies typically examined the
processing efficiency using behavioral measures.
Accordingly, our present study aimed to address this
critical gap in the literature.
We used a modified Multi-Source Interference Task
(MSIT, [22]) to assess attentional inhibitory control
related to anxiety. This task is a top-down control
task of selective attention, where a target number
and a distracter (an image of the emotional facial
expression) are presented. In behavioral studies, it
was found that faces capture attention more readily
than other visual stimuli such as pictures of musical
instruments or appliances [23]. In particular, faces
displaying a negative affect vs. neutral or happy
faces capture attention differentially [24]. Ebner [25]
showed that task-irrelevant faces increase the response
time in face-unrelated number trials using the MSIT.
Task-irrelevant faces also disrupt responses to face-
unrelated targets. Thus, we believe that the MSIT
is a valid method of assessing attentional inhibitory
control because attention is required to select the
target number, and inhibition is required to ignore the
task-irrelevant face.
The goal of our study was to understand how anxiety
impairs efficient inhibition control by assessing neural
mechanisms using direct measures (such as ERP
parameters) and to correlate this neural measure with
the behavioral response. We hypothesized several
outcomes. First, anxiety would be correlated with
cortical activity in the prefrontal regions (such as
the DLPFC) within the MSIT period. Second, higher
anxiety would be related to longer latencies of the
ERP components and longer response times due to
slowed target identification. Third, higher anxiety
would be associated with enhanced neural activation
in order to achieve a given level of the cognitive task
performance.
METHODS
Participants. Nineteen right-handed graduates
(10 men and 9 women; mean age 30 years) from
the Yonsei University participated in the study. All
participants were free from psychiatric or neurological
disorders. Data from three participants were excluded
because of excessive artifacts in the ERP data, leading
to a final sample of 54.
Participants completed the Korean version of the
State-Trait Anxiety Inventory (STAI, [26, 27]) prior
NEUROPHYSIOLOGY / НЕЙРОФИЗИОЛОГИЯ.—2014.—T. 46, № 5484
S. H. SEO, Ch. K. LEE, and S. K. YOO
to every session.
Multi-Source Interference Task. This task was
a modified version of the Multi-Source Interference
Task (MSIT, [22]). Participants were instructed to
maintain fixation on the center of a cross presented on
the monitor. A pair of face and number stimuli (three
digits around the nose area of the emotional face)
was presented for 1500 msec. Each pair consisted of
three numbers and one emotional face image. Pictures
were selected from the FACES 3.3.1 database at the
Max Planck Institute for Human Development (MPIB;
FACES ID: ssanghee) [28]. The pictures contained
a set of images of naturalistic faces of 144 young
men with each one displaying one of the following
six facial expressions, namely neutrality, sadness,
disgust, fear, anger, and happiness. Three digits, two
matching and one nonmatching, were presented, and
the digit that differed was the target number. Next, a
gray screen with the fixation cross was presented for
1500 msec.
Participants completed three sessions that were
conducted at the same time of day over the course of
three consecutive days. After taking a 10-min break
with their eyes open, the task was presented for 3
min to familiarize participants with the protocol prior
to the start of the experiment. As illustrated in Fig.
1, 144 fixation crosses and 144 pictures (6 facial
types presented 24 times) were each presented for a
period of 1.5 sec randomly during the experiment.
The three numbers (e.g., 1, 2, and 3) were presented
randomly against the background of the emotional
face. For example, if 232 was presented, the digit
3 would be the target number. Participants were asked
to press the target button using a number keypad (made
in the laboratory). For each participant, the response
time to press the target button was measured. Errors
included incorrect key pressing, missed key presses,
or a response time > 1500 msec.
EEG Recording and Data Reduction. EEG was
continuously recorded for each participant during the
task. Recording electrodes were positioned according
to the international 10–20 system of electrode
placement, including the earlobes. A ground electrode
was placed on the back of the neck (Iz), whereas
reference electrodes were placed on the right and
left ears (A1+A2). The use of two linked earlobes as
a reference conceptually provides a virtual reference
site in the middle of the head. The use of symmetrical
reference sites provides avoiding bias recordings
toward activity in one hemisphere [29]. Eye movement
artifacts were corrected using the ICA algorithm. All
electrode impedances were below 5 kΩ. The EEG
signals were amplified using a Biopac MP150 TM
system, band-pass filtered (0.1-100 Hz), and digitized
at a sampling rate of 1000 sec–1. The high-pass and low
pass filters for EEG signals were set to 0.5 and 100 Hz,
respectively. The 60 Hz notch filter was continuously
in use. Values of the amplitudes of ERP components
were obtained by stimulus-locked averaging from
0 to 1000 msec after baseline correction.
