Eur Arch Psychiatry Clin Neurosci

DOI 10.1007/s00406-016-0689-2

ORIGINAL PAPER

Neural correlates ofdistraction inborderline personality disorder

beforeand afterdialectical behavior therapy
DorinaWinter1 IngaNiedtfeld1 RuthSchmitt2 MartinBohus3,4
ChristianSchmahl1 SabineC.Herpertz2

Received: 10 September 2015 / Accepted: 13 March 2016

Springer-Verlag Berlin Heidelberg 2016

Abstract Neural underpinnings of emotion dysregula-
tion in borderline personality disorder (BPD) are charac-
terized by limbic hyperactivity and disturbed prefrontal
activity. It is unknown whether neural correlates of emo-
tion regulation change after a psychotherapy which has the
goal to improve emotion dysregulation in BPD, such as
dialectical behavioral therapy (DBT). We investigated dis-
traction as a main emotion regulation strategy before and
after DBT in female patients with BPD. Thirty-one BPD
patients were instructed to either passively view or memo-
rize letters before being confronted with negative or neu-
tral pictures in a distraction task during functional magnetic
resonance imaging. This paradigm was applied before and
after a 12-week residential DBT-based treatment program.
Christian Schmahl and Sabine C. Herpertz have contributed
equally to this work.
Electronic supplementary material The online version of this
article (doi:10.1007/s00406-016-0689-2) contains supplementary
material, which is available to authorized users.
* Dorina Winter
dorina.winter@zimannheim.de
1
Department ofPsychosomatic Medicine andPsychotherapy,
Central Institute ofMental Health, Medical Faculty
Mannheim/Heidelberg University, J5, 68159Mannheim,
Germany
2
Department ofGeneral Psychiatry, Medical Faculty
Heidelberg/Heidelberg University, Vostr. 2,
69115Heidelberg, Germany
3 the processing of aversive information in BPD is reflected
Institute ofPsychiatric andPsychosomatic Psychotherapy,
Central Institute ofMental Health, Medical Faculty
Mannheim/Heidelberg University, J5, 68159Mannheim,
Germany
4
Faculty ofHealth, University ofAntwerp, Antwerp, Belgium
We compared the DBT group to 15 BPD control patients,
who continued their usual, non-DBT-based treatment or
did not have any treatment, and 22 healthy participants.
Behaviorally, BPD groups and healthy participants did not
differ significantly with respect to alterations over time. On
the neural level, BPD patients who received DBT-based
treatment showed an activity decrease in the right inferior
parietal lobe/supramarginal gyrus during distraction from
negative rather than neutral stimuli when compared to both
control groups. This decrease was correlated with improve-
ment in self-reported borderline symptom severity. DBT
responders exhibited decreased right perigenual anterior
cingulate activity when viewing negative (rather than neu-
tral) pictures. In conclusion, our findings reveal changes in
neural activity associated with distraction during emotion
processing after DBT in patients with BPD. These changes
point to lower emotional susceptibility during distraction
after BPD symptom improvement.
Keywords Borderline personality disorder Emotion
regulation Dialectical behavior therapy Functional
magnetic resonance imaging Inhibition
Introduction
Emotion dysregulation is a core feature of borderline per-
sonality disorder [BPD; 1, 2]. BPD patients are particularly
sensitive and react especially emotionally to negative stim-
uli [35]. On the neural level, emotion dysregulation during
by hyper-reactivity of the amygdala and insula [6]areas
involved in emotion processing [7]. Also, BPD patients dis-
play lower activity in areas involved in cognitive emotion
regulation [8], such as the dorsolateral, medial, and orbital

13

prefrontal cortex and the anterior cingulate cortex [ACC;
911], which may be interpreted as difficulty to regulate
aversive responses via inhibitory pathways.
Neural correlates of emotion dysregulation in BPD
might change with therapeutically induced improvement
of emotion regulation. First studies suggest that amygdala
activity in response to passively viewing aversive stimuli
declines after BPD symptom improvement by psycho-
therapy [12, 13]. Yet they did not include a patient control
group nor examined explicit emotion regulation strategies,
such as distraction. In order to study alterations in neural
correlates of explicit forms of emotion regulation after BPD
symptom improvement, we aimed to improve BPD symp-
toms via dialectical behavior therapy [DBT; 1416]. DBT
is the currently best established psychosocial treatment for
BPD [17, 18] and has a strong focus on training emotion
regulation skills. Among others, distraction is taught as a
self-administered cognitive intervention to attenuate exag-
gerated levels of aversive emotions [14].
Distraction as an emotion regulation strategy has proven
to be very effective and comparatively fast [19, 20] in
reducing negative affect [2023], particularly when stimuli
are highly arousing [24]. On the neural level, healthy indi-
viduals performing a distraction task show higher activity
in a widespread fronto-parietal network including lateral
and medial prefrontal areas, the inferior parietal cortex,
and the ACC, as well as lower activity in emotion process-
ing areas such as the amygdala [21, 22, 2527]. In com-
parison with healthy controls, BPDpatients showed previ-
ously higher activity in the amygdala [28, 30] and insula
[28] during distraction from negative stimuli, indicating
stronger emotional responding. Individuals with BPD also
did not show the temporal and frontal response that healthy
controls did to distracting negative compared with neutral
stimuli [31] but showed higher inferior parietal lobe activ-
ity [30]. In the subgenual and perigenual parts of the ACC
and adjacent brain areas, the difference between negative
and neutral distracting stimuli was smaller in individuals
with BPD than in the healthy control group [10, 29, 31],
while evidence for differential activation of the dorsal ACC
was heterogeneous [10, 31, 34]. These findings point to
disturbed fronto-parietal, inhibitory functioning, which is
associated with enhanced insular and limbic, emotional
response during distraction in BPD patients [32, 33].
Based on these facts, our aim was to examine whether
neural correlates of distraction change after symptom
improvement by psychotherapy in a non-randomized con-
trolled trial. We acquired fMRI of female BPD patients
before and after 12weeks of residential DBT-based treat-
ment (BPD+DBT). As control groups, we included
BPD participants without DBT-based treatment (BPD-C),
and healthy controls (HC) at the beginning and end of a
12-week interval. We expected that BPD+DBT patients
13
Eur Arch Psychiatry Clin Neurosci
would display an increased use of a fronto-parietal emo-
tion regulation network and reduced limbic hyper-reactivity
during a distraction task after treatment.
Methods
Sample
We consecutively recruited BPD+DBT patients seeking
treatment at residential DBT programs in Mannheim and
Heidelberg. BPD-C and HC were recruited through adver-
tisement. Exclusion criteria for all participants were left-
handedness, traumatic brain injury, lifetime schizophrenia
or bipolar I disorder, mental or developmental disorders,
substance dependence during the last year, drug consump-
tion in the last 2months, current diagnosis of a severe
depressive episode, and benzodiazepine use. We included
only unmedicated patients and patients with the same
medication at each fMRI measurement. We excluded BPD
patients who had significant DBT skills training experience
prior to the study. Included BPD patients met DSM-IV
diagnosis for BPD, including affective instability and self-
injurious behavior. HC had no current or lifetime psychiat-
ric disorder and no psychotropic medication.
Trained clinical psychologists assessed BPD diagno-
ses using the International Personality Disorder Examina-
tion [IPDE; 35] and Axis I disorders using the Structured
Clinical Interview for DSM-IV [SCID-I; 36]. Accom-
panying each fMRI measurement, BPD symptom sever-
ity was assessed using an established clinical interview
(Zanarini Rating Scale for borderline personality disorder
[ZAN-BPD; 37]) and an established self-report measure
(Borderline Symptom List [BSL; 38]). Emotion regula-
tion difficulties were assessed by the Difficulties in Emo-
tion Regulation Scale [DERS; 39]. Additionally, we asked
the participants whether they used skills within the last
threedays and whether they considered this skill use as
successful (see Supplement for a detailed description).
The final sample comprised 31 BPD+DBT, 15 BPD-
C, and 22 HC. Table1 summarizes their demographic data
and Table2 their clinical characteristics. BPD groups and
HC did not differ significantly in age and education. The
number of medicated participants did not differ between
the BPD groups (BPD+DBT 67.7%, BPD-C 46.7%;
Fishers exact test: p =0.208). BPD groups did not dif-
fer significantly in the proportion of each medication type
per group, but revealed heterogeneities for some types
(Fishers exact tests): Of all BPD+DBT/BPD-C patients,
41.9/40.0% received selective serotonin reuptake inhibitors
(p =1.000), 22.6/0.0% serotoninnorepinephrine reup-
take inhibitors (p =0.078), 22.6/6.7% other antidepres-
sants (tricyclic, tetracyclic, or norepinephrinedopamine
Eur Arch Psychiatry Clin Neurosci

