Journal of Affective Disorders 341 (2023) 335–345

Contents lists available at ScienceDirect

Journal of Affective Disorders

journal homepage: www.elsevier.com/locate/jad

Review article

Resting-state functional magnetic resonance imaging alterations in

borderline personality disorder: A systematic review
Mahan Shafie a, Elnaz Shahmohamadi a, Giulia Cattarinussi b, c, Hossein Sanjari Moghaddam a, d,
Shahin Akhondzadeh d, Fabio Sambataro b, c, Chiara Moltrasio e, Giuseppe Delvecchio e, *
a
School of Medicine, Tehran University of Medical Sciences, Tehran, Iran
b
Department of Neuroscience (DNS), Padua Neuroscience Center, University of Padova, Padua, Italy
c
Padua Neuroscience Center, University of Padova, Padua, Italy
d
Psychiatry and Psychology Research Center, Roozbeh Hospital, Tehran University of Medical Sciences, Tehran, Iran
e
Department of Neurosciences and Mental Health, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milan, Italy

A R T I C L E I N F O A B S T R A C T

Keywords: Background: Borderline personality disorder (BPD) is a severe psychiatric disorder characterized by emotion
Borderline personality disorder dysregulation, impulsivity, and interpersonal disturbances. Several structural and functional neuroimaging ab­
Functional connectivity normalities have been described in BPD. In particular, resting-state functional magnetic resonance imaging (rs-
Resting-state
fMRI) studies have recently suggested various connectivity alterations within and between large-scale brain
ALFF
fALFF
networks in BPD. This review aimed at providing an updated summary of the evidence reported by the available
ReHo rs-fMRI studies in BPD individuals.
DC Methods: A search on PubMed, Scopus, and Web of Science was performed to identify rs-fMRI alterations in BPD.
Triple network model A total of 15 studies met our inclusion criteria.
Results: Overall, aberrant resting-state functional connectivity (rs-FC) within and between default mode network
(DMN), salience network (SN), and central executive network (CEN) were observed in BPD compared to healthy
controls, as well as selective functional impairments in bilateral amygdala, anterior and posterior cingulate
cortex, hippocampus, and prefrontal cortex.
Limitations: The observational design, small sample size, prevalence of females, high rates of concurrent
comorbidities and medications, and heterogeneity across imaging methodologies limit the generalizability of the
results.
Conclusions: The identification of altered patterns of rs-FC within and between selective brain networks,
including DMN, SN, and CEN, could further our knowledge of the clinical symptoms of BPD, and therefore, future
studies with multimodal methodologies and longitudinal designs are warranted to further explore the neural
correlates of this disorder.

1. Introduction comorbidity with other psychiatric disorders, including mood disorders,

anxiety disorders, substance abuse, along with other personality disor­
Borderline personality disorder (BPD) is a severe and common psy­ ders, which may mask underlying disorders and delay their diagnosis
chiatric disorder estimated to affect 0.5–5 % of the general population and treatment (Lieb et al., 2004). Although the number of studies on the
and 10–12 % of psychiatric patients (Ellison et al., 2018). The under­ potential neurobiological underpinnings of BPD has increased over the
lying psychopathology of BPD is characterized by a pervasive pattern of past years, its etiology has not been yet clearly understood and remains
emotion dysregulation, altered cognitive processing, including dissoci­ controversial.
ation, behavioral dysregulation and impulsivity, and interpersonal dis­ One of the most common findings identified through structural
turbances (Sayrs and Whiteside, 2006). BPD is commonly observed in magnetic resonance imaging (MRI) and diffusion tensor imaging (DTI)

* Corresponding author at: Department of Neurosciences and Mental Health, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, via F. Sforza 35, 20122
Milan, Italy.
E-mail address: giuseppe.delvecchio@policlinico.mi.it (G. Delvecchio).

https://doi.org/10.1016/j.jad.2023.09.001
Received 20 February 2023; Received in revised form 24 August 2023; Accepted 1 September 2023
Available online 4 September 2023
0165-0327/© 2023 Elsevier B.V. All rights reserved.
M. Shafie et al. Journal of Affective Disorders 341 (2023) 335–345

studies is an alteration in gray matter (GM) and white matter (WM) In this context, this review aimed at providing an updated summary of
volume in regions within the limbic system and in frontal areas (Grot­ the evidence reported by the available rs-fMRI studies that investigated
taroli et al., 2020; Yu et al., 2019). Also, meta-analytic evidence showed alterations in spontaneous brain activity and FC in BPD patients to
significant volumetric differences in various brain regions, including further our knowledge on the neurobiology of this disorder.
hippocampal, temporal, and pre-frontal regions, in patients with BPD
compared to healthy controls (HC) (Schulze et al., 2016; Yang et al., 2. Methods
2016). In addition, a meta-analysis exploring DTI abnormalities in BPD
identified lower WM integrity in the corpus callosum and fornix, which A literature review was conducted on PubMed, Scopus, and Web of
was also found to be associated with emotion dysregulation (Kelleher- Science to identify all relevant studies published until September 2022
Unger et al., 2021). Notably, these structural alterations in regions using the following keywords: (“borderline” OR “borderline personal­
within the prefrontal-limbic system were also confirmed by functional ity” OR “borderline personality disorder” OR “BPD”) AND (“resting-state
MRI (fMRI) studies, that suggested the association between brain ac­ functional magnetic resonance imaging” OR “resting-state functional
tivity/functional connectivity (FC) of these regions and core symptoms MRI” OR “resting-state fMRI” OR “resting-state functional connectivity”
of BPD, such as emotion dysregulation and interpersonal difficulties. OR “resting-state” OR “functional connectivity”). Additional articles
More in detail, a review of neuroimaging findings in BPD reported that were also reviewed by searching the reference lists of relevant articles.
the most consistent results were limbic hyperactivity and diminished Two authors (M.S. and E.S.) independently searched through a list of
recruitment of frontal regions, which could explain the key features of 151 articles. Only fMRI studies investigating rs-fMRI alterations in BPD
BPD, including emotion dysregulation, impulsivity, and interpersonal patients, including seed-based rs-FC, network-based rs-FC, spontaneous
disturbances (Krause-Utz et al., 2014b). Importantly, this evidence was brain activity, and local connectivity, with or without pharmacological
further confirmed by a subsequent meta-analysis (Schulze et al., 2016) treatments, were included. For the aim of this review, studies were
and a review (Van Zutphen et al., 2015). included if they met the following criteria: (1) original articles investi­
Moreover, several resting-state fMRI (rs-fMRI) studies have been gating rs-fMRI alterations in BPD patients; (2) patients who met the
conducted to clarify the neural underpinnings of a range of illnesses, Diagnostic and Statistical Manual of Mental Disorders (DSM) IV/IV-TR/
spanning from neurodegenerative to psychiatric disorders (Filippi et al., 5 criteria for BPD. The exclusion criteria were as follows: (1) the main
2019; Oathes et al., 2015; Prodoehl et al., 2014). Rs-fMRI measures diagnosis was not BPD; (2) studies that were not resting-state fMRI in­
spontaneous brain activity and patterns of resting-state FC (rs-FC) be­ vestigations (3) lack of a healthy control group. Our search resulted in
tween brain areas. This technique relies on spontaneous low-frequency 151 papers, and 82 were selected for full-text reading (Fig. 1). Eventu­
fluctuations in the blood oxygen level-dependent (BOLD) signal in the ally, the eligibility criteria were met by 15 rs-fMRI studies, whose
absence of any sensory or cognitive stimulus (Biswal et al., 1995). characteristics are reported in Table 1. To evaluate the quality of the
Specifically for BPD, a previous meta-analysis of rs-fMRI and positron included studies, we employed the Newcastle-Ottawa Scale (NOS), a
emission tomography (PET) investigations showed alterations in neural tool designed for appraising the quality of non-randomized studies
activity at rest in regions of the midline core and dorsal subsystem of the included in systematic reviews and meta-analyses (Wells et al., 2000)
default mode network (DMN) in BPD (Visintin et al., 2016). In partic­ (see Supplementary Materials).
ular, this study found an increased activity at rest in the medial pre­
frontal cortex (mPFC) and anterior cingulate cortex (ACC) and 3. Results
decreased activity in the right middle and inferior temporal gyrus, as
well as a less robust decreased activity at rest in the parahippocampus 3.1. Quality assessment
and the dorsolateral prefrontal cortex (DLPFC) in BPD compared to HC
(Visintin et al., 2016). Interestingly, it has been also suggested that The results of the risk assessment for the included studies are pre­
specific symptoms of BPD could be related to altered FC in a number of sented in the Supplementary Materials (Table S.1). In terms of fulfilling
brain regions. More in detail, previous rs-fMRI investigations described the NOS checklist, each study obtained a score of 3 or 4 stars within the
an association between impaired self-referential processes and selection domain, received at least one star in terms of comparability,
emotional dysregulation and the FC of the mPFC and posterior cingulate and received 3 stars on the outcome/exposure domain. Consequently, all
cortex (PCC), as well as between interpersonal disturbances and the FC the studies were classified as “good quality” according to the NOS
of the temporal cortex, which is a hub for social communication, social criteria. Notably, three studies (Das et al., 2014; Salvador et al., 2016;
cognition, and language processing (Andrews-Hanna et al., 2014). Traynor et al., 2021) achieved the lowest score of 7 out of 9.
Notably, cognitive deficits in BPD, such as difficulties in attention,
concentration, memory, and planning, seem to be associated with ab­ 3.2. Characteristics of the studies
normalities in the hippocampus, retrosplenial cingulate cortex, and
ventromedial PFC, which are part of the DMN system (Ruocco, 2005). In the selected studies, BPD patients presented a variety of comor­
These alterations have been also observed in a variety of psychiatric bidities, such as BD, lifetime or current MDD, generalized anxiety dis­
disorders that share similarities with BPD, including major depressive order (GAD), post-traumatic stress disorder (PTSD), social and specific
disorder (MDD), bipolar disorder (BD), and schizophrenia (SCZ) (Mar­ phobia, panic disorder, attention deficit hyperactivity disorder (ADHD),
tino et al., 2016; Sambataro et al., 2014, 2010). eating disorders, and alcohol and substance use disorders, with some
While structural studies have suggested the presence of widespread studies also examining patients with other personality disorders,
morphological alterations in several cortical and subcortical brain re­ including cluster A, B, and C personality disorders. Furthermore, pa­
gions among individuals with BPD, the use of fMRI holds the potential to tients were using various medications like antidepressants, anti-
enhance our understanding of the neural processes that underlie the psychotics, and mood stabilizers. We did not exclude these studies to
intense emotional manifestations associated with BPD. However, it is enhance the generalizability of our results, since comorbidities are
important to acknowledge that measuring brain activity during task commonly present in BPD. None of the matched HC in the selected
performance may inherently introduce a bias based on the specific studies had psychiatric disorders or used medications. Most of the
experimental conditions being tested (Visintin et al., 2016). This po­ studies explored the association of rs-fMRI measures with the severity of
tential bias can be mitigated with the use of rs-fMRI, which avoids the borderline symptoms using the Borderline Symptom List (BSL) (Bohus
confounding effects of task performance. During periods of resting state, et al., 2001), BPD Severity Index (Arntz et al., 2003), Diagnostic Inter­
not only can local variations in neural activity be assessed, but also view for Borderline - Revised (DIB-R) (Zanarini et al., 1989), and Clin­
changes in the FC between different brain regions (Biswal et al., 1995). ical Global Impression scale for BPD (CGI-BPD) (Pérez et al., 2007).