Primary ERP analysis was focused on the P100
and P300 components. The P100 is sensitive to
visual spatial attention toward emotional faces [30],
and the P300 component is associated with number
identification because this component is closely
related to the selective attentional requirements
of target identification [31]. Thus, the P100
(100-200 msec) and P300 (250-450 msec) components
were compared by analyzing the mean amplitudes and
peak latencies of these waves at the F3, F4, Cz, and Pz
positions after stimulus onset.
Statistical Analyses. To test DLPFC and ACC
involvement related to attentional inhibitory control
during MSIT, Pearson’s correlation analysis between
P100 and P300 components was performed. To test
the effects of anxiety on attentional inhibitory control,
the dependence between ERP parameters and anxiety
state was estimated using multiple regression where
the relationships of the state of anxiety (STAI score)
to each of the P100 amplitude, P300 amplitude, P100
latency, and P300 latency at each position (e.g., F3,
F4, Cz, and Pz) were examined. Also, to determine
the relationship between ERP indices and behavioral
responses, the relationship between the response time
1.5 sec
113 233 232 322 211
1.5 sec
7 (min)
Total 144 images, 144 cross
F i g. 1. Schematic representation of the
experimental design; participants were
instructed to press a keypad button at target
number as soon as possible.
Р и с. 1. Схема експерименту; тестовані
були інструктовані якнайшвидше
натиснути кнопку при пред’явленні
цільового числа.
NEUROPHYSIOLOGY / НЕЙРОФИЗИОЛОГИЯ.—2014.—T. 46, № 5 485
NEURAL CORRELATES RELATED TO ANXIETY IN ATTENTIONAL INHIBITION CONTROL
with the face type presented and ERP parameters was
tested using multiple regression analysis. Multiple
regression analysis was also used to examine the
relationship between ERP responses and the number of
errors. All statistical analyses were carried out using
SPSS 16.0 (SPSS Inc., USA).
RESULTS
ERP Responses to the Multi-Source Interference
Task. Attentional control has been mentioned as an
available mechanism that may mediate difficulties in
disengagement of attention from a threat or emotional
distracter [1]. Two stages of information processing,
automatic and strategic, are required under such
conditions [32, 33]. Automatic processing refers
to processing that occurs without intent, control, or
awareness, whereas strategic processing refers to that
which is intentional, controllable, and dependent on
awareness. The MSIT requires combined information
processing (both automatic and strategic) with
respect to the facial expression (automatic) and
target identification (strategic).The ERP responses at
presentations of target numbers and facial expressions
are shown in Fig. 2. Because facial expression receives
high attentional priority, the early-emerging P100
peak appears to be related to presentation of the task-
irrelevant facial expression, while the later-emerging
P300 peak appears to be associated with identification
of the target number.
ERP Responses and Attentional Inhibitory
Control Influenced by Anxiety. The purpose of this
analysis was to determine: (i) whether the MSIT is a
valid task for assessing emotion-related attentional
inhibitory control, and (ii) whether the ERP P300
–0.2
–0.2
0 0100 100200 200300 300400 400500
msec
Cz Pz
F4
F3
msec
500
–0.1
–0.1
0
0
0.1
0.1
0.2
0.2
μV
μV
F i g. 2. ERP waveforms induced by emotional
facial expression and target number.
Р и с. 2. Форми хвиль пов’язаних із подією
потенціалів, індукованих пред’явленням
зображень облич з емоціональними виразами
та цільових чисел.
Table 1. Correlation scores between the P100 and P300 ERP components
Т а б л и ц я 1. Кореляція між характеристиками компонентів Р100 та Р300 в складі пов’язаних із подією потенціалів
P100 amplitude P100 latency
Position F3 F4 Cz Pz F3 F4 Cz Pz
P300 amplitude
F3 0.171 0.298* 0.076 –0.083 –0.255 –0.136 –0.139 –0.195
F4 0.227 0.344* 0.112 –0.107 –0.279* –0.180 –0.189 –0.215
Cz 0.244 0.344* 0.240 0.060 –0.154 –0.096 –0.075 –0.160
Pz 0.172 0.289* 0.172 0.035 –0.351** –0.277* –0.161 –0.201
P300 latency
F3 0.587** 0.601** 0.561** 0.070 0.307* 0.338** 0.108 0.067
F4 0.598** 0.591** 0.540** 0.025 0.395** 0.383** 0.160 0.136
Cz 0.606** 0.602** 0.557** 0.045 0.319* 0.346* 0.111 0.079
Pz 0.626** 0.613** 0.613** 0.120 0.311* 0.288* 0.103 0.008
Footnote: ** P < 0.01, * P <0.05
NEUROPHYSIOLOGY / НЕЙРОФИЗИОЛОГИЯ.—2014.—T. 46, № 5486
S. H. SEO, Ch. K. LEE, and S. K. YOO
component specifically reflects emotion-related
attentional inhibitory control related to the anxiety
level. The DLPFC and ACC may be neural mechanisms
responsible for difficulties in disengaging attention
from an emotional distracter. Table 1 shows the
correlation between parameters of the P100 component
to facial expression and those of the P300 component
related to target identification. The P100 amplitudes
and latencies at F3 and F4 demonstrated positive
correlations with the P300 latencies at all positions.
The F3 and F4 sites correspond to the DLPFC area. In
addition, this result shows that the greater the neural
activation induced by facial expression, the greater the
ERP latency related to target identification.
To test whether the ERP latency to the target
number predicts inhibitory control related to anxiety
under MSIT conditions, multiple regression analysis
was performed using a backward method, where the
P100 and P300 latency at each position (in particular,
F3, F4, Cz, and Pz) served as predictors, and the
STAI score served as the dependent variable. If the
relationship between the ERP latency in the DLPFC
region (at F3 and F4 leads) and the anxiety level is
significant, the MSIT could be regarded as a valid
task. In particular, the P300 latency could be a neural
factor that reflects the delay in target identification,
suggesting that anxiety-related executive control failed
to inhibit attention to task-irrelevant face images. The
results showed that higher anxiety was associated
with increased awareness of task-irrelevant faces
(Table 2a), and that participants with a longer P300
latency in F3 and a shorter P300 latency in F4 reported
higher anxiety levels (Table 2b). The regression
models of the P100 (F = 15.285, df = 1, P < 0.001)
and the P300 (F = 6.452, df = 3, P = 0.001) latencies
were both statistically significant, while the Cz latency
of the P300 component did not reach the statically
significant level (P > 0.05). Negative (although weak)
correlation between the P100 Pz latency and the STAI
was observed (r = –0.195), suggesting that the higher
the anxiety state, the shorter the P100 latency at Pz,
indicating faster processing of emotional face images.
Additionally, the relationships of the F3 latency
(r = 0.250) and F4 latency (r = –0.210) with the STAI
score were also significant. Thus, the correlation
coefficient for the F3 latency was positive, indicating
that the longer the F3 latency, the higher the STAI
score. At the same time, the coefficient for the F4
latency was negative, suggesting that the shorter
the F4 latency, the higher the STAI score. Thus,
opposite characteristics of the P300 latency in two
DLPFC regions were associated with higher anxiety.
Consistent with predictions, the MSIT could be used
to evaluate the association between neural activation
in the DLPFC area and the state of anxiety.
Behavioral Responses Related to Attentional
Inhibitory Control. Behavioral data were analyzed
to determine whether a longer ERP latency is
accompanied by slower target identification. We
analyzed the relationship between the ERP latency and
response time to target identification. Additionally, we
tested a supposition that the type of facial expression
used as a distractor would affect target identification.
Multiple regression was used for averaged values of
the response time as the dependent variable and the
F3, F4, Cz, and Pz latencies of the P300 component
as predictors. The averaged response time was defined
as the mean of the response times to all types of target
Table 2. Associations of the STAI score with ERP latency variables using multiple linear regression analysis
Т а б л и ц я 2. Зв’язки між оцінками за STAI та латентними періодами компонентів пов’язаних із подією потенціалів,
визначені з використанням аналізу множинної лінійної регресії
(a) Latency of the P100 component
Variable M ± s.e.m. P value
Pz latency –0.195±0.05 0.000
(b) Latency of the P300 component
Variable M ± s.e.m. P value
F3 latency 0.250±0.067 0.000
F4 latency –0.210±0.066 0.002
Cz latency –0.083±0.045 0.071
Footnotes: s.e.m. are the standard errors; dependent variable is STAI score
NEUROPHYSIOLOGY / НЕЙРОФИЗИОЛОГИЯ.—2014.—T. 46, № 5 487
NEURAL CORRELATES RELATED TO ANXIETY IN ATTENTIONAL INHIBITION CONTROL
trials, regardless of the type of facial expression. The
regression model of the P300 latency (F = 7.629, d = 1,
P = 0.008) was significant. There was a rather strong
positive correlation between the P300 F3 latency and
average response time (r = 0.703), indicating that the
longer the P300 F3 latency, the greater the average
response time.
