Skip to main content
NCBI home page
Frontiers in Psychology logoLink to Frontiers in Psychology
. 2026 Aug 5;17:1853182. doi: 10.3389/fpsyg.2026.1853182

Personality functioning as a pathway from childhood trauma to substance use in clinical and community samples

Jasmin Brouschek 1,2, Elisa Renner 2, Raphael Wimmer 3, Zoe Zipper 3, Hannah Brössler 3, Jürgen Fuchshuber 4,5, Wolfgang Beiglböck 2, Human-Friedrich Unterrainer 1,3,6,7,8,*
PMCID: PMC13487721  PMID: 42621011

Abstract

Background

Traumatic childhood experiences are associated with an increased risk of substance abuse. It remains unclear to what extent this association is mediated by personality organization and mentalization ability across different populations.

Method

A structural equation model (SEM) with bootstrap estimates was calculated in a community sample (N = 498), and Bayesian path analysis with informed priors was calculated in an inpatient sample of individuals diagnosed with substance use disorder (SUD sample) (N = 68) to examine direct and indirect relationships between childhood trauma, personality organization, mentalization ability, and addictive behavior.

Results

In the community sample, structural personality impairments partially mediated the effect of childhood trauma on addictive behavior (indirect: β = 0.14, p < 0.001; direct: β = 0.19, p = 0.002), whereas mentalization showed no independent effect. In the SUD sample, childhood trauma was associated with structural impairments (β = 0.24), which, in turn, were associated with addictive behavior (β = 0.22).

Conclusion

These findings highlight the central role of structural personality impairments in the association between childhood trauma and addictive behavior. While structural deficits act as a partial mediator in the general population, they emerge as the strongest and most consistent correlate of addictive behavior in clinical SUD model.

Keywords: Bayesian modelling, childhood trauma, mentalization, personality structure, structural equation modeling, substance use

1. Introduction

Substance use disorders (SUDs) remain one of the greatest challenges in modern healthcare and are associated with significant socioeconomic costs and profound psychological suffering for those who are affected (American Psychiatric Association, 2022; United Nations Office on Drugs and Crime [UNODC], 2025; World Health Organization, 2025). Although neurobiological and behavioral correlates of SUD have been well researched, the underlying psychological instances that can make individuals vulnerable to addictive behavior have not been fully studied. A large number of studies show that traumatic childhood experiences, such as emotional or physical neglect/abuse or sexual abuse, can be predictors for the development of SUD (Dube et al., 2003; Felitti et al., 1998; Hyman et al., 2008; Teicher and Samson, 2016). Furthermore, the results from large meta-analyses show that the risk of addiction problems later in a person’s life increases with the number of stressful childhood experiences. Individuals with higher cumulative trauma exposure may be more likely to report substance abuse later in life (Hughes et al., 2017; Norman et al., 2012). Even when social factors and genetic influences are considered, childhood trauma remains consistently associated with an increased risk of later SUD. While the association between childhood trauma and SUD has been well established in the literature, the psychological mechanisms underlying this association can be studied further.

Theoretical frameworks, such as the self-medication hypothesis (Khantzian, 1985, 1997), propose that substance use may serve as a coping strategy for overwhelming emotional states related to traumatic experiences in early childhood, particularly when these frameworks had a negative impact on personal development (Costanzo et al., 2023; Hawn et al., 2020). In this context, impairments in fundamental psychological capacities involved in emotion regulation, self-perception, and interpersonal functioning may be particularly relevant (Gross, 2015). To better understand the underlying mechanisms of this trauma-related coping process, recent research has focused on two interrelated constructs: structural personality organization and mentalization.

Mentalization also shares the conceptual overlap with attachment security and emotion regulation as all three constructs involve the understanding and regulation of affective experiences in interpersonal contexts (Mikulincer and Shaver, 2019). More specifically, mentalization refers to the capacity to interpret one’s own and others’ behavior in terms of underlying mental states, such as feelings, intentions, desires, and beliefs. Within the context of SUD, impairments in mentalization may, therefore, be relevant because they can contribute to difficulties in affect regulation, interpersonal functioning, and the use of substances as a coping strategy (Fonagy et al., 2018; Luyten and Fonagy, 2015). The development of mentalization ability depends on secure attachment relationships, in which the child’s inner states are appropriately reflected by their caregivers, thereby gradually developing good reflective capabilities (Fonagy and Luyten, 2009). Abuse and neglect can hinder this development and can lead to epistemic mistrust and deficits in reflective ability (Fonagy et al., 2019). Underdeveloped mentalization skills can cause emotions to be experienced as overwhelming physical states, and a lack of self-regulation skills can lead to the use of substances to regulate these affects (Bateman and Fonagy, 2016, 2019; Thorberg and Lyvers, 2010; Khantzian, 1985, 1997). While mentalization deficits are frequently observed in SUD patients (Schindler, 2019), their specific role as a mediator, aside from general structural deficits, remains to be clearly identified.

Personality organization refers to broader structural capacities of personality functioning. Contemporary psychodynamic models increasingly suggest that mentalization may be hierarchically embedded within broader personality functioning. However, both constructs were included separately in the present analyses to examine whether mentalization explains variance in addictive behavior beyond general structural personality impairment (Bateman and Fonagy, 2019; Luyten and Fonagy, 2015; Kernberg, 1984). According to Kernberg (1984), personality organization encompasses dimensions such as identity integration, defense mechanisms, and reality testing. Broader psychodynamic conceptualizations additionally include aspects related to object relations, aggression regulation, and moral functioning. Consistent with this broader framework, the Inventory of Personality Organization (IPO) additionally contains subscales addressing aggression intensity and moral values. PO is seen as a key part of personality pathology, which manifests in both the integration of self-representations and the ability to maintain stable interpersonal relationships (Caligor et al., 2018). A well-integrated personality enables individuals to smoothly adjust and better manage and regulate their affects and impulses more adaptively and to respond more flexibly to stressful situations. Traumatic relational experiences may impair the development of these structural capacities. This may be associated with uncertainties in one’s own identity, and those affected more frequently resort to primitive defense mechanisms, such as splitting or denial, to cope with stressful feelings and fears (Kernberg et al., 2008; Zimmermann et al., 2013).

Research increasingly shows that structural deficits may present one possible mechanism underlying the association between trauma and different mental health disorders. They may act as risk factors and explain how trauma is related to various mental disorders, including addictive behaviors (Freier et al., 2022; Fuchshuber et al., 2018, 2019; Insel, 2014; Kampling et al., 2022; Krakau et al., 2021). Krakau et al. (2021) showed that impairments in personality functioning significantly explained the association between maltreatment in childhood and then a person’s mental health in adulthood. It was also found that structural personality deficits led to the relationship between different types of child maltreatment and the development of depression and anxiety symptoms in a large representative sample (Freier et al., 2022). Extending this mediation perspective, it was found that both epistemic trust and personality functions acted as mediating mechanisms linking negative childhood experiences to post-traumatic stress disorder and complex post-traumatic stress disorder, underscoring the developmental importance of structural impairments following early relational trauma (Kampling et al., 2022).

Fuchshuber and Unterrainer (2020) proposed a neuropsychoanalytic framework to investigate the hypothesis that trauma-related impairments in structural PO may be related to changes in the neural systems involved in stress regulation and reward processing. Consistent with this framework, a network analysis by Fuchshuber et al. (2025a), which was used to show the interrelationships between childhood trauma and PO, identified identity diffusion and primitive defense mechanisms as central nodes within the trauma-psychopathology network. In a different study, Fuchshuber et al. (2025b) also showed that PO not only explained the association between the relationship between childhood stress and paranoid thinking but also led to it, thus highlighting its role as a vulnerability and a deciding factor in the development of psychopathology. Developmental neurobiological studies further suggest that early traumatic experiences and early-life stress may be associated with alterations in frontolimbic networks and stress-sensitive neural systems (Kirsch et al., 2020; Teicher and Samson, 2016), which have been associated with difficulties in affect regulation, impulse control, and increased vulnerability to addictive behavior (Koob and Volkow, 2016).

2. The current study

This study builds on the same dataset used by Renner et al. (2025), who examined the relationship between addictive behavior, primary emotions, and dissociative symptoms in the same sample. The present study expands the theoretical framework by specifically examining the role of mentalization alongside structural organization. Indirect statistical pathways involving structural personality organization and mentalization in the association between childhood trauma and addictive behavior were examined. Two studies with different methodological approaches were conducted, tailored to the respective characteristics of the samples.

