MBSR vs aerobic exercise in social anxiety: fMRI of emotion regulation of negative self-beliefs.

Abstract

Mindfulness-based stress reduction (MBSR) is thought to reduce emotional reactivity and enhance emotion regulation in patients with social anxiety disorder (SAD). The goal of this study was to examine the neural correlates of deploying attention to regulate responses to negative self-beliefs using functional magnetic resonance imaging. Participants were 56 patients with generalized SAD in a randomized controlled trial who were assigned to MBSR or a comparison aerobic exercise (AE) stress reduction program. Compared to AE, MBSR yielded greater (i) reductions in negative emotion when implementing regulation and (ii) increases in attention-related parietal cortical regions. Meditation practice was associated with decreases in negative emotion and social anxiety symptom severity, and increases in attention-related parietal cortex neural responses when implementing attention regulation of negative self-beliefs. Changes in attention regulation during MBSR may be an important psychological factor that helps to explain how mindfulness meditation training benefits patients with anxiety disorders.

Full text

Page 1

MBSR vs aerobic exercise in social anxiety:
fMRI of emotion regulation of negative self-beliefs
Philippe Goldin, Michal Ziv, Hooria Jazaieri, Kevin Hahn, and James J. Gross
Department of Psychology, Stanford University, Stanford, CA, 94305, USA
Mindfulness-based stress reduction (MBSR) is thought to reduce emotional reactivity and enhance emotion regulation in patients with social anxiety
disorder (SAD). The goal of this study was to examine the neural correlates of deploying attention to regulate responses to negative self-beliefs using
functional magnetic resonance imaging. Participants were 56 patients with generalized SAD in a randomized controlled trial who were assigned to
MBSR or a comparison aerobic exercise (AE) stress reduction program. Compared to AE, MBSR yielded greater (i) reductions in negative emotion when
implementing regulation and (ii) increases in attention-related parietal cortical regions. Meditation practice was associated with decreases in negative
emotion and social anxiety symptom severity, and increases in attention-related parietal cortex neural responses when implementing attention
regulation of negative self-beliefs. Changes in attention regulation during MBSR may be an important psychological factor that helps to explain how
mindfulness meditation training benefits patients with anxiety disorders.
Keywords: social anxiety; emotion regulation; mindfulness; fMRI; attention
INTRODUCTION
Mindfulness meditation is a form of mental training that has existed
for more than two and a half millennia in the East and has recently
become integrated into Western health care approaches (Chiesa and
Serretti,2011).Mindfulness-Based
Kabat-Zinn, 1990), the format of mindfulness training that has
received the most empirical investigation, has been shown to reduce
symptoms of stress, depression and anxiety (Hofmann et al., 2010).
MBSR has been applied as an alternative clinical intervention for a
variety of psychological [e.g. anxiety and depression disorders (Allen
et al., 2006; Carmody, 2009; Chiesa and Serretti, 2009)] and chronic
medical problems (Merkes, 2010).
Recently, researchers have begun to examine the mechanisms of
action underlying MBSR. One candidate mechanism is emotion regu-
lation, a key feature of mood and anxiety disorders and a popular
target of many psychosocial clinical interventions (Campbell-Sills
and Barlow, 2007). Recent studies have found that emotion regulation
may (i) be related to dispositional mindfulness (Modinos et al., 2010)
and (ii) enhanced with MBSR (Goldin and Gross, 2010). In the present
study, we used functional magnetic resonance imaging (fMRI) to
investigate changes in the neural bases of emotion regulation in a
randomized controlled trial (RCT) of MBSR and aerobic exercise
(AE) for adults diagnosed with generalized social anxiety disorder
(SAD).
StressReduction(MBSR;
Mechanisms underlying MBSR
MBSR is associated with decreases in the habitual tendency to emo-
tionally react to and ruminate about transitory thoughts and physical
sensations (Teasdale et al., 2000; Ramel et al., 2004), stress, depression
and anxiety symptoms (Segal et al., 2002; Evans et al., 2008; Chiesa and
Serretti, 2009) and distorted self-view (Goldin et al., 2009c). MBSR is
also associated with increases in behavioral self-regulation (Lykins and
Baer, 2009), volitional orienting of attention (Jha et al., 2007) and
emotion regulation (Goldin and Gross, 2010; Modinos et al., 2010).
There are a variety of emotion regulation strategies (Gross, 2007)
with distinct neural circuitry and temporal features (Goldin et al.,
2008). However, little is known about how MBSR influences emotion
regulation abilities (Chambers et al., 2009). Preliminary evidence sug-
gests that MBSR may influence the ability to exert cognitive control
over negative rumination (Ramel et al., 2004), self-referential processes
(Goldin et al., 2009c) and attention allocation and regulation (Slagter
et al., 2008). Importantly, MBSR may be related to greater ability to
disengage attention from aversive emotional stimuli (Lutz et al., 2008);
thereby reducing emotional reactivity and enhancing the ability to
implement emotion regulation strategies.
MBSR and SAD
One clinical context for examining the effects of MBSR on emotion
regulation is SAD. SAD is a common, chronic and insidious psychiatric
condition characterized by intense fear of evaluation and avoidance of
social or performance situations (Stein and Stein, 2008). SAD involves
a distorted attentional focus on internal cues (e.g. negative thoughts
and self-imagery) and external cues (e.g. other’s facial expressions)
during social situations (Schultz and Heimberg, 2008), which main-
tains social anxiety symptoms by interfering with natural habituation
processes that lead to corrective learning (Heimberg and Becker, 2002).
Studies of patients with SAD have reported abnormal attentional
processes consisting of early hypervigilance followed by attentional
avoidance (i.e. reduced visual processing) of social threat (Mueller
et al., 2008). Imaging studies with patients with SAD have begun to
identify specific neural patterns related to these abnormal attentional
processes. Findings include hyperactivity in brain regions related to
immediate emotional reactivity (amygdala, insula) (Stein et al.,
2007) and insufficient recruitment of brain networks implicated in
cognitive reappraisal [dorsolateral prefrontal cortex (dlPFC), dorsome-
dial prefrontal cortex (dmPFC), dorsal anterior cingulate cortex
(dACC)] and attention regulation [posterior cingulate (PC)/precu-
neus, inferior parietal lobe (IPL), supramarginal gyrus] when
modulating responses to social threat (Goldin et al., 2009b) and nega-
tive self-beliefs (Goldin et al., 2009a).
Although effective cognitive–behavioral and pharmacological treat-
ments have been developed for adults (Heimberg and Becker, 2002),
investigating alternative approaches to help 30–40% of the patients
who do not respond to traditional interventions is warranted
Received 12 August 2011; Accepted 30 April 2012
The authors of this manuscript do not have any direct or indirect conflicts of interest, financial or personal
relationships or affiliations to disclose. This work was supported by the National Center for Complimentary and
Alternative Medicine Institutes of Health awarded to James Gross [Grant AT003644].
Correspondence should be addressed to Philippe Goldin, Department of Psychology, Jordan Hall, Bldg. 420,
Stanford, CA 94305-2130, USA, E-mail: pgoldin@stanford.edu
doi:10.1093/scan/nss054 SCAN (2012) 1of 8
? TheAuthor (2012).PublishedbyOxfordUniversityPress.For Permissions,pleaseemail:journals.permissions@oup.com