The response times in negative facial expression trials
were compared with those in neutral and positive facial
expression trials. We tested whether the response time
to the type of facial expression affected the P300 F3
and F4 latencies. As is shown in Table 3, all regression
models were significant. Correlations of the F3 and F4
latencies with the response time were positive, indicating
that the longer the F3 and F4 latencies, the slower the
response performance. Interestingly, the P300 F3 latency
significantly correlated with the average response time
for negative facial expression trials such as annoyance,
disgust, and fear. In contrast, the P300 F4 latency was
not significantly associated with the response time in any
of the trial types. Also, the P300 F4 latency significantly
correlated with the average response time to facial
expressions such as smiling, neutrality, and sadness,
while correlation of the P300 F3 latency was insignificant.
These results appear to be related to the lateralization of
the neural responses in the DLPFC region according to
the facial expression type. We focused on analyzing the
F3 latency because interference effects of negative facial
expressions compared to those of neutral or positive ones
were greater for individuals with high anxiety. Finally,
this result indicated that longer P300 F3 latencies were
related to anxiety and associated with longer response
times suggesting slower target identification.
ERP Activation Effect on the Processing
Efficiency. The processing efficiency theory argues
that higher anxiety is associated with the reduced
performance efficiency during a demanding task, with
more efforts and resources being expended to avoid
actual decrements in the effectiveness or accuracy. At
the neural level, the attentional control theory predicts
that high anxiety will produce increased prefrontal
cortical activation in order to achieve a given desirable
level of the cognitive task performance. We examined
whether high anxiety is associated with greater cerebral
activity. Multiple regression analysis was performed
using the STAI score as the dependent variable and
the F3, F4, Cz, and Pz amplitudes of the P100 and
P300 components as predictors. The regression model
for the P100 amplitude was clearly not significant
(F = 1.010, d = 2, P = 0.371), but the analogous model
for the P300 amplitude was rather close to the level
of significance (F = 2.511, d = 2, P = 0.091). Table 4
shows that the effects of anxiety were associated with
conflicting characteristics of the amplitude at the F3
and F4 positions. The correlation coefficient for the F3
amplitude was negative, suggesting that the lower the
F3 amplitude, the higher the STAI score. At the same
time, the coefficient for the F4 amplitude was positive,
Table 3. Associations of the response time to the face type using multiple regression analysis
Т а б л и ц я 3. Зв’язки між часом реакції та типом виразу обличчя, визначені з використанням аналізу множинної лінійної
регресії
Variable F3 latency ANOVA F4 latency ANOVA
Type of face M ± s.e.m. P value F P value M ± s.e.m. P value F P value
Annoy 0.767±0.263 0.005 8.512 0.005 NS
Disgust 0.636±0.254 0.015 6.286 0.015 NS
Fear 0.611±0.265 0.025 5.320 0.025 NS
Smile NS 0.750±0.276 0.009 7.353 0.009
Neutral NS 0.811±0.289 0.007 7.856 0.007
Sadness NS 0.779±0.274 0.006 8.075 0.006
Footnotes: Dependent variable is the average response time according to facial expression; NS, not significant.
Table 4. Associations of the STAI score with the ERP amplitude variables using multiple linear regression analysis
Т а б л и ц я 4. Зв’язки між оцінками за STAI та амплітудами компонентів пов’язаних із подією потенціалів, визначені з
використанням аналізу множинної лінійної регресії
Variable M ± s.e.m. P value
F3 amplitude 86.875 ± 43.918 0.053
F4 amplitude 90.164 ± 41.361 0.034
Footnote: Dependent variable is STAI score.
NEUROPHYSIOLOGY / НЕЙРОФИЗИОЛОГИЯ.—2014.—T. 46, № 5488
S. H. SEO, Ch. K. LEE, and S. K. YOO
indicating that the greater the F4 amplitude, the higher
the STAI score. Thus, the opposite characteristics of
the P300 latency were associated with higher anxiety
in the DLPFC region of two hemispheres.
Finally, these results showed that higher STAI
scores were associated with the ERP pattern having
longer latencies and lower amplitudes at F3, whereas
higher STAI score was associated with the shorter
latency and greater amplitude of ERP component at
F4.