In Study 1, a community sample from the general population was used to create a latent structural equation model (SEM). SEM enables the simultaneous evaluation of complex, multi-path relationships while strictly accounting for the measurement error through the use of latent variables (Kline, 2023). The present study aimed to investigate the extent to which trauma-related impairments in personality structure and mentalization ability mediate addictive behavior. To test these mediation paths, a bootstrapping procedure was used, which does not rely on the assumption that indirect paths are normally distributed (Cheung and Lau, 2008; Hayes, 2022). Personality structure was assessed using the Inventory of Personality Organization (IPO), mentalization ability using the Reflective Functioning Questionnaire (RFQ-6), and childhood trauma using the Childhood Trauma Questionnaire (CTQ). Addictive behavior was measured using the global cumulative risk score of the Alcohol, Smoking and Substance Involvement Screening Test (WHO-ASSIST). Building on these results from Study 1, Study 2 aimed to replicate and validate these associations in a sample of patients undergoing inpatient treatment for SUD (SUD sample) in an addiction treatment center in Vienna (Grüner Kreis Wien). This approach enabled the investigation of whether the psychological mechanisms identified in the community population remain stable in a sample of individuals diagnosed with SUD. The comparisons between the community and clinical samples were interpreted descriptively. Formal multigroup analyses or tests of measurement invariance were not conducted, as the primary aim of the study was not to directly compare latent structures across groups but rather to examine whether the observed associations could be descriptively replicated in a clinical SUD sample. To address the methodological challenges posed by the limited sample size (N = 68), a Bayesian path analysis framework was employed. This approach offers advantages, particularly with small samples, as it does not rely on the asymptotic properties of large samples (Depaoli, 2021; McNeish, 2016). In line with current recommendations for path modeling with small samples (Smid and Winter, 2020), informed priors (μ = 0.33, σ = 0.15) derived from the parameter expectations of Study 1 were included in the analysis to improve estimation stability. The inclusion of prior information in this way facilitates the precise estimation of complex signaling pathways, even under conditions of limited statistical power (van de Schoot et al., 2021). To assess the robustness of the SUD sample model, sensitivity analyses were conducted using weakly informative (μ = 0, σ = 0.5) and diffuse priors (μ = 0, σ = 10). Based on developmental and psychodynamic frameworks (Bateman and Fonagy, 2019; Kernberg, 1984), it was hypothesized that higher levels of childhood trauma in both samples would be consistently associated with greater impairments in structural personality organization and reduced mentalization abilities. The study aimed to investigate potential differences in the strength of direct and indirect pathways between the community and SUD groups.

3. Study 1

3.1. Materials and methods

3.1.1. Sample and procedure

Data collection was carried out via the online platform LimeSurvey1 and disseminated via social media channels as well as through the University of Graz (KFU) and Sigmund Freud University Vienna (SFU). The study was based on a convenience sample, with voluntary participation and no random sampling procedure. Therefore, the findings cannot be considered statistically representative of the general population. The data were collected between December 2023 and March 2024. The inclusion criteria were legal age (≥18 years) and sufficient knowledge of German. The study was approved by the ethics committees of Karl-Franzens-University Graz (KFU) and Sigmund Freud University of Vienna (SFU). All participants provided their informed consent before the survey began. A total of 971 participants completed the questionnaire. The assessment included multiple psychodynamic and clinical self-report measures comprising approximately 108 items, including questionnaires about primary emotions (BANPS, BANPS-L), sexual feelings and behavior (Berliner Lust Skala, BYAS), experience in close relationships (ECR-RD8), gaming activity (GDT), mental stress (BSI-18), dissociative symptoms (FDS), early childhood experiences (FEE), reflective functioning (RFQ), questions about the personality organization (IPO-16), and addictive behavior (WHO-ASSIST and questions referring to nicotine use). It required approximately 60–75 min to complete the questionnaire, given the length and emotionally demanding nature of several questionnaires. In addition to insufficient knowledge of German, participant fatigue, incomplete responding, and study discontinuation may have contributed to the comparatively high proportion of missing values. After excluding cases with a high proportion of missing values, the final sample comprised 498 participants aged between 18 and 88 years (M = 28.01, SD = 11.83; Mdn = 24). A majority of the respondents identified as female (73.5%), 115 as male (23.1%), and 17 as diverse (3.4%). On average, participants had 0.17 children (SD = 0.54; range = 0–4). A total of 75 participants (15.1%) reported having a diagnosed mental illness, with depression (7.8%), post-traumatic stress disorder (3.6%), and anxiety disorders (3.0%) being the most common. A total of 112 participants (22.5%) reported taking medication regularly, primarily antidepressants (5.2%), hormonal contraceptives (3.6%), and thyroid hormones (2.4%). In terms of addictive behavior, 53 individuals (10.6%) reported a current or past addiction, primarily nicotine addiction (6.6% of the total sample; 62.3% of individuals who reported an addiction). Other reported addictions included behavioral addictions (1.6%), eating disorders (1.0%), and alcohol abuse (0.8%).

3.1.2. Psychometric assessment

After providing their consent to the collection and processing of their personal data, the participants received a questionnaire on their sociodemographic data. This included questions about sex, age, marital status, level of education, language skills, medication, and psychiatric diagnoses.

3.1.2.1. Childhood trauma

Adverse childhood experiences were measured using the Childhood Trauma Questionnaire (Bernstein and Fink, 1998, Bernstein et al., 2003; German version: Klinitzke et al., 2012). Emotional, physical, and sexual abuse, as well as emotional and physical neglect, were measured. Higher scores indicate more severe traumatic experiences. In the present sample, the CTQ showed very good internal consistency (Cronbach’s α = 0.90).

3.1.2.2. Addictive behavior

To assess addictive behavior and substance involvement, the WHO-ASSIST was used (Humeniuk et al., 2010). The WHO-ASSIST is a screening instrument that evaluates lifetime substance use, abuse-related symptoms, and problems related to substance use, for 10 substance groups (such as “tobacco,” “alcohol,” “cannabis,” “cocaine,” “amphetamines,” “inhalants,” “sedatives,” “hallucinogens,” “opioids,” and “other drugs”). Internal consistency was high (Cronbach’s α = 0.94).

3.1.2.3. Mentalization

Mentalization ability was measured using the German short version of the Reflective Functioning Questionnaire (Fonagy et al., 2016; German adaptation according to Spitzer et al., 2021). In accordance with the recommendations of this validation study, six of the original eight items were included (item 4, “When I get angry, I say things I later regret,” and item 7, “I always know what I’m feeling,” were excluded). The questionnaire assesses the capacity to understand one’s own and others behavior in relation to mental states, such as thoughts, feelings, and intentions. Following the scoring guidelines, higher RFQ scores reflect lower mentalization ability (hypomentalizing), whereas lower scores indicate better mentalization capacity. The scale demonstrated satisfactory internal consistency (Cronbach’s α = 0.81).

3.1.2.4. Personality organization

Structural personality organization was evaluated using the Inventory of Personality Organization (Lenzenweger et al., 2001; German version: Zimmermann et al., 2013). The instrument measures the dimensions of identity diffusion, primitive defense mechanisms, and reality testing. Higher values indicate lower structural integration of the personality. In the present sample, the IPO showed good internal consistency (Cronbach’s α = 0.85).

3.1.3. Statistical analysis

SEM was conducted to examine the hypothesized relationships between childhood trauma, structural personality organization, mentalization ability, and addictive behavior. All analyses were performed in R (version 4.4.3; R Core Team, 2025) using the lavaan package (version 0.6-19; Rosseel, 2012). Parameters were estimated using maximum likelihood estimator (MLR). Model fit was evaluated according to established standards (Hu and Bentler, 1999; Kline, 2023). Acceptable model fit was defined as a comparative fit index (CFI) and a Tucker–Lewis index (TLI) of ≥0.90, a root mean square error of approximation (RMSEA) of ≤0.08 with an upper limit of the 90% confidence interval of ≤0.10, and a standardized root mean square residual (SRMR) of ≤0.08. The Akaike information criterion (AIC) was used for model comparison.

