Elsevier

Clinical Psychology Review

Volume 35, February 2015, Pages 10-18
Clinical Psychology Review

Content specificity of attention bias to threat in anxiety disorders: A meta-analysis

https://doi.org/10.1016/j.cpr.2014.10.005Get rights and content

Highlights

  • This meta-analysis of 37 samples examines content specificity of attention bias.

  • Results reveal a small but significant threat-bias specificity effect.

  • The effect is discussed in relation to neuro-cognitive models of attention bias.

  • The findings could bear implications for the design and stimulus choice in ABMT.

  • Stimuli congruent with the anxiety disorder may produce better results.

Abstract

Despite the established evidence for threat-related attention bias in anxiety, the mechanisms underlying this bias remain unclear. One important unresolved question is whether disorder-congruent threats capture attention to a greater extent than do more general or disorder-incongruent threat stimuli. Evidence for attention bias specificity in anxiety would implicate involvement of previous learning and memory processes in threat-related attention bias, whereas lack of content specificity would point to perturbations in more generic attention processes. Enhanced clarity of mechanism could have clinical implications for the stimuli types used in Attention Bias Modification Treatments (ABMT). Content specificity of threat-related attention bias in anxiety and potential moderators of this effect were investigated. A systematic search identified 37 samples from 29 articles (N = 866). Relevant data were extracted based on specific coding rules, and Cohen's d effect size was used to estimate bias specificity effects. The results indicate greater attention bias toward disorder-congruent relative to disorder-incongruent threat stimuli (d = 0.28, p < 0.0001). This effect was not moderated by age, type of anxiety disorder, visual attention tasks, or type of disorder-incongruent stimuli. No evidence of publication bias was observed. Implications for threat bias in anxiety and ABMT are discussed.

Introduction

Selective processing of threat has been studied widely and is thought to contribute to the etiology and maintenance of anxiety disorders. In particular, the tendency of anxious individuals to overly attend to threat stimuli has been documented with different attention tasks and in different types of anxiety (Bar-Haim, Lamy, Pergamin, Bakermans-Kranenburg, & van IJzendoorn, 2007). A causal relation between attention bias and anxiety has been demonstrated by experimentally inducing attention bias toward threat in healthy children and adults and recording associated elevations in stress reactivity (Eldar et al., 2008, Mathews and MacLeod, 2002). This research, underscoring the role of threat-related attention bias in anxiety has further informed the development of Attention Bias Modification Treatments (ABMT) designed to reduce anxiety through change in threat-related attention patterns (Bar-Haim, 2010, Mathews and MacLeod, 2002). Recent meta-analyses indicate that ABMT carries a significant small-to-medium effect size in anxiety reduction (Beard et al., 2012, Hakamata et al., 2010, Hallion and Ruscio, 2011, Mogoase et al., 2014).

Despite ample evidence for threat-related attention bias in anxiety, the underlying mechanisms supporting these biases remain largely unclear (Cisler and Koster, 2010, Heeren et al., 2013, Ouimet et al., 2009). One potentially relevant mechanistic aspect of threat-related attention bias in anxiety concerns the nature and the specific features of the threat stimuli that over capture the attention of anxious participants. Specifically, it is still unclear whether threat-related attention bias in anxiety is triggered by threats in general, regardless of their specific content, and thereby reflects perturbations in a generic neuro-behavioral mechanism, or alternatively evidenced exclusively, or more saliently, in relation to specific threat contents that are directly related to a participant's idiosyncratic angst.

Neuro-cognitive and cognitive models of anxiety emphasize different threat processing mechanisms that could contribute to threat-related attention bias and to the potential impact of threat's content on bias magnitude. For instance, some models emphasize the role of threat detection or initial response to threat as a major contributor to anxiety (Beck and Clark, 1997, Williams et al., 1988) This adaptive attentional function that facilitates detection of danger (LeDoux, 2000) is thought to be amplified in anxiety. In line with models emphasizing automatic capture of attention by minor threats in anxiety, amygdala hypersensitivity was shown among anxious but not healthy subjects, and the magnitude of amygdala engagement with threat stimuli positively correlated with both anxiety severity (Monk et al., 2008) and with attention bias toward threat (van den Heuvel et al., 2005). This view predicts initial, automatic threat evaluation, classifying incoming stimuli crudely as threatening or safe, without registering specific content (Mathews & Macleod, 1994). Such mechanism would allow relatively small impact for content specificity on threat-related attention bias in anxiety.

Other models refer to threat-related attention bias as a result of difficulty in regulation and allocation of attention (Bishop, 2007, Eysenck et al., 2007). For instance, it was found that anxious individuals showed poorer performance on attention control tasks with threat stimuli and less activation in the lateral pre-frontal cortex (LPFC) relative to non-anxious individuals (Bishop et al., 2004, Monk et al., 2006). The LPFC is thought to play a role in the regulation of amygdala activation in the presence of threat (Pine et al., 2009, Quirk and Mueller, 2008). Emphasizing general attention control dysfunction as a primary mechanism underlying threat-related attention bias alludes to a more generic aspect of threat-related attention bias in anxious individuals that may be less affected by the specific content of threat.

Other models of threat processing describe a potentially more elaborate threat evaluation process that impacts attention allocation to threat (e.g., Bar-Haim et al., 2007, Beck and Clark, 1997, Mogg and Bradley, 1998, Öhman, 1996). These models assume schema-driven processing that rely on associations to personal learning and memory, that may involve a content-specific aspect of attention bias, driven by specific threats that are idiosyncratically relevant to a person's anxiety type.

The current meta-analysis investigates the potential role of content specificity in threat-related attention bias in anxiety against the backdrop of the alternative of more generic perturbations in attention allocation that leave much less room to content specificity. In addition to broadening the conceptualization of the nature of attention bias to threat in anxiety, the examination of attention bias specificity also carries potential clinical relevance for the development of treatments that rely on presentation of different threat-related stimuli.

Attention bias specificity has typically been explored by testing whether disorder-congruent stimuli (e.g., socially relevant stimuli for social phobia or trauma-related stimuli for posttraumatic stress disorder) render larger threat-related attention bias than do general threat stimuli, or stimuli that are congruent with the threat content of a different anxiety disorder (i.e., disorder-incongruent stimuli). Most studies compare the magnitude of threat-related attention bias of disorder-congruent and disorder-incongruent stimuli using response times in classic visual attention tasks. However, despite considerable research on stimulus specificity, results from single studies are mixed.

For example, Foa, Feske, Murdock, Kozak, and McCarthy (1991) found attention bias specificity for trauma-related words, relative to other threat-word types in rape-victims suffering from post-traumatic stress disorder (PTSD). Additional studies also reported content specificity effects in adults with PTSD (Ashley et al., 2013, Buckley et al., 2002, