Elsevier

Psychiatry Research

Volume 150, Issue 3, 15 April 2007, Pages 265-275
Psychiatry Research

Psychophysiological ambulatory assessment of affective dysregulation in borderline personality disorder

https://doi.org/10.1016/j.psychres.2006.04.014Get rights and content

Abstract

Many experts now believe that pervasive problems in affect regulation constitute the central area of dysfunction in borderline personality disorder (BPD). However, data is sparse and inconclusive. We hypothesized that patients with BPD, in contrast to healthy gender and nationality-matched controls, show a higher frequency and intensity of self-reported emotions, altered physiological indices of emotions, more complex emotions and greater problems in identifying specific emotions. We took a 24-hour psychophysiological ambulatory monitoring approach to investigate affect regulation during everyday life in 50 patients with BPD and in 50 healthy controls. To provide a typical and unmanipulated sample, we included only patients who were currently in treatment and did not alter their medication schedule. BPD patients reported more negative emotions, fewer positive emotions, and a greater intensity of negative emotions. A subgroup of non-medicated BPD patients manifested higher values of additional heart rate. Additional heart rate is that part of a heart rate increase that does not directly result from metabolic activity, and is used as an indicator of emotional reactivity. Borderline participants were more likely to report the concurrent presence of more than one emotion, and those patients who just started treatment in particular had greater problems in identifying specific emotions. Our findings during naturalistic ambulatory assessment support emotional dysregulation in BPD as defined by the biosocial theory of [Linehan, M.M., 1993. Cognitive–Behavioral Treatment of Borderline Personality Disorder. The Guildford Press, New York.] and suggest the potential utility for evaluating treatment outcome.

Introduction

Borderline personality disorder (BPD) is characterized by severe deficits in interpersonal, cognitive, and emotional functioning (Lieb et al., 2004). Many experts now believe that the pervasive problems in affect regulation (often termed “emotion dysregulation”) represent the central area of dysfunction (Linehan, 1993, Coid, 1993, Corrigan et al., 2000, Skodol et al., 2002, Sanislow et al., 2002). According to the biosocial theory of Linehan (1993), emotion dysregulation in BPD comprises increased sensitivity to emotional stimuli, unusually strong reactions, the occurrence of complex emotions (more than one emotion simultaneously), and problems in identifying emotions. However, data actually demonstrating emotion dysregulation is sparse and inconclusive.

Support for the theory of emotion dysregulation is primarily based on subjective measures of emotion and experimental studies. A few studies that used multiple self-rating (diaries) over time reported a higher level of unpleasant affect in BPD patients, compared to a clinical control group (Stein, 1996) and psychologically healthy controls (Cowdry et al., 1991, Stiglmayr et al., 2001). Using self-rating questionnaires, Koenigsberg et al. (2002) found however no evidence of elevated affect intensity in BPD patients when compared with other personality-disordered individuals. Experimental studies that applied emotion-inducing techniques also supported greater emotional dysregulation in BPD: Levine et al. (1997) elicited a heightened intensity of negative emotions in BPD patients, and Herpertz et al. (1997) provided evidence of elevated baseline emotional activation among patients with self-injuries, as compared to healthy controls.

Studies using psychophysiological indicators of emotion have thus far failed to find a consistent pattern of affective dysregulation in BPD. Herpertz et al. (1999) exposed BPD patients and healthy controls to affect-inducing pictures but did not find any hyperreactivity among patients, either with respect to physiological indicators of emotion (heart rate, electromyogram or skin conductance) or to subjective ratings of emotion. In a similar study using functional MRI, Herpertz and colleagues (2001) found no differences in subjective emotions. However, BPD participants did show elevated blood flow in the amygdala. Likewise, Donegan et al. (2003) revealed greater left amygdala activation to emotional facial expression among patients with BPD, and Ebner-Priemer et al. (2005) revealed significantly higher startle response in BPD; enhanced startle response is caused by increased amygdala activation (e.g. by electrical stimulation), at least in animals (Davis et al., 1999). Unfortunately, neither Donegan et al. (2003) nor Ebner-Priemer et al. (2005) reported subjective emotional ratings.

