Diffusion Tensor Imaging Studies on Chinese Patients with
Social Anxiety Disorder
Changjian Qiu,1Chunyan Zhu,1,2Jingna Zhang,3,4Xiaojing Nie,1Yuan Feng,1
Yajing Meng,1Ruizhi Wu,1Xiaoqi Huang,5Wei Zhang,1and Qiyong Gong5
1Mental Health Center, West China Hospital of Sichuan University, Chengdu 610041, China
2The Seventh People’s Hospital of Hangzhou, Hangzhou 310013, China
3Key Laboratory for Neuroinformation of Ministry of Education, School of Life Science and Technology,
University of Electronic Science and Technology of China, Chengdu 610054, China
4Department of Medical Image, College of Biomedical Engineering, Third Military Medical University, Chongqing 400038, China
5Huaxi MR Research Center (HMRRC), Department of Radiology, West China Hospital of Sichuan University,
West China School of Medicine, Chengdu 610041, China
Correspondence should be addressed to Xiaoqi Huang; email@example.com and Wei Zhang; firstname.lastname@example.org
Received 20 October 2013; Revised 6 January 2014; Accepted 29 January 2014; Published 3 March 2014
Academic Editor: Yong He
Copyright © 2014 Changjian Qiu et al. This is an open access article distributedunder the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
The aim of this study was to explore white-matter disruption in social anxiety disorder (SAD) patients by using diffusion tensor
imaging (DTI) and to investigate the relationship between cerebral abnormalities and the severity of the symptoms. Eighteen SAD
(FA) and apparent diffusion coefficient (ADC) for each subject. We used voxel-based analysis to determine the differences of FA
and ADC values between the two groups with two-sample 푡-tests. The SAD patient showed significantly decreased FA values in the
SAD patients, we observed a significant negative correlation between FA values in the left insula and the total LSAS scores and a
positive correlation between the ADC values in the left inferior frontal gyrus and the total LSAS scores. Above results suggested
that white-matter microstructural changes might contribute to the neuropathology of SAD.
alcohol abuse, is well established [5, 6].
white matter of the left insula, left inferior frontal gyrus, left middle temporal gyrus, and left inferior parietal gyrus and increased
ADC values in the left insula, bilateral inferior frontal gyrus, bilateral middle temporal gyrus, and left inferior parietal gyrus. In
Social anxiety disorder (SAD) is a marked and persistent fear
of social or performance situations in which the person is
exposed to unfamiliar people or to possible scrutiny by oth-
ers. The situations provoke intense anxiety symptoms in the
patients, which are often experienced as somatic symptoms,
and, as a result, the individuals may avoid social situations.
SAD can be particularly disabling in some patients, leading
to reduced likelihood of employment, social isolation, func-
tional disability, and dissatisfaction with life and health .
Data obtained by the National Comorbidity Survey indicate
that the adult lifetime prevalence of SAD is 13.3% . Recent
research also indicated that the 12-month prevalence rate of
SAD was 2.48%∼7.9% and that the lifetime prevalence was
however, an increasing amount of the neuroscience literature
is being devoted to social functioning. The previous func-
tional magnetic resonance imaging (fMRI) studies on SAD
found altered brain function in SAD within the medial pre-
frontalcortex and the limbicregionswhich formedthe corti-
colimbic circuits, including the amygdala, hippocampus, and
cuits, plays a critical role in learning about environmental
predictors of threat and in attention and facial emotions in
SAD . Its ability to control fear responses to threatening
Hindawi Publishing Corporation
BioMed Research International
Volume 2014, Article ID 860658, 8 pages
2BioMed Research International
stimuli was regulated by the hippocampus and medial pre-
frontal cortex . In addition, the insular cortex is a pivotal
structure in the greater limbic lobe and plays a role in diverse
functions linked to emotion and memory . However, a
better understanding of the neurobiology of SAD would
require investigations at the microstructure or anatomical
connectivity level, especially in patients not yet exposed to
psychotherapy or psychiatric medications.
