Potential psychological processes linking mindfulness to borderline personality disorder symptoms: a systematic review and thematic synthesis

Abstract

Background

Borderline personality disorder (BPD) is characterized by pervasive instability in affect regulation, impulse control, and interpersonal functioning. Mindfulness has been increasingly examined as a key construct associated with BPD; however, existing research has primarily focused on direct associations rather than on the psychological processes that may account for this relationship.

Objective

This systematic review aimed to identify and synthesize empirical evidence on the potential psychological processes associated with the relationship between mindfulness and borderline personality disorder symptoms.

Methods

The review was conducted in accordance with PRISMA guidelines, and the protocol was registered with PROSPERO. A systematic search was performed across PsycINFO, PubMed, Scopus, Web of Science, EBSCO, and Google Scholar for studies published between 2010 and April 2026. Quantitative studies examining mindfulness in relation to BPD symptoms or related clinical outcomes were included. A total of 31 studies met the eligibility criteria and were included in the review. Data were extracted using a structured form, and findings were synthesized using a narrative and thematic approach due to methodological heterogeneity.

Results

The included studies identified several psychological processes that may contribute to the relationship between mindfulness and BPD symptoms, although the strength of evidence varied across study designs and analytical approaches. Emotion regulation emerged as the most consistently supported candidate pathway, alongside other potentially relevant processes, including mindfulness facets (e.g., nonjudging and acting with awareness), self-related processes (e.g., self-compassion), cognitive-control processes (e.g., decentering and attentional control), and interpersonal factors (e.g., attachment and rejection sensitivity). Behavioral outcomes such as nonsuicidal self-injury, aggression, and treatment dropout were also associated with these processes. The findings suggest that these candidate processes may operate as interconnected pathways rather than as isolated variables.

Conclusions

The literature suggests that studies examining the relationship between mindfulness and borderline personality disorder symptoms, including both direct associations and studies assessing potential psychological processes. However, existing studies remain fragmented and typically examine single pathways in isolation. This review provides a structured synthesis of these potential processes within a coherent conceptual framework, while highlighting the need for further longitudinal and process-focused research before firm mechanistic conclusions can be drawn.

Clinical trial number

Not applicable.

Introduction

Borderline personality disorder (BPD) is a complex and severe psychiatric condition characterized by pervasive instability in affect regulation, impulse control, interpersonal relationships, and self-image [37]. Core features of the disorder include emotional dysregulation, impulsivity, self-harming behaviors, and chronic feelings of emptiness, all of which contribute to significant functional impairment and increased risk of suicide [20, 39]. Given this clinical profile, recent research has increasingly focused on identifying modifiable psychological processes that may explain the development and maintenance of BPD symptoms and inform intervention strategies.

Although borderline personality disorder is traditionally diagnosed in clinical populations, borderline symptoms are increasingly examined across a broader spectrum that includes individuals with subclinical borderline traits. Individuals with subclinical features may experience difficulties in emotion regulation, impulsivity, interpersonal functioning, and self-image without meeting the full diagnostic criteria for BPD. Consistent with the diversity of samples included in the present review, studies have examined both clinically diagnosed BPD and individuals with elevated borderline traits, highlighting the importance of considering both populations when investigating the psychological processes associated with mindfulness and borderline pathology [28, 36].

Borderline personality disorder is estimated to affect approximately 0.7–2.7% of the general population, whereas substantially higher prevalence rates have been reported in psychiatric settings, reaching approximately 12% among psychiatric outpatients and 22% among psychiatric inpatients [22]. In contrast, subclinical borderline traits are considerably more common in community populations and may be present without meeting the full diagnostic criteria for BPD. Individuals with these traits often experience clinically relevant difficulties in emotion regulation, impulsivity, interpersonal functioning, and self-image. Accordingly, examining both clinically diagnosed BPD and subclinical borderline traits provides a broader understanding of the psychological processes associated with mindfulness across the continuum of borderline pathology.

Mindfulness has emerged as a key construct in this context. In the present review, mindfulness is conceptualized as a multidimensional construct that may be examined from different perspectives depending on the study design. Some studies assessed dispositional mindfulness, referring to an individual’s naturally occurring tendency to maintain present-moment awareness and nonjudgment in everyday life, whereas others investigated mindfulness meditation as an intervention designed to cultivate these capacities through structured training. In addition, several studies examined specific mindfulness facets, such as acting with awareness and nonjudging, rather than global mindfulness. Distinguishing these related but conceptually different approaches facilitates interpretation of the evidence synthesized in this review.

Conceptually, mindfulness refers to the capacity to attend to present-moment experiences with openness and nonjudgment [42]. Empirical evidence indicates that individuals with BPD exhibit significantly lower levels of mindfulness compared to healthy controls, particularly across facets such as acting with awareness and nonjudging [7, 29]. Furthermore, deficits in mindfulness have been consistently associated with core features of the disorder, including emotional instability, impulsivity, and maladaptive coping behaviors [2, 20]. Complementing this evidence, meta-analytic findings have demonstrated a moderate negative association between mindfulness and BPD symptoms across clinical and non-clinical samples [3].

Understanding the psychological processes associated with clinical improvement is essential for explaining how and why psychological interventions produce change. In psychotherapy research, a mediator is a variable that statistically accounts for part of the relationship between an intervention or predictor and an outcome, whereas a mechanism of change refers to the underlying process through which therapeutic change is produced [15, 16]. Establishing a mechanism therefore requires more than demonstrating a cross-sectional association or statistical mediation; it generally requires evidence of temporal precedence, experimental manipulation, specificity, and consistency across study designs [12, 15]. This distinction is important for mechanism-informed treatment development because identifying modifiable processes can help refine intervention components, clarify therapeutic targets, and determine which processes should be prioritized for specific clinical populations.

Within mindfulness-based approaches, proposed mechanisms of action span attentional, emotional, cognitive, bodily, and self-related domains. Hölzel et al. [13] identified attention regulation, body awareness, emotion regulation, and changes in self-perspective as central components of mindfulness-related change, whereas Tang et al. [41] emphasized attentional control, emotion regulation, and self-awareness as interrelated psychological and neurocognitive processes. These processes may operate sequentially or interactively, with changes in attention and present-moment awareness supporting subsequent improvements in emotional regulation, cognitive flexibility, self-processing, and behavioral responding. These theoretical models provide a basis for examining the potential processes through which mindfulness may be associated with BPD symptoms, while also underscoring the need to distinguish candidate processes from empirically established mechanisms of change. This process-based framework therefore supports the development of an integrative model that is responsive to the clinical characteristics of BPD without implying that all identified pathways have been causally established.

The candidate processes included in this review were selected because they represent theoretically plausible links between mindfulness and core features of BPD. Emotion regulation, impulse control, distress tolerance, and experiential avoidance were included because they reflect central difficulties in managing intense affect and behavioral responses. Decentering, attentional control, and specific mindfulness facets were included as cognitive and attentional capacities through which present-moment awareness may be associated with reduced automatic reactivity. Self-compassion, self-efficacy, and related self-processes were included because BPD is characterized by negative self-evaluation and unstable self-functioning. Attachment, rejection sensitivity, and related interpersonal processes were included because interpersonal instability is a defining feature of BPD. Behavioral outcomes such as self-harm, impulsive behavior, aggression-related responses, and treatment dropout were examined as clinically relevant outcomes potentially associated with these processes.

Building on this broader process-based framework, research on mindfulness and BPD has increasingly moved beyond the examination of direct associations toward the investigation of specific psychological pathways that may be associated with clinical symptoms and related outcomes. Among these mechanisms, emotion regulation has received the most empirical support. Longitudinal and intervention-based studies indicate that mindfulness-related skills and emotion-regulation capacities are associated with clinical outcomes, although the specific pattern of change varies across studies [14, 21]. Similarly, mediation analyses have highlighted the role of specific components such as impulse control and access to adaptive regulation strategies [1].