The measurement model included four latent variables (CTQ_lat, IPO_lat, RFQ_lat, and WHO_lat). Childhood trauma was represented by the five CTQ subscales: emotional abuse, physical abuse, sexual abuse, emotional neglect, and physical neglect. Personality Organization (PO) consisted of the three IPO subscales (identity diffusion, primitive defense mechanisms, and impaired reality testing). Mentalization was modeled using the six items of the RFQ-6. The items of the WHO-ASSIST were grouped into six content-defined subscales—frequency of use, craving, problems related to use (health, social, legal, or financial consequences), concern of family or friends, unsuccessful attempts to reduce use, and intravenous use. The model included direct paths from childhood trauma to personality structure, mentalization ability, and substance-related behavior, as well as indirect paths via IPO_lat and RFQ_lat. Indirect effects were estimated using bias-corrected bootstrapping (2,000 resamples; 95% confidence intervals), and effects were considered statistically significant if the confidence interval did not include zero. To ensure reproducibility, a fixed random seed was used (set.seed = 123). Psychopathological variables, such as self-reported psychiatric diagnoses, were used for descriptive characterization of the sample but were not included as covariates in the SEM as the primary aim was to test the theoretically specified pathways from childhood trauma to addictive behavior via personality organization and mentalization (Figure 1).

Figure 1.

Path diagram illustrating relationships among childhood trauma, personality organisation, mentalization, and addictive behaviors. Arrows indicate predictive directions, strengths (beta values), and associated measured variables for each latent construct.

Open in a new tab

Measurement model of Study 1: childhood trauma (CTQ_lat), personality organization (IPO_lat), mentalization ability (RFQ_lat), and level of substance use (WHO_lat).

3.2. Results

3.2.1. Correlations

Zero-order correlations between the variables are shown in Table 1. Childhood trauma (CTQ) showed a significant positive association with impaired personality structure (IPO; r = 0.41, p < 0.001) and reduced mentalization ability (RFQ; r = 0.34, p < 0.001). Childhood trauma was also positively associated with addictive behavior (WHO; r = 0.31, p < 0.001). Impaired personality structure demonstrated the strongest association with lower mentalization ability (r = 0.63, p < 0.001) and was moderately positively correlated with addictive behavior (r = 0.36, p < 0.001). Reduced mentalization ability demonstrated a weak but significant association with addictive behavior (r = 0.26, p < 0.001).

Table 1.

Descriptive statistics and zero-order correlations among the Study 1 variables.

Scale M SD CTQ IPO RFQ WHO
CTQ 41.04 14.82 1
IPO 29.76 8.99 0.41*** 1
RFQ 18.12 7.19 0.34*** 0.63*** 1
WHO 18.62 25.42 0.31*** 0.36*** 0.26*** 1
Open in a new tab

N = 498. ***p < 0.001.

3.2.2. SEM results: relationships between childhood trauma, personality structure, mentalization, and addictive behavior

First, a confirmatory factor analysis (CFA) including all four latent constructs was conducted to evaluate the measurement model. The CFA demonstrated an acceptable fit to the data (CFI = 0.928, TLI = 0.917, RMSEA = 0.064, 90% CI [0.059, 0.069], and SRMR = 0.050), indicating that the latent constructs were adequately represented by their observed indicators. Subsequently, the fully hypothesized SEM, including all theoretically proposed paths, was estimated. This initial structural model showed insufficient fit (CFI = 0.891, TLI = 0.874, RMSEA = 0.079 CI [0.074, 0.087], SRMR = 0.095, and AIC = 22,251.54). Examination of the structural paths revealed that the direct effect of mentalization (RFQ_lat) on the level of substance use (WHO_lat) was not statistically significant (β = 0.019, SE = 0.128, z = 0.259, and p = 0.796). Following the initial estimation, this non-significant path was removed to obtain a more parsimonious model, which was then re-estimated. The respecified final model demonstrated good fit to the data (CFI = 0.928, TLI = 0.916, RMSEA = 0.068, 90% CI [0.061, 0.074], SRMR = 0.050, and AIC = 22,644.481). Although the AIC increased slightly compared to the initial model, the respecified model was retained due to improved global fit indices and greater theoretical interpretability (Table 2). The final model supported the assumption that childhood trauma was associated with higher structural personality impairment, which, in turn, was related to increased addictive behavior. Mentalization did not show a direct association with addictive behavior. In the measurement model, the four latent constructs (CTQ_lat, IPO_lat, RFQ_lat, and WHO_lat) were specified with freely estimated covariances. This reflects the assumption that childhood trauma, structural personality functions, mentalization ability, and addictive behavior are interrelated but distinct domains. No causal relationships are implied at this level of measurement (see Tables 3, 4).

Table 2.

Model fit indices for the hypothesized and nested structural equation models.

Model CFI TLI RMSEA 90% CI RMSEA SRMR AIC
Hypothesized model 0.891 0.874 0.079 [0.074, 0.087] 0.095 22,251.54
Nested model 0.928 0.916 0.068 [0.061, 0.074] 0.050 22,644.481
Open in a new tab

The nested model excluded the non-significant structural path from mentalization ability (RFQ_lat) to substance-related risk behavior (WHO_lat); CFI = comparative fit index; TLI = Tucker–Lewis index; RMSEA = root mean square error of approximation; SRMR = standardized root mean square residual.

Table 3.

Standardized factor loadings of the measurement model.

Scale Subscale λ
CTQ Emotional abuse 0.85
Physical abuse 0.56
Sexual abuse 0.45
Emotional neglect 0.78
Physical neglect 0.62
IPO Identity diffusion 0.81
Primitive defenses 0.85
Reality testing 0.79
RFQ RFQ1 0.45
RFQ2 0.70
RFQ3 0.62
RFQ4 0.69
RFQ5 0.84
RFQ6 0.56
WHO Frequency 0.79
Craving 0.85
Problems related to use 0.81
Expressed concerns 0.83
Attempts to reduce use 0.84
Intravenous use 0.27
Open in a new tab
Table 4.

Direct, indirect, and total effects from the structural equation model.

Effect Path β p 95% CI
Direct effects CTQ_lat → IPO_lat 0.47 <0.001 [0.37, 0.57]
CTQ_lat → RFQ_lat 0.40 <0.001 [0.29, 0.50]
IPO_lat → WHO_lat 0.31 <0.001 [0.20, 0.41]
RFQ_lat → WHO_lat – >0.05 –
CTQ_lat → WHO_lat 0.19 0.002 [0.07, 0.30]
Indirect effects CTQ_lat → IPO_lat → WHO_lat 0.14 <0.001 [0.09, 0.20]
Total effect CTQ_lat → WHO_lat 0.33 <0.001 [0.23, 0.43]
Open in a new tab

3.2.3. Direct effects

The analysis demonstrated that childhood trauma was significantly associated with greater structural impairment of personality and reduced mentalization ability. Higher levels of childhood trauma predicted greater structural impairment (β = 0.47, 95% CI [0.37, 0.57], p < 0.001) and reduced mentalization ability (β = 0.40, 95% CI [0.29, 0.50], p < 0.001). Structural personality impairment was significantly positively associated with addictive behavior (β = 0.31, 95% CI [0.20, 0.41], p < 0.001). The direct effect of childhood trauma on addictive behavior was also significant (β = 0.19, 95% CI [0.07, 0.30], p = 0.002).

3.2.4. Indirect effects

The results showed a significant indirect effect of childhood trauma on addictive behavior via structural personality impairment (βind = 0.14, 95% CI [0.09, 0.20], p < 0.001). The direct effect of childhood trauma on addictive behavior remained significant. These results indicate a partial mediation. The total effect of childhood trauma on addictive behavior combining direct and indirect pathways was also significant (βtotal = 0.33, 95% CI [0.23, 0.43], p < 0.001).

4. Study 2

4.1. Materials and methods

4.1.1. Sample and procedure

The data for the SUD sample were collected during the same time period as those in Study 1. For this study, paper questionnaires were distributed to 75 patients at the Grüner Kreis organization. All participants were undergoing inpatient treatment for a diagnosed SUD. Seven individuals were excluded due to missing data. The questionnaires were exactly the same as those used in Study 1. The final SUD sample consisted of 68 participants aged between 19 and 72 years (M = 35.96, SD = 12.62). A majority of the participants were male (76.5%), while 23.5% were identified as female. In addition, a majority of the participants were single (70.6%); a small number reported being in a relationship (5.9%), married (10.3%), divorced (11.8%), or widowed (1.5%). Nearly half of the sample had completed professional or technical training (48.5%), followed by a compulsory school diploma (27.9%) and a high school diploma (19.1%). Only a small proportion of participants reported having an academic degree (≤1.5%). With regard to substance-related disorders, alcohol-related problems were reported most frequently (36.8%), followed by the use of stimulants (25.0%), opioids (17.6%), and benzodiazepines (11.8%). Polydrug use was reported by 13.2% of the sample. The inclusion criteria were a minimum age of 18 years and current or previous psychiatric or psychotherapeutic treatment.

