Psychotic-like cognitive biases in borderline personality disorder

Abstract

Whereas a large body of research has linked borderline personality disorder (BPD) with affective rather than psychotic disorders, BPD patients frequently display psychotic and psychosis-prone symptoms, respectively. The present study investigated whether cognitive biases implicated in the pathogenesis of psychotic symptoms, especially delusions, are also evident in BPD. A total of 20 patients diagnosed with BPD and 20 healthy controls were administered tasks measuring neuropsychological deficits (psychomotor speed, executive functioning) and cognitive biases (e.g., one-sided reasoning, jumping to conclusions, problems with intentionalizing). Whereas BPD patients performed similar to controls on standard neuropsychological tests, they showed markedly increased scores on four out of five subscales of the Cognitive Biases Questionnaire for Psychosis (CBQp) and displayed a one-sided attributional style on the revised Internal, Personal and Situational Attributions Questionnaire (IPSAQ-R) with a marked tendency to attribute events to themselves. The study awaits replication with larger samples, but we tentatively suggest that the investigation of psychosis-related cognitive biases may prove useful for the understanding and treatment of BPD.

Introduction

A. Stern (1938) coined the term “borderline” to denote a disorder at the border of neurosis and psychosis (Barnow et al., 2010). According to current psychiatric classification, the clinical picture of borderline personality disorder (BPD) is characterized by emotional dysregulation, fear of abandonment, disturbance in the individual’s sense of self, unstable interpersonal relationships, and impulsivity.

While studies point to a relationship between BPD with affective rather than schizophrenia spectrum disorders (e.g., Tyrer, 2009), a recent review (Barnow et al., 2010) concludes that approximately one out of four patients with BPD displays severe psychotic symptoms and three out of four patients show psychosis-like experiences and paranoid ideation. Research on the cognitive underpinnings of BPD has not yet arrived at a consistent cognitive profile. Reviews have rather consistently found deficits in executive functioning (LeGris and van Reekum, 2006, Ruocco, 2005). However, these are common in many mental disorders leaving their etiological importance and specificity for BPD equivocal (Moritz et al., 2002). In marked contrast and somewhat surprisingly given positive correlations between neurocognitive and social functioning, several studies found normal or even greater-than-normal performance of BPD patients on tests tapping social cognition/theory of mind which captures the ability to empathize with others (Arntz et al., 2009, Fertuck et al., 2009).

Quite consistently, dysfunctional beliefs and dichotomous thinking (DT) have been reported in BPD (Arntz et al., 2004, Butler et al., 2002), whereby patients with BPD seem to evaluate others in extreme but mixed (positive and negative) terms instead of “all good” versus ‘‘all bad” (Napolitano and McKay, 2007, Veen and Arntz, 2000). Moreover, patients seem to display DT only when confronted with BPD-specific stimuli (Veen & Arntz, 2000). In contrast, other studies failed to confirm substantial DT and reported a general negative evaluative style of others (Barnow et al., 2009, Sieswerda et al., 2005).

For the present study, we compared BPD and healthy participants on cognitive biases implicated in the pathogenesis of schizophrenia (Bell et al., 2006, Bentall et al., 2009, Freeman, 2007) such as abnormalities of attributional style and jumping to conclusions (i.e., decision-making on the basis of few information). Biases are deviances in the processing and appraisal of information rather than “cold” neuropsychological deficits (i.e., neuropsychological impairments reflecting neural dysfunctions not elicited by secondary influences (e.g., low effort)) and the two domains are at best weakly related (Moritz , Veckenstedt, et al., 2010, Moritz, Vitzthum, et al., 2010, Moritz, Woodward, et al., 2010): Biases such as over-confidence in errors and jumping to conclusions may appear in the absence of neuropsychological impairments and vice versa. Two driving forces motivated our investigation. First, in view of frequent psychotic symptoms in BPD (Barnow et al., 2010) it is worth investigating if cognitive biases putatively underlying schizophrenia delusions are to some degree also detectable in BPD. Secondly, the first author devised a metacognitive training (MCT) program for schizophrenia (for reviews see Moritz , Veckenstedt, et al., 2010, Moritz and Woodward, 2007) targeting specific cognitive biases in eight modules (e.g., jumping to conclusions). We received anecdotal reports from other researchers that the MCT was successful with BPD patients tentatively suggesting that cognitive biases seen in psychoses may also be relevant to both the understanding and treatment of BPD.

The Cognitive Biases Questionnaire for psychosis (CBQp; Peters et al., 2010) was administered which captures many cognitive biases typical of schizophrenia patients and also includes a subscale on DT. In addition, we assessed attributional style (Moritz, Veckenstedt, et al., 2010), which has also been linked to paranoia, especially a tendency to blame others for failure and to make monocausal inferences. Finally, two neuropsychological parameters were assessed: Speed of information processing and executive functioning. We hypothesized that borderline patients would display deficits in “cold” neuropsychological functions and pronounced cognitive biases on the CBQp. For the attributional styles questionnaire no directed hypothesis was made.