Table1Demographic variables and comorbid disorders in patients with borderline personality disorder undergoing dialectical behavior therapy
(BPD+DBT), patients with borderline personality disorder without dialectical behavior therapy (BPD-C), and healthy control participants (HC)
BPD+DBT BPD-C HC Statistics
(n=31) (n=15) (n=22)
M SD () M SD () M SD () p

Age (years) 28.17 7.47 25.73 5.90 27.95 8.34 F(2)=0.43 0.650
Years of education [n (%)] 22 = 9.57 0.284
<9years 1 3.2 0 0
9years 7 22.6 0 1 4.5
10years 11 35.5 6 40.0 12 54.5
13years 12 38.7 9 60.0 9 40.9
Number of Axis I comorbidities (mean) 1.68 1.12 1.93 0.96 0 0 t(44)=0.77 0.448

reuptake inhibitor; p=0.243), 22.6/0% neuroleptic medi- Distraction task

cation (p =0.078), 9.6/0% mood stabilizer/anticonvul-
sants (p=0.541), and 9.6/0% other medication (cyclopyr-
rolone, dopamine agonist, or opioid receptor antagonist;
p=0.541). Use of multiple types medication was allowed
and the number of medication types did not differ signifi-
cantly between both BPD groups (BPD+DBT M=1.29,
BPD-C M=0.47; 52 = 5.77; p=0.329).
The Supplement provides further information on recruit-
ment, patient flow, and comorbidities. The study was part
of a project on emotion regulation in BPD after DBT.
Dialectical behavior therapy
BPD+DBT patients participated in a well-established and
evaluated 12-week standard residential DBT-based treat-
ment [16, 4042]. The program comprised weekly skills
training groups (emotion regulation skills, mindfulness,
self-esteem, social competence) and individual treatment
twice a week. Treatment was conducted at two residential
units specialized in DBT at the two medical faculties Man-
nheim and Heidelberg of Heidelberg University, Germany.
Therapists were experienced Ph.D., M.D., and M.Sc. level
clinicians (psychologists and physicians) and were super-
vised regularly.
BPD control group
BPD-C patients continued the non-DBT treatment they
received, if any: 40% reported various forms of outpatient
psychotherapy, 6.6% reported residential crisis interven-
tion, 46.7% reported pharmacotherapy, 6.6% visited a
self-help group, and 33.3% did not report any treatment
during study participation. No BPD-C patient was waiting
for the DBT-based treatment.
In each trial, participants were instructed to either look at
(view) or memorize five consonants (e.g., XZKVT)
presented for 2000ms, followed by a negative or neutral
photographical picture (6000ms). We chose pictures from
standardized picture sets (International Affective Picture
System [43] and the Emotional Picture Set [44]). Negative
pictures were low in valence and high in arousal. Neutral
pictures had intermediate valence and low arousal.
After picture presentation, in the MEMORIZE condi-
tion, a probe letter was presented. The participants needed
to indicate as fast as possible via button press whether the
letter had been presented in the letter string before the
picture or not. In the VIEW condition, an O instructed
the participant to press a given button as fast as possi-
ble. In both conditions, the letter presentation ended with
the button press or after 2000ms, followed by a jittered
inter-trial interval of 30008000ms (mean 5500ms).
Each condition (MEMORIZE-NEGATIVE, MEMORIZE-
NEUTRAL, VIEW-NEGATIVE, VIEW-NEUTRAL)
comprised 18 trials. The order of conditions was pseudo-
randomized and fixed at each time point for each partic-
ipant in order to control for mood carryover effects and
to increase comparability between subjects. Participants
received 20 training trials to assure they were able to
complete the task.
The paradigm was repeated at each of the two measure-
ment time points (t1: first measurement/pre-treatment; t2:
second measurement 12th week after t1/post-treatment),
with different pictures matched in valence and arousal
across regulation conditions and sessions. Stimuli were
presented via digital video goggles (Resonance Technolo-
gies, Northridge, CA) using Presentation (nbs.neuro-bs.
com).