In addition to emotion regulation, a range of other mechanisms has been identified. Dialectical behavior therapy (DBT), a cognitive-behavioral treatment originally developed for individuals with borderline personality disorder, integrates mindfulness with skills training in emotion regulation, distress tolerance, and interpersonal effectiveness to address the core features of the disorder. Within this framework, distress tolerance has been shown to mediate treatment outcomes in DBT interventions [45], while cognitive processes such as decentering operate as sequential mediators linking mindfulness to reduced emotional dysregulation [33]. Attentional and executive control processes also appear to play a role, with effortful control partially explaining differences in mindfulness capacities among individuals with BPD [42]. Self-related processes represent another domain of interest, as self-compassion has been shown to mediate the relationship between mindfulness and both BPD severity and emotion dysregulation [31], and inner strengths have been linked to reduced self-harm through mediating pathways [39]. Furthermore, interpersonal and social-cognitive mechanisms—including attachment patterns and rejection sensitivity—have been associated with BPD outcomes through mindfulness-related processes [28, 43].

Despite this growing body of research, the literature remains conceptually and methodologically fragmented. Previous systematic reviews have primarily focused on the effectiveness of mindfulness-based interventions, particularly within DBT frameworks, demonstrating improvements in clinical outcomes but without systematically examining the mechanisms underlying these effects [17]. In contrast, meta-analytic studies have quantified the strength of the association between mindfulness and BPD symptoms but have not addressed the processes through which this relationship operates [3]. At the level of primary studies, most investigations examine single mediators—such as emotion regulation or self-compassion—in isolation, without integrating these processes into a broader explanatory framework [18, 19, 24].

In addition, substantial methodological variability further complicates synthesis across studies. Differences in sample characteristics, measurement instruments, and analytical approaches limit comparability, as some studies focus on clinically diagnosed BPD populations while others examine subclinical symptoms or mixed clinical groups [28, 36]. Similarly, outcome variables range from general symptom severity to specific behaviors such as nonsuicidal self-injury or treatment dropout [20, 40]. While these studies provide valuable insights, the absence of an integrative synthesis prevents a comprehensive understanding of how mindfulness operates across multiple psychological domains to influence BPD outcomes.

Accordingly, there is a clear need for a systematic and theoretically grounded synthesis that organizes the diverse psychological mechanisms identified in the literature into a coherent framework. Addressing this gap is essential for advancing theoretical understanding and for informing the development of mechanism-based interventions tailored to the complexity of BPD.

This review advances the existing literature by organizing findings on the relationship between mindfulness and borderline personality disorder symptoms into an integrative conceptual framework. Specifically, it synthesizes the potential psychological processes through which mindfulness may be associated with BPD-related outcomes, while distinguishing between direct associations and findings derived from process-focused analyses. Rather than treating emotion regulation, self-compassion, cognitive control, interpersonal vulnerability, and behavioral outcomes as isolated variables, the review conceptualizes them as potentially interrelated processes within a broader explanatory framework. In doing so, the review provides a structured typology of mechanisms and proposes a process-oriented model that clarifies how mindfulness may be linked to BPD symptoms through direct, indirect, sequential, and moderating pathways.

The contribution of this review is twofold. First, it systematically synthesizes the potential psychological processes examined in quantitative studies of mindfulness and BPD-related symptoms, while distinguishing direct-association studies from those employing mediation, moderation, longitudinal, or intervention-based analyses. Second, it organizes these findings into an integrative conceptual model of interacting intrapersonal, cognitive, interpersonal, and behavioral pathways. This model is intended as a theory-generating framework rather than as confirmation of established causal mechanisms.

Therefore, the present systematic review has two objectives. The primary objective is to identify and synthesize theoretically relevant psychological processes examined in quantitative studies of mindfulness and borderline personality disorder symptoms. The secondary objective is to integrate these findings into a coherent conceptual model that illustrates how these candidate processes may interact, while distinguishing evidence derived from direct associations from that obtained through process-focused analyses.

Method

This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A review protocol was developed before the study commenced to define the research objectives, eligibility criteria, search strategy, and procedures for data extraction and synthesis, thereby enhancing methodological rigor and transparency. The protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO; registration number: CRD420261385866).

Research question

The review question was structured using an observational framework consistent with the Population–Exposure–Outcome (PEO) approach, which is commonly applied in systematic reviews examining psychological processes and mechanisms. The population included individuals diagnosed with borderline personality disorder or exhibiting clinically relevant BPD symptoms across different settings. The exposure of interest was mindfulness, assessed either as a dispositional trait or through specific mindfulness facets using validated psychometric instruments. The outcomes of interest were borderline personality disorder symptoms and related clinical manifestations, with particular attention to psychological processes that may contribute to the relationship between mindfulness and these outcomes, including emotion regulation, self-compassion, distress tolerance, and cognitive control.

Based on this framework, the research question guiding the review was: What potential psychological processes have been examined in quantitative studies investigating the relationship between mindfulness and borderline personality disorder symptoms using validated psychometric measures?

Inclusion and exclusion criteria

Studies were eligible for inclusion if they met the following criteria: (1) publication as peer-reviewed journal articles employing a quantitative research design, including cross-sectional, correlational, longitudinal, quasi-experimental, or experimental designs; (2) publication in English between January 2010 and April 2026; (3) inclusion of participants diagnosed with borderline personality disorder or exhibiting clinically relevant BPD symptoms; (4) assessment of mindfulness as a dispositional trait, specific mindfulness facets, or through mindfulness-based interventions using psychometrically validated instruments; and (5) examination of the relationship between mindfulness and BPD symptoms or related clinical outcomes, including symptom severity, emotion dysregulation, self-harm, impulsivity, or theoretically relevant psychological processes. Both direct-association studies and studies formally testing mediation, moderation, sequential, longitudinal, or intervention-related pathways were eligible (e.g., emotion regulation, self-compassion, distress tolerance, cognitive control).

Studies were excluded if they relied exclusively on qualitative methods, lacked quantitative statistical analysis, did not empirically examine the relationship between mindfulness and borderline personality disorder or its associated psychological mechanisms, or did not employ validated psychometric measures of mindfulness. Additionally, dissertations, conference proceedings, book chapters, and other non–peer-reviewed sources were excluded to ensure methodological consistency and the inclusion of high-quality empirical evidence.

Search strategy and study selection

A systematic literature search was conducted to identify quantitative studies examining the relationship between mindfulness and borderline personality disorder symptoms, including both direct associations and studies assessing potential psychological processes. Searches were performed across multiple electronic databases, including PsycINFO, PubMed, Scopus, Web of Science, EBSCO, and Google Scholar.

The search strategy combined controlled vocabulary (e.g., MeSH terms) and free-text keywords related to the core constructs. The electronic search strategy was developed iteratively to maximize both sensitivity and specificity across databases. Boolean operators, truncation symbols, and database-specific indexing terms were adapted according to the requirements of each database. The primary search syntax combined mindfulness-related terms (“mindfulness,” “mindfulness meditation,” “mindfulness-based,” “mindful awareness”) with borderline personality disorder-related terms (“borderline personality disorder,” “BPD,” “emotionally unstable personality disorder”) using the operators AND and OR. An example of the search strategy applied in PubMed was as follows: (“mindfulness” OR “mindfulness-based” OR “mindfulness meditation” OR “mindful awareness”) AND (“borderline personality disorder” OR “BPD” OR “emotionally unstable personality disorder”). Searches were restricted to peer-reviewed studies published in English between January 2010 and April 2026. Where available, Medical Subject Headings (MeSH) and equivalent indexed subject terms were incorporated to improve retrieval accuracy. In addition, the reference lists of eligible studies and relevant review articles were manually screened to identify potentially eligible records not retrieved through the database searches. This citation-searching process did not identify any additional records (n = 0).

The search strategy was intentionally broad and did not require mechanism-specific terminology because potentially relevant studies may examine psychological processes without referring to them as mechanisms, mediation, or pathways in their titles, abstracts, or indexing terms. Eligibility and data extraction therefore captured both direct-association studies and studies employing mediation, moderation, sequential, longitudinal, or intervention-based analyses.