The Performance Effectiveness and Anxiety. We
also tested whether anxiety could adversely affect
the performance effectiveness. The latter index was
defined as the number of errors, which included cases
where the participant responded to the wrong target,
did not respond at all, or responded with a delay longer
than 1500 msec. We examined whether the ERP latency
to the target number predicted the smaller performance
effectiveness. Multiple regression was estimated using
a backward method of selection where the number of
errors served as the dependent variable, and values of
the P300 latency at each position (F3, F4, Cz, and Pz)
were the predictors. In addition, single regression was
calculated where the error count was the dependent
variable, and the STAI score was the predictor.
The multiple regression model of the P300 latency
(F = 1.450, d = 2, P = 0.244) and the single regression
model of STAI score (F = 0.006, d = 1, P = 0.939)
were both insignificant. These results indicate that
the level of anxiety does not considerably affect the
performance accuracy under MSIT conditions. The
regression analyses also showed that the state of
anxiety does not affect the performance effectiveness.
DISCUSSION
The goal of our study was to elucidate whether high
anxiety impairs attentional inhibitory control and
processing efficiency during the MSIT using ERP and
behavioral response measurements. A few findings
relevant to the initial hypotheses did emerge. First,
our results indicated that the MSIT can be used
as a valid technique for assessing emotion-related
attentional inhibitory control related to anxiety.
Second, the results also showed that the anxiety level
is specifically associated with the ERP P300 latency
in the DLPFC region. Higher anxiety was related to
a longer latency of this wave at F3 but to a shorter
latency at F4. Third, our results showed that the
response times were different according to the type of
the facial expression image used as a distracter in such
a way that the response time to negative facial stimuli
(e.g., annoyance, disgust, and fear) was affected by
the P300 latency at F3, whereas this time to other face
stimuli (e.g., neutral, smile, and sadness) was affected
by the P300 latency at F4. Finally, the results showed
that anxiety is oppositely associated with the ERP
P300 amplitude in the DLPFC region. Higher anxiety
was related to lower amplitude at F3 but to a greater
amplitude at F4.
In previous studies, several different tasks (e.g., an-
tisaccade, visual search, emotional spatial-cuing, and
dot probe tasks) were used to assess the effects of anxi-
ety on inhibition of distracting stimuli [7, 12, 13, 15,
16]. These studies showed that highly anxious vs. less
anxious individuals showed clearly impaired inhibition
control and lower processing efficiency. However, the
tasks used could not be claimed as process-pure ones
that primarily reflect a single underlying process [34].
In our study, we used the MSIT [22], a technique that
was not used previously to assess the effects of anxi-
ety on inhibition of distracting stimuli. A task that re-
quired the interplay of emotions and attention to mea-
sure emotion-related attentional inhibitory control was
needed in order to directly test any relationship be-
tween the above control and the anxiety level. Ebner
[25] showed that the MSIT can measure interference
effects of task-irrelevant faces on target numbers. This
study, however, did not examine neural responses on
the interference effects of task-irrelevant faces and on
the target number.
Classical neurocognitive models of executive atten-
tional control implicate the lateral PFC (DLPFC and
VLPFC) and ACC in executive attentional control and,
particularly, in the allocation of resources during con-
flicting or distracting situations [17-19]. If a signi-
ficant neural response is present within the DLPFC
and ACC areas during the MSIT, the task can be con-
sidered valid for measuring inhibition control of dis-
tracting stimuli. We found that the parameters of P300
at F3 and F4 in the DLPFC are related to the anxiety
state. Thus, the MSIT is a rather valid method.
It is important to assess attentional processes as
precisely and directly as possible in research on anx-
iety and performance. In previous studies, inhibi-
tion control and processing efficiency were examined
only indirectly, by behavioral measures. Ansari et al.
[12] showed that anxiety impairs attentional inhibi-
tory control; the authors measured the latency of eye
movements under distracting conditions. High-anxiety
compared with low-anxiety individuals showed longer
NEUROPHYSIOLOGY / НЕЙРОФИЗИОЛОГИЯ.—2014.—T. 46, № 5 489
NEURAL CORRELATES RELATED TO ANXIETY IN ATTENTIONAL INHIBITION CONTROL
latencies of eye movements to the target. It was also
found [13] that angry facial expressions were detected
faster in a crowd of happy expressions, whereas happy
faces were the slowest to be detected in a crowd of an-
gry faces. Further, Fox et al. [15, 16] found that indi-
viduals in a high-anxiety state needed a significantly
longer time to disengage attention from an angry face
in an emotional spatial-cuing task compared to those
with low anxiety. In general, these studies showed that
higher anxiety leads to slower processing when trying
to identify a target. In our study, the later-emerging
P300 component was found to be associated with the
number of correctly identified targets. If the latency of
the P300 component is greater, it may be interpreted as
slower identification of the target number. Interesting-
ly, we found that higher anxiety estimates positively
correlated with longer P300 latencies at F3 (left DLP-
FC) but not at F4 (right DLPFC), where shorter laten-
cies of P300 were observed. The results appear to in-
dicate lateraliztion of neural responses in the DLPFC
region during the MSIT.