13
 Eur Arch Psychiatry Clin Neurosci

Psychometric andbehavioral data analysis standard brain of the Montreal Neurological Institute
(MNI) space, and smoothing with a Gaussian kernel with a
To examine group differences in psychometric measures full width at half maximum of 9mm.
at t1 and in symptom alterations over time (GroupTime On single-subject level, we applied a general linear
interaction), we performed 32 analyses of vari- model (GLM) per time point using separate regressors for
ance (ANOVA) with the between-subject factor Group each experimental condition (MEMORIZE-NEGATIVE,
(BPD +DBT, BPD-C, HC) and the within-subject factor MEMORIZE-NEUTRAL, VIEW-NEGATIVE, VIEW-
Time (t1 vs. t2) samples for ZAN-BPD, BSL, and DERS NEUTRAL; starting with the instruction, ending with the
scores. picture presentation) as well as the two probe types (MEM-
Error rates were excluded from further behavioral anal- ORIZE or VIEW condition) and the six motion parameters
ysis, because participants had very low error rates, which as regressors of no interest. To correct for global signal
would limit interpretability of respective results. Reac- intensity variation and low-frequency fluctuations, a high-
tion time (RT) data of correct responses were log-trans- pass filter of 128-s cutoff was applied. In order to avoid a
formed due to left-skewed distribution of the raw data. four-factorial design at group level, valence conditions
Group differences in mean RTs at t1 were analyzed by a were contrasted at first level (NEGATIVE-NEUTRAL)
322-ANOVA with the between-subject factor Group for each regulation condition [MEMORIZE(NEGATIVE-
and the within-subject factors Valence (negative, neutral) NEUTRAL), VIEW(NEGATIVE-NEUTRAL)].
and Task (memorize, view). To test for RT alterations over For group level analyses, we used a full factorial
time, we calculated differences for negative versus neutral model with the independent factor Group (BPD+DBT,
pictures for each condition to avoid a four-factorial design BPD-C, HC) and the repeated measurement factors Time
and used a 322 ANOVA with the between-subject (t1, t2) and Task (memorize, view). We calculated the
factor Group and the within-subject factors Task and Time Group TimeTask interaction (F-contrast) in this
point (t1, t2). model. To identify areas exhibiting neural alterations over
Post hoc comparisons were performed using t tests time during distraction (MEMORIZE-VIEW) based on
(Bonferroni-corrected for multiple comparisons) if applica- DBT, we compared distraction before and after therapy
ble. All analyses were performed at a threshold of p<0.05 (TimeTask interaction) within the group of BPD+DBT
with SPSS Statistics 20 (IBM, USA). using t-tests for t1(MEMORIZE-VIEW) t2(MEMORIZE-VIEW) as
well as t2(MEMORIZE-VIEW) t1(MEMORIZE-VIEW). To ensure
MRI data acquisition that the identified regions were specific for BPD+DBT
and significantly weaker in the control groups, we
MRI scanning was conducted on a 3Tesla Siemens masked the resulting statistical maps with the difference
TRIO-Scanner (Siemens, Erlangen) with a 32-channel BPD+DBTBPD-C or BPD+DBT-HC of the respec-
head coil. Blood oxygen level-dependent (BOLD) sig- tive contrasts [45, account also used by, e.g., 26]. Fol-
nal was measured with 36 3-mm transversal slices cov- lowing Lieberman and Cunningham [46], all whole-brain
ering the entire brain using gradient-echo echo-planar second level analyses were performed with a threshold
imaging (EPI, T2-weighted contrast, FOV=192mm, of p<0.001, k>10 voxel, uncorrected for multiple com-
voxel size 333mm3, 6464 voxel matrix, flip parison at voxel level. For post hoc tests, we extracted
angle=80, TE=30ms, TR=2000ms). We collected first-level beta values from the first-level contrasts (NEG-
a high-resolution anatomical scan that served as an indi- ATIVE-NEUTRAL) and tested for significant differences
vidual template for functional data using three-dimensional over time for the view or memorize condition per group
magnetization prepared rapid acquisition gradient echo using SPSS.
(MPRAGE, T1-weighted contrast, FOV=256mm, voxel
size 111mm3, flip angle=15, TE=2.75ms, Responder analysis
TR=1570ms).
To examine particular differences between the patients who
fMRI data analysis responded to the DBT-based treatment and those who did
not respond to it, we used the Reliable Change Index [RCI;
FMRI data were analyzed with SPM8 (http://www.fil.ion. 47] with the criterion1.96 in the ZAN-BPD to classify
ucl.ac.uk/spm/). The first five volumes were discarded to responders. This resulted in 16 DBT responders (ZAN-
minimize T1 effects. EPI time series were preprocessed BPD = 12.00 5.72 SD) and 15 DBT non-responders
using slice time correction and spatial realignment to cor- (ZAN-BPD = 1.13 2.97 SD), with DBT respond-
rect for head motion, co-registration onto participants ers showing also a larger decrease in BSL (t(29) =2.58,
segmented high-resolution T1 scan, normalization to the p=0.015) but not DERS scores (t(29)=1.02, p=0.316)

13
 Eur Arch Psychiatry Clin Neurosci

when being compared to DBT non-responders. This latter

Table2Psychometric alterations in patients with borderline personality disorder undergoing dialectical behavior therapy (BPD+DBT), patients with borderline personality disorder without

10.28 <0.001 (),

0.026 (),
Tukey

14.98 <0.001 ,
Group differences in

Post
is in line with the result that DBT responders did not differ

hoc
change over time significantly in their increase in skills use or skills use suc-
cessfulness (p>0.288; data in Supplement). Further demo-

p
graphical and clinical characteristics are summarized in the

3.86
F Supplement. Above-mentioned contrasts were calculated
for DBT responders compared with DBT non-responders.
Tukey
Group differences at t1

79.54 <0.001 ,
75.65 <0.001 ,

96.64 <0.001 ,
Post
hoc

Exploratory correlation withalterations insymptom

severity
p

To explore the relation of the neuronal changes over time

with changes in symptom severity in BPD+DBT, we per-

BSL Borderline Symptom List, DERS Difficulties in Emotion Regulation Scale, ZAN-BPD Zanarini Rating Scale for borderline personality disorder
F

formed correlational analyses between the RCI of ZAN-

t1t2

0.358
0.657

0.183

BPD, BSL, and DERS scores and the main results of the
above-mentioned fMRI results. For this, peak voxel activ-
SD () p

ity differences were extracted from single-subject level

0.67
0.13

59.75 10.91

(contrast NEGATIVE-NEUTRAL) and correlated with the

increase (t2t1) or decrease (t1t2) over time.
0.41
0.18

Post hoc tests at a significance level of p<0.05: BPD+DBT>BPD+TAU, BPD+DBT>HC, BPD+TAU>HC

SD () M
t2

Results
0.69
0.12

12.03
(n=22)

Symptom severity
0.19a

0.23

61.64
HC

M
t1t2 t1

Table 2 summarizes psychometric data and analy-

0.803

0.068

0.830

ses, which are shown in Fig.1. There were significant

Group Time interactions in all psychometric meas-
SD () p

ures (ZAN-BPD: F(2,65) =10.28, p<0.001, 2 =0.24;

0.87

6.30

22.17 131.20 25.28

BSL: F(2,64) =3.86, p =0.026, 2 =0.11; DERS:

F(2,64) =14.98, p<0.001, 2 =0.32): BPD+DBT
2.19

13.07

showed a decrease in all psychometric measures between

SD () M
t2

t1 and t2, but there was no significant change over time in

dialectical behavior therapy (BPD-C), and healthy control participants (HC)

0.86

5.96

the BPD-C and HC groups. Comparing alterations over

time between groups, BPD+DBT displayed a larger
(n=15)

decrease in all psychometric measures than HC (prepost

<0.001 134.14a
BPD-C

2.22

15.93

between group effect sizes: BSL: d =0.77, ZAN-BPD:

M
t1

d =1.51, DERS: d =1.39). Compared with BPD-C, the

0.006

<0.001

BPD+DBT group showed significantly stronger decrease

t1t2

in DERS scores after therapy, as well as a trend (p<0.10)

SD () p

for BSL and ZAN-BPD. Effect sizes were in the usual

0.80

109.54 24.02
6.27

range for psychotherapies for BPD compared with control

groups (prepost between group effect size: BSL: d=0.57,
1.71

10.06

ZAN-BPD: d =0.58, DERS: d =1.20 [18]). BPD-C did

not differ from HC in symptom alterations over time (all
SD () M
t2

p>0.234).
0.65

135.33 23.72
6.09

There was also a main effect of Group in each of

BPD+DBT

Missing data: n1

the measures (ZAN-BPD: F(2,65) =74.36, p<0.001,

() Statistical trend
(n=31)

2 =0.70; BSL: F(2,64) =70.62, p<0.001, 2 =0.69;

2.10

16.81

DERS: F(2,64) =94.55, p<0.001, 2 =0.75):