oped MRI technique, allows for the examination of the inte-
grity of the white-matter microstructure and, thus, serves as
an important tool for mapping the anatomical connectivities
in humans . DTI measures the directionality and coher-
ence of water diffusion (as reflected by the degree of ani-
sotropy), which provides an estimate of the axonal organiza-
tion in the brain . The indices used to interpret DTI data
were fractional anisotropy (FA) and the apparent diffusion
coefficient (ADC). FA values reflect the directionality and
coherence of water self-diffusion. Tissues with highly regular
fibers have high anisotropy, whereas those with less regular
quantitative indicator of white-matter coherence or integrity,
with lower values signifying decreased structural connectiv-
ity in white matter. The ADC values reflect the degree of
apparent water diffusivity. Tissues without obstacles (such as
cerebrospinal fluid (CSF)) have high water diffusivity,
whereas those with obstacles (such as white matter) have low
white-matter tracts in SAD patients has been limited. There-
fore, the aim of the current study was to explore the differ-
ences in the white-matter connectivity in SAD patients and
healthy controls by using DTI. In addition, the current study
also explored the relation between the severity of SAD
symptoms and abnormalities in the white matter.
2. Methods and Materials
2.1. Subjects. The subjects were 18 adult patients with SAD
(according to the DSM-IV criteria) who were recruited from
were interviewed using the Structured Clinical Interview for
DSM-IV criteria Patient Version (SCID-I/P) , with addi-
tional probes from the self-administered Liebowitz social
and had LSAS scores ≥38 without psychiatric comorbidities
treatment. Eighteen healthy controls (HCs), matched for age,
area by poster advertisement and screened using the SCID-
I/P to rule out the presence of current diagnosis or past his-
tory of SAD/other axis I psychiatric disorders/axis II anti-
social or borderline personality disorders and had LSAS less
The exclusion criteria for SAD patients and HCs were as
follows: (1) any current or past serious medical or neurologi-
cal illness, including neurologic (Tourette’s syndrome, Hunt-
ington’s disease, Parkinson’s disease, encephalitis, stroke,
or other medical conditions. None of the patients had
received any pharmacological and/or psychotherapeutic
aneurysms, tumors, central nervous system infections, deg-
enerative brain diseases, or trauma), pulmonary, cardiac,
renal, hepatic, endocrine, or metabolic (including dehydra-
tion) disorders; prior psychosurgery or contraindications to
Interview for DSM-IV criteria); (4) a history of drug depen-
dence or abuse; (5) a history of psychiatric illness in first-
The study procedure and the involved risks were
explained to the subjects; all the subjects gave their written
informed consent according to the protocol approved by the
2.2. Image Acquisition. All MRI scans were performed using
a 3.0 Tesla GE Signa scanner with an eight-channel phased-
array head coil. A board-certified neuroradiologist reviewed
technique with 15 motion-probing gradient orientations. The
key data acquisition parameters for the DTI scan were as
number of diffusion gradient directions = 15; 푏 = 0 and
2.3. Image Processing. The FA and ADC maps were obtained
using DTI-Studio (Department of Radiology, Johns Hopkins
University School of Medicine, Baltimore, MD, USA; avail-
able at http://cmrm.med.jhmi.edu). Image analysis was per-
formed using SPM2 software (developed at the Wellcome
University College London), which was run on MATLAB7.0
(Mathworks, Sherborn, MA). Spatial normalization is an
essential preprocessing step in SPM-based analysis. The con-
T1-weighted and T2-weighted template images provided by
matrix = 128 × 128; field of view (FOV) = 24 × 24cm2; slice
thickness = 3mm; number of slices = 50; slice gap = 0mm;
1000s/mm2; total scan time = 6min and 47s.
were created using data from all the participants. Each 푏 = 0
and the mean images (푏 = 0 template, FA template, and ADC
istering each image with the customized FA and ADC
parameter was applied to the respective 푏 = 0, FA, and ADC
width at half-maximum (FWHM) isotropic Gaussian kernel,
template) were created. Then, all the FA and ADC maps in
native space were transformed into stereotactic space by reg-
2.4. Statistical Analysis. Differences between the demo-
graphic variables of the 2 groups were examined using in-
for each voxel of the FA and ADC values across the entire
brain. In these analyses, the statistical threshold was defined
BioMed Research International3
Table 1: Demographic characteristics (mean ± SD) of subjects.
