Psychodynamic Research
Mentalizing refers to the ability to make sense of one’s own and others’ actions by understanding the underlying thoughts, feelings, intentions, and beliefs that drive them. Although mentalizing is an evolutionarily selected, biologically grounded capacity (Tomasello, 2019; Fehlbaum et al., 2022; Gilead & Ochsner, 2021), it is also profoundly shaped by interpersonal environments and developmental experience (Luyten, Campbell, Allison, & Fonagy, 2020).
Initially, mentalizing theory—rooted in attachment research—emphasized how parental sensitivity and the caregiver’s capacity to mentalize the child confer protection or vulnerability to emotional distress across development. Over time, the framework has evolved into a broader developmental model. Contemporary mentalizing theory highlights that mentalizing is not formed solely in dyadic relationships but is influenced by family systems, peer interactions, and wider sociocultural contexts.
A central component of this expanded model is epistemic trust: the evolutionarily prewired capacity to treat others as trustworthy and relevant sources of knowledge. Epistemic trust both facilitates and is facilitated by mentalizing. Together, they support a salutogenic developmental process in which individuals learn effectively from their social worlds, remain open to new information, and adapt flexibly to changing circumstances (Luyten et al., 2020).
Importantly, disruptions to epistemic trust—whether in the form of epistemic mistrust (hypervigilance, hostility, suspicion) or epistemic credulity (excessive, uncritical openness)—are increasingly recognized as core features of many forms of complex mental health difficulties. These disruptions limit an individual’s ability to rely on others, to learn from experience, and to update maladaptive expectations. Empirical studies consistently link impaired epistemic trust to childhood adversity, insecure attachment, reduced mentalizing capacity, and symptoms across diagnostic categories including depression, anxiety, and personality disorder features (Asgarizadeh & Ghanbari, 2024; Benzi et al., 2023; Bincoletto et al., 2025; Campbell et al., 2021, 2025; Carone et al., 2025; Greiner et al., 2025; Hashemi et al., 2024; Kampling et al., 2022, 2024, 2025; Karagiannopoulou et al., 2024; Kumpasoglu et al., 2025; Liotti et al., 2023; Milesi et al., 2024; Rodriguez Quiroga et al., 2024; Szél et al., 2025; Weiland et al., 2024).
Mentalization-Based Treatment (MBT) was developed to address these vulnerabilities. MBT aims to strengthen and stabilize mentalizing so that individuals can regulate emotions more effectively, develop a more coherent and robust sense of self, and navigate relationships with greater flexibility (Bateman & Fonagy, 2016; Bateman, Fonagy, Luyten et al., 2023).
Therapy works by providing a relational context in which the therapist models a mentalizing stance—curiosity, openness, acknowledgment of uncertainty, and willingness to repair misunderstandings. Breakdowns in mentalizing are expected and are treated as opportunities for collaborative reflection rather than errors or failures. Over time, this shared attention to mental states helps the patient recover and maintain mentalizing even under stress.
Crucially, a consistent mentalizing stance fosters epistemic trust in the therapeutic relationship. As trust builds, individuals begin to treat the therapist as a reliable source of new, personally relevant information. The hope is that this renewed openness will extend beyond therapy, enabling individuals to re-engage in social learning and reconnect with others in their broader lives.
How Mentalization-Based Treatment (MBT) Developed and What We Know About Its Effectiveness
Mentalization-Based Treatment (MBT) was developed in the 1990s by Anthony Bateman and Peter Fonagy in response to a clear and urgent clinical need. At the time, mental health services in the UK—as in many other countries—were failing people with borderline personality disorder (BPD). Research painted a bleak picture: almost all individuals with BPD who sought help (97%) were treated in outpatient settings, often seeing multiple therapists—on average six—yet two- to three-year follow-ups showed that most made little progress or deteriorated (Lieb et al., 2004).
Against this backdrop, Bateman and Fonagy developed MBT as a practical, accessible, and attachment-informed psychotherapy that could be delivered by a wide range of mental health professionals. The first full MBT programme took place in a day-hospital setting, combining weekly individual sessions with group therapy three times per week within an 18-month partial hospitalisation model. Psychiatric nurses delivered much of the treatment, supported by close supervision to ensure fidelity (Bateman & Fonagy, 1999, 2000).
Since its original formulation, MBT has expanded far beyond its initial focus on BPD in adults. Adaptations now exist for:
· Children (Midgley et al., 2017)
· Adolescents (Sharp & Rossouw, 2024)
· Couples (Bleiberg, Safier, & Fonagy, 2022)
· Families (Asen, Bleiberg, & Fonagy, 2024)
MBT has also been applied to conditions including mood disorders, eating disorders, psychosis, and trauma-related difficulties, and to personality disorder presentations such as narcissistic, antisocial, and avoidant personality disorders (Bateman, Fonagy, Campbell, Luyten, & Debbané, 2023).
The Evidence Base for MBT
Research on MBT has grown substantially and now forms one of the most robust bodies of evidence within the treatment of BPD.
Meta-analytic findings show:
· Specialised psychotherapies—including MBT—outperform non-specialised treatments for BPD (Cristea, Kok, & Cuijpers, 2015).