M
t1

BPD +DBT and BPD-C did not differ significantly in

mean

BPD
DERS

ZAN-BPD (p=1.000), BSL (p=0.338), and DERS scores

ZAN-
BSL-

(p =0.181) and both BPD groups scored significantly

a

13

Fig.1Psychometric data and alterations over time. The graphs
depict changes in psychometric measures over time in patients with
borderline personality disorder (BPD) before and after dialectical
behavior therapy (BPD+DBT) as well as patients with BPD with-
out dialectical behavior therapy (BPD-C) and healthy control partici-
pants at the same time points. The graphs present data from (a) and
higher than HC in these measures (all p<0.001). Also,
there were main effects of Time in all of the measures
(t1>t2; ZAN-BPD: F(1,65)=20.14, p<0.001, 2=0.24;
BSL: F(1,64) =4.43, p =0.043, 2 =0.06; DERS:
F(1,64) =16.35, p<0.001, 2 =0.20). Additionally, the
BPD+DBT group showed an overall stronger increase in
skill use (F2,57 =20.19, p<0.001, 2 =0.415) and skill
use success (F2,56=25.13, p<0.001, 2=0.473; detailed
data in Supplement).
Behavioral data
Supplementary Table S6 presents the behavioral data.
At t1, there were neither significant group effects nor
Group Condition interactions, while experimental con-
ditions influenced RTs. There was an interaction of Task
and Valence (F(1,65)=15.26, p<0.001, 2=0.19), point-
ing to slower RTs for negative compared with neutral pic-
tures in the VIEW condition for (p<0.001) but no Valence
difference in the MEMORIZE condition (p =0.569).
There was a main effect of Valence (F(1,65) =5.871,
p<0.018, 2 =0.08; NEGATIVE>NEUTRAL)
and Task (F(1,65) =513.16, p<0.001, 2 =0.88;
MEMORIZE>VIEW).
Testing for alterations over time, we replicated main
effects of Task (F(1,65) =15.63, p<0.001, 2 =0.19;
MEMORIZE>VIEW) but found no group nor time effects
nor interactions (p>0.205).
The same analyses comparing DBT responders and
non-responders did not reveal significant group effects or
groupcondition interactions (ps >0.348).
fMRI data
Comparing BPD+DBT, BPD-C, and HC, there were no
suprathreshold clusters for the GroupTimeTask.
13
Eur Arch Psychiatry Clin Neurosci
an established self-report measure on BPD symptom severity (Bor-
derline Symptom List [BSL; 38]), b a standardized interview on BPD
symptom severity, the Zanarini Rating Scale for borderline personal-
ity disorder [ZAN-BPD; 37], and c self-reported emotion regulation
difficulties assessed by the Difficulties in Emotion Regulation Scale
[DERS; 39]
In BPD+DBT compared with BPD-C, the contrast
t1(MEMORIZE-VIEW) t2(MEMORIZE-VIEW) revealed a sig-
nificant cluster in the right inferior parietal lobe (IPL)/
supramarginal gyrus and the left supramarginal gyrus
(Table3). In BPD+DBT compared with HC, t1(MEMORIZE-
VIEW) t2(MEMORIZE-VIEW) also revealed a significant clus-
ter in the aforementioned right supramarginal gyrus.
The contrasts examining increases in activity differences
over time in BPD+DBT did not yield any significant
clusters. Post hoc analyses revealed a decrease in activ-
ity in BPD+DBT in the right IPL/supramarginal gyrus
in the MEMORIZE condition (t(30) = 2.24/3.28,
p =0.032/0.003) and an increase over time in the VIEW
condition (t(30)=3.14/2.17, p=0.004/0.038). No signifi-
cant time differences were found for BPD-C and HC (trend
for the increase in the right IPL in BPD-C (t(14) =1.81,
p =0.092); all other p>0.116; see Fig.2 for graphical
illustration). In the left supramarginal gyrus, BPD+DBT
and as a trend HC showed a decrease in differential
activity over time in the MEMORIZE (BPD+DBT:
t(30)=4.93, p<0.001; HC: t(21)=1.97, p=0.062)
but not in the VIEW condition, and there were also no
significant alterations over time in BPD-C in neither the
MEMORIZE nor the VIEW condition (all p>0.181).
Responder analysis
There was a GroupTimeTask interaction effect
in the right ventral inferior insula and the left cerebel-
lum. Post hoc analyses for the right ventral inferior insula
revealed an increase in differential activity (NEGATIVE-
NEUTRAL) in DBT non-responders in the VIEW condi-
tion (t(14) =3.78, p =0.002) and a trend for a decrease
in the MEMORIZE condition (t(14) =1.85, p =0.086),
while DBT responders did not show significant altera-
tions over time (p>0.131). For DBT responders compared
Eur Arch Psychiatry Clin Neurosci

Table3Alterations in neural correlates over time in patients with borderline personality disorder undergoing dialectical behavior therapy
(BPD+DBT), compared to patients with borderline personality disorder without DBT (BPD-C), and healthy control participants (HC)
Anatomical label BA Cluster size MNI Z value (peak voxel)
x y z

BPD+DBT [t1(MEMORIZE-VIEW)t2(MEMORIZE-VIEW)] masked by BPD+DBT [t1(MEMORIZE-VIEW)t2(MEMORIZE-VIEW)]>BPD-C [t1 (MEMORIZE-

VIEW)t2(MEMORIZE-VIEW)]
Inferior parietal lobe/ 40 101 51 55 43 3.90
Supramarginal gyrus 63 52 34 3.71
Supramarginal gyrus 40 27 60 55 28 3.45
BPD+DBT [t1(MEMORIZE-VIEW)t2(MEMORIZE-VIEW)] masked by BPD+DBT [t1(MEMORIZE-VIEW)t2(MEMORIZE-VIEW)]>HC [t1 (MEMORIZE-VIEW)t2
(MEMORIZE-VIEW)]
Supramarginal gyrus 40 11 63 52 34 3.74
BPD+DBT [t2 (MEMORIZE-VIEW)t1(MEMORIZE-VIEW)] masked by BPD+DBT [t2(MEMORIZE-VIEW)t1(MEMORIZE-VIEW)]>BPD-C [t2 (MEMORIZE-
VIEW)t1(MEMORIZE-VIEW)]
n.s.
BPD+DBT [t2 (MEMORIZE-VIEW)t1(MEMORIZE-VIEW)] masked by BPD+DBT [t2(MEMORIZE-VIEW)t1(MEMORIZE-VIEW)]>HC [t2(MEMORIZE-VIEW)t1
(MEMORIZE-VIEW)]
n.s.
DBT responders [t1 (MEMORIZE-VIEW)t2 (MEMORIZE-VIEW)] masked by DBT responders [t1 (MEMORIZE-VIEW)t2 (MEMORIZE-VIEW)]>DBT non-respond-
ers [t1 (MEMORIZE-VIEW)t2 (MEMORIZE-VIEW)]
n.s.
DBT responders [t2 (MEMORIZE-VIEW)t1 (MEMORIZE-VIEW)] masked by DBT responders [t2 (MEMORIZE-VIEW)t1 (MEMORIZE-VIEW)]>DBT non-respond-
ers [t2 (MEMORIZE-VIEW)t1 (MEMORIZE-VIEW)]
Cerebellum 19 6 58 38 3.48
Anterior cingulate cortex 32 20 6 35 5 3.39
9 23 8 3.25