14.11 ± 1.53
SAD (푁 = 18)
49.22 ± 40.17
HC (푁 = 18)
19.50 ± 8.50
푡 (df = 34)
SAD: social anxiety disorder; HC: healthy control; M: male; F: female.
Educational background was measured according to the years of education.
The severity of social anxiety was measured using the Liebowitz social anxiety scale.
Duration: the time from the beginning of the first episode to the time of assessment (measured in months).
22.72 ± 3.85
54.11 ± 11.90
21.78 ± 3.900.730.47
14.05 ± 2.04 0.927
Table 2: Differences between the FA and ADC values of the SAD patients and HCs.
Inferior frontal gyrus (L)
Inferior parietal gyrus (L)
Middle temporal gyrus (L)
Inferior frontal gyrus (L)
Inferior frontal gyrus (R)
Inferior parietal gyrus (L)
Middle temporal gyrus (L)
Middle temporal gyrus (R)
as a 푡 value above 2.44 (푃 < 0.01, uncorrected). We used
steps: firstly, ROIs were defined as regions in which the FA
and ADC values of the SAD patients were abnormal from
Talairach coordinates at the center of the cluster
MarsBar (http://marsbar.sourceforge.net) for extracting the
FA and ADC values in the ROIs (region of interest) by two
ROIs’ FA and ADC values and the score of LSAS in the SAD
patients. Correlations were calculated using Pearson’s corre-
values were extracted from FA or ADC map of each SAD
patient separately and investigated the correlations between
lation analysis, and a 푃 value less than 0.05 was considered
Age, sex, handedness, and education did not differ signifi-
cantly between the SAD patients and the HCs (Table 1). In
comparison to the HCs, the SAD patients had decreased FA
gyrus, left middle temporal gyrus, and left inferior parietal
gyrus (Figure 1; Table 2). The SAD patients also showed
increased ADC values in the left and right inferior frontal
etal gyrus, and left insula (Figure 2; Table 2). In addition, we
(푟 = −0.504, 푃 = 0.033) and a trend of negative correlation
between FA values and the left inferior frontal gyrus (푟 =
−0.414, 푃 = 0.087); there was a positive correlation between
the total LSAS scores of the SAD patients (푟 = 0.558, 푃 =
0.016) (Figure 3).
Few studies have examined the brain white matter in SAD
patients. The preliminary findings provided evidence of ab-
normal white-matter microstructure in SAD patients, as
study were as follows: (1) the SAD patients showed decreased
FA values in the left insula, left inferior frontal gyrus, left
middle temporal gyrus, and left inferior parietal gyrus and
showed increased ADC values in the left insula, left and right
inferior frontal gyrus, left and right middle temporal gyrus,
and left inferior parietal gyrus and a positive correlation
between the increased ADC values in the left inferior frontal
gyrus and the total LSAS scores.
These findings corroborate the findings of prior studies,
circuits of SAD patients [8–10]. In this study, significant
4BioMed Research International
Figure 1: Four clusters of significantly decreased FA values in the left insula cortex (a), left inferior frontal gyrus (b), left inferior parietal
gyrus (c), and left middle temporal gyrus (d) in the SAD patient group, in comparison to the corresponding values of the HCs. The clusters
are superimposed onto the images from a representative T1-weighted MRI study in Montreal Neurological Institute (MNI) space (Courtesy
of MRIcro, Chris Rorden).
white-matter abnormalities were observed on both sides of
SAD conducted using social anxiety imagery condition
noted following pharmacotherapy. A recent PET study
showed that the anticipatory anxiety in SAD subjects was
associated with decreased bilateral frontal activation .
Another recent SAD study revealed a significant positive
LSAS scores at the left frontal cortex . A review of the
plays a role in social cognition by integrating the afferents
from the posterior rostral MFC (involved in the monitoring
of action) and the orbitofrontal cortex (involved in the
monitoring of reward or punishment). The medial prefrontal
this study may indicate dysfunction in the cortical regions of
the SAD patients, which contribute directly to the etiology of
The insula is associated with strong emotional responses,
sensation , and might play a role in several anxiety dis-
orders . In this study, the FA and ADC values in the left
insula of the SAD patients were significantly lower and
higher, respectively, than the corresponding values in the
HCs. In addition, this result was also strengthened by the
independent finding of a correlation between the total LSAS
HCs, which paralleled the results of previous studies [10,
28]. Interestingly, there is evidence that brain activation in
response to threatening faces in SAD patients differs greatly
facial anger cues in SAD patients . The functional and
structural abnormalities in SAD patients may affect their
abnormalities were one of the neurobiological mechanisms.