In several studies, it was also suggested that high
anxiety slows down target identification [9]. The ERP
latencies were found to correlate with the response
time in the flanker task [35]. The relationship between
neural responses assessed by ERPs and the response
time provides clearer evidence of the relationship be-
tween anxiety and cognitive processing. Longer ERP
latencies may be related to slower target identification
and may consequently lead to longer response times.
We, when trying to determine whether the longer ERP
latency is accompanied by slowed target identification,
analyzed the relationship between the ERP latency and
response time and found that longer F3 and F4 laten-
cies were associated with longer response times. We
specially tested whether the type of facial expression
used as a distracter affects target identification. The
relationships between the F3 and F4 latencies of P300
and response times showed that the latter in the case
of negative face distracter types (e.g., annoyance, dis-
gust, and fear) were related only to the F3 latency,
while this time at other face distracter types (e.g., neu-
tral, smile, and sadness) was related exclusively to F4
latency. These results suggest that neural responses at
the left DLPFC are closely related to the behavioral
responses when negative facial expressions were pre-
sented. These neural responses reflect the difficulty in
disengaging attention from the negative face distracter.
The latency of P300 at F3 positively correlated with
the response time. Thus, the latency of P300 at F3 may
be a manifestation of emotion-related attentional in-
hibitory control related to anxiety, and this parameter
reflects impaired inhibition control.
The attentional control theory [1] suggests that high
anxiety impairs the processing efficiency to a greater
extеnt, as compared with the performance effectiveness.
Several studies showed that high anxiety is associated
with greater cerebral activity [36, 37]. Enhanced neu-
ral activation was assumed to reflect the use of greater
efforts and/or resources as a compensatory strategy to
achieve an adequate level of performance. However, trait
anxiety is also associated with reduced DLPFC involve-
ment when trying to inhibit distracter processing under
conditions of low perceptual loading in the absence of a
threat [9]. In our study, we found a slightly significant
relationship between the anxiety level and the amplitude
of P300 at F3 and F4. The reduced P300 amplitude at F3
and enhanced P300 amplitude at F4 are related to higher
anxiety. When examining the response time and P300 la-
tency related to anxiety in the left DLPFC (F3), we found
that the anxiety state is associated with reduced and de-
layed neural activation in the left DLPFC, reflecting de-
creased attentional inhibitory control. The result is con-
sistent with Bishop’s findings [9].
We also investigated the adverse effects of anxi-
ety on the performance effectiveness. The latter was
measured as the number of errors during the MSIT.
However, regression analysis showed no significant
influence of the anxiety state on the performance ef-
fectiveness. Anxiety did not affect the performance ac-
curacy, possibly due to the level of perceptual loading.
There was a very low error rate across all participants,
indicating that this task was relatively easy to perform
due to a low perceptual loading. If so, the low base
rate of errors may have precluded an adequate test of
the relationship between the performance effectiveness
and anxiety.
Notwithstanding these limitations, the results of our
study allow us to suggest that anxiety impairs emotion-
related attentional inhibitory control in the DLPFC
region, which was reflected in longer P300 latencies
and lower P300 amplitudes at F3 and a longer response
time to emotionally negative faces. The present ERP
finding can help to improve our understanding of
neural mechanisms involved in the effect of anxiety on
the central executive functions and, in particular, on
attentional inhibitory control. Further, these findings
may be used as an index to assess the cognition
performance for clinical and nonclinical anxiety
disorders. We believe that future studies will examine
whether these ERP findings extend to trait anxiety and
state of anxiety.
NEUROPHYSIOLOGY / НЕЙРОФИЗИОЛОГИЯ.—2014.—T. 46, № 5490
S. H. SEO, Ch. K. LEE, and S. K. YOO
Acknowledgments: This work was financially supported by
National Research Foundation of Korea (NRF) grant funded by
the Korean government (MEST; No. 2010-0026833).
All testing procedures were in accordance with the Helsinki
Declaration and with the ethical standards of the Responsible
Committee on Human Experimentation (Institutional Review
Board of the Severance Hospital). Written informed consent
was obtained from all persons included in the study.
The authors, S. H. Seo, Ch. K. Lee, and S. K. Yoo, confirm
that they have no conflict of interest.