Whole-brain fMRI data from group level analyses at a threshold of p<0.001, k>10 voxel uncorrected
n.s. not significant

with DBT non-responders, the contrast t2(MEMORIZE-
VIEW) t1(MEMORIZE-VIEW) showed a significant cluster in
the right perigenual ACC (pgACC) and the right cerebel-
lum. In the right pgACC, DBT responders showed a sig-
nificant decrease over time in the VIEW (t(15) = 4.14,
p=0.001) and a trend to an increase in the MEMORIZE
condition (t(15) =1.99, p =0.065), while DBT non-
responders did not show significant alterations over time
(p>0.515; see Fig.3).
compared with neutral pictures in the MEMORIZE con-
dition in the right supramarginal gyrus in BPDDBT
(r=0.412, p=0.021). The larger the decrease in symptom
severity in the ZAN-BPD over time, the larger the decrease
in neural activity in response to viewing negative com-
pared with neutral pictures in the right pgACC in the whole
BPDDBT group (r =0.362, p =0.046). There were no
further significant correlations.

Exploratory correlation withalterations insymptom Discussion

severity
We examined the correlation of the RCI of ZAN-BPD,
BSL, and DERS scores (more negative values indicating
higher symptom reduction) with the decrease in differential
activity in the MEMORIZE condition, the increase in the
VIEW condition in the right supramarginal gyrus ([x, y, z]:
63, 52, 34; peak voxel significant in comparison to both
control groups), and the decrease in differential activity in
the right pgACC in BPD+DBT.
The larger the decrease in symptom severity in the BSL,
the larger the decrease in neuronal activity to negative
We investigated whether neural correlates of distrac-
tion change in BPD patients after their BPD symptoms
improved. Patients in the BPD+DBT group reported
reduced difficulties in emotion regulation and increased
skill use compared with BPD-C as well as a reduction in
BPD symptom severity in the usual range for BPD psy-
chotherapies [18]. Also, BPD+DBT patients showed a
trend for a stronger reduction in borderline symptom sever-
ity than BPD-C. At the neural level, BPD+DBT patients
showed a particularly strong decrease over time in the
right supramarginal gyrus activity during distraction from

13

Fig.2Alterations in differential neural activity in the right supra-
marginal gyrus over time. For graphical illustration, beta values were
extracted from the right supramarginal gyrus (MNI-coordinate [x, y,
z]: 63, 52, 34) from the single-subject contrasts NEGATIVE-NEU-
TRAL. In patients with borderline personality disorder receiving dia-
lectical behavior therapy (BPD+DBT), differential neural activity
alters task specifically over time. This is not the case for patients with
borderline personality disorder without DBT treatment (BPD-C) and
healthy control participants (HC). The brain section provided depicts
the statistical contrast BPD+DBT [t1(MEMORIZE-VIEW)t2(MEMORIZE-
VIEW)] masked by BPD+DBT [t1(MEMORIZE-VIEW) t2(MEMORIZE-
VIEW)]>BPD-C [t1(MEMORIZE-VIEW) t2(MEMORIZE-VIEW)] at x =63
with a statistical threshold of p<0.001, uncorrected
negative rather than neutral stimuli. This decrease was cor-
related with reduced self-reported borderline symptom
severityin this group. Further, DBT responders exhibited
decreased right pgACC activity after treatment when view-
ing negative compared to neutral pictures and a trend for
increased activity in this brain region during distraction.
The supramarginal gyrus has been associated with emo-
tion regulation [25, 48, 49], including distraction [22,
26, 50]. Other functions of the right supramarginal gyrus
include (emotional) face processing [51, 52] and empathy
[5355]. BPD+DBT showed a decrease in right supra-
marginal gyrus activity compared with the control groups
during distraction from negative rather than neutral stimuli.
This was associated with reduced self-reported BPD symp-
tomatology. A reduction in supramarginal gyrus activity
after therapy in BPD+DBT may reflect less susceptibil-
ity to negative (rather than neutral) content during distrac-
tion after treatment and thus less necessity to differentially
recruit this brain region. Referring to Fig.2, it appears as
if the BPD+DBT group showed particularly elevated
supramarginal gyrus activity to negative compared with
neutral stimuli during distraction at baseline. Additional to
methodological considerations (see limitations), one could
13
Eur Arch Psychiatry Clin Neurosci
speculate whether this reflects a predictor of beneficial
psychotherapy outcome. Indeed, at least for self-reported
BPD symptom severity, higher baseline activity to nega-
tive compared with neutral stimuli during distraction was
associated with higher symptom reduction in BPD+DBT
(r = 0.378, p =0.036). Further, BPD+DBT also dis-
played a larger increase in right supramarginal gyrus activ-
ity during viewing negative rather than neutral stimuli com-
pared to HC and BPD-C. This was unrelated to any therapy
outcome, suggesting neither a beneficial nor adverse role of
this change. Acknowledging the role of the supramarginal
gyrus in (emotional) face processing [51, 52] and empathy
[5355], this finding may suggest that BPD+DBT exhib-
ited a more intensive processing of viewing negative com-
pared with neutral picturesafter DBT, e.g., by being more
able to get involved in the scenes depicted (e.g., mentalize
the pictured peoples mental states).
Comparing functional imaging data of DBT responders
to that of DBT non-responders revealed increased differ-
ential activity over time in the right ventral anterior insula
VIEW condition and a trend for a decrease in the MEMO-
RIZE condition in DBT non-responders. This region has
been associated with stimulus salience during affective and
emotional processing [5658]. Thus, our findings suggest
an increased salience of negative compared with neutral
stimuli over time in DBT non-responders when viewing
negative (rather than neutral) pictures. As DBT respond-
ers did not show these alterations, this may be interpreted
as increased salience of aversive stimuli after DBT in
