Introduction

Personality disorders (PDs) are prevalent in the general population worldwide with a point prevalence of 7.8% (Winsper et al., 2020). PDs are associated with detrimental impacts on social and occupational functioning (Skodol, 2018), and notable higher morbidity and mortality rates compared to individuals without PDs (Tyrer et al., 2021). Identifying factors contributing to personality-related dysfunction could help improve formulation and treatment for affected individuals.

Personality disorders are to date conceptualized and diagnosed within a categorical system through the Diagnostic and Statistical Manuals of Mental Disorders (American Psychiatric Association, 2013). However, researchers have raised several criticisms of this categorical approach. Personality disorders tend to co-occur, shows limited diagnostic reliability, temporal instability over time, and arbitrary diagnostic thresholds are often used (see Morey et al., 2022 for a review). A growing consensus among experts based on empirical evidence thus suggests that a dimensional rather than a categorical model, more accurately reflects human personality (Bernstein et al., 2007). Personality dysfunction, the hallmark of PDs is captured by the A criterion in dimensional models of PDs such as the Alternative DSM-5 Model for PDs (AMPD; American Psychiatric Association, 2013). Personality dysfunction refers to impairments in self and interpersonal functioning as well as general features and severity across PDs. Further, the B criterion entails maladaptive dimensional personality traits expressing individual differences or style of PDs such as “detachment” or “negative affectivity” (Hopwood et al., 2011; Mulder & Tyrer, 2019). More severe personality dysfunction is expected to be associated with a greater number of dimensional maladaptive traits and several studies have reported high correlations between criterion A and B factors (e.g., Anderson & Sellbom, 2018; Sleep et al., 2020). However, others have argued that potentially central factors for personality dysfunction beyond the B criteria are an important area of focus to better understand and alleviate personality dysfunction (Hopwood, 2018; Stricker et al., 2022).

Another longstanding model of general personality structure with an emphasis on dimensional factors is the Five-Factor Model (FFM; McCrae & Costa, 1985). This model incorporates normal-range personality traits (neuroticism, extraversion, openness, agreeableness, and conscientiousness) which overlaps with the B criteria of the AMPD (Widiger et al., 2018). In addition, research indicates they add value in explaining personality dysfunction above and beyond them (Gore & Widiger, 2013; Wright & Simms, 2014). Especially Neuroticism, Agreeableness, and Conscientiousness seem relevant showing moderate correlations with the A-criterion (McCabe et al., 2021). For example, healthy scores on agreeableness are associated with a range of positive outcomes such as social adaptiveness and work engagement (Wilmot & Ones, 2022). Thus, the inclusion of normal adaptive FFM personality traits in addition to maladaptive traits could provide more information about the totality of personality-functioning of an individual given that some level of these traits could also act as protective factors against dysfunction (Widiger et al., 2018).

In addition to trait-variables, state variables and other areas of functioning, such as comorbid psychiatric symptoms and psychosocial factors (e.g., leisure or work-related functioning), could contribute to personality dysfunction. Several studies have found personality dysfunction as assessed by self-report measures of criterion A to be moderate to strong and positively correlated with symptoms of anxiety and depression (Hutsebaut et al., 2016; Weekers et al., 2019) as well as general psychosocial functioning (e.g., Morey et al., 2013). Further disentanglement of personality dysfunction from general psychosocial impairment and psychopathology has therefore been encouraged (Morey et al., 2022).