· A larger meta-analysis of 87 studies identified MBT, schema therapy, and adapted DBT as producing above-average effect sizes, compared to below-averageoutcomes for treatment as usual (Rameckers et al., 2021).
· A 2020 Cochrane Review highlighted MBT and DBT as the two psychotherapies with the strongest empirical support for BPD (Storebø et al., 2020).
· Further reviews show medium to very large improvements across symptoms, interpersonal functioning, and educational or vocational outcomes (Volkert, Hauschild, & Taubner, 2019; Vogt & Norman, 2019).
The Landmark RCTs: Why MBT Stood Out
The first MBT randomised controlled trial evaluated the original 18-month day-hospital programme for BPD (Bateman & Fonagy, 1999, 2000). The results were striking:
· Dramatic reductions in suicide attempts and self-harm
· Fewer emergency and inpatient admissions
· Significant improvements in depression, anxiety, and interpersonal functioning
· Continued improvement during 18-month follow-up
The effect size was so large that the number needed to treat was approximately two—a level rarely seen in psychotherapy research.
Importantly, MBT was also cost-effective. It did not exceed the cost of standard care and produced long-term savings through reduced service use.
A long-term follow-up five years after treatment (eight years after therapy began) found enduring and substantial benefits compared with TAU (Bateman & Fonagy, 2008):
· Suicidality: 23% in MBT vs. 74% in TAU
· Retention of BPD diagnosis: 13% vs. 87%
· Lower service use, better global functioning, fewer medications, and higher employment rates
Strong support for MBT also comes from studies comparing it to structured clinical management (SCM). In an 18-month randomised controlled trial of intensive outpatient treatment, MBT showed clear superiority, particularly among individuals with multiple personality disorder diagnoses (Bateman & Fonagy, 2009, 2013). Across the trial, MBT produced greater reductions in suicidal and self-injurious behaviour and significantly fewer hospital admissions. Notably, the primary composite outcome—achieving six consecutive months free from suicidal behaviour, severe self-harm, or psychiatric hospitalisation—rose from 0% to 43% with SCM, but reached 73%with MBT.
Patients receiving MBT also demonstrated broader improvements across interpersonal, social, and mood-related domains. Long-term follow-up eight years later further underscored these advantages: 74% of participants originally treated with MBT were classified as recovered compared to 51%of those who had received SCM.
MBT has also shown significant benefits for individuals with comorbid antisocial personality disorder (ASPD). In outpatient settings, MBT led to reductions in aggression, paranoia and self-injury, and to improvements in mood, psychiatric symptoms, and social functioning (Bateman et al., 2016). A forensic-based trial further confirmed its effectiveness: individuals receiving MBT-ASPD had dramatically lower aggression scores on the Overt Aggression Scale-Modified compared to those receiving probation-as-usual (mean 90 [SD 126] vs. 186 [153]), representing a medium-to-large effect size of 0.74(Fonagy et al., 2025).
Findings from Denmark add further weight. In a randomised trial, MBT outperformed a lower-intensity manualised supportive group therapy on clinician-rated global functioning (Jørgensen et al., 2013), with gains persisting at 18-month follow-up (Jørgensen et al., 2014). Another Danish study using a partial-hospital model followed by group MBT delivered over approximately two years found strong improvements across functioning, hospital use, and employment outcomes, with additional gains observed up to two years post-treatment (Petersen et al., 2010).
Evidence from the Netherlands also supports MBT’s effectiveness. A multi-site RCT found that both day-hospital MBT and specialist treatment-as-usual (TAU) were beneficial, but MBT had substantially better retention—dropout rates were 9% for MBT versus 34%for TAU (Laurenssen et al., 2018). Another Dutch study of an 18-month manualised MBT programme for 45 individuals with severe BPD showed moderate-to-large improvements across symptoms and functioning, although firm conclusions are limited by the absence of a control group (Bales et al., 2012).
Naturalistic evidence strengthens these findings. A Norwegian comparison of MBT with psychodynamic group therapy suggested that higher severity predicted poorer outcomes in the psychodynamic condition but notin MBT, implying advantages for more complex cases (Kvarstein et al., 2019). A large multi-site comparison of day-hospital MBT (MBT-DH) and intensive outpatient MBT (MBT-IOP) followed patients for three years. Both formats produced large, durable improvements: 83% of patients improved in general symptoms and 97% improved in BPD-specific symptoms. No meaningful difference emerged between the two intensities, an important finding for service planning and cost-effectiveness (Smits et al., 2020).
Finally, MBT has been adapted for adolescents. Although the evidence is still developing—and a recent Cochrane Review notes that further rigorous trials are needed (Witt et al., 2021)—early results are promising. An RCT found that MBT-A significantly reduced self-harm and depression compared to TAU in self-harming adolescents (Rossouw & Fonagy, 2012). Pilot trials suggest benefits of combining group and individual formats (Beck et al., 2020; Bo et al., 2017). A naturalistic study of 118 adolescent inpatients with personality pathology also showed substantial improvements in general symptoms, personality-specific difficulties, and health-related functioning, with particularly strong gains for internalising problems (M. S. Jørgensen et al., 2021).