non-responders.
DBT responders exhibited decreased right pgACC activ-
ity during viewing of negative compared with neutral stim-
uli following therapy. DBT non-responders did not show
any changes in this area. The pgACC is thought to reflect
the automatic, attentional regulation of negative affect [59].
Thus, the decrease over time of pgACC activity to negative
(rather than neutral) stimuli in DBT responders may reflect
comparatively less impact of viewing negative in compari-
son with neutral stimuli in this group. In combination with
the finding of alterations in supramarginal gyrus activity,
our findings could hint at a shift from emotional to rather
cognitive processing during viewing aversive stimuli after
successful DBT, but not mere limbic habituation.
While the present study found alterations in subre-
gions of the fronto-parietal network associated with dis-
traction [21, 22, 2527], we did not find that BPD symp-
tom improvement was associated with reduced limbic
hyper-reactivity during distraction [12, 13]. If replicated,
this could suggest that only (parasympathetic) inhibi-
tory circuits but not (sympathetic) defensive mechanisms
are altered after symptom improvement [60, 61]. How-
ever, our findings could also suggest that distraction is an
automatized, effective, and overlearned emotion regulation
Eur Arch Psychiatry Clin Neurosci
Fig.3Alterations in differential activity in the right perigenual ante-
rior cingulate over time. For graphical illustration, beta values were
extracted from the right perigenual anterior cingulate [MNI-coor-
dinate (x, y, z): 6, 35, 5] from the single-subject contrasts NEGA-
TIVE-NEUTRAL. In patients with borderline personality disorder
responding to dialectical behavior therapy (DBT responders), differ-
ential neural activity alters task specifically over time. This is not the
case for patients with borderline personality being DBT non-respond-
ers (DBT non-responders). The brain section provided depicts the
statistical contrast DBT responders [t2(MEMORIZE-VIEW)t1(MEMORIZE-
VIEW)] masked by DBT responders [t2(MEMORIZE-VIEW)t1(MEMORIZE-
VIEW)]>DBT non-responders [t2(MEMORIZE-VIEW)t1(MEMORIZE-VIEW)]
at x=6 with a statistical threshold of p<0.001, uncorrected
strategy [19, 20], which already reduces limbic reactivity to
negative stimuli independent of symptom severity. Favor-
ing this explanation, we found no significant behavioral dif-
ference between distraction from negative compared with
neutral pictures at the first measurement, while there was
after viewing. Also, there were no behavioral group differ-
ences before treatment nor regarding alterations over time.
Focusing on less overlearned emotion regulation strategies
such as reappraisal may help clarify this issue.
Limitations
Even though the BPD groups were comparable in symp-
tom severity at the first measurement, BPD+DBT patients
received residential treatment, whereas BPD-C received
mainly outpatient treatment. In addition, as medication
may attenuate emotional responses in BPD [6], a limita-
tion of our study is that not all included BPD patients were
medication free. Since different combinations of medica-
tion subtypes were present in patients, it was also not possi-
ble to isolate the effect of a single agent. Importantly, medi-
cation was kept constant during the study period, and the
percentage of medicated individuals did not differ between
the BPD groups. In the future, studies will be most interest-
ingly that compare unmedicated patients on psychotherapy
with those on pharmacotherapy alone and those on the
combination of psychotherapy and pharmacotherapy.
To note, this study was not designed to test for DBT-spe-
cific effects on brain functioning. First, even though DBT
was delivered in specialized treatment units under regular
expert supervision, there were no DBT adherence ratings.
Second, this studys control group rather allows controlling
for time than for general psychotherapy effects. Overall,
a randomized controlled trial with a comparably intense
comparison treatment needs to address effect specificity
for DBT. Beyond that, we argue that our longitudinal fMRI
design has a strong benefit compared with earlier studies
on neural correlates of emotion processing in BPD as we
included a BPD-C control group in addition to a healthy
control group to control for effects of time.
Further, to increase sample homogeneity, results were
only obtained from females, which limits generalizabil-
ity to males. Also, we used a liberal significance thresh-
old for the fMRI data, and the findings would not survive
correction for multiple comparisons. Still, our significance
threshold is more conservative than in a previous study
[13]. One possible explanation is that we were only able
to collect data of 15 individuals in the BPD-C group, and
thus group size differences may have limited the statisti-
cal power of our study. In addition, we selected the respec-
tive distraction task specifically as being manageable for
all participants (via pilot study). Our task activated the
same neural network as reported in previous studies in
healthy controls (information in the Supplement) [21, 22,
2527]. However, a more difficult distraction task might
have revealed (stronger) group differences with respect to
behavioral and fMRI data at the first measurement as well
as stronger alterations over time.
Conclusions
This study addressed whether neuronal correlates of a com-
mon emotion regulation strategydistractionchange
after psychotherapy in BPD and related symptom improve-
ment. BPD patients displayed changes in areas associ-
ated with emotional distraction (supramarginal gyrus) and
automatic attentional control (pgACC). Further research
needs to examine whether this finding is specific to dis-
traction as one emotion regulation strategy among several
and how alterations develop in the long run in follow-up
examinations.
Acknowledgments We thank Indre Ciurlyte, Michael Rie, Clau-
dia Stief, and Miriam Wei for their help with data collection and
13