Although not included in the current proposal of the AMPD, “metacognition” has been demonstrated as an important marker for personality dysfunction (e.g., Semerari et al., 2014; D'Abate et al., 2020; Katznelson, 2014). The metacognitive model of psychological disorder (Wells, 2019; Wells & Matthews, 1994) places dysfunctional metacognitive beliefs as the central mechanism of disorder. Metacognitions, according to this model, are operationalized as beliefs about cognition further hypothesized to activate and sustain a Cognitive Attentional Syndrome (CAS; Wells, 2009) in response to spontaneous thoughts. The CAS consists of perseverative thinking (e.g., worry, self-criticism, rumination) inflexible self-attention/threat monitoring (e.g., looking for signs of rejection or dissatisfaction in social situations), and maladaptive coping behaviors (e.g., self-harm, avoidance, reassurance-seeking) which maintain and aggravate symptoms and impair reflexive self-regulatory capacity leading to impairment in functioning. Several domains of metacognitive beliefs have been identified (Wells & Cartwright-Hatton, 2004). For example, negative metacognitive beliefs such as beliefs about uncontrollability and danger of thoughts and thinking processes (e.g., “If I start to worry, I cannot stop”) can prevent disengagement from the CAS, and positive metacognitive beliefs (e.g.,”I need to worry in order to remain organized”) can motivate engagement in CAS activity. Hence, metacognitive beliefs can impact on personality functioning through activating unhelpful cognitive and interpersonal self-regulation strategies (i.e., the CAS) that backfire and negatively impact on self- and interpersonal functioning.

Metacognitive beliefs are elevated across various psychopathologies compared to healthy controls (Sun et al., 2017). Consistent with a role for dysfunctional metacognitive beliefs in personality functioning are previous findings that patients with PDs report significantly higher levels of dysfunctional metacognitive beliefs compared to patients without PDs (Spada et al., 2021). Furthermore, dysfunctional metacognitive beliefs correlate with interpersonal problems (Nordahl et al., 2021; Strand et al., 2018, 2023a, 2023c), trait-anxiety (Nordahl et al., 2019) and neuroticism (Bailey & Wells, 2015). In addition, Metacognitive therapy (MCT; Wells, 2009) which specifically targets dysfunctional metacognitive beliefs and unhelpful mental regulation strategies have shown a large effect on interpersonal problems in several treatment studies (e.g., Nordahl et al., 2016, 2018; Strand & Nordahl, 2024; Strand et al., 2023b). Thus, in line with the metacognitive model (Wells, 2019), empirical studies indicate a relationship between dysfunctional metacognitive beliefs and personality dysfunction, but to the best of the authors knowledge, these relationships have not previously been evaluated.

As identifying factors that may contribute to personality dysfunction has the potential to enhance our understanding of how personality problems should be formulated and treated, we aimed to evaluate a potential unique role for dysfunctional metacognitive belief domains in personality dysfunction when controlling the following relevant covariates: age and gender, maladaptive dimensional personality traits, FFM personality traits, and symptoms of anxiety, depression, and impairment in general psychosocial functioning. Our hypotheses were as follows: 1) We expected higher levels of personality dysfunction to be significantly and positively correlated with higher levels on the metacognitive belief domains, maladaptive personality traits, FFM personality traits (with an opposite direction for extroversion, openness, agreeableness, and conscientiousness), emotional distress symptoms and general psychosocial functioning. 2) We expected a unique and independent contribution from dysfunctional metacognitive beliefs in explaining variance in personality dysfunction above the covariates such that higher levels of metacognitive beliefs are associated with greater personality dysfunction. More specifically and in line with metacognitive theory, we expected that negative metacognitive beliefs about the uncontrollability and danger of worry would be important as this domain is the strongest correlate of maladaptive self-regulation and therefore of greatest significance to psychopathology in general (Wells, 2019).

Materials and Methods

Participants and Procedure

Participants were recruited using convenience sampling from the general population if they were 18 years old or above and could read Norwegian. No other inclusion or exclusion criteria were employed. The survey was conducted online and advertised through several social media platforms. The Regional Committees for Medical and Health Research Ethics (Ref nr: 467342) approved the study and it was also registered with the Norwegian Centre for Research Data (Ref nr: 686857). All participants had to provide informed consent before participating.