Social Cognitive and Affective Neuroscience Advance Access published August 27, 2012
at Stanford University on December 5, 2012
http://scan.oxfordjournals.org/
Downloaded from

Page 2

(Stein and Stein, 2008). Preliminary evidence suggests that MBSR may
be helpful for patients with SAD. MBSR has been found to improve
mood, functionality and quality of life in patients with SAD (Koszycki
et al., 2007). MBSR-related changes in self-referential processing and
attention regulation have also been observed in patients with SAD
(Goldin et al., 2009c; Goldin and Gross, 2010). More specifically,
during attention regulation, increased recruitment of posterior (par-
ietal and occipital) cortical attention-related brain regions from pre- to
post-MBSR was related to decreased anxiety in patients with SAD
(Goldin and Gross, 2010). While these initial studies are promising,
little is known about how MBSR impacts emotion regulation, how it
influences the neural bases of emotion regulation and how it compares
in the context of an RCT to other alternative and effective interven-
tions that have been shown to reduce stress and anxiety such as AE.
The present study
The goal of the present study was to investigate the differential effects
of MBSR vs AE (an alternative stress reduction intervention) in the
context of an RCT on the neural bases of emotion regulation of nega-
tive self-beliefs in patients with generalized SAD. To do this, we ob-
tained behavioral and neural measures of emotion regulation during
an fMRI experiment in patients with generalized SAD at baseline and
again after being randomly assigned to MBSR or AE.
We chose AE as an alternative intervention for several reasons. Like
MBSR, AE has been shown to improve both physical and mental health
(Penedo and Dahn, 2005), especially symptoms of stress, depression
and anxiety (Petruzzello et al., 1991; Stich, 1998; Stro ¨hle, 2009) in
clinical samples of mixed anxiety (e.g. Merom et al., 2007), panic dis-
order (e.g. Broocks et al., 1998; Dratcu, 2001) and generalized anxiety
disorder (e.g. Steptoe et al., 1989; McEntee and Halgin, 1999). Similar
to MBSR, it is a low-cost intervention that is readily available in the
community. Furthermore, as implemented in this study, individual
and group practice for both MBSR and AE were easy to match and
control. In patients with SAD, the effects of AE on the neural bases of
emotion regulation have not been examined or compared to MBSR in
an RCT.
Compared to AE, we expected MBSR to be associated with greater
(i) reduction of negative emotion and (ii) recruitment of posterior
cortical attention-related brain regions during attention regulation.
In the MBSR group, we expected meditation practice to be associated
with decreased negative emotion and increased brain responses in
attention-related posteriorcortical
post-MBSR.
regions from baselineto
METHODS
Participants
Participants were patients seeking treatment for SAD and who met
DSM-IV (Association, 1994) criteria for primary generalized SAD
based on the Anxiety Disorders Interview
DSM-IV-Lifetime version (DiNardo et al., 1994). Of the 316 individ-
uals assessed for eligibility, 260 were excluded due to not meeting study
criteria (173), other reasons (66) or declined to participate (21). The
remaining 56 participants were randomly assigned to either MBSR
(n¼31) or AE (n¼25) (Figure 1). Groups did not differ in gender,
age, ethnicity or education (Table 1). Complete fMRI assessments at
both baseline and post-intervention were available for 23 MBSR and
19 AE participants.
Participants passed an MRI safety screen and were excluded if they
had current pharmacotherapy or psychotherapy, history of medical
disorders or head trauma, or current psychiatric disorders other than
SAD, generalized anxiety disorder, major depressive disorder, obsessive
compulsive disorder, agoraphobia without a history of panic attacks or
Schedule for the
specific phobia. Participants were excluded if they had previously
completed an MBSR course or if they had a regular meditation practice
or AE regime (defined for our purposes as three or more times per
week, for >2 months).
In MBSR participants, current Axis I comorbidity included 10 with
generalized anxiety disorder, 5 with major depressive disorder, 3 with
specific phobia, 2 with dysthymia, 2 with panic disorder, 1 with agora-
phobia and 1 with obsessive–compulsive disorder. Past Axis I
comorbidity included nine with major depressive disorder, three
with eating disorders, one with panic disorder and one with dysthymia.
Fifteen participants reported past, non-cognitive–behavioral psycho-
therapy and seven reported past pharmacotherapy.
In AE participants, current Axis I comorbidity included eight with
generalized anxiety disorder, six with major depressive disorder, three
with dysthymia, two with specific phobia, two with panic disorder and
two with agoraphobia, Past Axis I comorbidity included two with
major depressive disorder, one with substance abuse, one with eating
disorder, 1 with obsessive–compulsive disorder and one with panic
disorder. Nine participants reported past, non-cognitive behavioral
psychotherapy and five reported past pharmacotherapy.
Procedure
All participants were recruited through web-based community listings
and referrals from mental health providers. Participants were told on
the phone that the study involved random assignment to two stress
reduction programs. After passing a telephone screening, potential
participants were administered the ADIS-IV-L (Di Nardo et al.,
1994) by a clinical psychologist (P.G. or M.Z.). We enrolled patients
with a principal diagnosis of generalized SAD (operationalized as
moderate or higher severity (?4 on a scale of 0–8) on the
ADIS-IV-L (i) Clinician’s Severity Rating for SAD and (ii) social fear
for five or more distinct social or performance situations in the SAD
section. All participants provided informed consent in accordance with
Stanford University Human Subjects Committee rules.
Participants were randomly assigned to 8 weeks of either MBSR or
AE using Efron’s biased coin randomization procedure (Efron, 1971),
which entails higher probability of allocation to the under-represented
treatment group throughout the duration of the clinical trial.
Randomization was done only after all baseline assessments (including
fMRI) were completed.
Participants were administered self-report clinical and well-being
measures (Jazaieri et al., 2012) and fMRI scanning assessments
before and after MBSR and AE.
Interventions
Mindfulness-based stress reduction
We administered the standard MBSR program consisting of eight,
weekly 2.5h group classes, a 1-day meditation retreat and daily
home practice. Participants were trained in formal meditation (i.e.
breath-focus, body scan, open monitoring), informal practice and
Hatha yoga. Daily logs were collected each week to measure group
and individual meditation and yoga practice. Participants attended
MBSR courses offered by seven different teachers in eight health care
settings throughout the San Francisco Bay Area. MBSR instructors had
15.7 years (SD¼4.1 years, range¼10–20 years) of MBSR teaching
experience. MBSR was offered at no-cost.
Aerobic exercise
We provided at no cost 2-month gym memberships to participants
randomized to AE. To match the individual and group components of
MSBR, participants in the AE intervention were required to complete
2 of 8 SCAN (2012)P.Goldinetal.
at Stanford University on December 5, 2012
http://scan.oxfordjournals.org/
Downloaded from

