Preliminary evidence for an emotion dysregulation model of generalized anxiety disorder

https://doi.org/10.1016/j.brat.2004.08.008Get rights and content

Abstract

Three studies provide preliminary support for an emotion dysregulation model of generalized anxiety disorder (GAD). In study 1, students with GAD reported heightened intensity of emotions, poorer understanding of emotions, greater negative reactivity to emotional experience, and less ability to self-soothe after negative emotions than controls. A composite emotion regulation score significantly predicted the presence of GAD, after controlling for worry, anxiety, and depressive symptoms. In study 2, these findings were largely replicated with a clinical sample. In study 3, students with GAD, but not controls, displayed greater increases in self-reported physiological symptoms after listening to emotion-inducing music than after neutral mood induction. Further, GAD participants had more difficulty managing their emotional reactions. Implications for GAD and psychopathology in general are discussed.

Introduction

Approximately 5% of people will suffer from generalized anxiety disorder (GAD) at some point in their lives (Kessler et al., 1994). GAD is associated with significant role impairment (Wittchen, Zhao, Kessler, & Eaton, 1994), increased health care utilization (Blazer, Hughes, & George, 1991), increased health care costs and decreased productivityy (Greenberg et al., 1999). However, despite the prevalence of GAD and the suffering associated with it, GAD remains poorly understood relative to other anxiety disorders and, as a result, has been more difficult to treat. When compared to the other anxiety disorders, far fewer investigations have examined the psychopathological mechanisms involved in GAD (Dugas, 2000).

Recently, theorists have begun to expand our understanding of GAD through the development of models that highlight the importance of worry. One of the most comprehensive accounts of the role of worry in GAD is Borkovec's avoidance theory (e.g., Borkovec, Alcaine, & Behar, 2004). Borkovec and colleagues have presented convincing empirical support for the notion that worry is a perseverative, cognitive activity that serves an avoidance function for persons with GAD. More specifically, Borkovec and colleagues argue that worry allows individuals to approach emotional topics at an abstract, conceptual level and, consequently, to avoid aversive images, autonomic arousal, and intense negative emotions in the short-run (for a review of numerous studies in support of this view, see Borkovec et al., 2004). However, over the long term, the individual is repeatedly confronted with the emotional material, frequently has a more intense experience of anxiety, and engages in repetitive worry to “dull” this experience. In doing so, the person again fails to fully confront the distressing stimuli, and emotional processing of aversive experiences is inhibited.

As described above, from the perspective of the avoidance theory of worry, worry facilitates avoidance of the imagery and physiological arousal associated with negative emotion. However, the nature of the emotional experience that prompts individuals with GAD to engage in frequent avoidance strategies such as worry has not been directly addressed. Although individuals with GAD may use worry to avoid distressing emotional experience, the theory does not explain why this experience is so aversive that it would need to be avoided. To understand this, the characteristics of the emotional experience that may prompt avoidance need to be explored. Further, emotion may play a larger role in GAD than specifically in relation to worry. Emotion and its dysregulation may be integral, yet largely unexplored, factors in the psychopathology of GAD and thus may have important implications for treatment (see Samoilov & Goldfried, 2000).

Conceptualizations of GAD may benefit from attention to advances in the fields of emotion theory (e.g., Ekman & Davidson, 1994), emotion regulation (e.g., Gross, 1998), and affective neuroscience (e.g., LeDoux, 1996). Contemporary theories of emotion emphasize its adaptive value (e.g., Gross, 1998). Theorists have argued that emotions are cues for readiness for action or “action tendencies” that work to establish, maintain, or disrupt relationships with particular internal and external environments that signify importance to the person (see Barlow, 2002). Emotion serves an information function, notifying individuals of the relevance of their concerns, needs, or goals in a given moment. Attention to the adaptive value to emotions may account for the resurgence of interest in the role of emotion in psychopathology and psychotherapy (e.g., Berenbaum, Raghavan, Le, Vernon, & Gomez, 2003; Greenberg, 2002; Kring & Bachorowski, 1999; Samoilov & Goldfried, 2000).