In all, 1418 individuals consented to participate and opened the survey portal. However, some did not answer the relevant questionnaires for the current study and were therefore excluded. Thus, the total sample consisted of 1278 individuals of whom 618 (48.4%) were male, and 660 (51.6%) were female with a mean age of 29.53 years (SD = 11.52, range = 18–77). Regarding civil status, 566 (44.4%) reported to be single, 256 (20.0%) in a romantic relationship, 419 (32.8%) were cohabitants or married, 34 (2.7%) reported to be separated or divorced, and 1 (0.1%) widowed. Two individuals did not provide information about their civil status. Concerning occupational status, 775 (60.6%) reported they were students, 367 (28.7%) were working, 16 (1.3%) were searching for work, 22 (1.7%) were on sick-leave, 80 (6.3%) reported receiving a work assessment allowance or a disability pension, and 18 (1.4%) were retired. In total 575 (45.1%) had high-school or below as their highest completed education, and 699 (54.8%) reported having a university degree of 3 or more years. Four individuals did not report their highest level of completed education. Regarding mental health problems, 417 (32.7%) reported having received a diagnosis of a mental disorder at some point in their life, but we did not have more specific information concerning diagnoses. Collecting data from the general population was deemed appropriate as a first test of our hypotheses especially given that the AMPD views personality dysfunction not only as relevant for individuals with PDs, but as something that affects individuals on a continuum. A key point with continuum models is that it is possible to experience personality difficulties without reaching proposed cut-offs for personality disorders. Hence, a link between metacognitive beliefs and personality functioning should also be evident in large convenience samples with variation in levels of personality functioning if they play a relevant role.

Measures

The Level of Personality Functioning Scale Brief Form 2.0 (LPFS-BF; Weekers et al., 2019) is a 12-item self-report measure of impairment in self (e.g., “I often do not know who I really am”) and interpersonal functioning (e.g., "my friendships or relationships usually do not last”). The items correspond with the 12 facets of the LPFS as delineated in Section III of DSM-5 (American Psychiatric Association, 2013) and measures Criterion A/level of personality functioning of the proposed alternative model for personality disorders. Items are rated on a scale from 1 (“completely untrue”) to 4 (“completely true”). It has shown good psychometric properties with internal consistencies of the total scale as being good to excellent (α above 0.70 across studies; Weekers et al., 2022). In the current study the internal consistency was good (α = 0.85, ω = 0.85).

The Personality Inventory for DSM-5 and ICD-11 – Brief Form Modified (PID5BF + M; Kerber et al., 2022) measures the six dimensional maladaptive personality traits from the DSM-5: Negative affectivity (e.g., “my emotional reactions are much stronger than those of others”, Detachment (e.g., “I keep myself distanced from people”), Antagonism (e.g., “I find it easy to take advantage of others”), Disinhibition (e.g., “I am very impulsive”), Psychoticism (e.g., “I have a lot of habits viewed as eccentric or odd to others”), and Anankastia (e.g., “It is important to me that things are done in a certain way”), each containing 3 facets with 2 items. Items are rated on a scale from 0 (“does not fit”) to 3 (“fits completely”). It has shown good psychometric properties regarding the latent structure, reliability, and criterion validity (Kerber et al., 2022). In the current study the internal consistency for the subdomains were acceptable to good: Negative Affectivity (α = 0.81, ω = 0.81) Detachment (α = 0.78, ω = 0.77), Antagonism (α = 0.73, ω = 0.74), Disinhibition (α = 0.74, ω = 74), Psychoticism (α = 0.76, ω = 0.76), and Anankastia (α = 0.83, ω = 0.83).

The Big-Five Inventory (BFI-10; Rammstedt & John, 2007) measures five personality traits: Neuroticism (e.g., “I get easily nervous”), Extraversion (e.g., “I’m outgoing and social”), Openness (e.g., “I have a rich imagination”), Agreeableness (e.g., “I generally trust other people”, and Conscientiousness (e.g., “I complete my work tasks in a thorough manner”), originating from the lexical tradition of research within the five-factor model of personality (e.g., Goldberg, 1992). It is an abbreviated version of the 44-item BFI (John & Srivastava, 1999) and items are rated on a scale from 1 (“strongly disagree”) to 5 (“strongly agree”). It has shown acceptable psychometric properties likened to full length versions of big-five factor measures (Rammstedt & John, 2007). The personality traits in the measure consists of only two items each.