Page 3

at least two individual AE sessions and one group AE session (other
than meditation or yoga) each week during the 8-week intervention.
Experimental task
A version of this task focused on cognitive reappraisal has been pub-
lished (Goldin et al., 2009a). The version used here implemented a
meta-cognitive attention regulation cued with a single word ‘observe’.
Thetaskconsisted of five
experimenter-composed neutral situation about washing a car that
was used to obtain baseline emotion ratings and fMRI BOLD signals
for reading neutral statements. Then, four participant-generated auto-
biographical social anxiety situations characterized by social anxiety,
humiliation and embarrassment were presented. For each situation,
participants composed a paragraph describing the events, thoughts
and feelings and five negative self-beliefs (NSBs) (e.g. I am incompe-
tent; others think I am not normal). Participants indicated their age at
the time of each situation and provided ratings, on a scales of 1 (not at
all) to 9 (very much), of the vividness of the memory, their experience
of shame at the time of the situation, as well as current shame,
disturbance, avoidance and frequency of talking about the situation.
situations.Thefirstwasan
Five situations were presented in a single run lasting 15min and 27s.
The sequence of instructions for the five situations was fixed: neutral,
react NSB, regulate NSB, react NSB and regulate NSB. Each situation
consisted of (i) an instruction to ‘react’ or ‘observe’ (6s), (ii) 16 sen-
tences (3s each) in white font against a black background describing
the situation, (iii) 10 NSBs (9s each) embedded in the unfolding story
in uppercase letters that flashed nine times (850ms on and 150ms off)
and (iv) a negative emotion rating after each NSB (3s) from 1¼not at
all to 3¼moderately to 5¼extremely negative (Figure 2). Negative
self-beliefs were flashed to maintain attention and appeared in white
font for neutral and react trials and green for reappraisal trials. The
task was programmed in Eprime (Psychology Software Tools, Inc) with
a button response pad positioned in the participant’s right hand inside
the magnet. Participants were trained prior to the baseline scan to react
to NSBs by reflecting on how the NSB represents something that is true
about the participant. Observing NSBs consisted of cultivating a pre-
sent moment, meta-cognitive attentional perspective that focused on
just noticing any thought, memory, emotion or sensation as it ap-
peared and dissolved of its own accord without grasping at or rejecting
any experience of body or mind.
Assessed for eligibility (n=316)
Excluded (n=260)
♦ Not meeting inclusion criteria (n=173)
♦ Other reasons (n=66)
♦ Declined to participate (n=21)
Analyzed (n=23)
♦ Excluded from analysis (n=0)
♦ Lost to follow-up (n=3)
Reasons: 1 too busy, 1 could not complete
assessments in allotted time, 1 not willing to
do fMRI again
Allocated to MBSR intervention (n=31)
♦ Received intervention (n=26)
♦ Did not receive allocated intervention (n=5)
Reasons: 1 attended 4 classes and was
uncomfortable with the class, 1 attended 2
classes and thought group format was too
intense, 1 attended 1 class and did not like the
content, 1 attended 0 classes and wanted to do
something else for treatment, 1 attended 0
classes because could no longer make class
time
♦ Lost to follow-up (n=4)
Reasons: 1 too busy, 1 could not complete
assessments in allotted time, 1 not willing to
complete, 1 no response
Allocated to AE intervention (n=25)
♦ Received intervention (n=23)
♦ Did not receive allocated intervention (n=2)
Reasons: 1 wanted to but too anxious to get to
gym, 1 could not find time to get to the gym as
it was too far away
Analyzed (n=19)
♦ Excluded from analysis (n=0)
Analysis
Follow-Up
Randomized (n=56)
•
31 MBSR
•
25 AE
Enrollment
Allocation
Fig. 1 Consolidated standards of reporting trials (CONSORTs) diagram for a RCT of MBSR vs AE.
MBSRvs AEin socialanxietySCAN (2012) 3 of 8
at Stanford University on December 5, 2012
http://scan.oxfordjournals.org/
Downloaded from

Page 4

Image acquisition
Imaging was performed on a General Electric 3-T Signa magnet with a
T2*-weighted gradient echo spiral-in/out pulse sequence (Glover and
Law, 2001) and a custom-built quadrature ‘dome’ elliptical bird cage
head-coil (GE Healthcare, Milwaukee, WI, USA). Head movement was
minimized using a bite-bar and foam padding. Across a single func-
tional run, 618 functional volumes were obtained from 22 sequential
axial slices (repetition time¼1500ms, echo time¼30ms, flip
angle¼608, field of view¼22cm, matrix¼64?64, single-shot, reso-
lution¼3.438mm2?4.5mm;
high-resolution anatomical scans were acquired using fast spin-echo
spoiled gradient recall (0.85942?1.5mm; field of view¼22cm,
frequency encoding¼256).
voxelvolume¼53.2mm3).3D
fMRI data preprocessing
Analysis of functional neuroimages (AFNI) software (Cox, 1996) was
used for preprocessing and statistical analysis. The functional run was
subjected to preprocessing steps to maximize signal-to-noise contrast.
This included an analysis of potential outliers, volume registration to a
base image, motion correction, 4mm3isotropic Gaussian spatial
smoothing, masking non-brain voxels, high-pass filtering (.011Hz),
linear detrending and conversion into BOLD signal percentage
change in each voxel. No volumes demonstrated motion in the x, y
or z directions in excess of ?0.8mm. There was no evidence of
stimulus-correlated motion, as assessed by correlations between
condition-specific reference functions and x, y, z motion correction
parameters (all P’s>0.48).
fMRI statistical analysis
Multiple-regression was implemented with AFNI 3dDeconvolve and
included baseline parameters to remove mean, linear and quadratic
trends and motion-related variance in the BOLD signal. Regressors
for the react NSBs, regulate NSBs and read neutral statements were
convolved with the gamma variate model (Cohen, 1997) of the hemo-
dynamic response function. Functional MRI BOLD signal intensity was
computed as percentage of signal change [(MR signal per voxel per
time point/mean MR signal in that voxel for the entire functional
run)?100].
Individual brain maps were converted to Talairach atlas space
(Talairach and Tournoux, 1988) and second-level group statistical
parametric maps were produced according to a random-effects
model. To investigate interaction effects, a 2 Group (MBSR, AE)?2
Time (pre, post) repeated-measures ANOVA (3dANOVA3 in AFNI)
was implemented for regulate vs react NSBs. Follow-up t-tests
examined within- and between-group effects.
To correct for multiple comparisons, we used 3dFWHMx to calcu-
late the intrinsic smoothness of the noise (i.e. residual in the BOLD
signal time series) for each subject. We then computed the group mean
FWHM (5.44151) and entered this value into 3dClustSim with a gray
matter mask generated by 3dAutomask. For a voxel-wise P¼0.004, 7.0
original size voxels (3.44?3.44?4.5mm) are required to achieve a
cluster-wise P-value of P¼0.05.
To examine the association of practice (meditation in MBSR and
exercise in AE) and whole-brain BOLD responses for the contrast of
regulate vs react NSB, we used AFNI 3dRegAna with the same thresh-
olding procedure noted above.
RESULTS
Preliminary analyses
Of the 31 participants, 5(16%) dropped from MSBR, and of the 25, 2
(8%) participants dropped during AE, a non-significant difference, ?2
(1, N¼56)¼0.84, P¼0.36. Between-group t-tests showed that the
average amount of practice time each week in the MBSR (M¼212,
SD¼104min) and AE (M¼196, SD¼86min) groups did not differ
[t(45)¼0.60, P>0.55].
Below, we report the differential effects of MBSR vs AE on negative
emotion ratings and neural responses when reacting to and regulating
negative self-beliefs. Within MBSR and AE groups, separately, we
report the association of practice (meditation or AE) with negative
emotion ratings during emotion regulation and with thresholded
whole-brain responses for the contrast of regulate vs react NSBs. The
effects of MBSR and AE on clinical and well-being measures in patients
with SAD is reported elsewhere (Jazaieri et al., 2012).
Emotional reactivity
Negative emotion
A 2 Group (MBSR, AE)?2 Time (baseline, post) repeated-measures
ANOVA for the negative emotion ratings yielded no interaction of
Instruction: REACT or OBSERVE
S1-3
NSB1
S4-5
NSB2
S6NSB3
S7-8
NSB4
S9-10
NSB5
S11-12
NSB6
S13
NSB7
S14
NSB8
S15
NSB9
S16
NSB10
S = Sentence = 3s each
NSB = 9s Negative Self-Belief +
3s Negative Emotion Rating
Fig. 2 Structure of one autobiographical social situation block with 10 trials embedded.
Table 1 Demographic characteristics for patients randomized to MBSR or AE
MBSR n¼31AE n¼25t-test or ?2
Females, n (%)
Age (M ?SD) (years)
Ethnicity, n (%)
Caucasian
Asian
Hispanic
Multiracial
Education (M?SD) (years)
Current axis-I comorbidity
Generalized anxiety disorder
Major depressive disorder
Dysthymia
Specific phobia
Panic disorder
Agoraphobia
Obsessive–compulsive disorder
Past axis-I comorbidity
Major depressive disorder
Dysthymia
Panic disorder
Obsessive–compulsive disorder
Substance abuse
Eating disorder
Past psychotherapy
Past pharmacotherapy
19 (61.3)
32.87?8.83
10 (40)
32.88?7.97
?2¼2.88
t¼0.42
?2¼0.43
13 (41.9)
14 (45.2)
3 (9.7)
1 (3.2)
16.40?2.00
10 (40)
11 (44)
1 (4)
3 (12)
16.84?2.64t¼0.34
?2¼0.88
10
5
2
3
2
1
1
8
6
3
2
2
2
0
?2¼1.49
9
1
1
0
0
3
15
7
2
0
0
1
1
1
9
5
?2¼1.84
?2¼1.56
Note. MBSR¼Mindfulness-Based Stress Reduction, AE¼aerobic exercise.
4 of 8 SCAN (2012)P.Goldinetal.
at Stanford University on December 5, 2012
http://scan.oxfordjournals.org/
Downloaded from