There are a number of possible explanations for these conflicting results. First, BPD patients without medication, who are often recruited for psychophysiological studies, are extremely rare (Zanarini et al., 2004) and may represent a healthier subgroup of BPD patients than medicated individuals. The same argument may apply to patients with BPD who are assessed after a long treatment program. Secondly, it is possible that the affect-induction methods used were insufficient to evoke affective dysregulation and that more personally relevant stimuli are preferable. Thirdly, most studies have been performed in a laboratory environment, and it is possible that the somewhat artificial conditions of the laboratory (e.g. insufficient time for adaptation, the psychological consequences of being observed and contrived experimental protocols) influenced the measurement of affective dysregulation.

One difficulty encountered in the investigation of physiological indices of emotions in ambulatory studies has been the teasing apart of emotional and physical influences. In order to address this problem, Myrtek (2004) developed an algorithm to partition that part of heart rate (HR) increase that does not directly result from physical or metabolic activity; he has termed this measure “additional heart rate" (aHr). The algorithm is based on two experimental findings: namely, that HR reactivity, elicited by a combination of mental, emotional and physical stressors, closely approximates the additive combination of the HR responses evoked by the individual factors (Myrtek and Spital, 1986, Roth et al., 1990), and that metabolic demand can be estimated with sensitive motion detectors (Myrtek, 2004).

This non-metabolic heart rate increase (aHr) appears to be a valid measure of emotional response and has now been validated in multiple studies including more than 1300 participants. For example, aHr in male students was found to be greater when watching erotic films compared to comedies and this applied not only during rest but also under varied levels of physical exercise, and that these physiological differences corresponded to self-ratings of excitement (Myrtek and Bruegner, 1996). This demonstrates that the algorithm is not only able to detect emotional events under basal conditions but also under conditions of activity. Effects on aHr have also been demonstrated for children watching television at home: both school-age boys (Myrtek et al., 1996b) and preschool boys and girls (Wilhelm et al., 1997) manifested greater aHr during films with action scenes. In the same study, girls showed increased aHr while watching commercials that targeted girls (e.g. Barbie), as compared to gender-neutral commercials (Wilhelm et al., 1997). Additionally, one investigation found that train drivers showed higher aHr during situations associated with heightened risk for accidents (Myrtek et al., 1994). aHr is increased during leisure-time activities compared to periods of monotonous work (Myrtek et al., 1999), and in social interactions as compared to being alone (Myrtek et al., 1995). This is consistent with the use of aHr as an emotional indicator. There are several indications that aHr is primarily under sympathetic control: beta-adrenergic blockade studies of HR (Langer, 1985), evidence that metabolically induced HR responses are mainly vagally mediated (Grossman, 1983, Grossman and Svebak, 1987, Watkins et al., 1998) and the missing correlation between aHr and heart rate variability (Myrtek et al., 1996a).

A large and growing literature has linked indices of reduced baseline vagal activity to disorders with dysregulated affective styles, like depression, anxiety and panic (Beauchaine, 2001) and estimates of high vagal activity to relaxation (Houtveen et al., 2002).

In order to address the question of emotional dysregulation in BPD, we employed psychophysiological ambulatory monitoring, which is also called ecological momentary assessment (Stone and Shiffman, 1994), to investigate both self-report measures (frequency and intensity of discrete emotions) and physiological indices of emotions (additional heart rate, spectral analysis indices of cardiac vagal activity) during 24 h of normal daily life. This approach circumvents several methodological limitations. Investigating naturally occurring symptoms in everyday life may render affect-inducing methods unnecessary and eliminates many of the problems associated with laboratory studies. We included patients who were currently in our treatment programs and did not alter their medication schedules; this was in order to provide a typical and experimentally unmanipulated sample of BPD patients in treatment, and therefore increase the generalizability of our findings.

According to the biosocial theory (Linehan, 1993), we specifically hypothesized that patients with BPD, in contrast to matched healthy controls, would 1) show a higher frequency and intensity of self-reports of emotion, 2) manifest greater levels of additional heart rate and lower levels of heart rate variability indices of autonomic control, 3) report more complex emotions, and 4) show greater difficulty in identifying discrete emotions during periods of reported emotional arousal.