The Generalized Anxiety Disorder 7 (GAD-7; Spitzer et al., 2006) measures symptoms of generalized anxiety (e.g., feeling nervous, anxious, or on edge) and was used in the current study to control for anxiety. It contains 7-items rated on a 4-point scale from 0 (“not at all”) to 4 (“nearly every day”). It has shown excellent internal consistency (α = 0.92; Spitzer et al., 2006), and in the current study the internal consistency was good (α = 0.89, ω = 0.89).

The Patient Health Questionnaire 9 (PHQ-9; Kroenke et al., 2001) measures depression symptoms based on the nine criteria for depression as specified in DSM-IV (e.g., feeling depressed or filled with hopelessness) on a scale from 0 (“not at all”) to 3 (“nearly every day”). This measure was thus employed in the current study to control for levels of depressive symptoms. Several studies have supported it as a valid instrument for measuring depression (Kroenke et al., 2010) with good internal consistency (α = 0.89; Kroenke et al., 2001). In the current study, the internal consistency was excellent (α = 0.90, ω = 0.90).

The Work and Social Adjustment Scale (WSAS; Mundt et al., 2002) is a self-report questionnaire with 5 items assessing impairments in social and occupational functioning rated on a scale from 0 (“not at all”) to 8 (“to a serious degree”). It assesses levels of impairment related to work, home management, social and private leisure, and close relationships (e.g., “Does your problems have consequences for family and extended relationships like initiating and maintaining close relationships to others, including those you live with”). Internal consistency has been found to be good (α = 0.80; Mundt et al., 2002). In the current study the internal consistency was excellent (α = 0.91, ω = 0.91).

The Metacognitions Questionnaire 30 (MCQ-30; Wells & Cartwright-Hatton, 2004) measures five categories of dysfunctional metacognitive beliefs and consists of 30 items rated from 1 (“do not agree”) to 4 (“agree very much”). It has demonstrated good psychometric properties (Wells & Cartwright-Hatton, 2004) and in the current study the internal consistency for the subscales were acceptable to good: positive metacognitive beliefs (e.g., “Worrying helps me avoid problems in the future”, α = 0.84, ω = 0.84), negative metacognitive beliefs (e.g., “When I start worrying, I cannot stop”, α = 0.87, ω = 0.87), cognitive confidence (e.g., “I have a poor memory”, α = 0.89, ω = 0.89), need for control (e.g., “Not being able to control my thoughts is a sign of weakness”, α = 0.76, ω = 0.76), and cognitive self-consciousness (e.g., “I monitor my thoughts”, α = 0.81, ω = 0.81).

Overview of Statistical Procedures

IBM SPSS version 27 was used to for the analyses. Tests of skewness, kurtosis, and multicollinearity were conducted with no values (e.g., tolerance statistics and variance inflation factors) exceeding recommended thresholds. Graphical exploration in SPSS between the predictors and outcome-variable did not indicate violation of the assumption of homoscedasticity. Basic relationships between the variables were first investigated with bivariate correlations. To test the incremental contribution of metacognitive beliefs to personality dysfunction, we conducted a multiple hierarchical regression analysis using the LPFS-BF as the dependent and controlling for age and gender on step 1, dimensional maladaptive personality traits in step 2, FFM personality traits in step 3, symptoms of generalized anxiety, depression, and impairment in general functioning in step 4, and finally metacognitive beliefs in step 5.

Results

Personality dysfunction showed weak to moderate positive associations with the metacognitive belief domains of which negative metacognitive beliefs showed the strongest correlation. Higher personality dysfunction showed weak but significant correlations with lower age and female gender, weak to moderate positive and significant associations with all of the dimensional maladaptive personality traits, negative and weak correlations with the expected big-5 personality traits Extraversion, Agreeableness, and Conscientiousness, a positive and moderate correlation with Neuroticism, no association with Openness, positive and moderate to strong correlations with symptoms of generalized anxiety, depression symptoms, and impairment in general functioning. The correlations are presented in Table 1.