The search covered studies published between January 2010 and April 2026. The lower date limit was selected to capture the contemporary evidence base that developed alongside the increasing use of mindfulness-related constructs and mindfulness-based approaches in research on BPD, while maintaining a sufficiently focused and methodologically comparable body of literature. The end date reflected the most recent search conducted for the review. To enhance retrieval sensitivity, both title/abstract searches and indexed subject headings were used where available. Given the broad scope and variability of results in Google Scholar, screening was limited to the first thirty pages of results sorted by relevance, consistent with established practice in systematic reviews.

All retrieved records were imported into a reference management system, and duplicate records were identified and removed before screening. The study selection process was conducted by five independent reviewers with expertise in clinical psychology and systematic review methodology.

Screening was performed in two stages. First, titles and abstracts were independently screened by two reviewers against the predefined inclusion and exclusion criteria. Studies that did not examine mindfulness in relation to borderline personality disorder or did not include quantitative data were excluded at this stage. Discrepancies between reviewers were resolved through discussion, and where necessary, a third reviewer was consulted to reach consensus.

Second, full-text articles were assessed for eligibility using the same procedure. Studies were excluded at this stage if they did not employ validated psychometric measures of mindfulness, did not examine borderline personality disorder or its associated symptoms, or did not provide empirical analysis of the relationship between the constructs of interest. During full-text assessment, studies were classified according to the type of evidence they provided, including direct associations, mediation or moderation analyses, longitudinal relationships, sequential pathways, and intervention-related changes.

To ensure consistency in screening decisions, a calibration exercise was conducted before full screening, during which the research team jointly reviewed a subset of studies. Inter-rater reliability was assessed using Cohen’s kappa coefficient, with agreement levels exceeding 0.80, indicating high consistency across reviewers.

Following this process, 31 studies met all eligibility criteria and were included in the review. The complete study selection process, including records identified through database and citation searching and the reasons for full-text exclusion, is presented in the PRISMA 2020 flow diagram (Fig. 1).

Fig. 1.

Flow diagram

Data extraction and analysis

Data extraction was conducted systematically for all studies included in the final review using a structured data extraction form developed in advance. The form was designed to align with the objectives of this review, which focused on identifying and synthesizing potential psychological processes associated with the relationship between mindfulness and borderline personality disorder symptoms.

Before full data extraction, a calibration process was undertaken to ensure consistency in coding decisions and interpretation of study variables. The extraction form was piloted on a subset of included studies, and minor refinements were made to improve clarity and standardization. Any uncertainties or discrepancies that arose during the extraction process were resolved through careful review and comparison of the original study reports to ensure accuracy and consistency.

The following data were extracted from each study: (1) bibliographic information (author(s) and year of publication); (2) country in which the study was conducted, where reported; (3) sample characteristics, including clinical status (e.g., diagnosed BPD, BPD symptoms/traits, or mixed clinical samples) and sample size; (4) study design (e.g., cross-sectional, longitudinal, quasi-experimental, or randomized controlled trial); (5) mindfulness construct assessed (e.g., dispositional mindfulness, mindfulness facets, or intervention-based mindfulness) and the corresponding validated measurement instruments used; (6) psychological outcomes and BPD-related variables assessed (e.g., symptom severity, emotion dysregulation, self-harm, impulsivity, or related constructs); (7) theoretically relevant mediating or moderating variables, including emotional-regulatory, cognitive-attentional, self-related, and interpersonal processe; and (8) the main statistical findings reported by the original authors, including the direction and nature of associations or mediation effects.

Only results as reported by the original studies were extracted, without reanalysis or transformation of the data. Studies were classified according to whether they examined direct associations or formally tested indirect, mediating, moderating, sequential, longitudinal, or intervention-related pathways. Emotion regulation emerged as the most consistently examined candidate pathway linking mindfulness to BPD-related outcomes.

Due to substantial heterogeneity in study designs, populations, measurement instruments, and analytical approaches, a quantitative synthesis or meta-analysis was not undertaken. Instead, the extracted data were synthesized using a descriptive and thematic approach. Studies were grouped according to the type of psychological process examined (e.g., emotion-regulation, self-related, cognitive-control, and interpersonal processes), allowing for structured comparison across studies.

The key characteristics of the included studies are presented in Table 1, and the thematic synthesis of the findings is reported in the Results section.

Table 1.

Characteristics of included studies examining psychological processes associated with mindfulness and borderline personality disorder symptoms