4.2. Statistical analysis

The analyses were performed in R (version 4.4.3; R Core Team, 2025) using the blavaan package (Merkle and Rosseel, 2018), which extends the lavaan framework for Bayesian structural equation modeling (Gabry et al., 2019; Rosseel, 2012). Model fit was evaluated using the posterior predictive p-value (PPP), where values of approximately 0.50 indicate an optimal fit between the observed data and the model-implied data (Gelman et al., 2013; Smid and Winter, 2020). A latent Bayesian SEM was specified equivalent to Study 1. A preliminary evaluation of the measurement model indicated weak and inconsistent factor loadings for the latent substance use construct (WHO_lat). Specifically, items reflecting active consumption behavior (e.g., intravenous use and consumption frequency) showed weak correlations with subjective items (e.g., craving and health problems), likely due to the abstinence requirement in the inpatient setting. Combined with the small sample size, this led to model convergence issues and poor fit (PPP = 0.001). Consequently, the analysis was shifted to a Bayesian path model using manifest variables to ensure robust estimation. The initial model including all parameters (CTQ, IPO, RFQ, and WHO) still showed a weak fit (PPP = 0.313). After removing the RFQ–WHO path, the fit improved only slightly (PPP = 0.321). When RFQ was excluded from the model entirely, the fit increased further (PPP = 0.480). The final model therefore retained CTQ, IPO, and WHO. To examine the robustness of these results and account for the limited data availability, a sensitivity analysis was performed comparing three different prior specifications. These included a model with default diffuse priors [N(0, 10)], a model with weakly informative priors [N(0, 0.5)], and a final model utilizing informed priors [N(0.33, 0.15)] derived from the empirical findings of Study 1. While all three path specifications demonstrated excellent global fit (PPP ranging from 0.480 to 0.519), the models relying on diffuse and weakly informative priors yielded wide credibility intervals that consistently crossed zero. The Bayesian path model with informed priors demonstrated a good overall fit to the data (PPP = 0.480).

5. Results

5.1. Direct effects

Using the informed priors, childhood trauma was significantly associated with structural personality impairment. The posterior mean for the unstandardized effect was B = 0.30 (PSD = 0.11, 95% CI [0.10, 0.51]), corresponding to a standardized effect of β = 0.24. Similarly, structural personality impairment significantly predicted addictive behavior of B = 0.23 (PSD = 0.10, 95% CI [0.04, 0.43], β = 0.22). In contrast to Study 1, the direct path from childhood trauma to addictive behavior was not significant, with the 95% credibility interval crossing zero with B = 0.17 (PSD = 0.11, 95% CI [−0.05, 0.39], β = 0.12).

5.2. Indirect effects

The analysis of the indirect effect showed a posterior mean of Bind = 0.07 (PSD = 0.04). However, the 95% credibility interval ranged from −0.01 to 0.15, slightly including the value zero. While the direction of the effect is consistent with Study 1 (βind = 0.05), it does not reach strict statistical significance in the smaller SUD sample. The total effect was positive, and its credibility interval excluded zero (Btotal = 0.24, 95% CI [0.01, 0.46], βtotal = 0.17), suggesting an overall association between trauma and addictive behavior in the SUD group (see also Table 5).

Table 5.

Bayesian path model with informed priors (SUD sample, N = 68).

Effect Path B Post. SD 95% CI β
Direct effects CTQ → IPO 0.30 0.11 [0.10, 0.51] 0.24
IPO → WHO 0.23 0.10 [0.04, 0.43] 0.22
CTQ → WHO 0.17 0.11 [−0.05, 0.39] 0.12
Indirect effect CTQ → IPO → WHO 0.07 0.04 [−0.01, 0.15] 0.05
Total effect CTQ → WHO 0.24 – [0.01, 0.46] 0.17
Open in a new tab

6. Discussion

The primary aim of this study was to identify psychological mechanisms linking childhood trauma to addictive behavior, with the focus on the mediating role of PO and mentalization. The results from both studies can be interpreted within the framework of developmental and psychodynamic theories. From this perspective, early adverse childhood experiences have been associated with impairments in self-regulation skills and increased vulnerability to psychopathology and addictive behavior (Felitti et al., 1998; Khantzian, 1985; Bateman and Fonagy, 2019). In both samples, childhood trauma was significantly associated with impairments in structural personality organization, which is also consistent with Kernberg’s (1984) object relations model that suggests that early relational trauma can impair the integration of positive and negative experiences with attachment figures and can lead to long-term structural impairments. This interpretation may also be in line with neurobiological findings. Trauma during sensitive developmental stages may be associated with changes in brain regions involved in affect regulation and impulse control, given that impaired impulse control has repeatedly been associated with an increased vulnerability for substance use disorders (Kirsch et al., 2020; Koob and Volkow, 2016; Teicher and Samson, 2016).

In Study 1, the mediation path was statistically significant, with structural impairments partially explaining the association between childhood trauma and addictive behavior, which is consistent with the assumption that personality structure may serve as a regulatory factor in the general population (Freier et al., 2022; Krakau et al., 2021). Lower levels of structural integration were associated with higher addictive behavior. Difficulties in structural personality integration may be associated with reduced affect regulation capacities and a greater tendency to use substances as a maladaptive coping strategy. This perspective is reflected in the self-medication hypothesis (Khantzian, 1997, 2013; Suh et al., 2008), which proposes that substance use following early traumatic experiences may, in part, serve as a means of regulating underlying structural impairments (Costanzo et al., 2023; Hawn et al., 2020).

In Study 2, childhood trauma was statistically associated with impairments in structural personality organization. These, in turn, were related to addictive behavior. From this perspective, substance misuse may be understood as one of the several possible expressions of impaired self-regulatory integration (Caligor et al., 2018; Levy et al., 2006). The absence of a direct association between childhood trauma and addictive behavior in Study 2 should be considered in the context of the study conditions. In contrast to Study 1, in which natural variations in substance use were observed, the SUD sample was assessed within a controlled inpatient setting focused on abstinence. Therefore, the WHO-ASSIST scores in Study 2 may reflect past patterns of substance misuse and subjective craving rather than acute addictive behavior. Given that participants were assessed within a structured inpatient rehabilitation setting characterized by abstinence requirements, restricted access to substances, and ongoing therapeutic support, variability in current substance use was likely reduced. Accordingly, the observed associations in the clinical sample may have been more strongly related to enduring psychological burden and structural vulnerabilities than to current substance use behavior itself.

This contextual factor may have weakened the observable direct relationship between trauma and addictive behavior within the statistical model. In Study 2, the direct association between childhood trauma and addictive behavior was not statistically significant. Similarly, the specific indirect pathway via structural personality organization did not achieve statistical significance although the credibility interval only slightly exceeded zero. It is important to point out that the overall effect of childhood trauma on addictive behavior remained statistically credible, suggesting a robust overall association. This difference compared to the community sample is likely due to methodological and sampling-related characteristics. The current findings suggest that, within the present model, structural personality impairment emerged as the strongest and most consistent correlate of addictive behavior in the clinical sample. This may indicate that, in individuals with more severe clinical presentations, substance-related burden is closely associated with broader impairments in personality functioning. In this scenario, the influence of childhood trauma may be reflected in current structural impairments, which were more strongly associated with addictive behavior in the present model. Studies of individuals with severe personality pathology and high trauma exposure have shown that structural personality impairments occur frequently. These impairments are associated with far-reaching difficulties in regulating cognition, emotions, and behavior (Sharp et al., 2011; Wright and Simms, 2016). This concept is also referred to in current literature as the “p-factor,” which suggests that severe structural impairments function as a central vulnerability for general psychopathology (Caspi et al., 2024). This is consistent with the recent paradigm shift in the ICD-11 and the “Alternative Model for Personality Disorders” in the DSM-5. According to this model, the general level of personality functioning is viewed as the transdiagnostic core of personality pathology (Bach and First, 2018; Hopwood et al., 2018; Zimmermann et al., 2019). In samples of individuals with severe general functional impairments, who often have multiple mental disorders and comorbid conditions, more specific associations, such as the association between trauma and addictive behavior, may be less clearly visible in statistical models (Caspi et al., 2024).