336
M. Shafie et al. Journal of Affective Disorders 341 (2023) 335–345

Identification of studies via databases and registers

Identification
Records identified through Records removed before
databases (PubMed, Scopus, screening:
Web of Science) (n = 151) Duplicate records removed (n
= 69)

Records excluded as they did not

Records screened
meet the inclusion criteria (n =
(n = 82)
49)
Screening

Reports excluded (n=8):

Reports assessed for eligibility No resting-state fMRI (n = 4)
(n = 23) Lack of a healthy control
group (n = 4)
Included

Studies included in review

(n = 15)

Fig. 1. PRISMA flow chart of selection of publications for inclusion in review.

Additionally, all of the studies also calculated the correlations between evaluate the rs-FC of pre-defined brain regions.
rs-fMRI alterations and psychometric measures, such as impulsivity
assessed by the Barratt Impulsiveness Scale (BIS) (Patton et al., 1995),
dissociative experiences assessed by the Dissociation Experience Scale 3.3. Network-based studies
(DES) (Bernstein and Putnam, 1986) or the Dissociative Tension Scale
(DTS) (Stiglmayr et al., 2010), emotion dysregulation assessed with the Wolf et al. (2011) reported aberrant rs-FC within the DMN and CEN
Toronto Alexithymia Scale (TAS) (Bagby et al., 1994), the Difficulties in using the Independent Component Analysis (ICA). More specifically,
Emotion Regulation Scale (DERS) (Gratz and Roemer, 2004), or the within the DMN, decreased rs-FC in the left cuneus and increased rs-FC
Cognitive Emotion Regulation Questionnaire (CERQ) (Garnefski and in the left insula and the left frontopolar cortex (FPC) were seen in the
Kraaij, 2007), interpersonal violence evaluated by the Inventory of sample of female BPD patients compared to female HC. Also, reduced rs-
Interpersonal Problems (IIP) (Barkham et al., 1996), and history of FC in the left inferior parietal lobule and right middle temporal gyrus
childhood maltreatment measured by the Childhood Trauma Ques­ within the CEN was observed (Wolf et al., 2011). Similarly, in another
tionnaire (CTQ) (Bernstein et al., 1998). Lastly, metacognitive abilities investigation that employed ICA, BPD patients showed lower rs-FC in
were assessed with the Metacognition Assessment Interview (MAI) the DMN and SN compared to HC. At the local level, lower rs-FC was
(Semerari et al., 2012). observed in the right PCC, while no significant difference was observed
With regards to rs-fMRI analyses, five studies performed network- in the SN (Quattrini et al., 2019). Consistent with these findings, Traynor
based analysis to examine the rs-FC within and between large-scale et al. (2021) analyzed rs-FC patterns in patients with personality dis­
brain networks, including the DMN, salience network (SN), and cen­ orders with a minimum diagnosis of BPD, although the majority of the
tral executive network (CEN). Additionally, one study explored degree sample had one or more additional personality disorders, within the
centrality (DC) and fractional amplitude of low frequency fluctuations DMN, SN, CEN, and fronto-limbic network. The authors found decreased
(fALFF), one amplitude of low frequency fluctuations (ALFF) and rs-FC of the left anterior parahippocampal gyrus with the precuneus/
regional homogeneity (ReHo) and one global brain connectivity (GBC) PCC within the DMN, and increase rs-FC of the DLPFC with the anterior
and ALFF to examine spontaneous brain activity and local connectivity. parahippocampal gyrus, amygdala, and hippocampus (Traynor et al.,
The results of these analyses were subsequently used to explored seed- 2021). Das et al. (2014) also reported reduced rs-FC between the SN and
based rs-FC. Lastly, seven studies used seed-based methodology to CEN and between the SN and precuneus in a sample of subjects with BPD
compared to HC, although it did not survive correction for multiple

337
M. Shafie et al. Journal of Affective Disorders 341 (2023) 335–345

Table 1
Selection of studies investigating resting-state fMRI alterations in borderline personality disorder.
Author Subjects: Rs-FC analysis Medications in Comorbidities Neuropsychologic Findings Significant associations
(year) Number BPD assessments between rs-FC and
(F/M) age neuropsychologic
[mean ± assessments
SD]

Baczkowski BPD: 48 Seed-based: Bilateral AD (31), AP (5), MDD, DYS, BPII, BPDSI, BSI, BPD- BPD vs HC: ↓ rs-FC of No significant
et al. (48/0) amygdalae (with whole MS (1), GAD, PD, CL, ITEC, DTS amygdala with right correlation
(2017) brain FC) anxiolytics (2) AGOPHOB, PHOB, ventral ACC and right
30.8 ± SOCPHOB, OCD, OFC; no significant
9.21 PTSD, SOM, ED, differences in rs-FC were
CPD: 21 SUD, IED, AVPD, obtained between
(21/0) DEPD, OCPD, medicated BPD vs non-
PAPD, DPD, PDP, medicated BPD
31.48 ± STPD, SZPD
11.8
HC: 39
(39/0)

28.67 ±
10.70
Balducci Cocaine Seed-based: right and AD (10), AP (5), MDD, DYS, PD, BIS, DERS, ASI, BPD (− ) cocaine (+): ↑ 12 significant
et al. (+) BPD left amygdalae and AC (12), SOCPHOB, PHOB, CGI-BPD rs-FC in left amygdala- correlations between rs-
(2018) (+): 20 (5/ DMN (medial PFC, anxiolytics (2), PTSD, GAD, ADHD, medial PFC; BPD (+) FC and neuropsychiatric
15) PCC, medial temporal unknown (2) SUD, ETOH, cocaine (− ): ↓ rs-FC in assessments (BIS. DERS):
cortex, and left amygdala-medial
31 ± 7 rostrolateral PFC) PFC; BPD (− ) cocaine Negative correlation
Cocaine (− ): ↓ rs-FC in right between rs-FC of left
(− ) BPD amygdala-left insula amygdala-medial PFC
(+): 10 (5/ and impulsivity (BIS-11
5) total score);
Negative correlation
31 ± 6 between rs-FC of right
Cocaine amygdala-left insula and
(+) BPD non-planning impulsivity
(− ): 19 (1/ (BIS-11 subscale)
18)

31 ± 6
HC: 20 (2/
18)