Table 1 Means With Standard Deviations and Bivariate Correlations Between the Variables (N = 1278)
Full size table

When predicting personality dysfunction, all steps of the regression were significant. In step 1, age and gender accounted for 2.3% of the variance where lower age and being female was associated with higher personality dysfunction. In step 2, all the dimensional maladaptive personality traits as a block accounted for an additional 60.9% of the variance, and all traits except for antagonism showed a significant and unique contribution to personality dysfunction. In this step, gender became non-significant as a predictor. In step 3, the big-5 personality traits accounted for an additional 1.7% of variance with only neuroticism, agreeableness, and conscientiousness making significant and unique contributions. In step 4, symptoms of generalized anxiety, depression and impairment in general functioning were entered and accounted for an additional 4.6% of the variance. All the three accounted for unique and independent variance. In this step, psychoticism and conscientiousness became non-significant. In the final step, metacognitive beliefs were entered and accounted for 1.4% additional variance with negative metacognitive beliefs and cognitive confidence showing significant and unique contributions. In summary, when controlling for the overlap between all the predictors, a lower age, and higher scores on negative affectivity, detachment, disinhibition, anankastia, neuroticism, agreeableness, symptoms of generalized anxiety and depression, impairment in general functioning, negative metacognitive beliefs and (lower) cognitive confidence remained significant predictors of personality dysfunction. When examining the relative strength of the individual predictors in this step, detachment showed the strongest unique relationship with personality dysfunction followed by depression symptoms, negative metacognitive beliefs and negative affectivity. The results from the regression analysis are presented in Table 2.

Table 2 Hierarchical Linear Regression Analysis with Level of Personality Functioning as the Dependent and Age/Gender, Dimensional Personality Traits, Big-5 Personality Traits, Symptoms of Anxiety, Depression, and Impairment of General Functioning, and Metacognitive Beliefs as Independent Variables (N = 1278)
Full size table

Discussion

We set out to evaluate the relationships between dysfunctional metacognitive beliefs and personality dysfunction derived from the AMPD. Even when controlling for age, gender, dimensional maladaptive personality traits, big-5 personality traits, symptoms of generalized anxiety and depression, and impairment in general functioning, negative metacognitive beliefs and lower cognitive confidence accounted for unique and independent variance in personality dysfunction, indicating that these metacognitive belief domains may play a role in personality dysfunction.

With the exception of openness, personality dysfunction significantly correlated with all the predictors in the expected directions showing moderate to strong correlations with symptoms of depression and anxiety, impairment in general functioning, negative metacognitive beliefs, cognitive confidence, metacognitive beliefs about need for control, negative affectivity, detachment, disinhibition, psychoticism, neuroticism, and anankastia. It further showed weak correlations with antagonism, extraversion, agreeableness, conscientiousness, positive metacognitive beliefs, and cognitive self-consciousness.

In the final step of the regression when controlling for the overlap between the predictors, younger age was associated with higher levels of personality dysfunction which is in line with previous findings (Bach et al., 2020). Of the maladaptive dimensional personality traits, negative affectivity, detachment, disinhibition, and anankastia showed independent contributions with detachment as the strongest predictor. Overall, the maladaptive personality traits showed the expected contribution to personality dysfunction which aligns with the amount of research implicating them as relevant (Anderson & Sellbom, 2018; Sleep et al., 2020). However, psychoticism and antagonism did not show a significant association to personality dysfunction when controlling the other covariates. Psychoticism has in previous research showed weak discriminant validity and been proposed to reflect nonspecific maladaptivity (Crego et al., 2015). Antagonism further measures traits such as callousness, grandiosity and manipulativeness and are thought to be related to PDs such as antisocial and narcissistic PD which have a rather low prevalence in the general population (Sleep et al., 2021) which could explain why this trait was not of particular importance to the outcome in the current study also reflected by the mean score and standard deviation in our sample. Regarding the FFM traits, neuroticism and agreeableness contributed significantly as independent predictors of personality dysfunction indicating higher levels of neuroticism may be a vulnerability factor and agreeableness a protective factor for personality dysfunction. This is in line with previous research demonstrating a link between neuroticism and personality dysfunction (McCabe et al., 2021) as well as a recent meta-analysis showing agreeableness to be associated with several favorable outcomes such as relational and work investment, social norm orientation and focus on social integration (Wilmot and Ones, 2022). Symptoms of anxiety, depression, and impairments in general functioning all contributed as significant and independent predictors. These factors have shown close relationships with personality dysfunction in previous research which is not surprising given the high prevalence of comorbidity between both anxiety and depressive disorders with PDs where the presence of PDs is also associated with indicators of chronicity such as duration of depressive episode, symptom severity, recurrence of depression and poor psychosocial functioning (Friborg et al., 2013; Hutsebaut et al., 2016; Van & Kool, 2018; Weekers et al., 2019). The association between personality dysfunction and impairment in general functioning have further been shown previously and indicates associations between different areas of functioning such as interpersonal and self-domains (personality dysfunction) to work, daily practical, leisure/social activities, and in maintaining close relationships (Morey et al., 2013).