In line with theories of epistemic mistrust (Fonagy et al., 2019; Fonagy et al., 2016; Kampling et al., 2022), Study 1 showed that a greater extent of childhood trauma is associated with lower mentalization ability. Explanatory models suggest that early trauma may impair trust in others and the capacity for self-reflection (Jurist, 2005). Although mentalization deficits were associated with childhood trauma, they did not show an independent association with addictive behavior once structural personality organization was considered in the model. This suggests that their statistical contribution could be partly explained by more comprehensive structural impairments. This lack of an independent statistical mediation pathway should not be seen as an indication of the irrelevance of the concept. Instead, it points to conceptual overlaps between mentalization and general personality structure (Zettl et al., 2020). From this perspective, the findings may support the assumption that mentalization represents a more specific psychological capacity operating within broader domains of personality functioning rather than a fully independent mechanism. Both instruments partly assess capacities related to self-reflection, affect regulation, and interpersonal functioning, which may have reduced the ability to detect unique effects of mentalization within the structural models (Marszał and Górska, 2015). It is possible that, in a unified model, the broader transdiagnostic construct of structural personality impairment takes precedence, causing the significance of mentalization ability to recede into the background (Luyten et al., 2020). Additionally, measurement constraints of the RFQ-6 short form may have contributed to the absence of distinct effects as the brief format may not sufficiently capture fluctuations between hypermentalizing and hypomentalizing processes frequently observed in addictive disorders (Choi-Kain and Gunderson, 2008; Taubner and Curth, 2013; Taubner and Volkert, 2016).

In the SUD sample, structural personality impairment emerged as the most consistent correlate of addictive behavior within the model. This is supported by other clinical studies, which show that hospitalized patients with personality impairment are more likely to engage in impulsive and addictive behaviors (Gratz and Tull, 2010; Weiss et al., 2012). These findings are compatible with theoretical and psychodynamic treatment models that emphasize personality structure as a central dimension in the treatment of severe psychopathology (Kernberg et al., 2008; Yeomans et al., 2015). From this perspective, addiction among individuals who have experienced trauma in the past can be understood not merely as a behavioral problem but as part of broader structural vulnerabilities. Interventions focusing on identity integration and the development of more adaptive coping mechanisms could therefore represent a transdiagnostic approach aimed at strengthening general self-regulation abilities (Unterrainer, 2025; Clarkin et al., 1999).

The findings of the present study are consistent with theoretical perspectives suggesting that the addictive behavior in the context of traumatic childhood experiences may, in part, reflect attempts to maintain emotional and psychological stability (Costanzo et al., 2023; Khantzian, 1997). From this perspective, abstinence alone may not entirely address the underlying structural vulnerabilities associated with SUD. Additional corrective emotional experiences and a therapeutic relationship are important and can serve as a stable framework for building trust and strengthening the capacity for self-reflection (Bateman and Fonagy, 2019; Yeomans et al., 2015). In this context, the present findings could support treatment approaches that focus not only on symptom reduction but also on the long-term development of psychological structures (Unterrainer, 2025).

7. Limitations and future perspectives

Despite a careful methodological approach, it is important to mention the limitations of this study. Although statistical mediation models were tested, the present analyses are based on cross-sectional data and the identified indirect pathways should therefore be interpreted as statistical associations rather than evidence of temporal or causal processes. Alternative causal relationships between the study variables cannot be ruled out. For example, addictive behavior itself may contribute to impairments in personality functioning or reflective capacity, and reciprocal associations between these constructs are also theoretically plausible (Kline, 2023). Although several model respecifications and sensitivity analyses were conducted, these primarily addressed statistical parsimony and the robustness of Bayesian estimates rather than substantively different theoretical models. Therefore, competing pathways, including reversed or reciprocal associations and alternative mediation structures, were not formally tested. Given the cross-sectional design, the directionality of these associations cannot be determined. Future longitudinal studies are needed to compare competing theoretical models and clarify causal relationships among childhood trauma, personality functioning, mentalization, and addictive behavior (Bateman and Fonagy, 2019; Schindler, 2019).

Regarding the data collection performed in Study 1, the recruitment via social media and university-related platforms may have introduced self-selection biases as individuals with greater interest in psychology, higher educational levels, or increased willingness to reflect on personal experiences may have been more likely to participate. Furthermore, the comparatively high exclusion rate may have affected the representativeness of the final sample. This may partly be attributable to the relatively long assessment procedure, which consisted of 108 questionnaire items and may have increased the likelihood of incomplete responses or study discontinuation. It is possible that individuals with higher symptom burden or lower concentration were somewhat less likely to complete the assessment procedure, potentially resulting in a final sample with slightly higher functioning. Relatedly, the present models did not account for potential confounding variables. Although psychiatric diagnoses and symptom-related variables, such as depression, anxiety, and post-traumatic stress symptoms, were available descriptively, these variables were not included in the structural analyses. Socioeconomic status and other clinically relevant factors were also not considered. Therefore, the observed associations between childhood trauma, personality organization, mentalization, and addictive behavior should be interpreted within the boundaries of the selected theoretical model. Future studies should include additional clinically relevant variables as covariates, such as depressive symptoms, post-traumatic stress symptoms, and socioeconomic status, to assess the robustness of the observed associations. Another restriction applies to the assessment of addictive behavior in the clinical sample. Given that the participants were undergoing inpatient treatment to achieve abstinence, their substance use was limited at the time of the survey. The WHO-ASSIST scores in this group therefore likely reflect the subjective burden of addiction, such as craving or past substance-related difficulties, rather than current use. Due to the inpatient setting, the participants likely differed less in their actual substance use than they would under naturalistic conditions. This restriction of variability may have weakened the observable association between childhood trauma and current addictive behavior within the statistical models. In naturalistic settings, traumatic experiences may be more directly associated with maladaptive coping through active substance misuse, particularly during periods of emotional distress or impaired self-regulation. However, within the structured inpatient environment, acute consumption behavior was externally restricted through abstinence requirements. Therefore, trauma-related differences may have been expressed less through current addictive behavior itself and more through persistent structural impairments, emotional dysregulation, or subjective craving. Additionally, childhood trauma was assessed retrospectively using self-report measures, which may be influenced by recall bias, current emotional state, or subjective reinterpretation of past experiences. Although the CTQ is a well-established instrument, retrospective assessments of childhood adversity may not fully correspond to objectively verifiable experiences (Bernstein et al., 2003).

All collected data were based on self-reports. These primarily reflect conscious assessments, while unconscious processes, particularly those related to self-regulation and emotional regulation, are only recorded to a limited extent (Panksepp and Biven, 2012). In this context, the use of self-reports poses a particular methodological challenge, as individuals with severe structural impairments may show reduced reflective capacities, potentially limiting the accuracy of self-report assessments of mentalization ability. For future studies, it might be beneficial to rely more heavily on methods based on expert-rated clinical interviews (e.g., structured interviews on personality functioning) in addition to self-report measures. In this study, both personality organization and mentalization ability can be assessed in a more nuanced and less biased manner. In particular, interview-based measures of reflective functioning or methods such as “Movie for the Assessment of Social Cognition” (Dziobek et al., 2006) may provide a more ecologically valid assessment of mentalization processes, especially in clinical populations characterized by impaired self-awareness, affect dysregulation, or unstable self-representations (Zettl et al., 2020).

In addition, the WHO-ASSIST combines multiple dimensions of substance involvement, including frequency of use, craving, social consequences, and previous attempts to reduce consumption, into a global cumulative risk score. While this allows for a broad assessment of addictive behavior, it may obscure potentially meaningful differences between substances and between current versus past substance-related difficulties. Particularly in the inpatient SUD sample, some indicators may have reflected retrospective burden or subjective craving rather than ongoing substance use behavior, which may have reduced the specificity of the construct within the statistical models. The comparatively weak factor loading of the intravenous use indicator in Study 1 likely reflects the characteristics of the community sample, in which intravenous substance use was reported only rarely. As a result, this indicator may have contributed less strongly to the overall latent substance use construct compared to broader indicators, such as craving or substance-related problems. Furthermore, the WHO-ASSIST assesses 10 different substance groups. Substances such as nicotine or alcohol can be linked with different psychological mechanisms. Therefore, future research could investigate whether different psychopathological patterns are associated with specific types of addictive behavior (Han et al., 2021; Koob and Volkow, 2016; Verdejo-García et al., 2008). If the individual classes of substances are considered separately, specific relationships, such as those with PO or mentalization, might become clearer and more distinct.