In addition, a recent study found that if the anterior insula
participated in anticipatory processing , then abnormal-
ities in the white matter of the insula may be interpreted as
neural mechanism for symptom of anticipatory anxiety in
Our study reported abnormalities in the white matter of
jects with SAD. Many studies have revealed that, in the tem-
poral cortex of monkeys and humans, the temporoparietal
junction, which is located primarily in the superior temporal
sulcus (STS) region, is activated by movements of the eyes,
mouth, hands, and body, suggesting that this junction is
BioMed Research International5
Figure 2: Six clusters of significantly increased ADC values in the left insula (a), left and right inferior frontal gyrus (b), left inferior parietal
gyrus (c), and left and right middle temporal gyrus (d) of the SAD patient group, in comparison to the corresponding values of the HCs. The
clusters are superimposed on the images from a representative T1-weighted MRI study in MNI space (Courtesy of MRIcro, Chris Rorden).
and personality traits of others [32, 33], and, more generally,
the social evaluation of others . Face recognition and
analysis of facial expression form an important part of every-
day interactions among humans. Previous studies have sug-
expressions and personality traits, resulting in the fear of
social interaction [35, 36]. A significant behavioral effect
while processing socially relevant stimuli, face processing in
particular, has been shown in previous studies. Specifically,
behavioral studies have reported that SAD patients tend to
judge neutral faces negatively , remember critical faces
pattern of eye movements than that used by the HCs .
negative or wary attitude. Straube et al. found that in a com-
parison of angry versus neutral faces in an implicit task,
activationof the STS region in the SAD patientswas stronger
than that in the HCs . These results suggest a specific pat-
tern of activity in the different parts of the distributed neural
system for face perception in SAD patients. Abnormalities in
the above-mentioned neuroanatomical regions may supply
in SAD patients.
It is particularly interesting that all of significantly
decreased FA regions occurred in the left hemisphere includ-
ing the left insula cortex, left inferior frontal gyrus, left infe-
patient group. The possible lateralization of emotion was
gdala, and insular cortex in a previous meta-analysis of neu-
roimaging studies . Our results were also consistent with
recent functional neuroimaging studies showing an overall
lateralization of cortex and limbic system activations to the
left hemisphere, particularly for corticolimbic circuits .
Further studies specifically designed to examine whether the
lateralization of the function and structure exists and how it
works in SAD patients would be helpful to clarify the disease
patient group was interviewed using the Structured Clinical
Interview for the Diagnosis of Axis-I Disorders, but they
were not classified into general SAD or specific SAD groups.
ity of the 2 groups for other potentially confounding factors,
such as socioeconomic status, was not assessed. Finally, it
needs to be noted that this study used a low significance
thresholdforawhole-brain,voxel-wiseanalysis(푃 < 0.01un-
6BioMed Research International
50 60 708090
Total LSAS scores
Left insula FA
R = −0.50447
P = 0.032764
50 60 7080 90
Total LSAS scores
R = −0.41472
P = 0.087033
Left inf. frontal FA
Total LSAS scores
Left inf. frontal ADC
R = 0.55817
P = 0.01607
Figure 3: (a) A cluster in the left insula region with a significant negative correlation between the FA values and the total LSAS scores. (b) A
cluster in the left inferior frontal region with a trend of a negative correlation between the FA values and total LSAS scores. (c) A cluster in
the left inferior frontal region with a significant positive correlation between the ADC values and the total LSAS scores.
that we were still able to perform a comparison of 2 reason-
maximize sensitivity and to use a less strict threshold so as
to avoid overlooking significant findings. Future studies with
larger sample size may help to see whether there would be
abnormalities. Despite these limitations, we believe that our
findings will contribute to the growing literature on the
imaging studies of anxiety disorders.