Emotion regulation, as a field of study, examines how individuals influence, control, experience, and express their emotions (Gross, 1998, p. 275). In discussing regulation of one's own emotions, Thompson (1990) stresses the importance of the restraint of emotion, as well as its maintenance and enhancement. Clearly, needs to diminish emotional arousal to work effectively or contain one's anger in a public setting are aspects of emotion regulation. However, investigators have more recently argued for the importance of accentuating both positive and negative emotional experiences to gain a greater understanding of goal pursuit (see Bonnano, 2001). Adaptive regulation of emotions has been found to relate to positive functioning and emotional health. Salovey, Mayer, and colleagues (e.g., Mayer, Salovey, Caruso, & Sitarenios (2001), Mayer, Salovey, Caruso, & Sitarenios (2003)) theorize that people differ in their ability to attend to, process, and act as a result of their emotions, which they have called emotional intelligence. Research has begun to accumulate that supports the functional benefits of emotional intelligence (e.g., Mayer et al., 2003).

If greater ability to manage emotions is associated with productivity and positive emotional health, then low levels of emotion regulatory ability should be associated with psychopathology and difficulties in adjustment. Indeed, investigators have applied findings from emotion research to clinical phenomena such as unipolar depression (e.g., Rottenberg, Kasch, Gross, & Gotlib, 2002), eating disorders (e.g., Westen & Harnden-Fischer, 2001), and borderline personality disorder (e.g., Wagner & Linehan, 1999). Rottenberg and Gross (2003) caution that, when looking at the relationship between emotion dysregulation and psychopathology, investigators need to recognize that regulation occurs dynamically throughout different points in the emotion generative process. Using this perspective as a base, Mennin and colleagues (for an introduction to this perspective, see Mennin, Heimberg, Turk, & Fresco, 2002; Mennin, Heimberg, Turk, & Carmin, 2004) have developed an emotion dysregulation model of GAD. Emotion may become dysfunctional in individuals with GAD through a rapid temporal process of dysregulation involving (1) heightened intensity of emotions; (2) poor understanding of emotions; (3) negative reactivity to one's emotional state (e.g., fear of emotion); and (4) maladaptive emotional management responses. Heightened emotional intensity coupled with an inadequate base of knowledge about emotions and discomfort with emotional experience may lead individuals with GAD to use maladaptive coping strategies to control and constrain their emotional experience to decrease this aversive state. Rather than processing an emotion through attention, understanding, and experiencing (Foa & Kozak, 1986), they may utilize worry and other intra- and interpersonal processes to avoid the distress associated with these emotions.

We have suggested that individuals with GAD have emotional reactions that occur more easily, quickly, and intensely than for most other people (i.e., heightened emotional intensity). They may frequently experience strong negative affect, which sometimes is elicited by situations that are not evocative to most other people. Consistent with the research showing that higher levels of emotional intensity are associated with more emotion expressivity (Gross & John, 1997), individuals with GAD may also express more of their emotions, especially negative emotions, than most other people. Being overly expressive of negative emotions on a regular basis may lead to criticism or rejection by others, which, in turn, may elicit high levels of negative affect in these emotionally sensitive individuals with GAD. Research suggests that individuals with GAD perceive their relationships with family, friends, and romantic partners as moderately to severely impaired (Turk, Mennin, Fresco, & Heimberg, 2000), and evidence is mounting that individuals with GAD may have interpersonal styles that contribute to the relationship problems that they perceive. Pincus and Borkovec (1994) and Eng and Heimberg (2004) found that the majority of individuals with GAD endorsed interpersonal styles that were best characterized as overly nurturant and intrusive.