Study (Author, Year)	Country	Sample Characteristics	N	Design	Analytical Pathway	Mindfulness /Intervention	BPD/ Outcome Measures	Processes Examined	Key Findings
Rueda et al. [28]	Spain	SUD patients with BPD traits	192	Cross-sectional	Simple mediation	Brief FFMQ	PAI-BPD scale	Attachment and dispositional mindfulness	Mindfulness fully mediated attachment–BPD severity
Amestoy et al. [1]	Canada	Students (BPD traits)	54	RCT	Multiple mediation within an RCT	KIMS	LPI, DERS, DTS	Impulse control, emotion-regulation strategies, acting with awareness, and distress tolerance	Impulse control mediated outcomes
Kresznerits et al. [20]	Hungary	Women with BPD	109	Cross-sectional	Direct association / multivariable prediction	Mindfulness skills	NSSI	Impulsivity, mindfulness skills, emotion-regulation strategies, and self-compassion	Mindfulness was not directly associated with NSSI frequency; impulsivity and lower self-compassion were principal correlates.
Schmid et al. [32]	Germany	Psychiatric inpatients with BPD	31	RCT	Intervention-related change within an RCT	Mindfulness scale	Emotion, self-efficacy	Mindfulness, affect, and self-efficacy outcomes	Short-term improvements in mindfulness, affect, and self-efficacy
Sripunya et al. [39]	Thailand	BPD symptoms	302	Cross-sectional	Multiple mediation	i-SBI	Self-harm	Meditation, equanimity, and the Five Precepts	Inner strengths mediated the relationship
Hood et al. [14]	Canada	BPD self-harm patients	240	RCT	Longitudinal association	KIMS	DERS, BSL	Mindfulness and emotion regulation	Primary analyses did not support mediation; supplemental analyses suggested ER mediation
Schmidt et al. [34]	Spain	BPD outpatients	75	Longitudinal	Longitudinal association	Mindfulness practice (DBT)	DERS	Mindfulness practice and emotion dysregulation	Mindfulness practice reduced emotion dysregulation
Casellas-Pujol et al. [5]	Spain	BPD outpatients	60	RCT	Intervention-related change within an RCT	mindfulness-related compassion practices	BSL-23, RDQ, WHOQOL-BREF, PHI	Self-compassion, self-criticism, well-being, and depressive symptoms	Contextual-compassion training improved well-being, and depressive symptoms
Bahmani et al. [2]	Iran	BPD vs. controls	149	Comparative	Direct group comparison	FFMQ	DERS, SCS	Emotion regulation, mindfulness facets, and self-compassion	BPD showed lower mindfulness and higher dysregulation
Lee et al. [21]	Australia	DBT participants (BPD traits)	57	Longitudinal	Longitudinal association	DBT mindfulness	Symptom scales	Mindfulness, emotion regulation, and interpersonal effectiveness	Improvement in emotion regulation was the only DBT skill uniquely associated with improvement in clinical outcomes.
Sampedro et al. [30]	Spain	BPD patients	38	Longitudinal (fMRI)	Intervention-related change	Mindfulness scales	DERS, BSL	Right anterior-insula functional connectivity and emotion dysregulation	Insula connectivity linked to ER deficits
Salgó et al. [29]	Hungary	BPD vs. controls	129	Case-control	Direct group comparison	FFMQ	CERQ, DERS, SCS	Emotion-regulation strategies, mindfulness facets, and self-compassion	BPD showed maladaptive ER patterns
Heath et al. [11]	USA	Outpatients	136	Longitudinal	Intervention-related change	DBT modules	ER scales	Emotion-regulation dimensions across DBT skills modules	Incremental improvement across modules
Schmidt et al. [33]	Spain	BPD patients	102	Intervention	Sequential mediation	FFMQ	BPD severity, DERS, EQ	Decentering and emotion dysregulation	Sequential mediation: mindfulness → decentering → ER
Stratton et al. [40]	Canada	BPD outpatients	42	RCT	Longitudinal association	Mindfulness scales	Dropout	Mindfulness and treatment dropout	Low mindfulness predicted dropout
Zeifman et al. [45]	Canada	BPD patients	84	RCT	Parallel mediation within an RCT	Mindfulness scales	Distress tolerance	Mindfulness and distress tolerance	Independent mediation effects
Seow et al. [36]	Australia	Mixed outpatient	102	Longitudinal	Longitudinal	DBT skills	Distress, BPD severity	DBT skill use, attitudes toward skills, and BPD symptom severity	Stronger effects in high BPD
Reyes-Ortega et al. [27]	Mexico	BPD patients	65	Comparative	Intervention-related change	Mindfulness skills	BPD severity	Experiential avoidance, emotion dysregulation, attachment, and awareness	Experiential avoidance decreased alongside improvements in clinical outcomes.
Mitchell et al. [24]	Australia	BPD adults	35	Longitudinal	Longitudinal association	FFMQ	Symptom severity	Mindfulness facets, particularly acting with awareness and nonjudging	Increases in nonjudging and acting with awareness were most strongly associated with clinical improvement.
Krantz et al. [19]	Canada	BPD self-harm	84	RCT	Mediation within an RCT	KIMS	NSSI	Mindfulness facets, particularly acceptance without judgment	Nonjudging mediated reduction in NSSI
Carmona i Farrés et al. [4]	Spain	BPD patients	70	RCT	Intervention-related change	FFMQ	BPD, DERS	Emotion dysregulation and impulsivity	Improved ER and reduced impulsivity
Domínguez-Clavé et al. [8]	Spain	Community sample including participants with borderline traits	45	Pre–post observational	Intervention-related change	FFMQ-SF	DERS, MSI-BPD	Changes in mindfulness facets, decentering, and emotion dysregulation	Reduced emotion dysregulation and changes in mindfulness-related capacities
Tortella-Feliu et al. [42]	Spain	BPD patients, meditators, and healthy non-meditators	630	Cross-sectional	Multiple mediation	FFMQ	EC scales	Effortful control: attentional, inhibitory, and activation control	EC partially mediated between-group differences in mindfulness facets
Scheibner et al. [31]	Germany	BPD + community	118	Cross-sectional	Mediation	FFMQ	DERS, SCS	Self-compassion and emotion dysregulation	Self-compassion mediated relationship
Selby et al. [35]	USA	Students with BPD	200	Cross-sectional	Multiple / reciprocal mediation	MAAS	SCID-II, RRS	Mindfulness and rumination	Bidirectional mediation between rumination and mindfulness
Soler et al. [38]	Spain	BPD patients	64	RCT	Intervention-related change	DBT mindfulness training	BIS-11	Impulsivity: response inhibition, delay of gratification, and time perception	Mindfulness training improved delay of gratification and time-perception indices
Elices et al. [9]	Spain	BPD patients	64	RCT	Intervention-related change	FFMQ	BPD, EQ	Decentering and mindfulness capacities	Increased decentering and reduced BPD symptoms
Kramer et al. [18]	Switzerland	BPD patients	41	RCT	Simple mediation within an RCT	DBT skills	Symptom scales	Assertive anger and emotion regulation	Assertive anger mediated improvement
Velotti et al. [43]	Italy	Clinical + community	468	Cross-sectional	Mediation	FFMQ	RSQ, DERS	Emotion dysregulation and rejection sensitivity	ER mediates rejection sensitivity
Wupperman et al. [44]	USA	Psychiatric inpatients	70	Cross-sectional	Simple mediation	MAAS	PAI-BOR	Mindfulness and dysregulated behavior	Mindfulness mediates BPD and dysregulated behavior
Peters et al. [26]	USA	Students with BPD	223	Cross-sectional	Moderation analysis	FFMQ	PAI-BOR	Nonjudging and present-moment awareness	Nonjudgment moderates awareness–BPD relationship

Note. BPD = Borderline Personality Disorder; BIS = Barratt Impulsiveness Scale; BSL = Borderline Symptom List; BSL-23 = 23-item Borderline Symptom List; CERQ = Cognitive Emotion Regulation Questionnaire; DBT = Dialectical Behavior Therapy; DERS = Difficulties in Emotion Regulation Scale; EC = effortful control; ER = emotion regulation; FFMQ = Five Facet Mindfulness Questionnaire; FFMQ-SF = Five Facet Mindfulness Questionnaire–Short Form; fMRI = functional magnetic resonance imaging; KIMS = Kentucky Inventory of Mindfulness Skills; MAAS = Mindful Attention Awareness Scale; MSI-BPD = McLean Screening Instrument for Borderline Personality Disorder; NSSI = nonsuicidal self-injury; PAI-BOR = Borderline Features Scale of the Personality Assessment Inventory; PAI-BPD = Borderline Personality Disorder scale of the Personality Assessment Inventory; PHI = Pemberton Happiness Index; RCT = randomized controlled trial; RDQ = Recovery Domains Questionnaire; RRS = Ruminative Responses Scale; RSQ = Rejection Sensitivity Questionnaire; SCID-II = Structured Clinical Interview for DSM-IV Axis II Personality Disorders; SCS = Self-Compassion Scale; SUD = substance use disorder; WHOQOL-BREF = World Health Organization Quality of Life–Brief Version

Quality assessment, risk of bias, and certainty of evidence

The methodological quality and risk of bias of the included studies were assessed using established, design-specific appraisal tools to ensure rigor and comparability across studies with different methodological designs.

Randomized controlled trials (RCTs) were evaluated using the Cochrane Risk of Bias Tool, which assesses bias across domains including the randomization process, deviations from intended interventions, missing outcome data, outcome measurement, and selective reporting. Observational studies, including cross-sectional and longitudinal designs, were assessed using the Newcastle-Ottawa Scale, which evaluates participant selection, comparability, and outcome/exposure assessment. For non-randomized intervention studies, risk of bias was additionally considered in line with principles from the ROBINS-I.

Each study was independently assessed by two reviewers, with discrepancies resolved through discussion or consultation with a third reviewer. Studies were categorized as having low, moderate, or high risk of bias based on the criteria of the respective tools. For studies assessed using the Newcastle-Ottawa Scale, scores of 7–9 were classified as low risk, 4–6 as moderate risk, and ≤ 3 as high risk of bias. For randomized studies, overall judgments followed the Cochrane framework (low risk, some concerns, high risk).

The results of the risk-of-bias assessment were incorporated into the synthesis process, with greater weight assigned to findings derived from studies with lower risk of bias, particularly longitudinal and randomized controlled designs.

In addition to methodological quality assessment, potential sources of publication and reporting bias were considered during the synthesis process. Given the absence of a meta-analytic design and the substantial heterogeneity across studies, formal statistical assessment of publication bias was not conducted. Nevertheless, several factors that may influence the available evidence base were taken into account. First, studies reporting significant associations between mindfulness-related processes and BPD outcomes may be more likely to be published than studies reporting null findings. Second, a considerable proportion of the included intervention studies were conducted within dialectical behavior therapy (DBT)-oriented frameworks, which may contribute to an overrepresentation of DBT-related conceptualizations of mindfulness and emotion regulation mechanisms. Third, variability in the reporting of mediation and moderation analyses across studies may have affected the comparability and interpretability of mechanistic findings. These considerations were taken into account when interpreting the consistency and strength of the evidence.