32 ± 8
Das et al. BPD: 14 ICA: DMN, SN, CEN AD (15), AP NR CTQ, BIS, BDI, BPD vs BD: ↑ rs-FC: 1) Negative correlation
(2014) (14/0) (right and left FPN), (12), MS (12) DASS, DERS SN-right FPN, 2) SN- between rs-FC of SN-
ventromedial PFC, precuneus, 3) SN- right FPN and
32.00 ± precuneus ventromedial PFC, and impulsivity scores
7.86 4) DMN-precuneus; BD
BD: 16 vs HC: ↑ rs-FC: DMN-
(16/0) precuneus and SN-
ventromedial PFC (did
35.63 ± not survive multiple
10.71 comparisons); BPD vs
HC: 13 HC: ↓ rs-FC: SN-
(13/0) precuneus and SN-right
FPN (did not survive
31.15 ± multiple comparisons)
11.07
Doll et al. BPD: 14 ICA: DMN, SN (insular AD (11), AP (8), MDD, PTSD, SUD, BDI, HAM-D, BSL- BPD vs HC: ↑ rs-FC NR
(2013) (13/1) network [insular MS (1), ETOH, ED, ADHD, 23, GAF within DMN, SN, and
cortex, amygdalae, anxiolytics (3), SOM CEN: ACC, PCC, medical
30.4 hippocampus]), CEN PS (1) frontal gyrus, bilateral
HC: 16 superior parietal lobe,
(15/1) posterior insula; ↓ rs-FC
within posterior SN, left
34.0 CEN: right hippocampus,
left superior frontal;
shifted rs-FC from CEN to
SN in BPD vs HC
Duque- BPD: 18 Seed-based: medial AD (17), AP (8), MDD, SOCPHOB, CTQ, MASC BPD vs HC: ↓ rs-FC Negative correlation
Alarcón (18/1) PFC, ACC, precuneus, AC (3), PHOB, PD, GAD, between limbic regions between rs-FC of the
et al. inferior MTG, anxiolytics (3) PTSD that play a role in MTG with primary motor
(2019) 31.17 ± amygdala, anterior and emotional and affective cortex, supplementary
9.5 posterior insular cortex regulation and social motor, and ACC and
responses; ↑ rs-FC emotional abuse score
between the medial PFC (did not survive multiple
(continued on next page)

338
M. Shafie et al. Journal of Affective Disorders 341 (2023) 335–345

Table 1 (continued )
Author Subjects: Rs-FC analysis Medications in Comorbidities Neuropsychologic Findings Significant associations
(year) Number BPD assessments between rs-FC and
(F/M) age neuropsychologic
[mean ± assessments
SD]

HC: 16 and the left superior comparisons); positive

(16/1) parietal lobe; no correlation between rs-
significant differences in FC of the left medial PFC
32.80 ± the FC were obtained with nucleus accumbens,
8.6 between non-depressed caudate, putamen
vs depressed BPD subcallosal cortex, and
OFC and sexual abuse
score
Krause-Utz BPD: 20 Seed-based: bilateral Free of PTSD, MDD, ED, BSL-95, CTQ, DES, BPD vs HC: ↑ rs-FC Positive correlation
et al. (20/0) amygdalae (within medication SUD, SOCPHOB, BDI, BIS, DERS, between the amygdala between rs-FC of left
(2014) medial temporal lobe within the last PHOB, PD AIS, PDS and the dorsal insula, amygdala–right
29.55 ± network), dorsal ACC 14 days (in case lateral OFC and putamen dorsolateral PFC and DES
7.74 (within SN), ventral of fluoxetine 28 ↓ negative rs-FC between scores; negative
HC: 17 ACC (within DMN) days) the dorsal ACC and PCC correlation between rs-
(17/0) and dorsomedial PFC/ FC of left amygdala with
paracingulate cortex a cluster in the occipital
27.53 ± ↑ negative rs-FC between lobe including the
8.58 left ventral ACC and area lingual gyrus,
V1 of the occipital intracalcarine cortex and
cortex, lingual gyrus, and fusiform gyrus and DES
cuneus scores
Lei et al. BPD: 40 ALFF and ReHo Treatment- None CES-D, STAI, ASQ, BPD vs HC: ↓ ALFF and NR
(2017) (20/20) Seed-based: right PCC naïve CTQ, AIS, BIS ReHo in the right PCC
and adjacent and adjacent precuneus;
25.23 ± precuneus BPD vs HC: ↑ rs-FC of the
3.17 right PCC with the left
HC: 35 MTG/ITG, the left
(15/20) fusiform gyrus, the left
medial SFG, and the
24.77 ± right MFG.
1.19 ↑ rs-FC of right
precuneus with the left
SFG, the left SMA, the
left ITG, the bilateral
MTG, the bilateral MCG,
the right triangular part
IFG, and the right orbital
part of IFG
Lei et al. BPD: 43 DC and fALFF Treatment- None CES-D, STAI, ASQ, BPD vs HC: ↑ DC in the Positive correlation
(2018) (21/22) Seed-based: left MTG naïve CTQ, AIS left MTG, the left between DC values of the
and right precuneus supramarginal gyrus and left precuneus and
25.19 ± adjacent angular gyrus, insecure attachment
3.06 the left inferior occipital scores; negative
HC: 39 gyrus, and the bilateral correlation between
(23/16) precuneus; ↓ fALFF in the fALFF values of the right
right PCC, partly PCC and insecure
24.79 ± extending to the adjacent attachment scores
1.19 precuneus; ↑ fALFF in the
left MTG, right lingual
gyrus, and bilateral
cuneus
↓ coupling strengths in
the left MTG and right
precuneus
Lei et al. BPD: 50 Seed-based: bilateral Treatment- None CES-D, STAI, BPD vs HC: ↑ rs-FC of the Negative correlation
(2019) (25/25) ACC naïve CERQ, PDQ ACC with the right MFG; between FC between the
↓ rs-FC with the left left ACC and corpus
25.33 ± MTG; ↓ rs-FC between collosum and CERQ-
2.93 both ACC sides and the negative subscale scores;
HC: 54 corpus collosum negative correlation
(33/21) between rs-FC between
the left ACC and right
24.83 ± MFG and CES-D scores
1.37
Quattrini BPD: 21 ICA: DMN, SN, CEN AD (7), AP (10), MDD, ED, PD, NPD, BIS, STAI, STAXI, BPD vs HC: ↓ rs-FC in the Positive correlation
et al. (14/7) MS (11), DEPD DERS, IIP, SCL-90- DMN, SN, and left CEN between rs-FC within the
(2019) anxiolytics (12) R, TAS (only DMN and SN DMN and anger-state
35 ± 9 survive correction for intensity and expression
HC: 14 (5/ multiple comparison); (STAXI-2/state anger
9) ↓rs-FC of the right PCC subscale), and
within DMN interpersonal aggression
38 ± 7 (IIP/aggression
(continued on next page)

339
M. Shafie et al. Journal of Affective Disorders 341 (2023) 335–345

Table 1 (continued )
Author Subjects: Rs-FC analysis Medications in Comorbidities Neuropsychologic Findings Significant associations
(year) Number BPD assessments between rs-FC and
(F/M) age neuropsychologic
[mean ± assessments
SD]

subscale); positive
correlation between rs-
FC of right PCC within
the DMN and expression
of verbal physical anger
(STAXI-2/anger
expression-out); positive
correlation between rs-
FC within the SN and
metacognitive abilities;
negative correlation
between rs-FC within the
SN and interpersonal
aggression
Salvador BPD: 60 GBC and ALFF NR NR DIB-R, BSL-23 BPD vs HC: ↑ GBC in the Positive correlation
et al. (60/0) Seed-based: left pregenual ACC, ↓ GBC in between ALFF of left
(2016) hippocampus, the right temporal lobe; ↑ hippocampus, amygdala,
32.12 ± amygdala ALFF in left and left putamen and
7.16 hippocampus, amygdala, BSL-23 (did not survive
HC: 60 and left putamen, ↓ ALFF multiple comparisons);
(60/0) in the occipital lobes, negative correlation
right precuneus, and between GBC of right
33.73 ± dorsal PCC; ↑ rs-FC of left temporal lobe and DIB-R
12.8 hippocampus and (did not survive multiple
amygdala with ACC comparisons)
Sarkheil BPD: 26 ICC AD (15), AP (4) MDD, ED BSL-95, BDI BPD vs HC: ↑ rs-FC Negative correlation
et al. (26/26) Seed-based: caudate caudate with ACC, between rs-FC of caudate
(2020) nucleus, left insula, medial PFC, with bilateral insula and
25.17 ± VAA (within the right paracingulate gyrus, cuneus/right lingual
7.28 Brodmann area 17) SMA, and left and right gyrus and BSL-95;
HC: 26 ventral striatum; ↑ rs-FC negative correlation
(26/26) left insular with bilateral between rs-FC of left
OFC, bilateral inferior insula with the bilateral
24.89 ± parietal lobule, mid- striatum and BSL-95
2.80 cingulate cortex
extending to dorsal ACC,
and the right precentral
gyrus
Traynor BPD: 45 Seed-based: NR AVPD, DEPD, BSL-23, LPFS BPD vs HC:↓rs-FC of the Positive correlation
et al. (35/10) frontolimbic network, OCPD, PPD, STPD, left anterior between rs-FC within
(2021) DMN, SN, CEN SZPD, NPD, HPD, parahippocampal gyrus frontolimbic network
27.5 ± 7.9 ASPD with the precuneus/PCC; and LPFS;
HC: 29 ↑ rs-FC within the TPJ (i.
(25/4) e., the left STS with the
bilateral angular gyrus),
25.6 ± 5.9 and of the left STS with
the dorsomedial PFC and
ventromedial PFC; ↑ rs-
FC of the dorsolateral
PFC with three closely
interconnected limbic
areas: anterior
parahippocampal gyrus,
amygdala, and
hippocampus
Wagner BPD: 33 Seed-based Free of ADHD, SUD, ETOH BDI, STAI, BIS, BPD vs HC: ↑ rs-FC of the NR
et al. (33/0) (neurotransmitter psychotropic WURS, BPAQ, locus coeruleus with
(2018) producing sites): locus medication for ADHD-CL only the dorsal ACC
26.7 ± 6.4 coeruleus, upper raphe at least two (before and after
HC: 33 nuclei (DRN and NCS), weeks prior to controlling for the
(33/0) midbrain nuclei (VTA study presence of ADHD
and SNpc) symptoms); ↑ rs-FC of the
26.4 ± NCS with the left
6.20 frontopolar cortex (after
controlling for
aggression); anti-
correlation pf the SNpc
to the left dorsolateral
PFC
Wolf et al. BPD: 17 ICA: DMN, CEN, right AD (n = 14), AP MDD, ED, SUD, BIS, BDI, HAM-D, BPD vs HC: ↓ rs-FC in the Positive correlation
(2011) (17/0) FPN, left FPN (n = 10), MS (n ETOH, ANX BSL-23, DTS left cuneus within DMN; between rs-FC of the
= 11) ↑ rs-FC in the left FPC frontopolar cortex and
(continued on next page)