This study showed several brain-lobe abnormalities in the
white matter of the SAD patient group. These findings were
parallel to those of previous studies that showed functional
abnormalities in these regions; the findings were also consis-
tent with the hypothesized role of these regions in the mod-
ulation of excessive limbic activity in anxiety disorders. The
left insula and the left inferior frontal gyrus in SAD patients
and the FA or ADC values, which may point to the defective
perception of self and others in these patients. Future studies
to the disruption of corticolimbic circuits in SAD patients.
Conflict of Interests
The authors declare that there is no conflict of interests
regarding the publication of this paper.
BioMed Research International7
Changjian Qiu and Chunyan Zhu have equal contribution to
This research was partially supported by the State Impor-
tant Basic Research Development Program (Program no.
2008CB517407), the National Natural Science Foundation of
China (Grant no. 81171488), and the National Key Technolo-
thank the MRI Laboratory of the West China Hospital for
processing the fMRI data.
 H.-U. Wittchen and E. Beloch, “The impact of social phobia on
quality of life,” International Clinical Psychopharmacology, vol.
11, supplement 3, pp. 15–23, 1996.
psychopathology of social anxiety disorder,” Biological Psychia-
try, vol. 51, no. 1, pp. 44–58, 2002.
 J. Sareen and M. Stein, “A review of the epidemiology and
approaches to the treatment of social anxiety disorder,” Drugs,
vol. 59, no. 3, pp. 497–509, 2000.
 R. G. Heimberg, “Assessment and diagnosis of social phobia in
the clinic and the community,” Psychological Medicine, vol. 33,
no. 4, pp. 583–588, 2003.
 W. D. Shields, “Effects of epilepsy surgery on psychiatric and
behavioral comorbidities in children and adolescents,” Epilepsy
and Behavior, vol. 5, supplement 3, pp. S18–S24, 2004.
 V. Cramer, S. Torgersen, and E. Kringlen, “Quality of life and
anxiety disorders: A Population Study,” Journal of Nervous and
Mental Disease, vol. 193, no. 3, pp. 196–202, 2005.
disorder, and specific phobia,” American Journal of Psychiatry,
vol. 164, no. 10, pp. 1476–1488, 2007.
 R. E. Cooney, L. Y. Atlas, J. Joormann, F. Eug` ene, and I. H.
Gotlib, “Amygdala activation in the processing of neutral faces
in social anxiety disorder: is neutral really neutral?” Psychiatry
Research, vol. 148, no. 1, pp. 55–59, 2006.
 K. L. Phan, D. A. Fitzgerald, P. J. Nathan, and M. E. Tancer,
severity of social anxiety in generalized social phobia,” Biologi-
cal Psychiatry, vol. 59, no. 5, pp. 424–429, 2006.
 T. Straube, I.-T. Kolassa, M. Glauer, H.-J. Mentzel, and W. H. R.
Miltner, “Effect of task conditions on brain responses to threat-
ening faces in social phobics: an event-related functional mag-
12, pp. 921–930, 2004.
 F. Schneider, U. Weiss, C. Kessler et al., “Subcortical correlates
of differential classical conditioning of aversive emotional reac-
tions in social phobia,” Biological Psychiatry, vol. 45, no. 7, pp.
 D. G. Amaral, “The primate amygdala and the neurobiology of
social behavior: implications for understanding social anxiety,”
Biological Psychiatry, vol. 51, no. 1, pp. 11–17, 2002.
 R. Yehuda and J. LeDoux, “Response variation following
trauma: a translational neuroscience approach to understand-
ing PTSD,” Neuron, vol. 56, no. 1, pp. 19–32, 2007.
 B. P. Shelley and M. R. Trimble, “The insular lobe of reil-its
anatamico-functional, behavioural and neuropsychiatric attri-
butes in humans—a review,” World Journal of Biological Psychi-
atry, vol. 5, no. 4, pp. 176–200, 2004.