С. Х. Сео1, Ч. К. Лі2, С. К. Йоу3
НЕЙРОННІ КОРЕЛЯТИ РІВНЯ ТРИВОЖНОСТІ В
ГАЛЬМІВНОМУ КОНТРОЛІ УВАГИ: ДОСЛІДЖЕННЯ
З РЕЄСТРАЦІЄЮ ПОВ’ЯЗАНИХ ІЗ ПОДІЄЮ
ПОТЕНЦІАЛІВ
1 Університет Кюнгнам, Тьюгсангнам (Республіка Корея).
2 Центр біоніки Корейського інституту наук і технології,
Сеул (Республіка Корея).
3 Медичний Коледж Університету Йонсей, Сеул
(Республіка Корея).
Р е з ю м е
Ми досліджували вплив рівня тривожності на гальмівний
контроль уваги з використанням реєстрації пов’язаних із
подією потенціалів (ППП) та вимірювання часу реакції.
19 тестованих виконували мультистимульний тест з інтер-
ференцією; як візуальні стимули застосовували зображен-
ня тризначних чисел та облич з емоціональними виразами.
Параметри Р100 та Р300 компонентів ППП та час реакції
брали до уваги для того, щоб визначити можливість асоці-
йованості параметрів ППП та поведінкових відповідей з рів-
нем тривожності. Підвищена тривожність була асоційована
з довшими латентними періодами та зниженою амплітудою
компонента Р300 у відведенні F3, тоді як такий рівень три-
вожності асоціювався з меншими величинами латентного
періоду цієї ж самої хвилі та її вищою амплітудою у відве-
денні F4. Більший латентний період Р300 у відведенні F3
особливо чітко корелював із часом відповіді на цільове чис-
ло на тлі облич із негативними виразами як відволікаючих
факторів.
REFERENCES
1. M. W. Eysenck, N. Derakshan, R. Santos, and G. Calvo,
“Anxiety and cognitive performance: Attentional control
theory,” Emotion, 7, 336-353 (2007).
2. J. M. G. Williams, F. N. Watts, C. MacLeod, and A. Mathews,
Cognitive Psychology and Emotional Disorders, Wiley,
Chichester, Great Britain (1997).
3. J. M. Cisler and E. H. W. Koster, “Mechanisms of attentional
biases towards threat in the anxiety disorders: an integrative
review,” Clin. Psychol. Rev., 30, No. 2, 203-216 (2010).
4. T. E. Moffitt, “The neuropsychology of conduct disorder,” Dev.
Psychopathol., 5, 135-151 (1993).
5. J. T. Nigg, “On inhibition/disinhibition in developmental
psychopathology: Views from cognitive and personality
psychology and a working inhibition taxonomy,” Psychol.
Bull., 126, 220-246 (2000).
6. M. W. Eysenck and N. Derakshan, “New perspectives in
attentional control theory,” Pers. Individ. Differ., 50, 955-960
(2011).
7. Y. Bar-Haim, D. Lamy, L. Pergamin, et al., “Threat-related
attentional bias in anxious and nonanxious individuals: A
meta-analytic study,” Psychol. Bull., 133, 1-24 (2007).
8. N. Derakshan and M. W. Eysenck, “Anxiety, processing
efficiency and cognitive performance: New developments from
attentional control theory,” Eur. Psychologist, 14, No. 2, 168-
176 (2009).
9. S. J. Bishop, “Trait anxiety and impoverished prefrontal
control of attention,” Nat. Neurosci., 12, No. 1, 92-98 (2009).
10. N. Derakshan, T. L. Ansari, L. Shoker, et al., “Anxiety and
inhibition: An investigation using the antisaccade task,” Exp.
Psychol., 56, 48-55 (2009).
11. P. E. Hallet, “Primary and secondary saccades to goals defined
by instructions,” Vis. Res., 18, 1279-1296 (1978).
12. T. L. Ansari and N. Derakshan, “The neural correlates of
impaired inhibitory control in anxiety,” Neuropsychologia, 49,
1146-1153 (2011).
13. A. Öhman, D. Lundqvist, and F. Esteves, “The face in the
crowd revisited: A threat advantage with schematic stimuli,”
J. Pers. Soc. Psychol., 80, 381-396 (2001).
14. C. H. Hansen and R. D. Hansen, “Finding the face in the
crowd: An anger superiority effect,” J. Pers. Soc. Psychol.,
54, 917-924 (1988).
15. E. Fox, R. Russo, R. J. Bowles, and K. Dutton, “Do threatening
stimuli draw or hold visual attention in sub-clinical anxiety?”