Section snippets

Subjects

139 subjects were initially recruited for this study. Ten controls were excluded due to psychiatric disorders. Twenty-three data sets had to be eliminated because of technical problems such as defect electrocardiogram leads, damage to accelerative sensors, or data loss in the palm top. Six data sets were aborted early on account of compliance problems and were therefore also excluded. Ultimately, 50 female patients meeting criteria for BPD and a comparison group of 50 female psychologically

Subjective emotions

The relative frequency of the first reported emotion (overall and for specific emotions) is shown in Fig. 1. BPD patients reported specific negative emotions (anxious, angry, sad, shame, and disgust) significantly more often than the HC group. Probabilities and effect sizes are shown in Table 2. Furthermore BPD individuals also reported significantly fewer positive emotions (happy and interest). Effect sizes for group differences of all specific positive and negative emotions range between

Discussion

Our results confirm the first hypothesis that BPD participants show a higher frequency and intensity of self-reported emotions. Patients with BPD reported more emotions overall than HCs; specifically, BPD patients reported more negative emotions and fewer positive emotions. BPD patients also reported greater intensity of negative emotions, but not positive emotions. Our data therefore replicate the findings of increased negative affect reported in diary and questionnaire studies by BPD

Acknowledgements

This research was supported by a grant (Bo 1487/4-1) from the German Research Society (DFG), a grant from the Royalty Research Fund, University of Washington, the Borderline Personality Disorder Research Foundation (BPDRF), and the Swiss National Foundation (scholarship). The authors thank J. Fahrenberg, M. Myrtek, F. Foerster, and P. Hüttner from the Psychophysiology Research Unit, University of Freiburg for all methodological, technical and statistical support, T. Thielgen, L. Detterer, J.

References (54)

  • T.P. Beauchaine

    Vagal tone, development, and Gray's motivational theory: toward an integrated model of autonomic nervous system functioning in psychopathology

    Development and Psychopathology

    (2001)
  • G.G. Berntson et al.

    Heart rate variability: origins, methods, and interpretive caveats

    Psychophysiology

    (1997)
  • D. Clark-Carter

    Doing Quantitative Psychological Research: From Design to Report

    (1998)
  • J.W. Coid

    An affective syndrome in psychopaths with borderline personality disorder?

    British Journal of Psychiatry

    (1993)
  • R.W. Cowdry et al.

    Mood variability: a study of four groups

    American Journal of Psychiatry

    (1991)
  • M. Davis et al.

    Neurophysiology and neuropharmacology of startle and its affective modulation

  • J. Fahrenberg

    Concurrent assessment of blood pressure, physical activity, and emotional state in natural settings

  • J. Fahrenberg et al.

    MONITOR: acquisition of psychological data by a hand-held PC

  • M.B. First et al.

    User's Guide for the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I) — Clinical Version

    (1997)
  • F. Foerster et al.

    Motion pattern and posture: correctly assessed by calibrated accelerometers

    Behavior Research Methods, Instruments, and Computers

    (2000)
  • P. Grossman

    Respiration, stress, and cardiovascular function

    Psychophysiology

    (1983)
  • P. Grossman et al.

    Prediction of tonic parasympathetic cardiac control using respiratory sinus arrhythmia: the need for respiratory control

    Psychophysiology

    (1991)
  • P. Grossman et al.

    Respiratory sinus arrhythmia as an index of parasympathetic cardiac control during active coping

    Psychophysiology

    (1987)
  • S.C. Herpertz et al.

    Affective responsiveness in borderline personality disorder: a psychophysiological approach

    American Journal of Psychiatry

    (1999)
  • S.C. Herpertz et al.

    Emotion in criminal offenders with psychopathy and borderline personality disorder

    Archives of General Psychiatry

    (2001)
  • J.H. Houtveen et al.

    Contribution of tonic vagal modulation of heart rate, central respiratory drive, respiratory depth and respiratory frequency to respiratory sinus arrhythmia during mental stress and physical exercise

    Psychophysiology

    (2002)
  • M. Ikawa et al.

    Effects of combination psychotropic drug treatment on heart rate variability in psychiatric patients

    Psychiatry and Clinical Neurosciences

    (2001)
  • Cited by (0)

    View full text