Page 5

group by time for negative emotion when reacting to NSBs
[F(2,37)¼1.24,
[F(2,37)¼24.48, P<0.001, partial ?2(?
group [F(2,37)¼0.80, P¼0.38] (Figure 3). To rule out story effects,
a paired t-test showed that there were no differential negative emotion
responses to NSBs in all 56 participants during baseline react situation
1 vs 2 [t(55)¼1.45, P¼0.15].
P>0.27].There wasamaineffectoftime
2
p) ¼0.40] and no effect of
Neural responses
A 2 Group (MBSR, AE)?2 Time (baseline, post) repeated-measures
ANOVA on BOLD responses for the contrast of react NSBs vs neutral
statements resulted in interactions of group by time in one region:
right ventrolateral PFC (Table 2). This effect was characterized by
baseline to post-intervention BOLD response decrease for MBSR [%
BOLD change¼?0.07 %; t(22)¼2.90, P¼0.008] and increase for AE
[% BOLD change¼0.06%; t(18)¼4.01, P¼0.001] groups. This sug-
gests that both groups had a similar pattern of change in brain re-
sponses with only one brain dissociation in the prefrontal cortex.
Emotion regulation
Negative emotion
To assess the effect of MBSR vs AE on emotion regulation of NSBs, we
conducted2 Group (MBSR,
repeated-measures ANOVA on percent reduction of negative emotion
when regulating NSBs. This analysis yielded an interaction of group by
time [F(2,37)¼4.99, P¼0.03, ?
[F(1,37)¼4.10, P¼0.05, ?
Follow-up t-tests revealed a significant increase in percent reduction of
negativeemotion frombaseline
[t(20)¼2.13, P<0.05, ?
(Figure 3). To rule out story effects, a paired t-test showed that there
were no differential negative emotion responses to NSBs in all 56
participants during baseline regulate situation 1 vs 2 [t(55)¼0.89,
P¼0.37].
Among MBSR participants, total meditation practice was associated
with greater percent reduction of negative emotion when implement-
ing emotion regulation at post-MBSR [r(22)¼0.48, P¼0.02]. Among
AE participants, there was no association of AE practice with percent
reduction of negative emotion during emotion regulation (P>0.69).
AE)?2 Time(baseline, post)
2
p¼0.12], a main effect of group
p¼0.10] and no effect of time (P>0.25).
2
to post-trainingfor MBSR
2
p¼0.19], but not AE [t(17)¼0.98, P>0.34]
Neural responses
A 2 Group (MBSR, AE)?2 Time (baseline, post) repeated-measures
ANOVA on whole-brain BOLD responses for the contrast of regulate
vs react NSBs yielded five significant interactions, including right su-
perior parietal lobule (SPL), two regions within the right IPL, left
lingual gyrus and cerebellar culmen (Table 2 and Figure 5).
Follow-up t-tests in these interaction brain regions revealed baseline
to post-MBSR BOLD response decreases in right posterior superior
temporal gyrus (STG), and bilateral lingual gyrus, and increases in
left IPL, right anterior IPL, right posterior IPL and right SPL. From
baseline to post-AE, paired t-tests showed BOLD response decreases in
right SPL and bilateral IPL and increases in right lingual gyrus.
Among MBSR participants, total amount of meditation practice was
associated with increased BOLD responses during regulate (vs react
NSBs) after MBSR training in MPFC, left IPL, left middle occipital
gyrus) (Table 3). There were no associations between meditation prac-
tice and react (vs regulate NSBs). Among AE participants, there were
no associations between amount of AE and BOLD responses for regu-
late vs react NSBs or react vs regulate NSBs (all P’s >0.10).
DISCUSSION
The goal of this study was to investigate how mindfulness meditation
and AE stress reduction training programs differentially influence be-
havioral and neural outcomes in the context of emotion regulation.
NEGATIVE EMOTION
1.0
1.5
2.0
2.5
3.0
3.5
4.0
MBSR
BASELINE
AE
BASELINE
MBSR
POST
AE
POST
NeutReact Regulate
Fig. 3 The effect of MBSR and AE on negative emotion ratings when reading neutral statements,
reacting and regulating negative self-beliefs. *P<0.05, **P<0.005. Error bars¼standard error of
the mean.
-35
-30
-25
-20
-15
-10
-5
0
*
% REDUCTION
NEGATIVE EMOTION
MBSR
BASELINE
AE
BASELINE
MBSR
POST
AE
POST
Fig. 4 Percent reduction in negative emotion during emotion regulation of negative self-beliefs at
baseline and post-MBSR and AE. Interaction of Group by Time: F(2,37)¼4.99, P¼0.03, ?
Error bars¼standard error of the mean. *P<0.05
2
p¼0.12.
Table 2 BOLD Responses for interaction of Group (MBSR, AE)?Time (pre, post) for react
negative self-beliefs vs read neutral statements and regulate vs react negative self-beliefs
Brain regions (Cluster #)BAxa
ya
za
Vol (mm3)F-value
React NSBs>read neutral
R ventrolateral prefrontal cortex
Read neutral>react NSBs
Regulate>react NSBs
Parietal lobes
R SPL 3
R IPL 2
R IPL 5
Occipital lobes
L lingual gyrus 4
Cerebellum
Culmen 1
React>regulate NSBs
10
none
2145552915.8
7
40
40
24
52
58
?72
?58
?41
53
50
46
426
479
372
8.2
10.5
13.7
18
?21
?68
?23729.3
0
?37
?9 7458.0
none
aTalairach coordinates for voxel with peak BOLD % signal change.
F?7.66,
els?3.438mm2?4.5mm), cluster-wise P<0.05.
BA¼Brodmann Area, BOLD¼blood oxygen dependent, L¼left, NSBs¼negative self-beliefs,
R¼right, Vol¼volume. The Cluster # after each brain region refers to Figure 5.
voxelP<0.004, minimumclustervolumethreshold
?372mm3
(7vox-
MBSRvs AEin socialanxiety SCAN (2012)5 of 8
at Stanford University on December 5, 2012
http://scan.oxfordjournals.org/
Downloaded from