for supporting the MRI measurements. This work was funded by the
German Research Foundation (C.S., SCHM 1526/8-2; S.C.H., HE
2660/7-2).
Compliance with ethical standards
Conflict of interest On behalf of all authors, the corresponding
author states that there is no conflict of interest.
Ethical statement The authors assert that all procedures contributing
to this work comply with the ethical standard of the relevant national
and institutional committees on human experimentation and with the
Helsinki Declaration of 1975, as revised in 2008. The Research Ethics
Board II of the University of Heidelberg, Germany, approved the study.
All participants were informed about the study procedures and gave
written informed consent (capacity assessed via clinical interview). We
registered the study as clinical trial (DRKS-ID=DRKS00000778).
References
1. Glenn CR, Klonsky ED (2009) Emotion dysregulation as a core
feature of borderline personality disorder. J Personal Disord
23(1):2028. doi:10.1521/pedi.2009.23.1.20
2. Schmahl C, Herpertz S, Bertsch K, Ende G, Flor H, Kirsch P, Lis
S, Meyer-Lindenberg A, Rietschel M, Schneider M, Spanagel R,
Treede R-D, Bohus M (2014) Mechanisms of disturbed emotion
processing and social interaction in borderline personality disor-
der: state of knowledge and research agenda of the German Clin-
ical Research Unit. Borderline Personal Disord Emot Dysregul
1(1):117. doi:10.1186/2051-6673-1-12
3. Wagner AW, Linehan MM (1999) Facial expression recogni-
tion ability among women with borderline personality disor-
der: implications for emotion regulation. J Personal Disord
13(4):329344. doi:10.1521/pedi.1999.13.4.329
4. Lynch TR, Rosenthal MZ, Kosson DS, Cheavens JS, Lejuez CW,
Blair RJ (2006) Heightened sensitivity to facial expressions of
emotion in borderline personality disorder. Emotion 6(4):647
655. doi:10.1037/1528-3542.6.4.647
5. Santangelo P, Reinhard I, Mussgay L, Steil R, Sawitzki G, Klein
C, Trull TJ, Bohus M, Ebner-Priemer UW (2014) Specificity of
affective instability in patients with borderline personality dis-
order compared to posttraumatic stress disorder, bulimia ner-
vosa, and healthy controls. J Abnorm Psychol 123(1):258272.
doi:10.1037/a0035619
6. Schulze L, Schmahl C, Niedtfeld I (2016) Neural correlates of
disturbed emotion processing in borderline personality disor-
der: a multimodal meta-analysis. Biol Psychiatry 79(2):97106.
doi:10.1016/j.biopsych.2015.03.027
7. Ochsner KN, Silvers JA, Buhle JT (2012) Functional imaging
studies of emotion regulation: a synthetic review and evolving
model of the cognitive control of emotion. Ann NY Acad Sci
1251:E1E24. doi:10.1111/j.1749-6632.2012.06751.x
8. Kohn N, Eickhoff SB, Scheller M, Laird AR, Fox PT, Habel U
(2014) Neural network of cognitive emotion regulationan ALE
meta-analysis and MACM analysis. Neuroimage 87:345355.
doi:10.1016/j.neuroimage.2013.11.001
9. Koenigsberg HW, Siever LJ, Lee H, Pizzarello S, New AS,
Goodman M, Cheng H, Flory J, Prohovnik I (2009) Neu-
ral correlates of emotion processing in borderline personal-
ity disorder. Psychiatry Res 172(3):192199. doi:10.1016/j.
pscychresns.2008.07.010
10. Minzenberg MJ, Fan J, New AS, Tang CY, Siever LJ (2007)
Fronto-limbic dysfunction in response to facial emo-
tion in borderline personality disorder: an eventrelated
13
Eur Arch Psychiatry Clin Neurosci
fMRI study. Psychiatry Res 155(3):231243. doi:10.1016/j.
pscychresns.2007.03.006
11. Schulze L, Domes G, Kruger A, Berger C, Fleischer M, Prehn
K, Schmahl C, Grossmann A, Hauenstein K, Herpertz SC
(2011) Neuronal correlates of cognitive reappraisal in borderline
patients with affective instability. Biol Psychiatry 69(6):564
573. doi:10.1016/j.biopsych.2010.10.025
12. Goodman M, Carpenter D, Tang CY, Goldstein KE, Avedon J,
Fernandez N, Mascitelli KA, Blair NJ, New AS, Triebwasser
J, Siever LJ, Hazlett EA (2014) Dialectical behavior therapy
alters emotion regulation and amygdala activity in patients with
borderline personality disorder. J Psychiatry Res 57:108116.
doi:10.1016/j.jpsychires.2014.06.020
13. Schnell K, Herpertz SC (2007) Effects of dialectic-behavioral-
therapy on the neural correlates of affective hyperarousal in bor-
derline personality disorder. J Psychiatry Res 41(10):837847.
doi:10.1016/j.jpsychires.2006.08.011
14. Linehan MM (1993) Cognitive-behavioral treatment of border-
line personality disorder. Guilford Press, New York
15. Linehan MM (1993) Skills training manual for treating border-
line personality disorder. Guilford Press, New York
16. Bohus M (2004) Effectiveness of inpatient dialectical behav-
ioral therapy for borderline personality disorder: a con-
trolled trial. Behav Res Ther 42(5):487499. doi:10.1016/
s0005-7967(03)00174-8
17. Kliem S, Kroger C, Kosfelder J (2010) Dialectical behavior ther-
apy for borderline personality disorder: a meta-analysis using
mixed-effects modeling. J Consult Clin Psychol 78(6):936951.
doi:10.1037/a0021015
18. Stoffers JM, Vollm BA, Rucker G, Timmer A, Huband N, Lieb
K (2012) Psychological therapies for people with borderline
personality disorder. Cochrane Database Syst Rev 8:CD005652.
doi:10.1002/14651858.CD005652.pub2
19. Thiruchselvam R, Blechert J, Sheppes G, Rydstrom A, Gross JJ
(2011) The temporal dynamics of emotion regulation: an EEG
study of distraction and reappraisal. Biol Psychol 87(1):8492.
doi:10.1016/j.biopsycho.2011.02.009
20. Schonfelder S, Kanske P, Heissler J, Wessa M (2013) Time
course of emotion-related responding during distraction and
reappraisal. Soc Cogn Affect Neurosci 9(9):13101319.
doi:10.1093/scan/nst116
21. McRae K, Hughes B, Chopra S, Gabrieli JD, Gross JJ, Ochsner
KN (2010) The neural bases of distraction and reappraisal. J
Cogn Neurosci 22(2):248262. doi:10.1162/jocn.2009.21243
22. Kanske P, Heissler J, Schonfelder S, Bongers A, Wessa M (2011)
How to regulate emotion? Neural networks for reappraisal and
distraction. Cereb Cortex 21(6):13791388. doi:10.1093/cercor/
bhq216
23. Lieberman MD, Inagaki TK, Tabibnia G, Crockett MJ (2011)
Subjective responses to emotional stimuli during labeling, reap-
praisal, and distraction. Emotion 11(3):468480. doi:10.1037/
a0023503
24. Sheppes G, Meiran N (2007) Better late than never? On the
dynamics of online regulation of sadness using distraction and
cognitive reappraisal. Personal Soc Psychol Bull 33(11):1518
1532. doi:10.1177/0146167207305537
25. Erk S, Kleczar A, Walter H (2007) Valence-specific regulation
effects in a working memory task with emotional context. Neu-
roimage 37(2):623632. doi:10.1016/j.neuroimage.2007.05.006
26. Dorfel D, Lamke JP, Hummel F, Wagner U, Erk S, Walter H
(2014) Common and differential neural networks of emotion reg-
ulation by detachment, reinterpretation, distraction, and expres-
sive suppression: a comparative fMRI investigation. Neuroim-
age. doi:10.1016/j.neuroimage.2014.06.051
27. Kellermann TS, Sternkopf MA, Schneider F, Habel U, Turetsky
BI, Zilles K, Eickhoff SB (2011) Modulating the processing of
Eur Arch Psychiatry Clin Neurosci
emotional stimuli by cognitive demand. Soc Cogn Affect Neuro-
sci 7(3):263273. doi:10.1093/scan/nsq104
28. Krause-Utz A, Oei NY, Niedtfeld I, Bohus M, Spinhoven P,
Schmahl C, Elzinga BM (2012) Influence of emotional distrac-
tion on working memory performance in borderline personal-
ity disorder. Psychol Med 42(10):21812192. doi:10.1017/
S0033291712000153
29. Silbersweig D, Clarkin JF, Goldstein M, Kernberg OF, Tuescher
O, Levy KN, Brendel G, Pan H, Beutel M, Pavony MT, Epstein