340
M. Shafie et al. Journal of Affective Disorders 341 (2023) 335–345

Table 1 (continued )
Author Subjects: Rs-FC analysis Medications in Comorbidities Neuropsychologic Findings Significant associations
(year) Number BPD assessments between rs-FC and
(F/M) age neuropsychologic
[mean ± assessments
SD]

28.6 ± 8.0 and the left insula within BIS and BSL-23; positive
HC: 16 DMN; ↓ rs-FC in the left correlation between rs-
(16/0) inferior parietal lobule FC of the insula and DTS;
and the right MTG within negative correlation
27.2 ± 7.3 CEN between rs-FC of the
cuneus and DTS

AC: anticonvulsant, ACC: anterior cingulate cortex, AD: antidepressants, ADHD-CL: attention deficit hyperactivity disorder checklist, AGOPHOB: agoraphobia, AIS:
Affect Intensity Scale, ALFF: amplitude of low frequency fluctuations, AP: antipsychotics, ASI: addiction severity index, ASPD: antisocial personality disorder, ASQ:
Attachment Style Questionnaire, AVPD: avoidant personality disorder, BD: bipolar disorder, BDI: Beck depression inventory, BIS: Barratt impulsiveness scale, BPAQ:
Buss-Perry aggression questionnaire, BPD: borderline personality disorder, BPD-CL: BPD checklist, BPDSI: BPD severity index, BSI: brief symptom inventory, BSL:
borderline symptom list, CEN: central executive network, CPD: cluster C personality disorder, CERQ, cognitive emotion regulation questionnaire, CES-D, center for
epidemiologic studies depression scale, CGI-BPD: clinical global impression scale for BPD, CTQ: childhood trauma questionnaire, DASS: depression anxiety stress
scales, DC: degree centrality, DEPD: dependent personality disorder, DERS: difficulties in emotion regulation scale, DES: dissociation experience scale, DIB: Diagnostic
Interview for Borderlines, DMN, default mode network, DPD: depressive personality disorder, DRN: dorsal raphe nucleus, DTS: dissociation tension scale, DYS:
dysthymic disorder, ED: eating disorder, ETOH: alcohol abuse, fALFF: fractional amplitude of low frequency fluctuations, FPN: frontoparietal network, GAD,
generalized anxiety disorder, GAF: Global Assessment of Functioning Score, GBC: global brain connectivity, HAM-D: Hamilton rating scale for depression, HC: healthy
control, HPD: histrionic personality disorder, ICA: independent component analysis, ICC: intrinsic connectivity contrast, IED: intermitted explosive disorder, IFG:
inferior frontal gyrus, IIP: inventory of interpersonal problems, ITEC: Interview for Trauma Events in Childhood, ITG: inferior temporal gyrus, LPFS: levels of per­
sonality functioning scale, MASC: movie for the assessment of social cognition, MCG: middle cingulate gyri, MDD: major depressive disorder, MFG: middle frontal
gyrus, MS: mood stabilizers, MTG: middle temporal gyrus, NCS: nucleus centralis superior, NPD: narcissistic personality disorder, NR: not reported, OCD: obsessive
compulsive disorder, OCPD: obsessive-compulsive personality disorder, OFC: orbitofrontal cortex, PAPD: passive-aggressive personality disorder, PCC: posterior
cingulate cortex, PD: panic disorder, PDQ: personality diagnostic questionnaire, PDS: post-traumatic stress diagnostic scale, PFC: prefrontal cortex, PHOB: specific
phobia, PPD: paranoid personality disorder, PS: psychostimulants, PTSD: post-traumatic stress disorder, ReHo: regional and homogeneity, rs-FC: resting-state func­
tional connectivity, SCL-90-R: symptoms check list-90-R, SD: standard deviation, SFG: superior frontal gyrus, SMA: supplementary motor area, SN: salience network,
SNpc: substantia nigra-pars compacta, SOCPHOB: social phobia, SOM: somatoform disorder, STAI: state-trait anxiety inventory, STAXI: state-trait anger expression
inventory, STPD: schizotypal personality disorder, STS: superior temporal sulcus, SUD: substance abuse disorder (other than alcohol), SZPD: schizoid personality
disorder, TAS: Toronto alexithymia scale, TPJ: temporoparietal junction, VAA: visual association area, VTA: ventral tegmental area, WURS: Wender-Utah rating scale.

comparisons (Das et al., 2014). On the other hand, an ICA investigation 3.4. Spontaneous brain activity and local connectivity studies
that focused on the triple network, including the DMN, SN, and CEN,
reported increased rs-FC within the DMN, SN, and CEN, in particular in In the study by Salvador et al. (2016), the authors employed global
the ACC, PCC, medial frontal gyrus, bilateral superior parietal lobe, and brain connectivity (GBC) and ALFF to assess levels of covariability be­
posterior insula, and decreased rs-FC in the right hippocampus and left tween brain areas and spontaneous brain activity. Compared to HC,
superior frontal gyrus within the posterior SN and left CEN. Moreover, increased GBC values were observed in a single cluster located in the
the inter-network analysis revealed shifted connectivity from CEN to SN pregenual anterior cingulate, while decreased GBC was noted in the
among BPD patients (Doll et al., 2013). Lastly, Sarkheil et al. (2020) right temporal lobe. Additionally, elevated ALFF was observed in a
employed a data-driven method by intrinsic connectivity contrast to cluster located in the left hippocampus and amygdala, and in a cluster
identify brain regions with deviant rs-FC patterns in BPD, revealing a located in the left putamen, while reduced ALFF was observed in the
stronger rs-FC in BPD compared to HC within the caudate nucleus, left occipital lobe, right precuneus, and the dorsal PCC. Significant corre­
insula and the visual association area within the right Brodmann area 17 lations were observed between ALFF and BSL, as well as between GBC
(Sarkheil et al., 2020). and DIB-R within the identified clusters (Salvador et al., 2016). In Lei
Various correlations between the rs-FC metrics of these networks and et al. (2017), which used ALFF and ReHo to evaluate spontaneous brain
neuropsychological variables emerged from the reviewed studies. Wolf function and local connectivity, decreased ALFF in the right PCC with
et al. (2011) reported positive correlations of rs-FC of the left FPC with adjacent precuneus and decreased ReHo in the right precuneus and
the BIS and BSL scores. In addition, the DTS also showed a positive adjacent PCC were detected (Lei et al., 2017). In a subsequent investi­
correlation with rs-FC of the left insula and a negative correlation with gation by the same research group, a combination of DC and fALFF was
rs-FC of the cuneus (Wolf et al., 2011). In Quattrini et al. (2019), the rs- used. The outcomes revealed that altered DC and fALFF values in BPD
FC of the DMN was positively correlated with anger-state intensity and were distributed mainly in the DMN, and DC-fALFF coupling strengths
expression evaluated with the STAXI-2 and with interpersonal aggres­ were decreased in the left middle temporal gyrus and right precuneus. In
sion (IIP), while the rs-FC of the right PCC was positively correlated with addition, the authors reported a positive correlation between insecure
aggression expression of verbal physical anger (STAXI-2). Furthermore, attachment scores and left precuneus DC and a negative correlation with
the authors reported a positive correlation between the rs-FC within the fALFF of the right PCC in the BPD group (Lei et al., 2018).
SN and metacognitive abilities (MAI) and a negative correlation be­
tween the rs-FC within the SN and interpersonal aggression (IPP) 3.5. Seed-based studies
(Quattrini et al., 2019). In Traynor et al. (2021), a correlation was found
between increased intralimbic rs-FC and the severity of dimensional self- The amygdala, ACC, PCC, and PFC were the most common seeds
interpersonal impairment. However, no correlations were found be­ used in the reviewed studies. Specifically, Baczkowski et al. (2017)
tween borderline symptom severity and patterns of rs-FC (Traynor et al., compared rs-FC in a sample of subjects with BPD, subjects with cluster C
2021). Lastly, a negative correlation was observed between reduced rs- personality disorder and HC before and after a cognitive emotion
FC of SN-CEN and the scores of impulse control difficulties (Das et al., regulation task. Using bilateral amygdala as seeds, the study showed
2014). decreased amygdala rs-FC with right ventral ACC and right orbitofrontal
cortex (OFC) at baseline in the BPD group. After the emotion regulation