Page 6

The target was present moment focused meta-cognitive attention regu-
lation in patients with generalized SAD when they interacted with their
own negative self-beliefs embedded in autobiographical social anxiety
related situations.
Emotional reactivity
MBSR and AE resulted in equivalent reduction of self-reported nega-
tive emotional reactivity to NSBs. This converges with a prior study
showing MBSR-related reduction in negative emotional reactivity to
NSBs in patients with SAD (Goldin and Gross, 2010). This is the first
study to demonstrate that AE reduces emotional reactivity to NSBs in
patients with SAD.
The contrast of react NSBs vs read neutral statements yielded an
interaction in the right ventrolateral PFC, characterized by baseline to
post-intervention BOLD signal decrease for MBSR and increase for AE.
Right ventrolateral PFC has been implicated in cognitive control
processes including cognitive reappraisal of emotion (Ochsner and
Gross, 2005), response inhibition (Aron et al., 2004) and expressive
suppression (Goldin et al., 2008). This pattern of brain activity changes
may be related to lesser automatic cognitive control of emotion
experience while reacting to NSBs in MBSR.
Emotion regulation
MBSR, but not AE, resulted in a significant reduction in negative
emotion to NSBs when implementing the attentional emotion regula-
tion strategy deployed in the experimental task. This converges with
prior findings of MBSR-related reduction in negative emotion when
implementing breath-focused attention regulation of NSBs in patients
with SAD (Goldin and Gross, 2010). The attention regulation strategy
used here is a meta-cognitive attentional focus referred to as ‘open
monitoring’ or ‘receptive awareness’ in various contemplative trad-
itions. The degree of reduction in negative emotional reactivity with
attentional emotion regulation at post-MBSR is similar to what has
been reported during cognitive reappraisal of NSBs in patients with
SAD (Goldin et al., 2009a). This suggests that distinct emotion regu-
lation strategies are effective in patients with SAD.
Practice dose effects were observed for MBSR, but not AE.
Meditation practice during MBSR was associated with lesser negative
emotion during regulation of NSBs. The set of mindfulness meditation
skills taught in MBSR may impact core features of SAD, namely, social
fear, behavioral avoidance and exaggerated emotional reactivity by
promoting a more flexible and adaptive attentional engagement with
core negative self-beliefs. This is likely a result of the extensive training
in recognizing habitual patterns of thinking and interpreting in con-
junction with flexible implementation of meta-cognitive attention
regulation. The absence of a practice effect in the AE group suggests
-0.18
-0.15
-0.12
-0.09
-0.06
-0.03
0
0.03
0.06
0.09
0.12
0.15
0.180 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0. . . . . . . . . . . . . . . .1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 18 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8
z=-9
MBSRAE
% SIGNAL CHANGE
12345
1
z=-3
4
z=45z=52
2
5
3
Fig. 5 BOLD responses for interaction of Group (MBSR, AE)?Time (Pre, Post) for regulate vs react negative self-beliefs contrast. Percent signal change¼Change in BOLD signal from pre- to post-intervention.
Interaction of Group by Time: F(2,40)?7.66, P¼0.004, Error bars¼standard error of the mean. 1¼Culmen, 2¼Right IPL, 3¼Right SPL, 4¼Left Lingual Gyrus, 5¼Right IPL.
Table 3 Association between total amount of meditation and whole-brain BOLD signal
for observe vs react NSBs at post-MBSR in 23 patients with SAD
Brain regionsBAx y za
Vol (mm3)t-valueR2
Observe>react
Frontal lobes
L medial prefrontal cortex
Parietal lobes
L IPL
Occipital lobes
L Middle occipital gyrus
React >observe–none
6
?10 ?10 507293.950.43
40
?52 ?37 3910004.890.53
19/31
?28 ?68 228914.10 0.44
aTalairach coordinates for voxel with peak t-value.
t-value threshold?3.14, voxel P<0.005, minimum cluster volume threshold?372mm3(7
voxels?4.5mm?3.438 mm2), cluster P<0.05. BA¼Brodmann area, x, y, z¼Talairach and
Tournoux coordinates, R2¼variance explained.
6 of 8SCAN (2012) P.Goldinetal.
at Stanford University on December 5, 2012
http://scan.oxfordjournals.org/
Downloaded from