Given the methodological heterogeneity of the included studies, the diversity of study designs, and the absence of quantitative pooling, the certainty of evidence was interpreted using a qualitative confidence-based approach rather than a formal GRADE evaluation. The interpretation of evidence strength was informed by several considerations, including study design, consistency of findings across studies, methodological quality, replication of mechanisms across independent samples, and the degree to which the reported findings supported mechanistic inference. Consequently, statements regarding the relative strength of evidence were intended to reflect the overall consistency and convergence of the literature rather than formal quantitative certainty ratings.

Emotion regulation was supported by the strongest and most consistent body of evidence across the included studies. This rating reflects converging findings across randomized and longitudinal studies with relatively low risk of bias.

Evidence concerning mindfulness-facet processes (e.g., nonjudging and acting with awareness) was judged to be of moderate confidence, as evidence was generally consistent but largely derived from observational and mediation-based studies.

Evidence concerning self-related processes, particularly self-compassion, was also judged to be of moderate confidence, reflecting consistent associations across studies but limited longitudinal and experimental confirmation.

Evidence concerning cognitive-control processes, including decentering and attentional control, was judged to be of low-to-moderate confidence due to a smaller number of studies and variability in measurement approaches.

Evidence concerning interpersonal and social-cognitive processes (e.g., attachment and rejection sensitivity) was judged to be of low confidence, as evidence was primarily indirect and based on cross-sectional designs with limited causal inference.

Behavioral outcomes (e.g., nonsuicidal self-injury, aggression, treatment dropout) were rated as moderate certainty, given that multiple studies identified indirect pathways through established mechanisms such as emotion regulation and impulsivity.

Overall, the certainty of evidence indicates that Emotion regulation represents the most consistently supported candidate process, whereas other mechanisms require further longitudinal and experimental research to strengthen causal inference and clarify their role within an integrated framework.

A detailed summary of the risk of bias assessments is presented in Table 2.

Table 2.

Risk of bias assessment of included studies using standardized tools

Study	Design	Assessment Tool	Score / Judgment	Overall Risk of Bias
Rueda et al. [28]	Cross-sectional	Newcastle-Ottawa Scale	6/9	Moderate
Amestoy et al. [1]	RCT	Cochrane RoB 2	Low risk	Low
Kresznerits et al. [20]	Cross-sectional	Newcastle-Ottawa Scale	6/9	Moderate
Schmid et al. [32]	RCT	Cochrane RoB 2	Some concerns	Some concerns
Sripunya et al. [39]	Cross-sectional	Newcastle-Ottawa Scale	5/9	Moderate
Hood et al. [14]	RCT	Cochrane RoB 2	Low risk	Low
Casellas-Pujol et al. [5]	RCT	Cochrane RoB 2	Low risk	Low
Schmidt et al. [34]	Longitudinal	Newcastle-Ottawa Scale	7/9	Low
Bahmani et al. [2]	Comparative	Newcastle-Ottawa Scale	6/9	Moderate
Lee et al. [21]	Longitudinal	Newcastle-Ottawa Scale	7/9	Low
Sampedro et al. [30]	Longitudinal	Newcastle-Ottawa Scale	6/9	Moderate
Salgó et al. [29]	Case-control	Newcastle-Ottawa Scale	6/9	Moderate
Heath et al. [11]	Longitudinal	Newcastle-Ottawa Scale	7/9	Low
Schmidt et al. [33]	RCT secondary analysis	RoB 2	Some concern	Some concern
Stratton et al. [40]	RCT	Cochrane RoB 2	Some concerns	Some concerns
Zeifman et al. [45]	RCT	Cochrane RoB 2	Low risk	Low
Seow et al. [36]	Longitudinal	Newcastle-Ottawa Scale	6/9	Moderate
Reyes-Ortega et al. [27]	Comparative	ROBINS-I	Moderate	Moderate
Mitchell et al. [24]	Longitudinal	ROBINS-I	Moderate	Moderate
Krantz et al. [19]	RCT	Cochrane RoB 2	Low risk	Low
Carmona i Farrés et al. [4]	RCT	Cochrane RoB 2	Low risk	Low
Domínguez-Clavé et al. [8]	Observational	Newcastle-Ottawa Scale	5/9	Moderate
Tortella-Feliu et al. [42]	Cross-sectional	Newcastle-Ottawa Scale	6/9	Moderate
Scheibner et al. [31]	Cross-sectional	Newcastle-Ottawa Scale	6/9	Moderate
Selby et al. [35]	Cross-sectional	Newcastle-Ottawa Scale	5/9	Moderate
Soler et al. [38]	RCT	Cochrane RoB 2	Low risk	Low
Elices et al. [9]	RCT	Cochrane RoB 2	Low risk	Low
Kramer et al. [18]	RCT	Cochrane RoB 2	Some concerns	Some concerns
Velotti et al. [43]	Cross-sectional	Newcastle-Ottawa Scale	6/9	Moderate
Wupperman et al. [44]	Cross-sectional	Newcastle-Ottawa Scale	5/9	Moderate
Peters et al. [26]	Cross-sectional	Newcastle-Ottawa Scale	6/9	Moderate

Note. RCT = randomized controlled trial. NOS scores: 7–9 = low risk, 4–6 = moderate risk, ≤ 3 = high risk

Data synthesis

The synthesis was informed by the quality assessment of the included studies, with particular attention to study design and risk of bias. This review synthesized quantitative evidence from observational, longitudinal, and experimental studies examining the relationship between mindfulness and borderline personality disorder (BPD) symptoms, with particular emphasis on the psychological mechanisms underlying this association. Given that the included studies varied in design, including cross-sectional, longitudinal, and randomized controlled trials, the synthesis focused on reported associations, mediation pathways, and mechanism-based findings rather than solely on intervention outcomes.

Due to substantial heterogeneity in study populations, measurement instruments, study designs, and analytical approaches, a meta-analytic synthesis was not undertaken. The included studies employed a wide range of validated measures of mindfulness (e.g., FFMQ, KIMS) and assessed diverse BPD-related outcomes (e.g., symptom severity, emotion dysregulation, self-harm, impulsivity). In addition, the studies examined different mediating and moderating variables, often using distinct statistical models such as mediation, moderated mediation, and path analysis. This variability precluded the calculation of a pooled effect size.

Accordingly, a narrative and thematic synthesis approach was applied. Extracted data were organized into structured summary tables (Table 1) to facilitate comparison across studies. The synthesis focused on identifying and categorizing the psychological mechanisms examined in the literature, including emotion regulation processes, self-related processes (e.g., self-compassion), cognitive control mechanisms, and interpersonal factors. Particular attention was given to whether mindfulness was associated with BPD outcomes directly or indirectly through mediating pathways.

This approach allowed for a systematic comparison of findings across diverse methodological contexts while preserving the integrity of the original studies. Rather than aggregating results into a single statistical estimate, the synthesis highlights patterns in the mechanisms linking mindfulness to BPD symptoms and identifies areas of convergence and divergence in the empirical literature.

Results

The included studies were organized into five thematic pathways: emotion-regulation and impulse-control mechanisms, mindfulness-facet mechanisms, self-related mechanisms, interpersonal and social-cognitive mechanisms, and behavioral-outcome mechanisms. Some studies contributed to more than one pathway because they examined overlapping mechanisms.

Emotion-regulation and impulse-control mechanisms

Across the 31 included studies, emotion regulation emerged as the most consistently examined candidate pathway linking mindfulness to BPD-related outcomes. Evidence from comparative and cross-sectional designs indicated that individuals with BPD or elevated BPD symptoms exhibit lower levels of mindfulness alongside greater emotional dysregulation relative to control or comparison groups [2, 29, 43, 44]. This pattern was complemented by mixed-sample studies examining mindfulness-related capacities, effortful control, and DBT skill use in relation to emotional and psychological outcomes [36, 42].