341
M. Shafie et al. Journal of Affective Disorders 341 (2023) 335–345

task, there was an increase in amygdala rs-FC with the bilateral insula, raphe nuclei, ventral tegmental area, and substantia nigra. Their ana­
left caudate, bilateral middle frontal gyrus, right middle temporal gyrus, lyses showed a stronger rs-FC from the noradrenergic locus coeruleus to
left postcentral gyrus, right supplementary motor area, and left superior the ACC, and a stronger rs-FC between the serotonergic nucleus centralis
and inferior parietal lobule (Baczkowski et al., 2017). On the contrary, superior and the FPC. Furthermore, patients showed a diminished
the study performed by Krause-Utz et al. (2014), which explored rs-FC negative correlation from the substantia nigra-pars compacta to the left
patterns in treatment-naïve BPD patients with a history of interper­ DLPFC (Wagner et al., 2018).
sonal trauma, defined the amygdala and the dorsal and ventral ACC as
seeds to investigate rs-FC patterns. The authors found that, compared to 4. Discussion
HC, BPD patients showed increased rs-FC between the amygdala and the
insula, OFC, and putamen. Also, they reported diminished negative rs- In this review, we presented an updated overview of the current
FC between the dorsal ACC, PCC, and regions of the dorsomedial PFC, evidence on rs-fMRI abnormalities in BPD patients to further clarify the
as well as increased negative rs-FC between the ventral ACC and medial neurobiological bases and the neural mechanisms involved in this dis­
occipital regions. They also investigated the correlations between the order. The results of this study support the evidence that BPD is asso­
DES score and amygdala rs-FC patterns, and observed a correlation with ciated with various rs-FC disturbances within and between brain
the rs-FC between the amygdala and the DLPFC and fusiform gyrus networks, including the DMN, SN, CEN, and limbic regions.
(Krause-Utz et al., 2014a). In the study by Salvador et al. (2016), in In line with the findings of Visintin et al. (2016), several changes in
addition to the GFC and ALFF analyses, the authors selected the left the DMN were described. This evidence is not surprising, as previous
hippocampus-amygdala as seed and reported a hyperconnectivity of the neuroimaging evidence indicated that this network is involved in
left hippocampus-amygdala complex with a small cluster located in the various higher cognitive and behavioral functions, including internal
ACC (Salvador et al., 2016). Similarly, in an investigation by Sarkheil mentalizing, autobiographic memory, and monitoring of cognitive and
et al. (2020), in addition to a hypothesis-free whole-brain rs-FC study, affective states, abilities often found altered in BPD (Buckner et al.,
the caudate nucleus, left insula, and visual association area were used as 2008; Spreng et al., 2009). Notably, within the DMN, the PCC, hippo­
seeds for rs-FC analysis. The authors found increased rs-FC of the campus, and angular gyrus have been reported to be associated with
caudate with the ACC, mPFC, paracingulate gyrus, supplementary motor retrieval of episodic, autobiographical and semantic memory, and spe­
area, and left and right ventral striatum, as well as increased FC of left cific nodes in the mPFC seem to play a role in internal and external
insular with bilateral OFC, bilateral inferior parietal lobule, mid- cognitive processes, decision making, and emotion regulation (Menon,
cingulate cortex extending to dorsal ACC, and the right precentral 2011). Specifically, the PCC is a hub region included in several net­
gyrus. The authors further explored the relationship between rs-FC and works, such as the SN and the CEN, therefore, alterations in this area
the severity of BPD symptoms using the BSL questionnaire, and observed might partially explain some deficits, like the reflexive dysfunctions
a negative correlation between the rs-FC of the caudate and left insula commonly seen in BPD patients (Leech and Sharp, 2014). In addition,
and BPD symptoms severity (Sarkheil et al., 2020). previous studies also reported altered precuneus FC in BPD, which might
In the investigation by Balducci et al. (2018), the rs-FC of the lead to deficits in establishing and maintaining attention (Schultz and
amygdala and the DMN of subjects with BPD and cocaine dependence Lennert, 2009), which in turn could be related to dissociative symptoms
was explored. The analyses showed a significant interaction effect be­ in BPD. Although alterations in rs-FC of DMN regions were one of the
tween left amygdala and mPFC rs-FC, and between the right amygdala most replicated findings across studies, results were heterogeneous.
and left insula rs-FC. A number of correlations were also obtained be­ Indeed, while some investigations reported increased rs-FC within the
tween the significant clusters and BIS and DERS sub-scores (Balducci DMN (Doll et al., 2013; Krause-Utz et al., 2014a; Wolf et al., 2011),
et al., 2018). Duque-Alarcón et al. (2019) examined the effects of others observed a pattern of reduced rs-FC between DMN regions
childhood maltreatment on rs-FC in 18 BPD patients and 15 HC, defining (Quattrini et al., 2019; Traynor et al., 2021). These inconsistencies
the seeds based on previous BPD literature and regions associated with might be due to the heterogeneity in the clinical features of BPD and in
mentalization, located mainly in the DMN. The authors reported the methodologies of the reviewed studies.
increased rs-FC between the mPFC and the left superior parietal lobe and Moreover, the rs-FC changes within the SN highlighted by the
hypoconnectivity of limbic regions. By exploring the association be­ reviewed studies might also explain some of the behavioral and cogni­
tween clinical measures and rs-FC parameters, they reported a positive tive features of BPD. SN is involved in metacognitive processes and some
correlation between sexual abuse score and rs-FC of the left mPFC with complex functions, including communication, social behavior, and self-
nucleus accumbens, striatum, and OFC (Duque-Alarcón et al., 2019). In awareness, through the integration of somatosensory, emotional, and
Lei et al. (2017), ALFF and ReHo analyses led to the selection of the right cognitive information (Mauchnik and Schmahl, 2010; Menon and
PCC and adjacent precuneus as seeds. With the seed placed in the right Uddin, 2010). Notably, the insula, a key node of the SN, seems to be
PCC, BPD patients showed higher rs-FC with the left middle/inferior implicated in somatosensory perception and monitoring of affective
temporal gyrus(MTG/ITG), left fusiform gyrus, left superior frontal states (Ploghaus et al., 1999). Interestingly, a study on BPD reported
gyrus (SFG), and right middle frontal gyrus (MFG), while with the seed abnormal activation of the insula in BPD patients during affective and
placed in right precuneus, BPD patients presented higher rs-FC with the pain regulation, ultimately suggesting a probable contribution of this
left SFG, left supplementary motor area, left ITG, bilateral MTG, bilat­ region to emotional regulation, pain perception, and dissociative pro­
eral middle cingulate gyrus, and right inferior frontal gyrus (IFG) (Lei cesses (Niedtfeld et al., 2010). Therefore, this could represent a possible
et al., 2017). Furthermore, in an integrated study of rs-FC and proba­ neural mechanism underlying the self-injury behaviors commonly seen
bilistic fiber tracking to explore rs-FC of the ACC, the same research in these patients (Ludäscher et al., 2010). Similarly, the ACC, which is a
group showed increased rs-FC between ACC and right MFG, as well as major component of SN, is implicated in the monitoring of behavior,
decreased rs-FC with the left MTG in a group of BPD patients compared decision-making, and processes related to self-evaluation in social con­
to HC. Moreover, lower rs-FC between ACC and the corpus callosum was texts (Lockwood and Wittmann, 2018; Rüsch et al., 2010). Some in­
observed in BPD. Correlation analyses showed that CERQ-negative vestigations also reported the role of ACC in switching between
subscale scores correlated negatively with the rs-FC between the left networks and reallocating cognitive resources, suggesting its involve­
ACC and corpus callosum and depressive symptoms evaluated with the ment in a variety of cognitively demanding tasks (Menon and Uddin,
Epidemiologic Studies Depression Scale (CES-D) correlated negatively 2010; Sridharan et al., 2008). Moreover, previous imaging studies
with the rs-FC between the left ACC and right MFG (Lei et al., 2019). exploring neural correlates of BPD through task-based activation, re­
Finally, Wagner et al. (2018) investigated the rs-FC of neurotransmitter- ported aberrant activity within the SN, specifically in the ACC and PCC
producing areas in patients with BPD, including the locus coeruleus, (Holtmann et al., 2013; King-Casas et al., 2008; Koenigsberg et al., 2009;