Page 7

that improvement in clinical and well-being measures with AE in
patients with SAD (Jazaieri et al., 2012) is likely related to mechanisms
other than meta-cognitive attention regulation, such as reduction in
autonomic reactivity or anxiety sensitivity. However, this has yet to be
investigated in the context of SAD.
Brain responses from the contrast of regulate vs react NSBs provide
evidence for the role of posterior parietal attention-related brain
regions in elucidating how MBSR trains attention processes in patients
with SAD. The brain response interactions revealed a dissociation
characterized by MBSR increases and AE decreases in three posterior
parietal attention-related brain regions: right anterior IPL, right pos-
terior IPL and right SPL. These brain regions have been implicated in
attentional alerting to a stimulus (Fan et al., 2005). Enhanced recruit-
ment of these brain regions following MBSR may reflect greater atten-
tional engagement (rather than avoidance or distraction) with the
emotional probe. MBSR-related increased activation in similar parietal
cortex regions has been reported with breath-focused attention vs
distraction in patients with SAD (Goldin and Gross, 2010).
Of note, similar to the pattern with self-reported negative emotion,
therewerepracticedoseeffectsonbrainresponsesforMBSR,butnotAE,
specifically for the observe emotion regulation condition. Greater
amount of meditation practice was related to greater BOLD responses
for regulate vs react NSBs in brain regions implicated in cognitive
(MPFC) and visual attention (IPL, middle occipital gyrus). This pattern
of association suggests that MBSR enhancement of psychological flexi-
bility may depend on the amount of meditation practice and functional
integration of somatic–attention–cognitive control brain systems.
Implications
While MBSR has been shown to reduce symptoms of stress, anxiety
and depression, this RCT of MBSR vs AE provides evidence that
changes in attention regulation may be related to well-established
beneficial effects of mindfulness meditation. Specifically, MBSR, but
not AE, was associated with increased ability to downregulate negative
emotional reactivity to their own NSBs during the observe condition
during which participants were attempting to implement one type of
meta-cognitive attention regulation. Importantly, this indicates that
specific emotion regulation strategies with distinct neural architectures
can be effective in the same clinical sample. For example, increases in
cognitive reappraisal, which involves logical reasoning, linguistic
processing and modification of meaning, has been shown to mediate
reduction of SAD symptoms (Goldin et al., 2012). This study shows
that a meta-cognitive attention regulation strategy, which does not rely
on logical reasoning, linguistic processing and modification of mean-
ing, may also be effective in SAD patients.
Limitations
The current study is limited to inferences about a single emotion regu-
lation strategy, (present moment, meta-cognitive attention regulation)
for only one type of SAD-related emotional probe (NSBs). A direct
comparison of multiple emotion regulation strategies for different
anxiety-inducing probes will be important to consider in future
studies. This study used a fixed order of only four autobiographical
situations. Future studies may benefit from using a random sequence
of many more types of autobiographical life situations.
In the present study, the AE active comparison training was useful
for matching both individual and group practice components of
MBSR. However, given that MBSR provides multiple tools for learning
the skill of mindfulness, it may be helpful to add an active training
program that also teaches a new non-meditative mental skill, for ex-
ample, visualization as a tool for enhancing memory ability. Future
studies will also benefit from adding a waitlist control group.
While this study investigated brain changes before and after MBSR,
to address the issue of optimal treatment matching, future analyses
should determine if behavioral and neural signals during emotion
regulation at baseline can predict who will maximally benefit from
MBSR or AE (i.e. moderator analysis). To better understand how
MBSR and AE work, an analysis of underlying mechanisms might
investigate whether changes in BOLD responses from baseline to
post-intervention predict longer term clinical status (i.e. mediator
analysis). Specifically, future studies may consider investigating
whether and how changes in emotion regulation strategy during treat-
ment are related to clinical outcomes.
FUNDING
National Center for Complimentary and Alternative Medicine
Institutes of Health awarded to James Gross [Grant AT003644].
CONFLICT OF INTEREST
None declared.
REFERENCES
Allen, N.B., Chambers, R., Knight, W. (2006). Mindfulness-based psychotherapies: a review
of conceptual foundations, empirical evidence and practical considerations. Australian
& New Zealand Journal of Psychiatry, 40, 285–94.
Aron, A.R., Robbins, T.W., Poldrack, R.A. (2004). Inhibition and the right inferior frontal
cortex. Trends in Cognitive Science, 8, 170–7.
Association, A.P. (1994). Diagnostic and Statistical Manual of Mental Disorders. 4 edn..
Washington, DC: American Psychiatric Association.
Broocks, A., Bandelow, B., Pekrun, G., et al. (1998). Comparison of aerobic exercise,
clomipramine, and placebo in the treatment of panic disorder. American Journal of
Psychiatry, 155, 603–9.
Campbell-Sills, L., Barlow, D.H. (2007). Incorporating emotion regulation into conceptu-
alizations and treatments of anxiety and mood disorders. In: Gross, J.J., editor.
Handbook of Emotion Regulation. New York: Guilford, pp. 542–59.
Carmody, J. (2009). Evolving conceptions of mindfulness in clinical settings. Journal of
Cognitive Psychotherapy, 23, 270.
Chambers, R., Gullone, E., Allen, N.B. (2009). Mindful emotion regulation: an integrative
review. Clinical Psychology Review, 29, 56072.
Chiesa, A., Serretti, A. (2009). Mindfulness-based stress reduction for stress management in
healthy people: a review and meta-analysis. Journal of Alternative and Complementary
Medicine, 15, 593–600.
Chiesa, A., Serretti, A. (2011). Mindfulness based cognitive therapy for psychiatric
disorders: a systematic review and meta-analysis. Psychiatry Research, 187, 441–53.
Cohen, M.S. (1997). Parametric analysis of fMRI data using linear systems methods.
Neuroimage, 6, 93–103.
Cox, R.W. (1996). AFNI: software for analysis and visualization of functional magnetic
resonance neuroimages. Computers and Biomedical Research, 29, 162–73.
Di Nardo, P.A., Brown, T.A., Barlow, D.H. (1994). Anxiety Disorders Interview Schedule for
DSM-IV: Lifetime version (ADIS-IV-L). San Antonio, TX: Psychological Corporation.
DiNardo, P.A., Brown, T.A., Barlow, D.H. (1994). Anxiety Disorders Interview Schedule for
DSM-IV: Lifetime version (ADIS-IV-L). New York: Oxford University Press.
Dratcu, L. (2001). Physical exercise: an adjunctive treatment for panic disorder? European
Psychiatry, 16, 372–4.
Efron, B. (1971). Forcing a sequential experiment to be balanced. Biometrika, 58, 403.
Evans, S., Ferrando, S., Findler, M., Stowell, C., Smart, C., Haglin, D. (2008).
Mindfulness-based cognitive therapy for generalized anxiety disorder. Journal of
Anxiety Disorders, 22, 716–21.
Fan, J., McCandliss, B.D., Fossella, J., Flombaum, J.I., Posner, M.I. (2005). The activation of
attentional networks. Neuroimage, 26, 471–9.
Glover, G.H., Law, C.S. (2001). Spiral-in/out BOLD fMRI for increased SNR and reduced
susceptibility artifacts. Magnetic Resonance in Medicine, 46, 515–22.
Goldin, P., Ziv, M., Jazaieri, H., Werner, K., Kraemer, H., Heimberg, R.G., Gross, J.J.
(2012). Cognitive Reappraisal Self-Efficacy Mediates the Effects of Individual
Cognitive-Behavioral Therapy for Social Anxiety Disorder. Journal of Consulting and
Clinical Psychology, May 14, 2012, No Pagination Specified. doi:10.1037/a0028555.
Goldin, P.R., Gross, J.J. (2010). Effects of mindfulness-based stress reduction (MBSR) on
emotion regulation in social anxiety disorder. Emotion, 10, 83.
Goldin, P.R., Manber-Ball, T., Werner, K., Heimberg, R., Gross, J.J. (2009a). Neural mech-
anisms of cognitive reappraisal of negative self-beliefs in social anxiety disorder.
Biological Psychiatry, 66, 1091–9.
MBSRvs AEin socialanxietySCAN (2012)7 of 8
at Stanford University on December 5, 2012
http://scan.oxfordjournals.org/
Downloaded from