Although Bayesian analysis using informed prior distributions helped stabilize the estimates despite the sample size, the relatively small clinical sample may also have reduced statistical stability and limited the generalizability of the observed associations. In addition, the heterogeneous composition of the SUD sample and the comparatively low proportion of female participants could limit the generalizability of the results. The markedly different gender distributions across the two samples, with a higher proportion of men in the SUD sample and a higher proportion of women in the community sample, may limit the comparability and generalizability of the findings, particularly given known gender differences in trauma exposure, emotional processing, and addictive behavior (e.g., Belfrage et al., 2023; Langeland and Olff, 2024). Furthermore, the inpatient clinical sample consisted exclusively of treatment-seeking individuals undergoing abstinence-oriented rehabilitation. The findings may therefore not generalize to individuals with SUD in outpatient settings, active substance use phases, or non-treatment populations. In addition, both samples were recruited exclusively from Austria living populations, which may limit the cross-cultural generalizability of the results to healthcare systems, sociocultural environments, or substance use patterns in other regions. The findings highlight the importance of larger and more diagnostically differentiated samples to better understand the complex relationships between trauma, structural personality organization, mentalization, and the development of addiction (Hughes et al., 2017; Vonderlin et al., 2018).

8. Conclusion

The present findings emphasize the relevance of early traumatic experiences in the context of addictive. In the community sample, structural personality impairment explained part of the association between childhood trauma and addictive behavior. However, this statistical mediation pattern was not fully replicated in the SUD sample. Although patients reported higher levels of trauma exposure and structural impairments, the association with current substance use was less pronounced. This could be partly due to the influence of the inpatient treatment environment and an overall higher level of psychopathological distress. The results suggest that early traumatic experiences are closely associated with structural aspects of personality functioning. Therefore, interventions should focus on structural integration, emotional regulation, and reflective capacity in addition to symptom relief.

Funding Statement

The author(s) declared that financial support was not received for this work and/or its publication.

Edited by: David Titelman, Karolinska Institutet (KI), Sweden

Reviewed by: Tamara Fischmann, International Psychoanalytic University Berlin, Germany

Ewa Wojtynkiewicz, Kazimierz Wielki University, Poland

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics statement

The studies involving humans were approved by the ethics committees of the University of Graz (GZ. 39/5/63 ex 2023/24) and Sigmund Freud University Vienna (HD65DXEGC54CWD90938), Austria. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

Author contributions

JB: Investigation, Formal analysis, Data curation, Methodology, Writing – original draft, Writing – review & editing. ER: Investigation, Writing – review & editing. RW: Investigation, Writing – review & editing. ZZ: Investigation, Writing – review & editing. HB: Writing – review & editing. JF: Conceptualization, Methodology, Project administration, Validation, Writing – review & editing. WB: Supervision, Writing – review & editing. H-FU: Conceptualization, Project administration, Supervision, Validation, Writing – review & editing.

Conflict of interest

JF and H-FU would like to note that they were employed by the Grüner Kreis organization at the “Center of Integrative Addiction Research (CIAR)”) at the time the study was conducted.

The remaining author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that Generative AI was used in the creation of this manuscript. During the preparation of this work the authors used GPT 5.2 in order to improve the language of the manuscript. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the published article. DeepL Translate was used to further refine the English language.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