342
M. Shafie et al. Journal of Affective Disorders 341 (2023) 335–345

Minzenberg et al., 2007), which may explain some symptoms of BPD, could underlie impulsivity, one of the key features of BPD (Haber and
including emotional and social dysfunctions. Knutson, 2010). The amygdala is also connected with the PCC, which is
Furthermore, the rs-FC of CEN, which is known to subserve emotion a key node of the DMN and has been implicated in self-referential pro­
regulation (Buhle et al., 2014), was found to be altered in BPD subjects, cessing and autobiographical memory (Buckner and Carroll, 2007),
further supporting that emotion dysregulation may partially be a abilities found to be impaired in BPD patients (Baczkowski et al., 2017;
consequence of disturbed communication of brain regions within the Veer et al., 2011). Moreover, the reviewed studies consistently reported
CEN. Also, considering the central role of CEN in decision-making alterations in the rs-FC of the hippocampus (Doll et al., 2013; Salvador
through integrating information from the environment with internal et al., 2016; Traynor et al., 2021). As the hippocampus is mainly
representations (Vincent et al., 2008), abnormalities in rs-FC of the CEN implicated in episodic and autobiographical memory, disturbed hippo­
might underlie the impulsive behaviors often observed in BPD patients. campal rs-FC might be associated with the dysfunction of self-referential
In addition, the decreased prefrontal and temporoparietal rs-FC processes in BPD (Lanius et al., 2010). The role of the hippocampus in
within the attention network observed by the reviewed studies could BPD is also supported by studies reporting that hippocampal rs-FC is
be also related to attention deficits and learning difficulties in BPD pa­ associated with impulse control (Schreiner et al., 2019) and the fear
tients, since it has been hypothesized that these symptoms are associated processing (Kruse et al., 2018). Furthermore, the involvement of the
with a disturbed communication of prefrontal and posterior tempor­ hippocampus in interpersonal processing could also underlie the inter­
oparietal attentional systems (Smith et al., 2009). Additionally, the PFC personal and social disturbances often observed in BPD (Buchheim et al.,
is involved in a variety of cognitive and social functions, including 2008).
behavioral inhibition, affective regulation, intentional thought process, The results of the reviewed studies should be interpreted considering
and interpersonal connections (Den Ouden et al., 2005; Saxe, 2006), all some limitations, which may limit the generalizability of the findings.
abilities reported to be impaired in BPD subjects (Goethals et al., 2005; First, due to the complex nature of BPD, our current knowledge of the
Juengling et al., 2003; New et al., 2007). disorder is still limited, and further neuroimaging studies are warranted
Interestingly, most of the reviewed studies supported the idea of an to generalize the putative neurobiological basis of the disorder. Second,
aberrant interaction between DMN, SN, and CEN in BPD. This is in line several investigations exclusively included female BPD patients or
with the triple network model, which suggests that several neuropsy­ included a relatively small sample size, which could therefore negatively
chiatric disorders are associated with the aberrant organization and affect the results. Third, many patients with BPD presented other psy­
functioning of these three networks. Briefly, according to this model, the chiatric comorbidities, which may affect the consistency of the results
CEN shows a hyperactivation during cognitive or affective tasks, the across different studies. Fourth, the reviewed studies included clinically
DMN shows decreased activation, and the SN acts as an interface be­ heterogenous subjects taking psychotropic medications or with a history
tween the DMN and the CEN (Menon, 2011). Crucially, deficits in of substance and alcohol abuse, which may have a direct effect on fMRI
engagement and disengagement of these three core networks play a findings. Therefore, further studies on subjects without psychiatric
significant role in several neuropsychiatric disorders, including affective comorbidities, with similar illness duration and psychotropic medica­
and psychotic disorders (Menon, 2011). Indeed, one of the most prom­ tions are needed to clarify whether the observed alterations in brain FC
inent neurobiological models of BPD indicates that impaired emotion are specifically associated with BPD. Finally, most of the studies had an
regulation processes, as well as other symptoms of BPD such as impul­ observational cross-sectional design and differed in terms of methodo­
sivity and social disturbances, may be associated with connectivity ab­ logical approaches and analyses used. Therefore, future more homoge­
normalities between these networks (Mauchnik and Schmahl, 2010). nous studies with longitudinal design are needed to support the current
This idea has been also suggested in other disorders, such as SCZ and knowledge.
MDD, in which specific psychopathological symptoms have been found In conclusion, this review provided an updated overview of the rs-FC
to be associated with changes in the interaction between DMN, SN, and patterns within and between brain networks in BPD, including the DMN,
CEN (Hamilton et al., 2011; Manoliu et al., 2013). SN, and CEN. Notably, although various alterations in the rs-FC of
Furthermore, the rs-FC deficits in subcortical regions within the amygdala, PFC, ACC, PCC, and hippocampus were reported in these
limbic network, especially amygdala and hippocampus, reported by the studies, these observations present some inconsistencies among studies.
reviewed studies (Baczkowski et al., 2017; Balducci et al., 2018; Doll To address this issue, further studies with multimodal methodologies
et al., 2013; Krause-Utz et al., 2014a; Salvador et al., 2016; Traynor and longitudinal designs with larger sample sizes are warranted to
et al., 2021) are in line with the results of a recent meta-analysis, clarify the neural activity and FC at rest of this disorder.
showing a dysfunction of the amygdala and ACC in BPD during the
processing of emotional stimuli (Degasperi et al., 2021), which might Role of the funding source
ultimately explain the behavioral, emotional, and cognitive deficits
often observed in BPD patients. Notably, the amygdala is strictly con­ GD was partially supported by grants from the Ministry of Health
nected to other limbic regions, including ACC and insula, and is pri­ (GR-2019-12369100). GC was supported by a grant from Fondazione
marily implicated in emotional responses (Rolls, 2019). In this regard, a Cassa di Risparmio di Padova e Rovigo (CARIPARO). The study was
number of studies reported increased rs-FC of the amygdala with other partially supported by the Italian Ministry of Health (ricerca corrente
structures (Balducci et al., 2018; Krause-Utz et al., 2014a), while some 2023 to CM).
studies reported hypoconnectivity (Baczkowski et al., 2017; Duque-
Alarcón et al., 2019). Thus, rs-FC alterations of the amygdala could CRediT authorship contribution statement
reflect the well-known clinical features of BPD, including affective hy­
perarousal and emotional reactions, which are the basis of dysfunctional Mahan Shafie: Writing – original draft, Conceptualization. Elnaz
BPD behavioral patterns (Baczkowski et al., 2017; Krause-Utz et al., Shahmohamadi: Writing – original draft. Giulia Cattarinussi: Writing
2014a; O’Neill and Frodl, 2012). Similarly, the amygdala is extensively – original draft. Hossein Sanjari Moghaddam: Writing – review &
connected with the mPFC and some studies suggested that the impaired editing. Shahin Akhondzadeh: Writing – review & editing. Fabio
inhibitory control of the mPFC on the amygdala could contribute to Sambataro: Writing – review & editing, Supervision. Chiara Moltrasio:
hyperactivity and the exaggerated response of the amygdala, which may Writing – review & editing. Giuseppe Delvecchio: Writing – review &
in turn explain the dysfunctions in the emotion regulation process in editing, Supervision.
BPD patients (Baczkowski et al., 2017; Kamphausen et al., 2013).
Additionally, the rs-FC of the amygdala with the OFC and putamen,
which play an important role in reward processing and reinforcement,

