Introduction

One important aspect of self is evaluating relevance of environmental and mental events to self. Self-relevance (SR) is defined as top-down mechanism [1] with an impact on memory, perception, decision making and goal-directed behavior [2]. In the brain, a large-scale supramodal system has been proposed by neuroimaging studies to mediate the appraisal of self-relevant content, which includes the cortical midline structures and areas in the ventromedial prefrontal cortex [1,3,4] as well as several other regions including insula, basal ganglia and amygdala [5].

Clinical observations have reported aberrant judgements of SR in psychopathological conditions. Attribution of negative affects to the self has been acknowledged to be a characteristic of depressive syndromes [6] and the attainment of more positive interpretations for self has been posited as a desirable goal in clinical treatments of dysphoric individuals [7]. A dysfunctional differentiation between self-relevant and -irrelevant information has been suggested to underlie positive symptoms of schizophrenia thorough aberrant attribution of salience [8].

BPD is a severely impairing mental disorder, which affects 2–5% of adult population [9]. Individuals with this disorder suffer from affective dysregulation, heightened impulsivity, an unstable self and difficulties in interpersonal functioning [10], leading to a limited quality of life for the afflicted individuals and serious impact on mental health services. Impairments of the self is considered to be a central area of difficulty in borderline personality disorder (BPD) [11,12], which highlights the relevance of SR investigations to our understanding of BPD psychopathology.

In fact, modulatory role of SR evaluations on cognitive and emotional processing of information and especially on attention-emotion interactions in individuals with BPD has been already proposed [13], which is expected to have an influence on the goal-directed behavior. Based on a meta-analysis of attention bias studies in BPD, personally relevant negative words drive stronger attentional biases, as compared to the other negative material [14]. Recent evidence suggests that in individuals with BPD, the relevance to self can generate negative biases in information processing, which might enhance the disapproving view of themselves [15]. An alternated electrocortical reactivity has also been shown during processing of negative self-relevant information in females with BPD [16].

Altogether, this evidence suggests a modulatory role of SR evaluations on cognitive and emotional processing of information, which is critical in ruminations and dysfunctional behaviours in BPD. An increased tendency to attribute events to themselves have been shown in BPD [17]. To date, however, there has been limited empirical research investigating the judgments of SR in this clinical population. Especially the interaction of maladaptive attention and evaluation bias toward negative affects [18–21] and the increased intensity of negative emotions [22] in BPD has not been investigated yet.

In the present study, we relied on a challenge method by presenting written affective statements, which are phrased in third-person and attempted to probe if valence of the stimuli has an effect on the attribution of SR in BPD and control groups. The study assessed two hypotheses about BPD patient population. First, it was hypothesized that, negative sentences would trigger a higher SR. An interconnection between the appraisals of affect and SR is expected based on the widely recognized Linehan’s model for BPD [23] that suggests the emotional dysregulation to be responsible for formation of a disturbed self and other features of BPD. IQ and education was also based on the Linehan’s theory, supposing deficits of appropriate emotion regulation strategies to underlie maladaptive cognitions and behaviours [24]. Among the common emotion regulation strategies, cognitive reappraisal (changing the way one thinks about potentially emotion-eliciting events) is associated with healthier patterns of affect-related functioning as compared to expressive suppression (changing the way one behaviorally responds to emotion-eliciting events) [25]. We therefore expected a correlation between deficiencies in cognitive reappraisal and SR processing in BPD. The third hypothesis was that the severity of disorder-related symptoms would be related to the strength of the maladaptive processing of SR. Given the convoluted structure of the BPD symptoms, disturbances of the self are expected to interrelate with other difficulties of the BPD patients.

To examine the above hypotheses and derive clinical applications from the results of the research questions, we chose an approach that merged an experimental behavioral paradigm and BPD symptom assessment based on standardized questionnaires. We first investigated whether a maladaptive SR can be detected in BPD by an assessment task, and then related the intensity of SR to the severity of the BPD symptoms and emotion regulation strategies in patients.