Longitudinal and intervention-based studies generally associated mindfulness-related skills, mindfulness capacities, or DBT module participation with improvements in emotion regulation and related psychological outcomes [11, 21, 34]. In BPD-focused studies that assessed symptom outcomes, mindfulness-related or regulatory changes were also associated with reductions in BPD symptom severity [4, 14, 33, 45]. However, Hood et al. [14] did not confirm mediation in the primary analyses, although supplemental analyses suggested a possible emotion-regulation pathway. These findings were observed across different treatment formats, suggesting a comparatively consistent pathway across study designs.

At a more specific level, multiple studies identified distinct components of emotion regulation as mediators of this relationship. Schmidt et al. [33] demonstrated a sequential mediation pathway in which mindfulness training increased decentering, which in turn reduced emotion dysregulation and BPD symptoms; similar increases in decentering were also reported in intervention studies by Elices et al. [9]. Amestoy et al. [1] identified impulse control and access to adaptive emotion-regulation strategies as significant mediators, while Soler et al. [38] and Kresznerits et al. [20] highlighted the role of impulsivity as a closely related regulatory component influencing behavioral outcomes such as self-harm.

Additional evidence from observational, experimental, and neurobiological studies further supported the centrality of emotion regulation processes. Domínguez-Clavé et al. [8] reported changes following ayahuasca administration, whereas Schmid et al. [32] found short-term improvements after animal-assisted treatment; Sampedro et al. ; [30] linked insula connectivity to emotion-regulation deficits. Complementing these findings, Reyes-Ortega et al. [27] showed that reductions in experiential avoidance were linked to improved outcomes, indicating overlap between regulatory and avoidance-related processes.

Finally, Kramer et al. [18] extended the emotion-regulation pathway by demonstrating that changes in assertive anger—a specific regulatory expression—mediated symptom improvement following DBT-informed skills training. Collectively, these findings support emotion regulation as a central candidate pathway associated with mindfulness and BPD outcomes, including impulse control, decentering, experiential avoidance, and anger regulation.

Mindfulness-facet mechanisms

The findings indicate that mindfulness should not be conceptualized as a unitary construct, as multiple studies demonstrated that specific facets are differentially associated with BPD outcomes. Across the included studies, nonjudging and acting with awareness emerged as the most consistently relevant facets [19, 24, 26]. Krantz et al. [19] showed that acceptance without judgment mediated reductions in nonsuicidal self-injury, while Mitchell et al. [24] found that increases in acting with awareness and nonjudging were most strongly associated with reductions in BPD symptoms, depression, and psychological distress. In addition, Peters et al. [26] demonstrated that nonjudgment moderated the relationship between awareness and BPD symptoms, further highlighting the interactive role of specific facets.

Evidence from intervention and observational studies also supported facet-specific mechanisms. Domínguez-Clavé et al. [8] reported changes in several mindfulness facets following ayahuasca administration. Similarly, Amestoy et al. [1] identified acting with awareness as a significant mediator of treatment outcomes. Complementing these findings, Elices et al. [9] demonstrated increases in decentering following mindfulness-based intervention, suggesting overlap between specific facets and cognitive-control processes.

Additional studies indicated that facet-level differences are partly explained by underlying regulatory capacities. Tortella-Feliu et al. [42] showed that effortful control partially mediated differences in mindfulness facets between individuals with BPD and controls. Scheibner et al. [31] further demonstrated that mindfulness was indirectly linked to BPD symptoms through self-compassion, indicating that the effects of specific facets may operate through broader self-related processes.

Overall, these findings suggest that facet-level distinctions—particularly nonjudging and acting with awareness—represent candidate processes through which mindfulness is associated with BPD outcomes, rather than reflecting a uniform effect of global mindfulness.

Self-related mechanisms

Self-related processes emerged as a distinct mechanism pathway linking mindfulness to BPD outcomes across the included studies. Among these, self-compassion was the most consistently examined construct. Scheibner et al. [31] demonstrated that self-compassion mediated the relationship between mindfulness and both BPD symptom severity and emotion dysregulation. This pattern was supported by comparative studies showing that individuals with BPD exhibit lower levels of self-compassion relative to controls [2, 29]. In addition, Kresznerits et al. [20] reported that reduced self-compassion was associated with increased frequency of nonsuicidal self-injury, indicating its relevance for behavioral outcomes.

Further evidence suggests that self-compassion operates within a broader network of self-related vulnerabilities. For example, cross-sectional findings from Velotti et al. [43] and Wupperman et al. [44] indicated that deficits in mindfulness are linked to maladaptive self-processing through emotion dysregulation and dysregulated behavior, indirectly supporting the role of self-related mechanisms.

Other self-related constructs were also identified. Sripunya et al. [39] found that inner strengths mediated the relationship between feelings of emptiness and self-harm, highlighting the role of adaptive self-resources. Similarly, Schmid et al. [32] reported improvements in mindfulness, affect, and self-efficacy following intervention among Psychiatric inpatients with BPD, suggesting that changes in perceived self-capacity may accompany mindfulness-related improvements.

Overall, these findings indicate that mindfulness is associated with BPD outcomes through multiple self-related processes, with self-compassion representing the most consistently supported mediator, alongside constructs such as self-efficacy and inner strengths.

Interpersonal and social-cognitive mechanisms

The included studies indicate that mindfulness-related mechanisms are also linked to interpersonal and social-cognitive processes in BPD. Rueda et al. [28] found that mindfulness fully mediated the association between secure attachment and borderline personality severity among SUD outpatients.

Additional evidence supports indirect pathways through emotion regulation. Velotti et al. [43] demonstrated that emotion dysregulation mediated the relationship between mindfulness deficits and rejection sensitivity, suggesting that mindfulness may influence interpersonal vulnerability by affecting regulatory capacity. Complementing this, findings from Tortella-Feliu et al. [42] suggest that deficits in effortful control—relevant to both cognitive and interpersonal functioning—partially explain mindfulness-related differences in BPD, pointing to overlap between cognitive-control and social-cognitive mechanisms.

Moderation effects were also observed within this pathway. Seow et al. [36] reported that DBT skill use and positive attitudes toward skills were more strongly associated with improvements among individuals with higher BPD symptom severity, suggesting that baseline symptom levels may condition the effectiveness of mindfulness-related mechanisms in treatment contexts.

Finally, Reyes-Ortega et al. [27] showed that reductions in experiential avoidance, alongside increases in mindfulness skills, were associated with improvements in BPD outcomes, indicating that avoidance-related processes may represent an additional mechanism linking mindfulness to interpersonal and clinical functioning.

Overall, these findings suggest that mindfulness is connected to interpersonal and social-cognitive processes through both mediating and moderating pathways, often operating in conjunction with emotion-regulation and cognitive-control mechanisms.

Behavioral-outcome mechanisms

Several studies examined behavioral outcomes closely associated with BPD, including nonsuicidal self-injury (NSSI), self-harm, aggression, and treatment dropout. Evidence from mediation analyses indicated that these outcomes are indirectly linked to mindfulness through specific psychological mechanisms. Krantz et al. [19] demonstrated that acceptance without judgment mediated reductions in NSSI, highlighting the role of nonjudging as a protective factor. Similarly, Kresznerits et al. [20] found that higher frequency of NSSI was associated with increased impulsivity and reduced self-compassion, with mindfulness operating indirectly through its associations with these risk factors.

Additional findings support the role of self-related and regulatory mechanisms in behavioral outcomes. Sripunya et al. [39] reported that inner strengths mediated the relationship between feelings of emptiness and self-harm, indicating that adaptive self-resources may buffer against harmful behaviors. Consistent with this, Soler et al. [38] reported improvements in specific behavioral components of impulsivity, particularly delay of gratification and time perception, but not in trait impulsivity or response inhibition.

Aggression-related outcomes were examined by Kramer et al. [18], who found that changes in assertive anger mediated symptom improvement following DBT-informed skills training, suggesting that specific forms of anger regulation are relevant to behavioral change. In addition, Stratton et al. [40] identified lower mindfulness as a predictor of treatment dropout, indicating that mindfulness may also play a role in treatment engagement and adherence.