343
M. Shafie et al. Journal of Affective Disorders 341 (2023) 335–345

Declaration of competing interest Goethals, I., Audenaert, K., Jacobs, F., van den Eynde, F., Bernagie, K., Kolindou, A.,
Vervaet, M., Dierckx, R., van Heeringen, C., 2005. Brain perfusion SPECT in
impulsivity-related personality disorders. Behav. Brain Res. 157, 187–192.
None. Gratz, K.L., Roemer, L., 2004. Multidimensional assessment of emotion regulation and
dysregulation: development, factor structure, and initial validation of the difficulties
Acknowledgments in emotion regulation scale. J. Psychopathol. Behav. Assess. 26, 41–54.
Grottaroli, M., Delvecchio, G., Bressi, C., Moltrasio, C., Soares, J.C., Brambilla, P., 2020.
Microstructural white matter alterations in borderline personality disorder: a
None. minireview. J. Affect. Disord. 264, 249–255.
Haber, S.N., Knutson, B., 2010. The reward circuit: linking primate anatomy and human
imaging. Neuropsychopharmacology 35, 4–26.
Appendix A. Supplementary data Hamilton, J.P., Furman, D.J., Chang, C., Thomason, M.E., Dennis, E., Gotlib, I.H., 2011.
Default-mode and task-positive network activity in major depressive disorder:
Supplementary data to this article can be found online at https://doi. implications for adaptive and maladaptive rumination. Biol. Psychiatry 70, 327–333.
Holtmann, J., Herbort, M.C., Wüstenberg, T., Soch, J., Richter, S., Walter, H., Roepke, S.,
org/10.1016/j.jad.2023.09.001. Schott, B.H., 2013. Trait anxiety modulates fronto-limbic processing of emotional
interference in borderline personality disorder. Front. Hum. Neurosci. 7, 54.
References Juengling, F.D., Schmahl, C., Heßlinger, B., Ebert, D., Bremner, J.D., Gostomzyk, J.,
Bohus, M., Lieb, K., 2003. Positron emission tomography in female patients with
borderline personality disorder. J. Psychiatr. Res. 37, 109–115.
Andrews-Hanna, J.R., Smallwood, J., Spreng, R.N., 2014. The default network and self-
Kamphausen, S., Schröder, P., Maier, S., Bader, K., Feige, B., Kaller, C.P., Glauche, V.,
generated thought: component processes, dynamic control, and clinical relevance.
Ohlendorf, S., van Elst, L.T., Klöppel, S., 2013. Medial prefrontal dysfunction and
Ann. N. Y. Acad. Sci. 1316, 29–52.
prolonged amygdala response during instructed fear processing in borderline
Arntz, A., van den Hoorn, M., Cornelis, J., Verheul, R., van den Bosch, W.M., de Bie, A.J.,
personality disorder. World J. Biol. Psychiatry. 14, 307–318.
2003. Reliability and validity of the borderline personality disorder severity index.
Kelleher-Unger, I., Tajchman, Z., Chittano, G., Vilares, I., 2021. Meta-analysis of white
J. Personal. Disord. 17, 45–59.
matter diffusion tensor imaging alterations in borderline personality disorder.
Baczkowski, B.M., van Zutphen, L., Siep, N., Jacob, G.A., Domes, G., Maier, S.,
Psychiatry Res. Neuroimaging 307, 111205.
Sprenger, A., Senft, A., Willenborg, B., Tüscher, O., 2017. Deficient
King-Casas, B., Sharp, C., Lomax-Bream, L., Lohrenz, T., Fonagy, P., Montague, P.R.,
amygdala–prefrontal intrinsic connectivity after effortful emotion regulation in
2008. The rupture and repair of cooperation in borderline personality disorder.
borderline personality disorder. Eur. Arch. Psychiatry Clin. Neurosci. 267, 551–565.
Science 1979 (321), 806–810.
Bagby, R.M., Parker, J.D.A., Taylor, G.J., 1994. The twenty-item Toronto Alexithymia
Koenigsberg, H.W., Siever, L.J., Lee, H., Pizzarello, S., New, A.S., Goodman, M.,
Scale—I. Item selection and cross-validation of the factor structure. J. Psychosom.
Cheng, H., Flory, J., Prohovnik, I., 2009. Neural correlates of emotion processing in
Res. 38, 23–32.
borderline personality disorder. Psychiatry Res. Neuroimaging 172, 192–199.
Balducci, T., Gonzalez-Olvera, J.J., Angeles-Valdez, D., Espinoza-Luna, I., Garza-
Krause-Utz, A., Veer, I.M., Rombouts, S., Bohus, M., Schmahl, C., Elzinga, B.M., 2014a.
Villarreal, E.A., 2018. Borderline personality disorder with cocaine dependence:
Amygdala and anterior cingulate resting-state functional connectivity in borderline
impulsivity, emotional dysregulation and amygdala functional connectivity. Front.
personality disorder patients with a history of interpersonal trauma. Psychol. Med.
Psychol. 9, 328.
44, 2889–2901.
Barkham, M., Hardy, G.E., Startup, M., 1996. The IIP-32: a short version of the Inventory
Krause-Utz, Annegret, Winter, D., Niedtfeld, I., Schmahl, C., 2014b. The latest
of Interpersonal Problems. Br. J. Clin. Psychol. 35, 21–35.
neuroimaging findings in borderline personality disorder. Curr. Psychiatry Rep. 16,
Bernstein, E.M., Putnam, F.W., 1986. Development, Reliability, and Validity of a
1–13.
Dissociation Scale.
Kruse, O., León, I.T., Stalder, T., Stark, R., Klucken, T., 2018. Altered reward learning and
Bernstein, D.P., Fink, L., Handelsman, L., Foote, J., 1998. Childhood trauma
hippocampal connectivity following psychosocial stress. Neuroimage 171, 15–25.
questionnaire. In: Assessment of Family Violence: A Handbook for Researchers and
Lanius, R.A., Vermetten, E., Loewenstein, R.J., Brand, B., Schmahl, C., Bremner, J.D.,
Practitioners.
Spiegel, D., 2010. Emotion modulation in PTSD: clinical and neurobiological
Biswal, B., Yetkin, F.Z., Haughton, V.M., Hyde, J.S., 1995. Functional connectivity in the
evidence for a dissociative subtype. Am. J. Psychiatry 167, 640–647.
motor cortex of resting human brain using echo-planar MRI. Magn. Reson. Med. 34,
Leech, R., Sharp, D.J., 2014. The role of the posterior cingulate cortex in cognition and
537–541.
disease. Brain 137, 12–32.
Bohus, M., Limberger, M.F., Frank, U., Sender, I., Gratwohl, T., Stieglitz, R.-D., 2001.
Lei, X., Zhong, M., Liu, Y., Jin, X., Zhou, Q., Xi, C., Tan, C., Zhu, X., Yao, S., Yi, J., 2017.
Borderline Symptom List (BSL) [Database record]. APA PsycTests. https://doi.org/
A resting-state fMRI study in borderline personality disorder combining amplitude of
10.1037/t13890-000.
low frequency fluctuation, regional homogeneity and seed based functional
Buchheim, A., Erk, S., George, C., Kächele, H., Kircher, T., Martius, P., Pokorny, D.,
connectivity. J. Affect. Disord. 218, 299–305.
Ruchsow, M., Spitzer, M., Walter, H., 2008. Neural correlates of attachment trauma
Lei, X., Liao, Y., Zhong, M., Peng, W., Liu, Q., Yao, S., Zhu, X., Tan, C., Yi, J., 2018.
in borderline personality disorder: a functional magnetic resonance imaging study.
Functional connectivity density, local brain spontaneous activity, and their coupling
Psychiatry Res. Neuroimaging 163, 223–235.
strengths in patients with borderline personality disorder. Front. Psychol. 9, 342.
Buckner, R.L., Carroll, D.C., 2007. Self-projection and the brain. Trends Cogn. Sci. 11,
Lei, X., Zhong, M., Zhang, B., Yang, H., Peng, W., Liu, Q., Zhang, Y., Yao, S., Tan, C.,
49–57.
Yi, J., 2019. Structural and functional connectivity of the anterior cingulate cortex in
Buckner, R.L., Andrews-Hanna, J.R., Schacter, D.L., 2008. The brain’s default network:
patients with borderline personality disorder. Front. Neurosci. 13, 971.
anatomy, function, and relevance to disease. Ann. N. Y. Acad. Sci. 1124, 1–38.
Lieb, K., Zanarini, M.C., Schmahl, C., Linehan, M.M., Bohus, M., 2004. Borderline
Buhle, J.T., Silvers, J.A., Wager, T.D., Lopez, R., Onyemekwu, C., Kober, H., Weber, J.,
personality disorder. Lancet 364, 453–461.
Ochsner, K.N., 2014. Cognitive reappraisal of emotion: a meta-analysis of human
Lockwood, P.L., Wittmann, M.K., 2018. Ventral anterior cingulate cortex and social
neuroimaging studies. Cereb. Cortex 24, 2981–2990.
decision-making. Neurosci. Biobehav. Rev. 92, 187–191.
Das, P., Calhoun, V., Malhi, G.S., 2014. Bipolar and borderline patients display
Ludäscher, P., Valerius, G., Stiglmayr, C., Mauchnik, J., Lanius, R.A., Bohus, M.,
differential patterns of functional connectivity among resting state networks.
Schmahl, C., 2010. Pain sensitivity and neural processing during dissociative states
Neuroimage 98, 73–81.
in patients with borderline personality disorder with and without comorbid
Degasperi, G., Cristea, I.A., Di Rosa, E., Costa, C., Gentili, C., 2021. Parsing variability in
posttraumatic stress disorder: a pilot study. J. Psychiatry Neurosci. 35, 177–184.
borderline personality disorder: a meta-analysis of neuroimaging studies. Transl.
Manoliu, A., Riedl, V., Doll, A., Bäuml, J.G., Mühlau, M., Schwerthöffer, D., Scherr, M.,
Psychiatry 11, 314.
Zimmer, C., Förstl, H., Bäuml, J., 2013. Insular dysfunction reflects altered between-
Den Ouden, H.E.M., Frith, U., Frith, C., Blakemore, S.-J., 2005. Thinking about
network connectivity and severity of negative symptoms in schizophrenia during
intentions. Neuroimage 28, 787–796.
psychotic remission. Front. Hum. Neurosci. 7, 216.
Doll, A., Sorg, C., Manoliu, A., Wöller, A., Meng, C., Förstl, H., Zimmer, C.,
Martino, M., Magioncalda, P., Huang, Z., Conio, B., Piaggio, N., Duncan, N.W.,
Wohlschläger, A.M., Riedl, V., 2013. Shifted intrinsic connectivity of central
Rocchi, G., Escelsior, A., Marozzi, V., Wolff, A., 2016. Contrasting variability
executive and salience network in borderline personality disorder. Front. Hum.
patterns in the default mode and sensorimotor networks balance in bipolar
Neurosci. 7, 727.
depression and mania. Proc. Natl. Acad. Sci. 113, 4824–4829.
Duque-Alarcón, X., Alcalá-Lozano, R., González-Olvera, J.J., Garza-Villarreal, E.A.,
Mauchnik, J., Schmahl, C., 2010. The latest neuroimaging findings in borderline
Pellicer, F., 2019. Effects of childhood maltreatment on social cognition and brain
personality disorder. Curr. Psychiatry Rep. 12, 46–55.
functional connectivity in borderline personality disorder patients. Front. Psychol.
Menon, V., 2011. Large-scale brain networks and psychopathology: a unifying triple
10, 156.
network model. Trends Cogn. Sci. 15, 483–506.
Ellison, W.D., Rosenstein, L.K., Morgan, T.A., Zimmerman, M., 2018. Community and
Menon, V., Uddin, L.Q., 2010. Saliency, switching, attention and control: a network
clinical epidemiology of borderline personality disorder. Psychiatr. Clin. 41,
model of insula function. Brain Struct. Funct. 214, 655–667.
561–573.
Minzenberg, M.J., Fan, J., New, A.S., Tang, C.Y., Siever, L.J., 2007. Fronto-limbic
Filippi, M., Spinelli, E.G., Cividini, C., Agosta, F., 2019. Resting state dynamic functional
dysfunction in response to facial emotion in borderline personality disorder: an
connectivity in neurodegenerative conditions: a review of magnetic resonance
event-related fMRI study. Psychiatry Res. Neuroimaging 155, 231–243.
imaging findings. Front. Neurosci. 13, 657.
New, A.S., Hazlett, E.A., Buchsbaum, M.S., Goodman, M., Mitelman, S.A., Newmark, R.,
Garnefski, N., Kraaij, V., 2007. The cognitive emotion regulation questionnaire. Eur. J.
Trisdorfer, R., Haznedar, M.M., Koenigsberg, H.W., Flory, J., 2007.
Psychol. Assess. 23, 141–149.
Amygdala–prefrontal disconnection in borderline personality disorder.
Neuropsychopharmacology 32, 1629–1640.

344
M. Shafie et al. Journal of Affective Disorders 341 (2023) 335–345

Niedtfeld, I., Schulze, L., Kirsch, P., Herpertz, S.C., Bohus, M., Schmahl, C., 2010. Affect Schulze, L., Schmahl, C., Niedtfeld, I., 2016. Neural correlates of disturbed emotion
regulation and pain in borderline personality disorder: a possible link to the processing in borderline personality disorder: a multimodal meta-analysis. Biol.
understanding of self-injury. Biol. Psychiatry 68, 383–391. Psychiatry 79, 97–106.
Oathes, D.J., Patenaude, B., Schatzberg, A.F., Etkin, A., 2015. Neurobiological signatures Semerari, A., Cucchi, M., Dimaggio, G., Cavadini, D., Carcione, A., Battelli, V., Nicolò, G.,
of anxiety and depression in resting-state functional magnetic resonance imaging. Pedone, R., Siccardi, T., D’Angerio, S., Ronchi, P., Maffei, C., Smeraldi, E., 2012. The
Biol. Psychiatry 77, 385–393. https://doi.org/10.1016/j.biopsych.2014.08.006. development of the Metacognition Assessment interview: instrument description,
O’Neill, A., Frodl, T., 2012. Brain structure and function in borderline personality factor structure and reliability in a non-clinical sample. Psychiatry Res. 200,
disorder. Brain Struct. Funct. 217, 767–782. 890–895.
Patton, J.H., Stanford, M.S., Barratt, E.S., 1995. Factor structure of the Barratt Smith, S.M., Fox, P.T., Miller, K.L., Glahn, D.C., Fox, P.M., Mackay, C.E., Filippini, N.,
impulsiveness scale. J. Clin. Psychol. 51, 768–774. Watkins, K.E., Toro, R., Laird, A.R., 2009. Correspondence of the brain’s functional
Pérez, V., Barrachina, J., Soler, J., Pascual, J.C., Campins, M.J., Puigdemont, D., et al., architecture during activation and rest. Proc. Natl. Acad. Sci. 106, 13040–13045.
2007. The clinical global impression scale for borderline personality disorder Spreng, R.N., Mar, R.A., Kim, A.S.N., 2009. The common neural basis of autobiographical
patients (CGI-BPD): a scale sensible to detect changes. Actas Esp. Psiquiatr. 35, memory, prospection, navigation, theory of mind, and the default mode: a
229–235. quantitative meta-analysis. J. Cogn. Neurosci. 21, 489–510.
Ploghaus, A., Tracey, I., Gati, J.S., Clare, S., Menon, R.S., Matthews, P.M., Rawlins, J.N. Sridharan, D., Levitin, D.J., Menon, V., 2008. A critical role for the right fronto-insular
P., 1999. Dissociating pain from its anticipation in the human brain. Science 1979 cortex in switching between central-executive and default-mode networks. Proc.
(284), 1979–1981. Natl. Acad. Sci. 105, 12569–12574.
Prodoehl, J., Burciu, R.G., Vaillancourt, D.E., 2014. Resting state functional magnetic Stiglmayr, C., Schimke, P., Wagner, T., Braakmann, D., Schweiger, U., Sipos, V.,
resonance imaging in Parkinson’s disease. Curr. Neurol. Neurosci. Rep. 14, 1–8. Fydrich, T., Schmahl, C., Ebner-Priemer, U., Kleindienst, N., 2010. Development and
Quattrini, G., Pini, L., Pievani, M., Magni, L.R., Lanfredi, M., Ferrari, C., Boccardi, M., psychometric characteristics of the dissociation tension scale. J. Pers. Assess. 92,
Bignotti, S., Magnaldi, S., Cobelli, M., 2019. Abnormalities in functional connectivity 269–277.
in borderline personality disorder: correlations with metacognition and emotion Traynor, J.M., Wrege, J.S., Walter, M., Ruocco, A.C., 2021. Dimensional personality
dysregulation. Psychiatry Res. Neuroimaging 283, 118–124. impairment is associated with disruptions in intrinsic intralimbic functional
Rolls, E.T., 2019. The cingulate cortex and limbic systems for emotion, action, and connectivity. Psychol. Med. 1–11.
memory. Brain Struct. Funct. 224, 3001–3018. Van Zutphen, L., Siep, N., Jacob, G.A., Goebel, R., Arntz, A., 2015. Emotional sensitivity,
Ruocco, A.C., 2005. The neuropsychology of borderline personality disorder: a meta- emotion regulation and impulsivity in borderline personality disorder: a critical
analysis and review. Psychiatry Res. 137, 191–202. review of fMRI studies. Neurosci. Biobehav. Rev. 51, 64–76.
Rüsch, N., Bracht, T., Kreher, B.W., Schnell, S., Glauche, V., Il’yasov, K.A., Ebert, D., Veer, I.M., Oei, N.Y.L., Spinhoven, P., van Buchem, M.A., Elzinga, B.M., Rombouts, S.A.
Lieb, K., Hennig, J., Saur, D., 2010. Reduced interhemispheric structural R.B., 2011. Beyond acute social stress: increased functional connectivity between
connectivity between anterior cingulate cortices in borderline personality disorder. amygdala and cortical midline structures. Neuroimage 57, 1534–1541.
Psychiatry Res. Neuroimaging 181, 151–154. Vincent, J.L., Kahn, I., Snyder, A.Z., Raichle, M.E., Buckner, R.L., 2008. Evidence for a
Salvador, R., Vega, D., Pascual, J.C., Marco, J., Canales-Rodríguez, E.J., Aguilar, S., frontoparietal control system revealed by intrinsic functional connectivity.
Anguera, M., Soto, A., Ribas, J., Soler, J., 2016. Converging medial frontal resting J. Neurophysiol. 100, 3328–3342.
state and diffusion-based abnormalities in borderline personality disorder. Biol. Visintin, E., De Panfilis, C., Amore, M., Balestrieri, M., Wolf, R.C., Sambataro, F., 2016.
Psychiatry 79, 107–116. Mapping the brain correlates of borderline personality disorder: a functional
Sambataro, F., Blasi, G., Fazio, L., Caforio, G., Taurisano, P., Romano, R., di Giorgio, A., neuroimaging meta-analysis of resting state studies. J. Affect. Disord. 204, 262–269.
Gelao, B., lo Bianco, L., Papazacharias, A., 2010. Treatment with olanzapine is Wagner, G., Krause-Utz, A., de la Cruz, F., Schumann, A., Schmahl, C., Baer, K.-J., 2018.
associated with modulation of the default mode network in patients with Resting-state functional connectivity of neurotransmitter producing sites in female
schizophrenia. Neuropsychopharmacology 35, 904–912. patients with borderline personality disorder. Prog. Neuro-Psychopharmacol. Biol.
Sambataro, F., Wolf, N.D., Pennuto, M., Vasic, N., Wolf, R.C., 2014. Revisiting default Psychiatry 83, 118–126.
mode network function in major depression: evidence for disrupted subsystem Wells, G.A., Shea, B., O’Connell, D., Peterson, J., Welch, V., Losos, M., Tugwell, P., 2000.
connectivity. Psychol. Med. 44, 2041–2051. The Newcastle-Ottawa Scale (NOS) for Assessing the Quality of Onrandomized
Sarkheil, P., Ibrahim, C.N., Schneider, F., Mathiak, K., Klasen, M., 2020. Aberrant Studies in Meta-analyses.
functional connectivity profiles of brain regions associated with salience and reward Wolf, R.C., Sambataro, F., Vasic, N., Schmid, M., Thomann, P.A., Bienentreu, S.D.,
processing in female patients with borderline personality disorder. Brain Imaging Wolf, N.D., 2011. Aberrant connectivity of resting-state networks in borderline
Behav. 14, 485–495. personality disorder. J. Psychiatry Neurosci. 36, 402–411.
Saxe, R., 2006. Uniquely human social cognition. Curr. Opin. Neurobiol. 16, 235–239. Yang, X., Hu, L., Zeng, J., Tan, Y., Cheng, B., 2016. Default mode network and
Sayrs, J., Whiteside, U., 2006. Borderline personality disorder. In: Practitioner’s Guide to frontolimbic gray matter abnormalities in patients with borderline personality
Evidence-based Psychotherapy, pp. 151–160. disorder: a voxel-based meta-analysis. Sci. Rep. 6, 1–10.
Schreiner, M.W., Klimes-Dougan, B., Mueller, B.A., Nelson, K.J., Lim, K.O., Cullen, K.R., Yu, H., Meng, Y., Li, X., Zhang, C., Liang, S., Li, M., Li, Z., Guo, W., Wang, Q., Deng, W.,
2019. Neurocircuitry associated with symptom dimensions at baseline and with 2019. Common and distinct patterns of grey matter alterations in borderline
change in borderline personality disorder. Psychiatry Res. Neuroimaging 290, personality disorder and bipolar disorder: voxel-based meta-analysis. Br. J.
58–65. Psychiatry 215, 395–403.
Schultz, J., Lennert, T., 2009. BOLD signal in intraparietal sulcus covaries with Zanarini, M.C., Gunderson, J.G., Franken-burg, F.R., Chauncey, D.L., 1989. The revised
magnitude of implicitly driven attention shifts. Neuroimage 45, 1314–1328. diagnostic interview for borderlines: discriminating BPD from other Axis II disorders.
J. Personal. Disord. 3, 10–18.

345