Page 8

Goldin, P.R., Manber, T., Hakimi, S., Canli, T., Gross, J.J. (2009b). Neural bases of social
anxiety disorder: emotional reactivity and cognitive regulation during social and
physical threat. Archives of General Psychiatry, 66, 170–80.
Goldin, P.R., McRae, K., Ramel, W., Gross, J.J. (2008). The neural bases of emotion
regulation: reappraisal and suppression of negative emotion. Biological Psychiatry, 63,
577–86.
Goldin, P.R., Ramel, W., Gross, J.J. (2009c). Mindfulness meditation training and
self-referential processing in social anxiety disorder: behavioral and neural effects.
Journal of Cognitive Psychotherapy, 23, 242–56.
Gross, J.J., editor. (2007). The Handbook of Emotion Regulation. New York, NY: Guilford
Press.
Heimberg, R.G., Becker, R.E. (2002). Cognitive-Behavioral Group therapy for Social Phobia:
Basic Mechanisms and Clinical Strategies. New York: Guilford Press.
Hofmann, S.G., Sawyer, A.T., Witt, A.A., Oh, D. (2010). The effect of mindfulness-based
therapy on anxiety and depression: a meta-analytic review. Journal of Consulting and
Clinical Psychology, 78, 169.
Jazaieri, H., Goldin, P., Werner, K., Ziv, M., Heimberg, R., Gross, J.J. (2012). A randomized
clinical trial of mindfulness-based stress reduction versus aerobic exercise for social
anxiety disorder. Journal of Clinical Psychology, 68, 715–731.
Jha, A.P., Krompinger, J., Baime, M.J. (2007). Mindfulness training modifies subsystems of
attention. Cognitive Affective and Behavioral Neuroscience, 7, 109–19.
Kabat-Zinn, J. (1990). Full Catastrophe Living: Using the Wisdom of Your Body and Mind to
Face Stress, Pain, and Illness. New York: Dell Publishing.
Koszycki, D., Benger, M., Shlik, J., Bradwejn, J. (2007). Randomized trial of a
meditation-based stress reduction program and cognitive behavior therapy in general-
ized social anxiety disorder. Behavior Research and Therapy, 45, 2518–26.
Lutz, A., Slagter, H.A., Dunne, J.D., Davidson, R.J. (2008). Attention regulation and
monitoring in meditation. Trends in Cognitive Sciences, 12, 163–9.
Lykins, E.L.B., Baer, R.A. (2009). Psychological functioning in a sample of long-term
practitioners of mindfulness meditation. Journal of Cognitive Psychotherapy, 23, 226.
McEntee, D.J., Halgin, R.P. (1999). Cognitive group therapy and aerobic exercise in the
treatment of anxiety. Journal of College Student Psychotherapy, 13, 39–58.
Merkes, M. (2010). Mindfulness-based stress reduction for people with chronic diseases.
Australian Journal of Primary Health, 16, 200–10.
Merom, D., Phongsavan, P., Wagner, R., et al. (2007). Promoting walking as an adjunct
intervention to group cognitive behavioral therapy for anxiety disorders: a pilot group
randomized trial. Journal of Anxiety Disorders, 22, 959–68.
Modinos, G., Ormel, J., Aleman, A. (2010). Individual differences in dispositional mind-
fulness and brain activity involved in reappraisal of emotion. Social Cognitive and
Affective Neuroscience, 5, 369–77.
Mueller, E.M., Hofmann, S.G., Santesso, D.L., Meuret, A.E., Bitran, S., Pizzagalli, D.A.
(2008). Electrophysiological evidence of attentional biases in social anxiety disorder.
Psychological Medicine, 39, 1141–1152.
Ochsner, K.N., Gross, J.J. (2005). The cognitive control of emotion. Trends in Cognitive
Science, 9, 242–9.
Penedo, F.J., Dahn, J.R. (2005). Exercise and well-being: a review of mental and physical
health benefits associated with physical activity. Current Opinion in Psychiatry, 18,
189–93.
Petruzzello, S.J., Landers, A.C., Hatfield, B.D., Kubitz, K.A., Salazar, W. (1991).
A meta-analysis on the anxiety-reducing effect of acute and chronic exercise: outcomes
and mechanisms. Sports Medicine, 11, 143–82.
Ramel, W., Goldin, P.R., Carmona, P.E., McQuaid, J.R. (2004). The effects of mindfulness
meditation on cognitive processes and affect in patients with past depression. Cognitive
Therapy and Research, 28, 433.
Schultz, L.T., Heimberg, R.G. (2008). Attentional focus in social anxiety disorder: potential
for interactive processes. Clinical Psychology Review, 28, 1206–21.
Segal, Z.V., Williams, J.M.G., Teasdale, J.D. (2002). Mindfulness Based Cognitive
Therapy for Depression: A New Approach to Preventing Relapse. New York:
Guildford Press.
Slagter, H.A., Lutz, A., Greischar, L.L., Nieuwenhuis, S., Davidson, R.J. (2008). Theta phase
synchrony and conscious target perception: impact of intensive mental training. Journal
of Cognitive Neuroscience, 21, 1536–49.
Stein, M.B., Simmons, A.N., Feinstein, J.S., Paulus, M.P. (2007). Increased amygdala and
insula activation during emotion processing in anxiety-prone subjects. American Journal
of Psychiatry, 164, 318–27.
Stein, M.B., Stein, D.J. (2008). Social anxiety disorder. Lancet, 371, 1115–25.
Steptoe, A., Edwards, S., Moses, J., Mathews, A. (1989). The effects of exercise training on
mood and perceived coping ability in anxious adults from the general population.
Journal of Psychosomatic Research, 33, 537–47.
Stich, F.A. (1998). A meta-analysis of physical exercise as a treatment for symptoms of
anxiety adults from the general population. Psychosomatic Research, 33, 537–47.
Stro ¨hle, A. (2009). Physical activity, exercise, depression and anxiety disorders. Journal of
Neural Transmission, 116, 777–84.
Talairach, J., Tournoux, P. (1988). Co-planar Stereotaxic Atlas of the Human Brain. New
York: Thieme.
Teasdale, J.T., Segal, Z.V., Williams, J.M.G., Ridgeway, V.A., Soulsby, J.M., Lau, M.A.
(2000). Prevention of relapse/recurrence in major depression by mindfulness-based
cognitive therapy. Journal of Consulting and Clinical Psychology, 68, 615–23.
8 of 8 SCAN (2012) P.Goldinetal.
at Stanford University on December 5, 2012
http://scan.oxfordjournals.org/
Downloaded from