Overall, these findings suggest that mindfulness is associated with behavioral outcomes in BPD indirectly through mechanisms such as nonjudging, impulsivity, self-compassion, inner strengths, and anger regulation, rather than through direct effects on behavior.

Summary of findings across thematic pathways

To facilitate comparison across the five thematic pathways, Table 3 summarizes the principal processes examined, the main pattern of findings, and the relative strength and limitations of the available evidence. The pathways are not mutually exclusive, as several studies examined processes that operated across emotional, cognitive, self-related, interpersonal, and behavioral domains.

Table 3.

Summary of findings across the five thematic pathways

Thematic pathway	Principal processes examined	Summary of findings	Nature and strength of evidence
Emotion-regulation and impulse-control mechanisms	Emotion dysregulation, impulse control, distress tolerance, decentering, experiential avoidance, effortful control, and anger regulation	Lower mindfulness was generally associated with greater emotion dysregulation and impulsivity. Longitudinal and intervention studies linked increased mindfulness with improved emotion regulation and lower BPD symptoms. Process analyses implicated impulse control, adaptive regulation strategies, decentering, and related regulatory processes.	This was the most consistently supported candidate pathway across cross-sectional, longitudinal, mediation, and intervention studies, although causal evidence remains limited.
Mindfulness-facet mechanisms	Nonjudging, acting with awareness, observing, nonreactivity, and present-moment awareness	Nonjudging and acting with awareness were the facets most consistently associated with lower BPD symptoms and related distress. Nonjudging mediated reductions in nonsuicidal self-injury, acting with awareness was identified as a mediator of treatment-related change, and nonjudging moderated the association between awareness and BPD symptoms.	Evidence was generally consistent but was derived predominantly from observational, mediation, and small intervention studies. Findings support facet-specific rather than uniform effects of global mindfulness.
Self-related mechanisms	Self-compassion, self-efficacy, inner strengths, and maladaptive self-processing	Self-compassion mediated associations between mindfulness and both BPD symptom severity and emotion dysregulation. Lower self-compassion was also associated with greater nonsuicidal self-injury. Improvements in self-efficacy and the mediating role of inner strengths provided additional evidence for adaptive self-related processes.	Evidence was moderate and most consistent for self-compassion. Evidence for self-efficacy and inner strengths was based on fewer studies and requires replication in longitudinal and experimental designs.
Interpersonal and social-cognitive mechanisms	Attachment, rejection sensitivity, experiential avoidance, symptom severity as a moderator, and attitudes toward DBT skills	Mindfulness mediated the association between secure attachment and BPD severity, whereas emotion dysregulation mediated the association between mindfulness deficits and rejection sensitivity. Baseline BPD severity also moderated associations between DBT skill use and psychological outcomes.	Evidence was limited and primarily indirect or cross-sectional. Findings suggest that interpersonal processes may operate through emotion-regulation and cognitive-control pathways rather than independently.
Behavioral-outcome mechanisms	Nonsuicidal self-injury, self-harm, impulsive behavior, aggression-related processes, and treatment dropout	Behavioral outcomes were associated with nonjudging, impulsivity, self-compassion, inner strengths, and assertive anger. Distress tolerance was examined separately as a treatment-related mediator, and lower mindfulness predicted dropout.	Evidence was heterogeneous across mediation and intervention studies. Nonsuicidal self-injury and impulsivity were most frequently examined; aggression and dropout were each examined in individual studies.

As shown in Table 3, emotion regulation and impulse control constituted the most extensively examined and consistently supported candidate pathway. The remaining pathways provided complementary evidence that the association between mindfulness and BPD-related outcomes may depend on specific mindfulness facets, adaptive self-related capacities, and interpersonal or social-cognitive processes. Behavioral outcomes, including nonsuicidal self-injury, self-harm, anger-related responses, and treatment dropout, were generally examined as outcomes associated with these processes. However, the evidential basis differed across pathways, with stronger support derived from longitudinal, mediation, and intervention studies and more limited mechanistic inference from cross-sectional associations.

Discussion

Overview of the findings

The present systematic review aimed to identify and synthesize potential psychological processes associated with the relationship between mindfulness and borderline personality disorder symptoms. The findings identified candidate processes across emotional, cognitive, self-related, interpersonal, and behavioral domains, including emotion regulation, decentering, mindfulness facets, self-compassion, effortful control, attachment-related processes, and behavioral outcomes [14, 19, 28, 31, 33, 42]. However, the strength of evidence varied because some studies formally tested indirect or sequential pathways, whereas others reported direct or cross-sectional associations. Accordingly, the findings support a process-oriented conceptualization of mindfulness in BPD but do not uniformly establish causal mechanisms, consistent with methodological criteria distinguishing statistical mediation from demonstrated mechanisms of change [12, 15, 16].

As shown in Fig. 2, the synthesized evidence suggests that mindfulness may be linked to BPD symptoms through a network of interacting candidate processes. Emotion regulation appears to represent the most consistently supported pathway. Longitudinal and intervention-based studies associated mindfulness-related skills or DBT module participation with improvements in emotion regulation, while several BPD-focused studies also linked these changes to reductions in BPD symptom severity [11, 14, 21, 33, 34]. Specific mindfulness facets, particularly nonjudging and acting with awareness, may function as proximal processes associated with regulatory and behavioral outcomes [19, 24, 26]. Self-related findings included self-compassion, self-efficacy, and inner strengths [20, 31, 32, 39]. Cognitive-control processes such as decentering and effortful control may support subsequent changes in emotion regulation [9, 33, 42], whereas attachment and rejection sensitivity findings suggest that mindfulness-related processes may also extend to interpersonal functioning [28, 43]. Behavioral outcomes, including nonsuicidal self-injury and treatment dropout, appear to represent more distal outcomes associated with these processes [19, 20, 40].

Fig. 2.

Integrative mechanism-based framework linking mindfulness to BPD symptoms

Psychological processes associated with mindfulness in BPD

Among the identified candidate processes, emotion regulation emerged as the most consistently supported candidate pathway. Across observational, longitudinal, and intervention-based studies, Across observational, longitudinal, and intervention-based studies, greater mindfulness or improvements in mindfulness-related skills were generally associated with more adaptive emotion-regulation or clinical outcomes [2, 21, 29, 34]. In several BPD-focused studies, changes in mindfulness-related or regulatory processes were also associated with reductions in BPD symptoms [14, 33]. This pattern is consistent with Linehan’s biosocial model of BPD, which conceptualizes pervasive emotion dysregulation as arising from transactions between biological emotional vulnerability and an invalidating environment [6, 23]. The reviewed evidence further suggests that this pathway involves specific regulatory components, including impulse control, access to adaptive regulation strategies, distress tolerance, and experiential avoidance [1, 27, 38, 45]. Emotion regulation should nevertheless be described as the most strongly supported candidate pathway rather than as a definitively established causal mechanism.

The prominence of emotion regulation is also consistent with process models of mindfulness that identify attention regulation, body awareness, emotion regulation, and changes in self-related processing as interacting components of mindfulness-related change [13, 41]. The present synthesis extends these general frameworks to the BPD literature by indicating that attentional and cognitive-control processes may precede or interact with emotion-regulation changes, although the temporal ordering of these processes has been tested directly in only a limited number of studies [9, 33, 42].

Furthermore, the identified pathways are consistent with theoretical accounts of mindfulness and DBT, in which mindfulness is proposed to support attentional awareness, emotion regulation, and reduced automatic reactivity [13, 23, 41]. The current findings extend these accounts by indicating how cognitive-control, self-related, and interpersonal processes may operate alongside emotion regulation.

Beyond the thematic organization of candidate processes, the findings indicate substantial differences in the type and strength of evidence supporting each pathway. Emotion regulation was examined across cross-sectional, longitudinal, mediation, and intervention studies, providing greater convergence than the other pathways [14, 21, 33, 34]. In contrast, self-related processes were supported mainly by cross-sectional, comparative, and mediation findings [2, 20, 31], whereas interpersonal processes were represented by a smaller number of primarily cross-sectional studies [28, 43]. These differences support greater confidence in emotion regulation as a candidate pathway but do not establish it as a causal mechanism.