 K. O. Lim and J. A. Helpern, “Neuropsychiatric applications of
DTI—a review,” NMR in Biomedicine, vol. 15, no. 7-8, pp. 587–
 S. Mori and J. Zhang, “Principles of diffusion tensor imaging
51, no. 5, pp. 527–539, 2006.
for patients), 1996.
ties of self-report and clinician-administered formats,” Psycho-
logical Medicine, vol. 31, no. 6, pp. 1025–1035, 2001.
ropsychopharmacology, vol. 31, no. 10, pp. 2243–2253, 2006.
 M. Tillfors, T. Furmark, I. Marteinsdottir, and M. Fredrikson,
“Cerebral blood flow during anticipation of public speaking in
pp. 1113–1119, 2002.
no. 5, pp. 1251–1256, 2008.
 D. M. Amodio and C. D. Frith, “Meeting of minds: the medial
frontal cortex and social cognition,” Nature Reviews Neurosci-
ence, vol. 7, no. 4, pp. 268–277, 2006.
 J. LeDoux, “Fear and the brain: where have we been, and where
are we going?” Biological Psychiatry, vol. 44, no. 12, pp. 1229–
 M. L. Phillips, A. W. Young, C. Senior et al., “A specific neural
substrate for perceiving facial expressions of disgust,” Nature,
vol. 389, no. 6650, pp. 495–498, 1997.
 B. Wicker, C. Keysers, J. Plailly, J.-P. Royet, V. Gallese, and G.
Rizzolatti, “Both of us disgusted in My insula: the common
pp. 655–664, 2003.
 H. D. Critchley, S. Wiens, P. Rotshtein, A.¨Ohman, and R. J.
Dolan, “Neural systems supporting interoceptive awareness,”
Nature Neuroscience, vol. 7, no. 2, pp. 189–195, 2004.
 M. P. Paulus and M. B. Stein, “An insular view of anxiety,” Bio-
logical Psychiatry, vol. 60, no. 4, pp. 383–387, 2006.
generalized social phobia: A Functional Magnetic Resonance
NeuroImage, vol. 45, no. 3, pp. 976–983, 2009.
 F. Castelli, F. Happ´ e, U. Frith, and C. Frith, “Movement and
mind: a functional imaging study of perception and interpreta-
tion of complex intentional movement patterns,” NeuroImage,
vol. 12, no. 3, pp. 314–325, 2000.
8BioMed Research International Download full-text
 A. D. Engell and J. V. Haxby, “Facial expression and gaze-dir-
ection in human superior temporal sulcus,” Neuropsychologia,
vol. 45, no. 14, pp. 3234–3241, 2007.
 J. S. Winston, R. N. A. Henson, M. R. Fine-Goulden, and R. J.
Dolan, “fMRI-adaptationreveals dissociableneuralrepresenta-
tions of identity and expression in face perception,” Journal of
Neurophysiology, vol. 92, no. 3, pp. 1830–1839, 2004.
11, pp. 1803–1814, 2007.
 T. Allison, A. Puce, and G. McCarthy, “Social perception from
vol. 4, no. 7, pp. 267–278, 2000.
 F. Saboonchi, L.-G. Lundh, and L.-G.¨Ost, “Perfectionism and
self-consciousness in social phobia and panic disorder with
 K. Horley, L. M. Williams, C. Gonsalvez, and E. Gordon, “Face
ial expressions in social phobia,” Psychiatry Research, vol. 127,
no. 1-2, pp. 43–53, 2004.
 E. C. Winton, D. M. Clark, and R. J. Edelmann, “Social anxiety,
fear of negative evaluation and the detection of negative emo-
tion in others,” Behaviour Research and Therapy, vol. 33, no. 2,
pp. 193–196, 1995.
 F. Saboonchi, L.-G. Lundh, and L.-G.¨Ost, “Perfectionism and
self-consciousness in social phobia and panic disorder with
 T. D. Wager, K. L. Phan, I. Liberzon, and S. F. Taylor, “Valence,
gender, and lateralization of functional brain anatomy in emo-
age, vol. 19, no. 3, pp. 513–531, 2003.
 A. Hahn, P. Stein, C. Windischberger et al., “Reduced resting-
state functional connectivity between amygdala and orbito-
frontal cortex in social anxiety disorder,” NeuroImage, vol. 56,
no. 3, pp. 881–889, 2011.