Both negative metacognitive beliefs and (lower) cognitive confidence made unique and independent contributions to personality dysfunction beyond all the controlled variables. These findings are in line with the metacognitive model which suggests that dysfunctional metacognitive beliefs (and in particular negative metacognitive beliefs) are universal mechanisms of psychological disorder (Wells, 2019; Wells & Matthews, 1994). Furthermore, with the inclusion of all the covariates (e.g., emotional distress symptoms, general functioning, neuroticism), this observation is likely not a result of more general relationships between dysfunctional metacognitive beliefs and psychopathology. Personality dysfunction is indicative of impairments related to interpersonal and self-functioning, so how can we understand that metacognitive beliefs are of relevance in this context?

According to the metacognitive model, interpersonal functioning and problems can be understood as part of or resulting from the CAS (Wells, 2009). Interpersonal behaviors and difficulties are in themselves top-down controlled self-regulatory strategies, or resulting from maladaptive cognitive regulation (i.e., if worrying is perceived as uncontrollable, external means of control such as reassurance seeking is used to control worrying and prevent potential harm from excessive worry). Furthermore, the CAS is cognitively demanding as it taxes cognitive capacity and may impair taking the perspectives of others, or it may even shape one’s view of others (i.e., the CAS influence self-, other-, and world-view). Difficulties in “mentalization” (Bateman & Fonagy, 2016) could therefore be related to dysfunctional metacognitive beliefs rather than reflecting a general social competency deficit or capability. Previous research also supports a role for dysfunctional metacognitive beliefs to interpersonal functioning. Negative metacognitive beliefs and low cognitive confidence have been found to predict interpersonal problems even after controlling for symptoms of anxiety and depression, adult attachment, and FFM personality traits (Nordahl et al., 2021). Negative metacognitive beliefs have also been found to predict poorer social functioning (Bright et al., 2018). Further, treatment studies indicate that metacognitive change achieved through MCT (Wells, 2009) are associated with large effects on interpersonal problems in patients with Generalized Anxiety Disorder (Nordahl et al., 2018), Major Depressive Disorder (Strand et al., 2018, 2023b), Social Anxiety Disorder (Nordahl et al., 2016), as well as on symptoms of Borderline Personality Disorder (Nordahl & Wells, 2019). A recent study found that change in metacognitive beliefs were the only unique predictor of improvement in interpersonal problems in patients with Social Anxiety disorder above and beyond change in social phobic cognitions whilst controlling change in symptoms of social anxiety (Strand et al., 2023c).