An additional consideration concerns the heterogeneity of the included clinical populations. Although the review synthesized findings across studies involving diagnosed BPD patients, individuals with subclinical borderline traits, and mixed psychiatric samples, these groups may differ in symptom severity, emotional instability, and interpersonal dysfunction. Consequently, the identified mechanisms may not operate uniformly across all populations. For example, mechanisms related to impulsivity, self-harm, and emotion dysregulation may be more pronounced in clinically diagnosed BPD samples, whereas mindfulness-related cognitive or self-related processes may present differently in subclinical populations. This variability should be considered when interpreting the generalizability of the findings and highlights the need for future studies directly comparing mechanisms across levels of borderline pathology.

Beyond emotion regulation, the findings identified several additional potential psychological pathways. Self-compassion, conceptualized as a balanced and noncritical response to personal distress, was associated with mindfulness, emotion dysregulation, BPD severity, and nonsuicidal self-injury [2, 20, 25, 29, 31]. These findings suggest that adaptive self-relating may contribute to the association between mindfulness and BPD outcomes, although longitudinal confirmation remains limited. Cognitive processes, particularly decentering—the capacity to view thoughts and feelings as transient mental events—and effortful control were also associated with mindfulness-related changes [9, 10, 33, 42]. This interpretation is consistent with models proposing that mindfulness supports the observation of thoughts and emotions with reduced automatic identification and reactivity [13, 41].

Interpersonal and social-cognitive processes also emerged as potentially relevant pathways, although the evidence base was comparatively limited. Mindfulness mediated the association between attachment-related variables and BPD severity in one study involving individuals with substance use disorder and BPD traits [28], while emotion dysregulation was implicated in associations involving mindfulness and rejection sensitivity in another cross-sectional study [43]. These findings suggest that mindfulness-related processes may extend beyond intrapersonal regulation to interpersonal functioning; however, replication in clinically diagnosed BPD samples and longitudinal designs is required before stronger conclusions can be drawn.

In addition, several studies highlighted behavioral outcomes associated with these mechanisms, including nonsuicidal self-injury, self-harm, aggression, and treatment dropout. These outcomes appear to be linked to underlying processes such as impulsivity, distress tolerance, and self-compassion, further supporting the view that mindfulness exerts its effects through interconnected pathways that span emotional, cognitive, and behavioral domains.

Integrative conceptual model

Taken together, the findings suggest that mindfulness may be linked to BPD symptoms through interrelated candidate processes operating across emotional, cognitive, self-related, and interpersonal domains. Emotion regulation appears to represent the central pathway within this system, whereas other mechanisms, including self-compassion, cognitive control, and interpersonal processes, operate in parallel or sequentially. This pattern extends beyond previous research that has largely examined these mechanisms independently and highlights the importance of integrative models capable of accounting for their interaction.

Although the reviewed studies examined diverse psychological variables, the findings suggest that these mechanisms are unlikely to operate independently in clinical contexts. Rather, the evidence indicates that mindfulness-related processes may influence BPD symptoms through overlapping and mutually reinforcing pathways that involve emotional, cognitive, self-related, and interpersonal domains simultaneously. This pattern may partly explain the variability observed across studies, as different investigations often focused on isolated mechanisms despite substantial conceptual overlap between them.

From a theoretical perspective, the present findings contribute to the literature by moving beyond association-based and intervention-focused approaches toward a mechanism-oriented framework. Previous systematic reviews have primarily examined the effectiveness of mindfulness-based interventions, while meta-analytic studies have quantified the strength of the relationship between mindfulness and BPD symptoms. However, these approaches do not clarify the psychological processes through which mindfulness may be associated with clinical outcomes. The current synthesis addresses this limitation by organizing the identified mechanisms into a coherent structure, thereby providing a more comprehensive understanding of the processes involved.

The integrative framework proposed in this review can be conceptualized as a multi-pathway model in which mindfulness may be associated with BPD symptoms through interacting candidate processes operating at different levels. At the proximal level, mindfulness facets and cognitive-control processes influence attentional and regulatory capacities. At the central level, emotion regulation functions as the primary mechanism through which these processes impact symptom expression. At the secondary level, self-related and interpersonal mechanisms contribute to broader patterns of psychological functioning. Finally, behavioral outcomes represent distal manifestations of these interacting pathways.

This multi-level structure supports a mechanism hierarchy in which different processes operate in a coordinated and interdependent manner rather than as isolated variables. The present review synthesizes fragmented empirical findings into a mechanism-oriented framework linking mindfulness to borderline personality disorder symptoms. The evidence indicates that mindfulness is associated with BPD outcomes through interacting pathways involving emotion regulation, mindfulness facets, self-related processes, cognitive-control mechanisms, and interpersonal functioning. Collectively, these findings support the need for future research and intervention models that address these mechanisms as interconnected rather than isolated processes.

Clinical implications and limitations

From a clinical perspective, the findings generate hypotheses for intervention development and refinement rather than definitive treatment recommendations. Given the convergence of evidence concerning emotion regulation, mindfulness-based components may warrant closer evaluation in relation to impulse control, distress tolerance, and access to adaptive regulation strategies [1, 14, 45]. The findings concerning self-compassion, decentering, and effortful control further suggest that future interventions and dismantling studies should examine whether targeting multiple coordinated processes produces greater benefit than targeting a single pathway [31, 33, 42]. This is particularly relevant to dialectical behavior therapy, in which mindfulness and emotion-regulation skills are central treatment components, although the specific processes responsible for clinical change require further experimental testing.

Despite these contributions, several limitations should be acknowledged. First, the included studies exhibited substantial heterogeneity in terms of sample characteristics, study design, measurement instruments, and analytical approaches. This variability limited the ability to directly compare findings across studies and precluded the use of meta-analytic synthesis. Second, many studies relied on cross-sectional designs, which restrict the ability to draw causal inferences regarding the directionality of the observed relationships. Third, although some studies examined mediation models, the overall evidence base remains limited in terms of comprehensive, multi-mechanism analyses. Finally, differences in how mindfulness and BPD symptoms were operationalized across studies may have contributed to variability in findings.

Future research

Future research should aim to address these limitations by employing longitudinal and experimental designs that allow for stronger causal inference and by examining multiple mechanisms within integrated models. There is also a need for greater consistency in measurement approaches and for studies that directly compare the relative contribution of different mechanisms. In addition, future work should consider the interaction between intrapersonal and interpersonal processes to better capture the complexity of BPD.

Conclusions

Overall, this review synthesizes the available evidence into an integrative framework suggesting that mindfulness is associated with BPD symptoms through interacting emotional, cognitive, self-related, and interpersonal processes. Although emotion regulation emerged as the most consistently supported candidate pathway, further longitudinal and experimental research is needed to establish the temporal and causal relationships among these processes.

Author contributions

Dr. Aamer Aldbyani conceptualized the study, developed the original research idea, and served as the project director. He led the overall design of the systematic review, supervised all methodological stages, coordinated the review process, contributed to data screening and extraction, interpreted the findings, and drafted the manuscript. Ms. Chunhong Shen and Ms. Zhai Shuna contributed to literature screening, data extraction, data checking, and organization of extracted information, and participated in the revision of the manuscript. Naif Al-Barha contributed to the study screening, verification of extracted data, and critical review of the manuscript. Dr. Afnan Alhimaidi contributed to data screening, data verification, synthesis support, and critical revision of the manuscript. All authors reviewed and approved the final version of the manuscript.

Funding

The author(s) declared that they didn’t receive any financial support for this work.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethical approval

Ethical approval was not required for this study, as it is a systematic review of previously published literature and did not involve the collection of primary data from human participants. All procedures were conducted in accordance with accepted standards for systematic reviews and scholarly research integrity.

Consent to participate

Not applicable.

Consent for publication

The authors declare that they have no competing financial or non-financial interests.

Competing interests

The authors declare no competing interests.