Individuals with GAD may also have difficulty identifying primary emotions such as anger, sadness, fear, disgust, and joy and instead experience their emotions as undifferentiated, confusing, and overwhelming (i.e., poor understanding of emotions). In this way, persons with GAD may be unable to access and utilize the adaptive information conveyed by their emotions. Difficulty identifying and describing emotions has been related to various forms of psychopathology (see Taylor, Bagby, Parker, & Alexander, 1997). Given strong emotional responses and a poor understanding of them, individuals with GAD may experience emotions as aversive and become anxious when they occur (i.e., negative reactivity to emotions). Associated reactions may include extreme hypervigilance for threatening information and activation of negative beliefs about emotions.

We also hypothesize that individuals with GAD have difficulty knowing when or how to enhance or diminish their emotional experience in a manner that is appropriate to the environmental context (i.e., maladaptive emotional management). Given the intensity of their emotion and both their lack of skills for utilizing emotions and their negative reactions to them, we suggest that individuals with GAD turn to a variety of management approaches that are maladaptive. Typically, dysregulation is viewed in terms of faulty control mechanisms. However, there may be other types of dysregulation that occurs in individuals with GAD. Cicchetti, Ackerman, and Izard (1995) suggest that problems of emotional regulation can be divided into two categories. The first involves difficulties in modulation of emotional experience and/or expression; the second involves frequent or automatic attempts to control or suppress emotional experience or expression. In the first scenario, the person experiences emotions with great intensity but is unable to adequately modulate the experience (e.g., self-soothe, inhibit emotional expression). In the second scenario, the person engages in control strategies that prevent emotion from being experienced. Berenbaum et al. (2003) propose a similar dichotomy in their discussion of two emotional regulation disturbances, which they term emotional hyperreactivity and hyporeactivity.

Mennin, Heimberg, Turk, & Fresco (2002), Mennin, Turk, Heimberg, & Carmin (2004) argue that both types of deficits characterize GAD. For example, intense emotions that are misunderstood and misinterpreted as dangerous may contribute to the problems that individuals with GAD have in modulating their emotional experience and expression. Furthermore, Mennin and colleagues propose that individuals with GAD over-engage control mechanisms, decreasing emotional experience through avoidance and blunting. This latter form of dysregulation has been discussed extensively by Hayes and colleagues (see Hayes, Strosahl, & Wilson, 1999) in their examination of experiential avoidance. Experiential avoidance refers to one's difficulty engaging internal experiences such as emotions, thoughts, images, and sensations. Often this unwillingness stems from efforts to control and diminish the experience of pain associated with these internal events. However, paradoxically, these efforts to control often lead to further and stronger levels of discomfort, and may be an integral component of many forms of psychopathology (Hayes et al., 1999). Borkovec's avoidance model of worry can be viewed within such an emotion regulatory framework. Worry can be viewed as a cognitive control strategy employed in attempts to “fix” the regulatory problems associated with subjectively jarring emotional experience.

The following three investigations aim to provide preliminary support for this emotion dysregulation perspective on GAD. The first study is an initial investigation of difficulties in regulating emotions among college students with and without GAD. The ability of emotion dysregulation to predict the presence of GAD was also assessed. The second study was a replication of the first, comparing treatment-seeking patients with GAD to community controls. Finally, an experimental study is presented that examined whether self-reported physiological responses to negative mood induction, regardless of subjective awareness of emotions, are greater in individuals with GAD than control participants. Further, participants’ abilities to manage this induced mood experience were examined.