Methods

Results

Discussion

To our knowledge, this represents the first study to explicitly tackle the judgements of SR in BPD by showing a deviancy in comparison to a healthy sample and, most importantly, a relationship between SR and BPD symptom severity. We designed an experimental paradigm to examine the experience of SR by exposing a variety of sentences containing a third-person personal pronoun to the participants and engaging them in an assessment task. The affect rating did not differentiate the groups. Both BPD and HC groups were able to distinguish the positive, negative and neutral categories of sentences in this paradigm. Our results revealed that, in assessing hypothetical statements, BPD patients excessively attribute negative statements to self. Within our BPD sample, patients with higher symptom severity showed an increased tendency to judge the negative contents as related to self. To examine if the SR deviation is unique to the BPD pathology or is driven by co-existing depression, we conducted a partial correlation analysis between BSL-95 and SR of negative sentences while controlling for BDI-II score. A positive correlation between SR and symptom severity emerged after controlling for BDI-II scores, verifying that this difference was not an artifact of depressive severity.

These findings contribute to understanding the link between the disturbed self and the clinical presentations of BPD. Most people endorse more positive features than negative features for the self, which promotes experiencing more favorable psychological and social outcomes [35]. In BPD, deviations from such self-enhancing tendencies are likely to affect the sense of well-being. Negative self-relevant cues are likely to contribute to a heightened instability of self-image and in this way further trigger downstream behavioral outcomes such as impulsive and maladaptive behavior. Although the results confirmed o Our second hypothesis our hypothesis regarding a bias toward the self-relevance of negative contents in BPD, it should be considered that some of it might be explained by the effect of other psychopathological changes, like disturbed mood, on evaluative judgements of patients.

One possible contributing factor to disturbances of self is an impact of emotional processes such as valence appraisals. Emotions guide us to navigate in our complex environment. In pathological states, however, emotions may lead to maladaptive cognitive functions, as illustrated for example by exaggerated attentional bias to threat in anxiety [36], preferential memory for negative events in depression [37] and negative bias in discrimination of negative and neutral facial expressions in BPD [38,39]. Neuroimaging investigations have already suggested dysfunctionality in the neural circuitry responsible for evaluation of emotional salience and conflict control [40]. Our stimuli set, therefore, consisted of sentences of different valence categories. The negative valence augmented the attribution of relevance to self in BPD, indicating a role for emotional appraisals in self-evaluative processes in this clinical population. It is possible that the negative contents trigger SR through salience mechanisms. Another proposal is that the enhanced SR for negative contents, as measured by our paradigm, derives from emotion regulation deficits in the BPD sample. Support for this proposal comes from our data, which showed a correlation between SR bias and deficits of cognitive reappraisal strategies as measured by ERQ. This is in line with Linehan’s model for BPD [23], which postulates that an underlying vulnerability to emotional dysregulation is the basis for evolution of a disturbed self-image and interpersonal impairments of individuals affected by BPD.

Another principal finding that emerged from the correlation analysis was that the SR bias was higher in patients with more severe symptoms. A severely affected patient suffers from a biased SR far more than a mildly affected individual. The implication is that the SR bias is relevant to the BPD pathology and cannot be generalized to the unspecific characteristics of the clinical sample. Self-relevance biases along with other self-related pathologies might contribute to impairments of the self-concept in BPD [41]. We argue that a disruption of the SR evaluation system in BPD may be relevant to the troubled self-image and likewise other typical difficulties in BPD like affective stability and interpersonal problems in BPD patients.

Our findings about increased attribution of negative statements to self in BPD fits very well into the attribution style theory [42], which focuses on the causal interpretation of events regarding to the self or others by the individuals. A framework for understanding psychopathological deviations such as depressive mood [43] and persecutory delusions [44] has been already suggested based on the attribution style theory. In respect of BPD, an enhanced internal attributions for negative events has been already shown [45].

In sum, it appears that the construct of self-relevance and other self-evaluative processes in BPD deserves further attention. It might be quite elucidating to refer to a negative SR as a basis for maladaptive interrelations of emotion and self. We suggest that the investigation of the material encoded in a self-relevant manner may prove useful for the understanding and treatment of BPD. Cognitive interventions should focus on modification of a biased attribution of negative contents to self, using affective training material.

Limitations

We would like to acknowledge several limitations of our study, which should be considered in the interpretation of the results. Only female subjects were included, which limits the generalizability of the findings. IQ and education were not included as between-group matching variables. Most of the BPD patients were medicated, which may have affected some group differences. Regarding the stimuli, the effect of individual biography and pre-exposures may be more relevant for participants with BPD and the possible confounds associated with endorsed self-relevant statements were not explicitly examined.

Supporting information