Turning to the self-functioning domain, the metacognitive model offers insight into potential mechanisms and functions of the self-concept itself. Wells (2019) suggests that objective meta-awareness (i.e., sense of cognition) based on metacognitive procedural knowledge or system commands can be specified as “I” or “me” within online processing transforming meta-awareness into self-awareness, enabling us to take a subjective stance in relation to cognition. Thus, the self as a context for cognition provides us with the opportunity to make cognition the subject of our motivations or goals to be used in example for motivational or goal setting processes. However, if the self relies on a metacognitive system configuration, dysfunctional metacognition could lead to experience of an inconsistent and non-adaptive “self”, thus impairing a sense of identity and self-directedness. Furthermore, maladaptive self-regulation strategies (i.e., the CAS) resulting from dysfunctional metacognitions can maintain or strengthen negative self-beliefs and beliefs about others/the world (Wells, 2009). For example, self-criticism, inflexible self-attention in interpersonal contexts, and avoidance of close relationships are likely to verify convictions about low self-worth (and perception of others as superior). One study found a unidirectional and preceding contribution from metacognitions to negative self-beliefs in longitudinal data (Nordahl et al., 2022) and negative metacognitive beliefs and lower cognitive confidence have been found to be associated with low self-esteem (Kolubinski et al., 2019).

There are several potential implications based on the results. First, if dysfunctional metacognitive beliefs play a role in personality dysfunction, they could be targeted in treating individuals with PDs. They are potentially more amenable to change compared with other traits or individual differences which are not formulated within a theoretical framework with specific directions for how they can be modified. Dysfunctional metacognitive beliefs are relevant for a host of sustaining processes and behaviors implicated across disorders and several studies have shown transdiagnostic and trans-symptomatic changes following MCT (e.g., Callesen et al., 2020; Hagen et al., 2017; Nordahl et al., 2018; Strand et al., 2023b), which is especially relevant for individuals with PDs given the high level of comorbidity (Lenzenweger et al., 2007). This view is also supported by preliminary findings indicating that it is the general severity (criterion A) component in PDs that tend to change whereas personality style tends to be relatively stable (Wright et al., 2016), which could indicate that individuals can improve despite prototypical traits possibly less important for pathology. Second, MCT which directly aims to modify dysfunctional metacognitive beliefs might be especially suitable for treating PDs which in addition to the aforementioned offers a framework for conceptualizing, formulating and targeting phenomenon’s that are commonly observed/experienced by individuals with PDs such as resistance to change, chronicity of maladaptive coping strategies (including interpersonal strategies), and an unstable sense of self. In support of this, among patients with PDs being treated for primary major depression, 10 sessions of MCT led 40–60% of the patients to no longer meet the diagnostic criteria for a PD post-treatment following assessment by independent assessors (Hagen et al., 2017; Hjemdal et al., 2017). MCT has further demonstrated promising results for individuals with Borderline Personality Disorder (Nordahl & Wells, 2019), and in changing levels of neuroticism, a hypothesized underlying vulnerability factor, in patients with GAD (Kennair et al., 2021).

A strength in the current study was a large sample size providing sufficient statistical power to control several relevant factors in investigating the unique relationships between metacognitive beliefs and personality dysfunction. However, the study cannot draw causal inferences given the cross-sectional design. The study further relied exclusively on self-report measures and used several short versions. However, all the measures are well validated and frequently used. It is not given that the current results are generalizable to clinical settings or patients with PDs as the current sample was gathered at convenience and although we did collect self-reported levels of lifetime received mental health diagnosis, we did not enquire into which diagnoses that had been received specifically. Future studies could further investigate identified subdomains of personality dysfunction (self and interpersonal) and their associations with more specific domains of metacognitive beliefs employing longitudinal designs in specific clinical populations.

Conclusions

Dysfunctional metacognitive beliefs uniquely correlate with personality dysfunction even when controlling for a range of well-established predictors. Negative metacognitive beliefs and lower cognitive confidence emerged as independent predictors among the metacognitive belief domains and may be particularly relevant for formulation and treatment of personality dysfunction. It might be that targeting common mechanisms of psychological disorder (i.e., metacognition) is a more easily accessible and appropriate approach to reduce personality dysfunction than targeting personality traits and other individual differences. Furthermore, metacognitive change might influence emotional distress symptoms, psychological vulnerability, and personality dysfunction in parallel, but studies are needed to evaluate the effectiveness of MCT for those with PDs and complex disorder presentations.