Section snippets

Study 1: preliminary examination in an undergraduate sample

This initial study examined the relationships among components of emotion, its dysregulation and GAD. We hypothesized that individuals with GAD, in comparison to control participants, would demonstrate heightened intensity of emotions (i.e., greater emotion impulse strength, greater emotional expressivity), poorer understanding of emotions (i.e., less clarity of emotions, greater difficulty identifying and describing emotions), more negative reactivity to emotions (i.e., greater attention to

Study 2: application to a clinical sample

A sample of treatment-seeking GAD patients was compared to a sample of individuals from the local community on the same indices of emotion dysregulation. However, the BEQ was not administered as part of the battery at the time of this study, so it was not included in this replication. As in study 1, it was hypothesized that individuals with GAD, compared to the control group, would have greater difficulty on the three dimensions of emotion dysregulation assessed. Further, it was hypothesized

Preliminary analyses

GAD patients (M=33.88, SD=12.86) did not differ significantly from control participants (M=30.05, SD=10.38) in age [t (95)=1.62, ns]. Both the GAD (80.0%) and control (73.2%) samples consisted primarily of Caucasian individuals [χ2(4, N=101)=4.76, ns]. Both GAD patients (73.8%) and control participants (75.9%) were predominantly single and never married [χ2(4, N=97)=2.26, ns]. However, the GAD group included a significantly higher percentage of women than the control group [χ2 (1, N=97)=7.10, p<

Study 3: emotion dysregulation in response to induced negative mood

This study sought to experimentally investigate whether elicitation of emotion would lead to increases in reported somatic distress and subsequent difficulties in understanding, accepting, and managing this emotional experience in individuals with GAD compared to controls. The purpose of this study was twofold. First, by experimentally inducing negative emotions, negative reactions and dysregulatory responses would concern a specific episode of mood rather than asking individuals to reflect on

General discussion

This series of studies provides initial evidence for the presence of emotion dysregulation in individuals with GAD. Specifically, individuals who met criteria for GAD reported greater intensity of emotional experience and a greater tendency to express negative emotions than control individuals. Further, they also displayed marked difficulties in their ability to identify, describe, and clarify their emotional experiences. Individuals with GAD also displayed greater negative reactivity than

Douglas Mennin is now at Yale University. Cynthia Turk is now at LaSalle University. David Fresco is now at Kent State University. Portions of this paper were presented at the annual meeting the Association for the Advancement of Behavior Therapy in November 2000, 2001 and at the Anxiety Disorders Association of America Annual Meeting, 2001. Study 3 is part of Douglas Mennin's dissertation. He would like to thank his dissertation committee for their time and contributions (Richard G. Heimberg,

References (98)

  • J.D. Mayer et al.

    An emerging understanding of the reflective (meta-) experience of mood

    Journal of Research in Personality

    (1994)
  • D.R. McLeod et al.

    Somatic symptoms of anxietycomparison of self-report and physiological measures

    Biological Psychiatry

    (1986)
  • T.J. Meyer et al.

    Development and validation of the Penn State Worry Questionnaire

    Behaviour Research and Therapy

    (1990)
  • M.G. Newman et al.

    Preliminary reliability and validity of the Generalized Anxiety Disorder Questionnaire-IVa revised self-report diagnostic measure of generalized anxiety disorder

    Behavior Therapy

    (2002)
  • K.V. Petrides et al.

    On the dimensional structure of emotional intelligence

    Personality and Individual Differences

    (2000)
  • A.M. Ruscio

    Delimiting the boundaries of generalized anxiety disorderdifferentiating high worriers with and without GAD

    Journal of Anxiety Disorders

    (2002)
  • A.W. Staats et al.

    The paradigmatic behaviorism theory of emotions

    Clinical Psychology Review

    (1990)
  • R.A. Steer et al.

    Further evidence for the validity of the Beck Anxiety Inventory with psychiatric outpatients

    Journal of Anxiety Disorders

    (1993)
  • K.E. Williams et al.

    Are emotions frightening? An extension of the fear of fear construct

    Behaviour Research and Therapy

    (1997)
  • Abel, J. L. (1994). Alexithymia in an analogue sample of generalized anxiety disorder and non-anxious matched controls....
  • American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.) Washington,...
  • D.H. Barlow

    Anxiety and its disorders

    (2002)
  • A.T. Beck et al.