  1. American Psychiatric Association (2022). Diagnostic and Statistical Manual of mental Disorders. 5th ed., text rev. Edn. Washington, DC: American Psychiatric Publishing. [Google Scholar]
  2. Bach B., First M. B. (2018). Application of the ICD-11 classification of personality disorders. BMC Psychiatry 18:351. doi: 10.1186/s12888-018-1908-3, [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bateman A. W., Fonagy P. (2016). Mentalization-Based Treatment for Personality Disorders: A Practical Guide. Oxford: Oxford University Press. [Google Scholar]
  4. Bateman A. W., Fonagy P. (2019). Handbook of Mentalizing in Mental Health Practice. 2nd Edn. Washington, DC: American Psychiatric Publishing. [Google Scholar]
  5. Belfrage A., Mjølhus Njå A. L., Lunde S., Årstad J., Fodstad E. C., Lid T. G., et al. (2023). Traumatic experiences and PTSD symptoms in substance use disorder: a comparison of recovered versus current users. Nordic Stud. Alcohol Drugs 40, 61–75. doi: 10.1177/14550725221122222, [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bernstein D. P., Fink L. (1998). Childhood Trauma Questionnaire: A Retrospective Self-Report. San Antonio, TX: Psychological Corporation. [Google Scholar]
  7. Bernstein D. P., Stein J. A., Newcomb M. D., Walker E., Pogge D., Ahluvalia T., et al. (2003). Development and validation of a brief screening version of the childhood trauma questionnaire. Child Abuse Negl. 27, 169–190. doi: 10.1016/S0145-2134(02)00541-0, [DOI] [PubMed] [Google Scholar]
  8. Caligor E., Kernberg O. F., Clarkin J. F., Yeomans F. E. (2018). Psychodynamic Therapy for Personality Pathology: Treating Self and Interpersonal Functioning. Washington, DC: American Psychiatric Association Publishing. [Google Scholar]
  9. Caspi A., Houts R. M., Fisher H. L., Danese A., Moffitt T. E. (2024). The general factor of psychopathology (p): choosing among competing models and interpreting p. Clin. Psychol. Sci. 12, 53–82. doi: 10.1177/216770262211478, [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Cheung G. W., Lau R. S. (2008). Testing mediation and suppression effects of latent variables: bootstrapping with structural equation models. Organ. Res. Methods 11, 296–325. doi: 10.1177/1094428107300343 [DOI] [Google Scholar]
  11. Choi-Kain L. W., Gunderson J. G. (2008). Mentalization: ontogeny, assessment, and application in the treatment of borderline personality disorder. Am. J. Psychiatry 165, 1127–1135. doi: 10.1176/appi.ajp.2008.07081360, [DOI] [PubMed] [Google Scholar]
  12. Clarkin J. F., Yeomans F. E., Kernberg O. F. (1999). Psychotherapy for Borderline Personality. New York, NY: John Wiley & Sons Inc. [Google Scholar]
  13. Costanzo A., Santoro G., Schimmenti A. (2023). Self-medication, traumatic reenactments, and dissociation: a psychoanalytic perspective on the relationship between childhood trauma and substance abuse. Psychoanal. Psychother. 37, 443–466. doi: 10.1080/02668734.2023.2272761 [DOI] [Google Scholar]
  14. DePaoli S. (2021). Bayesian Structural Equation Modeling. 1st Edn. New York, NY: Guilford Publications. [Google Scholar]
  15. Dube S. R., Felitti V. J., Dong M., Giles W. H., Anda R. F. (2003). The impact of adverse childhood experiences on health problems: evidence from four birth cohorts dating back to 1900. Prev. Med. 37, 268–277. doi: 10.1016/S0091-7435(03)00123-3, [DOI] [PubMed] [Google Scholar]
  16. Dziobek I., Fleck S., Kalbe E., Rogers K., Hassenstab J., Brand M., et al. (2006). Introducing MASC: a movie for the assessment of social cognition. J. Autism Dev. Disord. 36, 623–636. doi: 10.1007/s10803-006-0107-0, [DOI] [PubMed] [Google Scholar]
  17. Felitti V. J., Anda R. F., Nordenberg D., Williamson D. F., Spitz A. M., Edwards V., et al. (1998). Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: the adverse childhood experiences (ACE) study. Am. J. Prev. Med. 14, 245–258. doi: 10.1016/S0749-3797(98)00017-8, [DOI] [PubMed] [Google Scholar]
  18. Fonagy P., Gergely G., Jurist E. L. (2018). Affect Regulation, Mentalization and the Development of the Self. London: Routledge. [Google Scholar]
  19. Fonagy P., Luyten P. (2009). A developmental, mentalization-based approach to the understanding and treatment of borderline personality disorder. Dev. Psychopathol. 21, 1355–1381. doi: 10.1017/S0954579409990198, [DOI] [PubMed] [Google Scholar]
  20. Fonagy P., Luyten P., Allison E., Campbell C. (2019). Mentalizing, epistemic trust and the phenomenology of psychotherapy. Psychopathology 52, 94–103. doi: 10.1159/000501526, [DOI] [PubMed] [Google Scholar]
  21. Fonagy P., Luyten P., Moulton-Perkins A., Lee Y.-W., Warren F., Howard S., et al. (2016). Development and validation of a self-report measure of mentalizing: the reflective functioning questionnaire. PLoS One 11:e0158678. doi: 10.1371/journal.pone.0158678, [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Freier A., Kruse J., Schmalbach B., Zara S., Werner S., Brähler E., et al. (2022). The mediation effect of personality functioning between different types of child maltreatment and the development of depression/anxiety symptoms–a German representative study. J. Affect. Disord. 299, 408–415. doi: 10.1016/j.jad.2021.12.020, [DOI] [PubMed] [Google Scholar]
  23. Fuchshuber J., Gruber M., Feichtinger K., Klauser M., Parth K., Kapusta N., et al. (2025b). Personality organization, childhood adversity, and paranoid thinking: moderating and mediating pathways. Psychopathology 58, 393–404. doi: 10.1159/000547518, [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Fuchshuber J., Hiebler-Ragger M., Kresse A., Kapfhammer H.-P., Unterrainer H. F. (2018). Depressive symptoms and addictive behaviors in young adults after childhood trauma: the mediating role of personality organization and despair. Front. Psych. 9:318. doi: 10.3389/fpsyt.2018.00318, [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Fuchshuber J., Hiebler-Ragger M., Kresse A., Kapfhammer H.-P., Unterrainer H.-F. (2019). The influence of attachment styles and personality organization on emotional functioning after childhood trauma. Front. Psych. 10:643. doi: 10.3389/fpsyt.2019.00643, [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Fuchshuber J., Unterrainer H.-F. (2020). Childhood trauma, personality, and substance use disorder: the development of a neuropsychoanalytic addiction model. Front. Psych. 11:531. doi: 10.3389/fpsyt.2020.00531, [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Fuchshuber J., Unterrainer H. F., Kapusta N. D., Blüml V. (2025a). Mapping childhood trauma and personality organization: a network approach. Psychoanal. Psychol. 42, 181–190. doi: 10.1037/pap0000549 [DOI] [Google Scholar]
  28. Gabry J., Simpson D., Vehtari A., Betancourt M., Gelman A. (2019). Visualization in Bayesian workflow. J. R. Stat. Soc. A. Stat. Soc. 182, 389–402. doi: 10.1111/rssa.12378 [DOI] [Google Scholar]
  29. Gelman A., Carlin J. B., Stern H. S., Dunson D. B., Vehtari A., Rubin D. B. (2013). Bayesian data analysis. 3rd Edn. Boca Raton, FL: CRC Press. [Google Scholar]
  30. Gratz K. L., Tull M. T. (2010). The relationship between emotion dysregulation and deliberate self-harm among inpatients with substance use disorders. Cogn. Ther. Res. 34, 544–553. doi: 10.1007/s10608-009-9268-4, [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Gross J. J. (2015). Emotion regulation: current status and future prospects. Psychol. Inq. 26, 1–26. doi: 10.1080/1047840X.2014.940781 [DOI] [Google Scholar]
  32. Han B., Jones C. M., Einstein E. B., Powell P. A., Compton W. M. (2021). Use of medications for alcohol use disorder in the United States: results from the 2019 National Survey on Drug Use and Health. JAMA Psychiatry 78, 922–924. doi: 10.1001/jamapsychiatry.2021.1271, [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Hawn S. E., Cusack S. E., Amstadter A. B. (2020). A systematic review of the self-medication hypothesis in the context of posttraumatic stress disorder and comorbid problematic alcohol use. J. Trauma. Stress. 33, 699–708. doi: 10.1002/jts.22521, [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Hayes A. F. (2022). Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach. New York, NY: Third Edn The Guilford Press. [Google Scholar]
  35. Hopwood C. J., Kotov R., Krueger R. F., Watson D., Widiger T. A., Althoff R. R., et al. (2018). The time has come for dimensional personality disorder diagnosis. Personal. Ment. Health 12, 82–86. doi: 10.1002/pmh.1408, [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Hu L.-T., Bentler P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Struct. Equ. Model. 6, 1–55. doi: 10.1080/10705519909540118 [DOI] [Google Scholar]
  37. Hughes K., Bellis M. A., Hardcastle K. A., Sethi D., Butchart A., Mikton C., et al. (2017). The effect of multiple adverse childhood experiences on health: a systematic review and meta-analysis. Lancet Public Health 2, e356–e366. doi: 10.1016/S2468-2667(17)30118-4, [DOI] [PubMed] [Google Scholar]
  38. Humeniuk R., Henry-Edwards S., Ali R., Poznyak V., Monteiro M. G. (2010). The Alcohol, Smoking and Substance Involvement Screening Test (ASSIST): Manual for Use in Primary Care. Geneva: World Health Organization. [Google Scholar]
  39. Hyman S. M., Paliwal P., Chaplin T. M., Mazure C. M., Rounsaville B. J., Sinha R. (2008). Severity of childhood trauma is predictive of cocaine relapse outcomes in women but not men. Drug Alcohol Depend. 92, 208–216. doi: 10.1016/j.drugalcdep.2007.08.006, [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Insel T. R. (2014). The NIMH research domain criteria (RDoC) project: precision medicine for psychiatry. Am. J. Psychiatry 171, 395–397. doi: 10.1176/appi.ajp.2014.14020138, [DOI] [PubMed] [Google Scholar]
  41. Jurist E. L. (2005). Mentalized affectivity. Psychoanal. Psychol. 22, 426–444. doi: 10.1037/0736-9735.22.3.426 [DOI] [Google Scholar]
  42. Kampling H., Kruse J., Lampe A., Nolte T., Hettich N., Brähler E., et al. (2022). Epistemic trust and personality functioning mediate the association between adverse childhood experiences and posttraumatic stress disorder and complex posttraumatic stress disorder in adulthood. Front. Psych. 13:919191. doi: 10.3389/fpsyt.2022.919191, [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Kernberg O. F. (1984). Severe Personality Disorders: Psychotherapeutic Strategies. New Haven, CT: Yale University Press. [Google Scholar]
  44. Kernberg O. F., Yeomans F. E., Clarkin J. F., Levy K. N. (2008). Transference focused psychotherapy: overview and update. Int. J. Psychoanal. 89, 601–620. doi: 10.1111/j.1745-8315.2008.00046.x, [DOI] [PubMed] [Google Scholar]