J. Exp. Psychol.: General, 130, 681-700 (2001).
16. E. Fox, R. Russo, and K. Dutton, “Attentional bias for threat:
Evidence for delayed disengagement from emotional faces,”
Cogn. Emot., 16, 355-379 (2002).
17. M. Botvinick, T. Braver, D. Barch, et al., “Conflict monitoring
and cognitive control,” Psychol. Rev., 108, No. 3, 624-652
(2001).
18. J. Duncan and A. M. Owen, “Common regions of the human
frontal lobe recruited by diverse cognivtive demands,” Trends
Neurosci., 23, 475-483 (2000).
19. E. K. Miller and J. D. Cohen, “An integrative theory of
prefrontal cortex function,” Ann. Rev. Neurosci., 24, 67-202
(2001).
20. J. G. Kerns, J. D. Cohen, A. W. MacDonald, et al., “Anterior
cingulate conflict monitoring and adjustments in control,”
Science, 303, 1023-1026 (2004).
21. S. Durston, M. C. Davidson, K. M. Thomas, et al., “Parametric
manipulation of conflict and response competition using rapid
mixed-trial event-related fMRI,” NeuroImage, 20, 2135-2141
(2003).
22. G. Bush and L. M. Shin, “The multi-source interference task:
An fMRI task that reliably activates the cingulo-frontal-
parietal cognitive/attention network,” Nature Protocols, 1,
308-313 (2006).
NEUROPHYSIOLOGY / НЕЙРОФИЗИОЛОГИЯ.—2014.—T. 46, № 5 491
NEURAL CORRELATES RELATED TO ANXIETY IN ATTENTIONAL INHIBITION CONTROL
23. R. Jenkins, A. M. Burton, and A. W. Ellis, “Long-term effects
of convert face recognition,” Cognition, 86, 43-52 (2002).
24. R. Palermo and F. Rhodes, “Are you always on my mind?
A review of how face perception and attention interact,”
Neuropsychologia, 45, 75-92 (2007).
25. N. C. Ebner and M. K. Johnson, “Age-group differences in
interference from young and older emotional faces,” Cogn.
Emot., 24, No. 7, 1095-1116 (2010).
26. C. D. Spielberger, R. L. Gorsuch, and R. F. Lushene, Manual
for the State-Trait Anxiety Inventory, Consulting Psychologists
Press, Palo Alto (1970).
27. J. T. Kim, Relationship between Trait Anxiety and Sociability:
Focused on STAI of Spielberger [Dissertation], Graduate
School of Korea Univ., Seoul (1978).
28. N. C. Ebner, M. Riediger, and U. Lindenberger, “FACES-A
database of facial expressions in young, middle-aged, and
older women and men: Development and validation,” Behav.
Res. Methods (2009).
29. D. Hagemann, E. Naumann, and J. F. Thayer, “The quest for
the EEG reference revisited: A glance from brain asymmetry
research,” Psychophysiology, 38, 847-857 (2001).
30. E. M. Mueller, S. G. Hofmann, D. L. Santesso, et al.,
“Electrophysiological evidence of attentional biases in social
anxiety disorder,” Psychol. Med., 39, No. 7, 1141-1152 (2009).
31. S. A. Hillyard and D. L. Wood, “Electrophysiological analysis
of human brain function,” in: Handbook of Behavioral
Neurology, M. S. Gazzaniga (ed.), Plenum, New York (1979),
pp. 345-378.
32. A. Moors and J. D. Houwer, “Automaticity: a theoretical and
conceptual analysis,” Psychol. Bull., 132, 297-326 (2006).
33. R. M. Shiffrin and W. Schneider, “Controlled and automatic
human information processing: Perceptual learning, automatic
attending and a general theory,” Psychol. Rev., 84, 127-190
(1977).
34. S. B. Hutton and U. Ettinger, “The antisaccade task as
a research tool in psychopathology: A critical review,”
Psychophysiology, 43, 302-313 (2006).
35. J. R. Folstein and C. Van Petten, “Influence of cognitive
control and mismatch on the N2 component of the ERP: A
review,” Psychophysiology, 45, 152-170 (2008).
36. S. Righi, L. Mecacci, and M. P. Viggiano, “Anxiety, cognitive
self-evaluation and performance: ERP correlates,” J.
Anxiety Disord., 23, 1132-1138 (2009).
37. A. N. Savostyanov, A. C. Tsai, M. Liou, et al., “EEG correlates
of trait anxiety in the stop-signal paradigm,” Neurosci. Lett.,
449, 112-116 (2009).
|