J, Lenzenweger MF, Thomas KM, Posner MI, Stern E (2007)
Failure of frontolimbic inhibitory function in the context of neg-
ative emotion in borderline personality disorder. Am J Psychiatry
164(12):18321841. doi:10.1176/appi.ajp.2007.06010126
30. Prehn K, Schulze L, Rossmann S, Berger C, Vohs K, Fleischer
M, Hauenstein K, Keiper P, Domes G, Herpertz SC (2013)
Effects of emotional stimuli on working memory processes in
male criminal offenders with borderline and antisocial personal-
ity disorder. World J Biol Psychiatry 14(1):7178. doi:10.3109/1
5622975.2011.584906
31. Wingenfeld K, Rullkoetter N, Mensebach C, Beblo T, Mertens
M, Kreisel S, Toepper M, Driessen M, Woermann FG (2009)
Neural correlates of the individual emotional Stroop in border-
line personality disorder. Psychoneuroendocrinology 34(4):571
586. doi:10.1016/j.psyneuen.2008.10.024
32. Dolcos F, Wang L, Mather M (2014) Current research and
emerging directions in emotioncognition interactions. Front
Integr Neurosci 8:83. doi:10.3389/fnint.2014.00083
33. Carretie L (2014) Exogenous (automatic) attention to emotional
stimuli: a review. Cogn Affect Behav Neurosci 14(4):12281258.
doi:10.3758/s13415-014-0270-2
34. Winter D, Krause-Utz A, Lis S, Chiu CD, Lanius RA, Schriner F,
Bohus M, Schmahl C (2015) Dissociation in borderline personal-
ity disorder: disturbed cognitive and emotional inhibition and its
neural correlates. Psychiatry Res 233(3):339351. doi:10.1016/j.
pscychresns.2015.05.018
35. Loranger AW (1999) International Personality Disorder Exami-
nation (IPDE): DSM-IV and ICD-10 modules. Psychological
Assessment Resources, Odessa
36. Wittchen HU, Wunderlich U, Gruschwitz S (1997) SKID. Struk-
turiertes Klinisches Interview fr DSM-IV Achse I. Hogrefe,
Gttingen
37. Zanarini MC, Vujanovic AA, Parachini EA, Boulanger JL,
Frankenburg FR, Hennen J (2003) Zanarini Rating Scale for
Borderline Personality Disorder (ZAN-BPD): a continuous
measure of DSM-IV borderline psychopathology. J Personal
Disord 17(3):233242. doi:10.1521/pedi.17.3.233.22147
38. Bohus M, Limberger MF, Frank U, Chapman AL, Kuhler T,
Stieglitz RD (2007) Psychometric properties of the Border-
line Symptom List (BSL). Psychopathology 40(2):126132.
doi:10.1159/000098493
39. Gratz KL, Roemer L (2004) Multidimensional assessment of
emotion regulation and dysregulation: development, factor struc-
ture, and initial validation of the difficulties in emotion regula-
tion scale. J Psychopathol Behav 26(1):4154. doi:10.1007/
s10862-008-9102-4
40. Bohus M, Haaf B, Stiglmayr C, Pohl U, Boehme R, Linehan MM
(2000) Evaluation of inpatient dialectical-behavioral therapy for
borderline personality disordera prospective study. Behav Res
Ther 38(9):875887. doi:10.1016/S0005-7967(99)00103-5
41. Bohus M, Wolf M (2009) Interaktives Fertigkeitentraining fr
Borderline-Strungen. Schattauer Verlag, Stuttgart
42. Kleindienst N, Limberger MF, Schmahl C, Steil R, Ebner-Priemer
UW, Bohus M (2008) Do improvements after inpatient dialectial
behavioral therapy persist in the long term? A naturalistic follow-
up in patients with borderline personality disorder. J Nerv Ment
Dis 196(11):847851. doi:10.1097/NMD.0b013e31818b481d
43. Lang PJ, Bradley MM, Cuthbert BN (1997) International Affec-
tive Picture System (IAPS): technical manual and affective rat-
ings. The Center for Research in Psychophysiology, University
of Florida, Gainesville
44. Wessa M, Kanske P, Neumeister P, Bode K, Heissler J, Schn-
felder S (2010) EmoPics: subjektive und psychophysiologische
Evaluationen neuen Bildmaterials fr die klinisch-bio-psycholo-
gische Forschung. Z Kl Psych Psychoth 39(Suppl. 1/11):77
45. Volman I, Toni I, Verhagen L, Roelofs K (2011) Endogenous
testosterone modulates prefrontalamygdala connectivity dur-
ing social emotional behavior. Cereb Cortex 21(10):22822290.
doi:10.1093/cercor/bhr001
46. Lieberman MD, Cunningham WA (2009) Type I and Type II
error concerns in fMRI research: re-balancing the scale. Soc
Cogn Affect Neurosci 4(4):423428. doi:10.1093/scan/nsp052
47. Jacobson NS, Truax P (1991) Clinical significance: a sta-
tistical approach to defining meaningful change in psy-
chotherapy research. J Consult Clin Psychol 59(1):1219.
doi:10.1037/0022-006X.59.1.12
48. Winecoff A, Clithero JA, Carter RM, Bergman SR, Wang L,
Huettel SA (2013) Ventromedial prefrontal cortex encodes
emotional value. J Neurosci 33(27):1103211039. doi:10.1523/
jneurosci.4317-12.2013
49. Ochsner KN, Ray RD, Cooper JC, Robertson ER, Chopra S,
Gabrieli JD, Gross JJ (2004) For better or for worse: neural
systems supporting the cognitive down- and up-regulation of
negative emotion. Neuroimage 23(2):483499. doi:10.1016/j.
neuroimage.2004.06.030
50. Goldstein M, Brendel G, Tuescher O, Pan H, Epstein J, Beu-
tel M, Yang Y, Thomas K, Levy K, Silverman M, Clarkin J,
Posner M, Kernberg O, Stern E, Silbersweig D (2007) Neu-
ral substrates of the interaction of emotional stimulus process-
ing and motor inhibitory control: an emotional linguistic go/
no-go fMRI study. Neuroimage 36(3):10261040. doi:10.1016/j.
neuroimage.2007.01.056
51. Carvajal F, Rubio S, Serrano JM, Rios-Lago M, Alvarez-Linera
J, Pacheco L, Martin P (2013) Is a neutral expression also a neu-
tral stimulus? A study with functional magnetic resonance. Exp
Brain Res 228(4):467479. doi:10.1007/s00221-013-3578-1
52. Radua J, Phillips ML, Russell T, Lawrence N, Marshall N,
Kalidindi S, El-Hage W, McDonald C, Giampietro V, Bram-
mer MJ, David AS, Surguladze SA (2010) Neural response
to specific components of fearful faces in healthy and schizo-
phrenic adults. Neuroimage 49(1):939946. doi:10.1016/j.
neuroimage.2009.08.030
53. Mitchell JP (2008) Activity in right temporo-parietal junction is
not selective for theory-of-mind. Cereb Cortex 18(2):262271.
doi:10.1093/cercor/bhm051
54. Silani G, Lamm C, Ruff CC, Singer T (2013) Right supramar-
ginal gyrus is crucial to overcome emotional egocentricity bias in
social judgments. J Neurosci 33(39):1546615476. doi:10.1523/
jneurosci.1488-13.2013
55. Saxe R, Kanwisher N (2003) People thinking about think-
ing people. The role of the temporo-parietal junction in the-
ory of mind. Neuroimage 19(4):18351842. doi:10.1016/
S1053-8119(03)00230-1
56. Kurth F, Zilles K, Fox PT, Laird AR, Eickhoff SB (2010) A link
between the systems: functional differentiation and integration
within the human insula revealed by meta-analysis. Brain Struct
Funct 214(56):519534. doi:10.1007/s00429-010-0255-z
57. Touroutoglou A, Hollenbeck M, Dickerson BC, Feldman Bar-
rett L (2012) Dissociable large-scale networks anchored in
the right anterior insula subserve affective experience and
attention. Neuroimage 60(4):19471958. doi:10.1016/j.
neuroimage.2012.02.012
13

58. Craig AD (2009) How do you feelnow? The anterior insula and
human awareness. Nat Rev Neurosci 10(1):5970. doi:10.1038/
nrn2555
59. Phillips M, Ladouceur C, Drevets W (2008) A neural model of
voluntary and automatic emotion regulation: implications for
understanding the pathophysiology and neurodevelopment of
bipolar disorder. Mol Psychiatry 13(9):829857. doi:10.1038/
mp.2008.65
13
Eur Arch Psychiatry Clin Neurosci
60. Thayer JF (2006) On the importance of inhibition: central and
peripheral manifestations of nonlinear inhibitory processes in
neural systems. Dose-Response 4(1):221. doi:10.2203/dose-
response.004.01.002.Thayer
61. Thayer JF, Brosschot JF (2005) Psychosomatics and psy-
chopathology: looking up and down from the brain. Psy-
choneuroendocrinology 30(10):10501058. doi:10.1016/j.
psyneuen.2005.04.014