    An inventory for measuring clinical anxietypsychometric properties

    Journal of Consulting and Clinical Psychology

    (1988)
  • A.T. Beck et al.

    Cognitive therapy of depression

    (1979)
  • H. Berenbaum et al.

    A taxonomy of emotional disturbances

    Clinical PsychologyScience and Practice

    (2003)
  • D.G. Blazer et al.

    Generalized anxiety disorder

  • G. Bonnano

    Self-regulation of emotions

  • T.D. Borkovec et al.

    Avoidance theory of worry and generalized anxiety disorder

  • T.A. Brown et al.

    Anxiety Disorders Interview Schedule for DSM-IV (ADIS-IV)

    (1994)
  • T.A. Brown et al.

    Reliability of DSM-IV anxiety and mood disordersimplications for the classification of emotional disorders

    Journal of Abnormal Psychology

    (2001)
  • D. Cicchetti et al.

    Emotions and emotion regulation in developmental psychopathology

    Development and Psychopathology

    (1995)
  • J. Cohen

    Statistical power analysis for the behavioral sciences

    (1988)
  • P.A. DiNardo et al.

    Anxiety Disorders Interview Schedule for DSM-IVlifetime version (ADIS-IV-L)

    (1994)
  • Eng, W., & Heimberg, R. G. (2004). Interpersonal correlates of generalized anxiety disorder. Submitted for...
  • W.W. Gaver et al.

    Play it again, Samon liking music

    Cognition and Emotion

    (1987)
  • R.A. Gould et al.

    Cognitive-behavioral treatmentsa meta- analytic review

  • L.S. Greenberg

    Emotion-focused therapycoaching clients to work through feelings

    (2002)
  • P.E. Greenberg et al.

    The economic burden of anxiety disorders in the 1990s

    Journal of Clinical Psychiatry

    (1999)
  • J.J. Gross

    The emerging field of emotion regulationan integrative review

    Review of General Psychology

    (1998)
  • J.J. Gross et al.

    Revealing feelings: facets of emotional expressivity in self-reports, peer ratings, and behavior

    Journal of Personality and Social Psychology

    (1997)
  • J.J. Gross et al.

    Emotion elicitation using films

    Cognition and Emotion

    (1995)
  • J.J. Gross et al.

    Emotion regulation and mental health

    Clinical PsychologyScience and Practice

    (1995)
  • Hayes, S. C., Bissett, R. T., Strosahl, K., Follette, W. C., Polusney, M. A., Pistorello, J., Toarmino, D., Batten, S....
  • S.C. Hayes et al.

    Acceptance and commitment therapyan experiental approach to behavior change

    (1999)
  • E.W. Herron

    The multiple affect adjective check lista critical analysis

    Journal of Clinical Psychology

    (1969)
  • R.C. Kessler et al.

    Impairment in pure and comorbid generalized anxiety disorder and major depression at 12 months in two national surveys

    American Journal of Psychiatry

    (1999)
  • R.C. Kessler et al.

    Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United Statesresults from the National Comorbidity Survey

    Archives of General Psychiatry

    (1994)
  • A.M. Kring et al.

    Emotion and psychopathology

    Cognition and Emotion

    (1999)
  • Cited by (0)

    Douglas Mennin is now at Yale University. Cynthia Turk is now at LaSalle University. David Fresco is now at Kent State University. Portions of this paper were presented at the annual meeting the Association for the Advancement of Behavior Therapy in November 2000, 2001 and at the Anxiety Disorders Association of America Annual Meeting, 2001. Study 3 is part of Douglas Mennin's dissertation. He would like to thank his dissertation committee for their time and contributions (Richard G. Heimberg, Philip C. Kendall, Jerome Resnick, Alan Sockloff, Thomas D. Borkovec, and Brian P. Marx). Thanks also to Jon Rottenberg for his suggestions for revising this manuscript.

    View full text