  45. Khantzian E. J. (1985). The self-medication hypothesis of addictive disorders: focus on heroin and cocaine dependence. Am. J. Psychiatry 142, 1259–1264. doi: 10.1176/ajp.142.11.1259, [DOI] [PubMed] [Google Scholar]
  46. Khantzian E. J. (1997). The self-medication hypothesis of substance use disorders: a reconsideration and recent applications. Harv. Rev. Psychiatry 4, 231–244. doi: 10.3109/10673229709030550, [DOI] [PubMed] [Google Scholar]
  47. Khantzian E. J. (2013). Addiction as a self-regulation disorder and the role of self-medication. Addiction 108, 668–669. doi: 10.1111/add.12004, [DOI] [PubMed] [Google Scholar]
  48. Kirsch D. E., Nemeroff C. B., Lippard E. T. C. (2020). Early life stress and substance use disorders: underlying neurobiology and pathways to adverse outcomes. Advers. Resil. Sci. 1, 29–47. doi: 10.1007/s42844-020-00005-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Kline R. B. (2023). Principles and Practice of Structural Equation Modeling. New York, NY: Guilford publications. [Google Scholar]
  50. Klinitzke G., Romppel M., Häuser W., Brähler E., Glaesmer H. (2012). The German version of the childhood trauma questionnaire (CTQ): psychometric characteristics in a representative sample of the general population. Psychother. Psychosom. Med. Psychol. 62, 47–51. doi: 10.1055/s-0031-1295495, [DOI] [PubMed] [Google Scholar]
  51. Koob G. F., Volkow N. D. (2016). Neurobiology of addiction: a neurocircuitry analysis. Lancet Psychiatry 3, 760–773. doi: 10.1016/S2215-0366(16)00104-8, [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Krakau L., Löffler-Stastka H., Unterrainer H.-F., Blüml V. (2021). Personality functioning as a mediator of adult mental health following child maltreatment. J. Affect. Disord. 291, 126–134. doi: 10.1016/j.jad.2021.05.006, [DOI] [PubMed] [Google Scholar]
  53. Langeland W., Olff M. (2024). Sex and gender in psychotrauma research. Eur. J. Psychotraumatol. 15:2358702. doi: 10.1080/20008066.2024.2358702, [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Lenzenweger M. F., Clarkin J. F., Kernberg O. F., Foelsch P. A. (2001). The inventory of personality organization: psychometric properties, factorial composition, and criterion relations with affect, aggressive dyscontrol, psychosis proneness, and self-domains in a nonclinical sample. Psychol. Assess. 13, 577–591. doi: 10.1037/1040-3590.13.4.577, [DOI] [PubMed] [Google Scholar]
  55. Levy K. N., Meehan K. B., Kelly K. M., Reynoso J. S., Weber M., Clarkin J. F., et al. (2006). Change in attachment patterns and reflective function in a randomized control trial of transference-focused psychotherapy for borderline personality disorder. J. Consult. Clin. Psychol. 74, 1027–1040. doi: 10.1037/0022-006X.74.6.1027, [DOI] [PubMed] [Google Scholar]
  56. Luyten P., Campbell C., Allison E., Fonagy P. (2020). The mentalizing approach to psychopathology: state of the art and future directions. Annu. Rev. Clin. Psychol. 16, 297–325. doi: 10.1146/annurev-clinpsy-071919-015355, [DOI] [PubMed] [Google Scholar]
  57. Luyten P., Fonagy P. (2015). The neurobiology of mentalizing. Personal. Disord. Theory Res. Treat. 6, 366–379. doi: 10.1037/per0000117, [DOI] [PubMed] [Google Scholar]
  58. Marszał M., Górska D. (2015). The regulative function of mentalization and mindfulness in borderline personality organization. Curr. Issues Personal. Psychol. 3, 51–63. doi: 10.5114/cipp.2015.50208 [DOI] [Google Scholar]
  59. McNeish D. (2016). On using Bayesian methods to address small sample problems. Struct. Equ. Model. 23, 750–773. doi: 10.1080/10705511.2016.1186549 [DOI] [Google Scholar]
  60. Merkle E. C., Rosseel Y. (2018). Blavaan: Bayesian structural equation models via parameter expansion. J. Stat. Softw. 85, 1–30. doi: 10.18637/jss.v085.i0430505247 [DOI] [Google Scholar]
  61. Mikulincer M., Shaver P. R. (2019). Attachment orientations and emotion regulation. Curr. Opin. Psychol. 25, 6–10. doi: 10.1016/j.copsyc.2018.02.006, [DOI] [PubMed] [Google Scholar]
  62. Norman R. E., Byambaa M., De R., Butchart A., Scott J., Vos T. (2012). The long-term health consequences of child physical abuse, emotional abuse, and neglect: a systematic review and meta-analysis. PLoS Med. 9:e1001349. doi: 10.1371/journal.pmed.1001349, [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Panksepp J., Biven L. (2012). The Archaeology of Mind: Neuroevolutionary Origins of Human Emotions. New York, NY: W. W. Norton & Company. [Google Scholar]
  64. R Core Team (2025). R: A Language and Environment for Statistical Computing (Version 4.4.3). Vienna: R Foundation for Statistical Computing. [Google Scholar]
  65. Renner E., Wimmer R., Zipper Z., Brössler H., Brouschek J., Fuchshuber J., et al. (2025). Affects, dissociative symptoms and addictive behaviours: insights from clinical and community samples. J. Affect. Disord. 370, 481–488. doi: 10.1016/j.jad.2024.11.031, [DOI] [PubMed] [Google Scholar]
  66. Rosseel Y. (2012). Lavaan: an R package for structural equation modeling. J. Stat. Softw. 48, 1–36. doi: 10.18637/jss.v048.i02 [DOI] [Google Scholar]
  67. Schindler A. (2019). Attachment and substance use disorders—theoretical models, empirical evidence, and implications for treatment. Front. Psych. 10:727. doi: 10.3389/fpsyt.2019.00727, [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Sharp C., Carolyn H. A., Fonagy P. (2011). Get them before they get you: trust, trustworthiness, and social cognition in boys with and without externalizing behavior problems. Dev. Psychopathol. 23, 647–658. doi: 10.1017/S0954579410000003 [DOI] [PubMed] [Google Scholar]
  69. Smid S. C., Winter S. D. (2020). Dangers of the defaults: a tutorial on the impact of default priors when using Bayesian SEM with small samples. Front. Psychol. 11:611963. doi: 10.3389/fpsyg.2020.611963, [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Spitzer C., Zimmermann J., Brähler E., Euler S., Wendt L., Müller S. (2021). Die deutsche Version des Reflective Functioning Questionnaire (RFQ): Eine teststatistische Überprüfung in der Allgemeinbevölkerung. Psychother. Psychosom. Med. Psychol. 71, 124–131. doi: 10.1055/a-1234-6317, [DOI] [PubMed] [Google Scholar]
  71. Suh J. J., Ruffins S., Robins C. E., Albanese M. J., Khantzian E. J. (2008). Self-medication hypothesis: connecting affective experience and drug choice. Psychoanal. Psychol. 25, 518–532. doi: 10.1037/0736-9735.25.3.518 [DOI] [Google Scholar]
  72. Taubner S., Curth C. (2013). Mentalization mediates the relation between early traumatic experiences and aggressive behavior in adolescence. Psihologija 46, 177–192. doi: 10.2298/PSI1302177T [DOI] [Google Scholar]
  73. Taubner S., Volkert J. (2016). Mentalisierungsbasierte Therapie für Adoleszente (MBT-A). Gottingen: Vandenhoeck & Ruprecht. [Google Scholar]
  74. Teicher M. H., Samson J. A. (2016). Annual research review: enduring neurobiological effects of childhood abuse and neglect. J. Child Psychol. Psychiatry 57, 241–266. doi: 10.1111/jcpp.12507, [DOI] [PMC free article] [PubMed] [Google Scholar]
  75. Thorberg F. A., Lyvers M. (2010). Attachment in relation to affect regulation and interpersonal functioning among substance use disorder in patients. Addict. Res. Theory 18, 464–478. doi: 10.3109/16066350903254783 [DOI] [Google Scholar]
  76. United Nations Office on Drugs and Crime (2025). World Drug Report 2025. New York, Vienna: United Nations Publication. [Google Scholar]
  77. Unterrainer H. F. (2025). Addiction, attachment, and the brain: a focused review of empirical findings and future directions. Front. Hum. Neurosci. 19:1625880. doi: 10.3389/fnhum.2025.1625880, [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. van de Schoot R., Depaoli S., King R., Kramer B., Märtens K., Tadesse M. G., et al. (2021). Bayesian statistics and modelling. Nat Rev Methods Primer 1:1. doi: 10.1038/s43586-020-00001-2 [DOI] [Google Scholar]
  79. Verdejo-García A., Lawrence A. J., Clark L. (2008). Impulsivity as a vulnerability marker for substance-use disorders: review of findings from high-risk research, problem gamblers and genetic association studies. Neurosci. Biobehav. Rev. 32, 777–810. doi: 10.1016/j.neubiorev.2007.11.003, [DOI] [PubMed] [Google Scholar]
  80. Vonderlin R., Kleindienst N., Alpers G. W., Bohus M., Lyssenko L. (2018). Dissociation in victims of childhood abuse or neglect: a meta-analytic review. Psychol. Med. 48, 2467–2476. doi: 10.1017/S0033291718000740, [DOI] [PubMed] [Google Scholar]
  81. Weiss N. H., Tull M. T., Viana A. G., Anestis M. D., Gratz K. L. (2012). Impulsive behaviors as an emotion regulation strategy: examining associations between intolerance of emotional distress and impulsive behaviors among substance dependent inpatients. J. Anxiety Disord. 26, 453–458. doi: 10.1016/j.janxdis.2012.01.007, [DOI] [PMC free article] [PubMed] [Google Scholar]
  82. World Health Organization (2025). World Mental Health Today: Latest Data. Geneva: World Health Organization. [Google Scholar]
  83. Wright A. G. C., Simms L. J. (2016). Stability and fluctuation of personality disorder features in daily life. J. Abnorm. Psychol. 125, 641–656. doi: 10.1037/abn0000169, [DOI] [PMC free article] [PubMed] [Google Scholar]
  84. Yeomans F. E., Clarkin J. F., Kernberg O. F. (2015). Transference-Focused Psychotherapy for Borderline Personality Disorder: A Clinical Guide. Arlington, VA: American Psychiatric Publishing. [Google Scholar]
  85. Zettl M., Volkert J., Vögele C., Kubera K., Taubner S. (2020). Mentalization and criterion a of the alternative model for personality disorders: results from a clinical and nonclinical sample. Personal. Disord. Theory Res. Treat. 11, 191–201. doi: 10.1037/per0000356, [DOI] [PubMed] [Google Scholar]
  86. Zimmermann J., Benecke C., Hörz-Sagstetter S., Dammann G., Kernberg O. F. (2013). Klinische Diagnostik der Persönlichkeitsorganisation mit dem IPO-16: Validierung und Normierung an einer deutschsprachigen Stichprobe. Diagnostica 59, 3–16. doi: 10.1026/0012-1924/a000074 [DOI] [Google Scholar]
  87. Zimmermann J., Kerber A., Rek K., Hopwood C. J., Krueger R. F. (2019). A brief but comprehensive review of research on the alternative model for personality disorders. Curr. Psychiatry Rep. 21:92. doi: 10.1007/s11920-019-1079-z [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Articles from Frontiers in Psychology are provided here courtesy of Frontiers Media SA

RESOURCES