Personality and the bipolar spectrum: normative and classification data for the Eysenck Personality Questionnaire–Revised

Abstract

Background

Personality traits dispose individuals toward particular affective states and may therefore have an important role in the etiology of affective disorders in particular. Despite being one of the most widely used and well-researched personality instruments, few studies have studied bipolar spectrum disorders using the revised Eysenck Personality Questionnaire (EPQ-R) (Eysenck HJ, Eysenck SBG. The Eysenck Personality Questionnaire–Revised. Sevenoaks: UK; Hodder & Stoughton, 1992).

Methods

The EPQ-R was administered to 50 bipolar patients, 50 unipolar patients, and 50 controls matched on age and sex. Participants in clinical groups were euthymic, and participants in the control groups were screened for symptoms of depression.

Results

The EPQ-R scores were most effective at discriminating unipolar patients from controls, such that unipolar patients were higher on neuroticism and lower on extraversion. Bipolar patients showed a similar personality profile to, but were not clearly distinguished from, unipolar patients.

Conclusions

This research provides preliminary normative data for the EPQ-R that complement previous theoretical and empirical work in this area and suggests the usefulness of this tool in a clinical setting.

Introduction

The bipolar spectrum encompasses syndromes of mood dysregulation ranging from unipolar depression to bipolar disorder [1]. Unipolar depression, sometimes described as clinical depression, is a state of sadness and despair reaching such intensity as to disrupt one's social functioning and daily living [2]. Bipolar disorder, formerly known as manic depression, is characterized by extreme and unpredictable moods; sufferers experience states of mania or hypomania, which alternate over time with clinical depression and normal mood lability [2]. The spectrum approach suggests that such disturbances in mood differ from “normal” affective variation by degree, rather than according to some discrete criterion. One might therefore expect that such syndromes can be located within the space described by taxonomies of affect-relevant individual differences, such as personality inventories. This may suggest a role for personality testing, using mainstream instruments such as the revised Eysenck Personality Questionnaire (EPQ-R) [3], in clinical assessment.

The linking of normal variation in cognition, emotion, and behavior (ie, personality) with abnormal variation (ie, psychopathology) has a long tradition in psychology [4], [5], [6]; but it is only relatively recently that a trait-based approach to clinical psychology has gathered discernable momentum [7], [8], [9]. Personality traits are thought to directly influence affective states and thereby explain long-term interindividual patterns of emotion and mood [10], [11], [12]. There is good evidence to suggest that this may extend to affective disorders such as the bipolar spectrum. Negative affect and symptoms of depression are predicted by traits such as neuroticism and anxiety [13], [14], [15]. Other traits such as impulsivity have been associated with the experience of intense positive emotion and hypomania in nonclinical samples [16] and elevated mood occurring in bipolar disorder [17]. There has also been noted comorbidity between bipolar disorder and borderline personality disorder [18], the latter of which is associated with emotional reactivity (as opposed to mood dysregulation per se). Not surprisingly, a number of studies have now demonstrated reliable differences in personality over bipolar spectrum disorders [19], [20]. However, few if any of these have adopted a classification approach to evaluate the diagnostic power of standard personality tests; and few have used the EPQ-R. The purpose of this research is to therefore (a) provide preliminary normative data for bipolar spectrum disorders using the EPQ-R and (b) evaluate the potential of this mainstream personality measure to distinguish between (ie, probalistically classify) bipolar patients, unipolar patients, and a group of normal controls.

The EPQ-R is one of the most widely used and best validated instruments in the personality literature.¹As such, the value of normative data for bipolar spectrum disorders using this instrument should be readily appreciated. The EPQ-R consists of 3 factors, including Extraversion and Neuroticism, which feature in essentially all personality models and are thought to reflect primary sources of personality variation [21]. The third factor, Psychoticism, is often divided into 2 or more factors within other taxonomies [22]. Several features of the EPQ-R suggest its potential usefulness as an aid to clinical assessment, particularly in the case of mood disorders. First, the Neuroticism scale, in its original form, was developed according to its ability to distinguish patients with affective disturbances from normal controls [23]. Second, Extraversion and Neuroticism have been consistently associated with the intensity of and proneness toward positive and negative mood states [24], [25], [26]. Although these associations are not a priori predictions from Eysenck's theory, they follow more or less directly from the view that Extraversion and Neuroticism are functional outcomes of brain reward and punishment systems [12], [27]. The reward system is thought to mediate the experience of positive affect in response to incentive, and the punishment system is thought to mediate the experience of negative affect in response to threat. Although bipolar disorder has not been clearly addressed in this literature, unipolar depression has been causally linked with punishment sensitivity [28], [29], which is thought to manifest in personality space as neurotic-introversion (specifically, 2 parts high Neuroticism to 1 part low Extraversion) [30].

To our knowledge, only 1 study has compared scores on the EPQ-R across individuals diagnosed with bipolar disorder and unipolar depression [31]. This cross-sectional Polish study found that unipolar patients had higher neuroticism scores and lower extraversion scores than normal controls. Bipolar patients were also higher on neuroticism and lower on extraversion compared with controls, but not to the same extent as unipolar patients. Psychoticism did not differentiate any groups in the study. These results agree with findings using very early versions of Eysenck's Extraversion and Neuroticism scales [32], [33]. They are also consistent with the suggestion that depression may be linked with a punishment system manifesting as trait neurotic introversion. The fact that bipolar patients were also similarly distinguished from controls possibly reflects the depressive phases that occur in this disorder, but is nevertheless surprising for the following reasons. First, the manic phases of bipolar disorder—consisting of euphoric emotional highs—are suggestive of an overactive reward system, whose footprint in personality space is thought to comprise high neuroticism and extraversion [30], and possibly also high psychoticism [34]. Second, a positive association with psychoticism and extraversion might be predicted from the known role of impulsivity—which is typically positively related to one or both of these personality factors [35]—in bipolar disorder [17], [36]. As the early Eysenckian scales used in other research did not include a measure of Psychoticism [32], [33], it remains to be seen if the lack of association between this trait and the bipolar spectrum is reliable.

In this study, we endeavored to assess a medium-sized sample of bipolar and unipolar patients (n = 100) on the original English version of the EPQ-R. It was predicted that unipolar patients would have higher neuroticism and lower extraversion scores than controls, whereas bipolar patients would have higher neuroticism, extraversion, and psychoticism scores. It was further predicted that EPQ-R scores would successfully classify unipolar, bipolar, and control participants at a level significantly above chance, suggesting the usefulness of this instrument for clinical assessments of affective disorders.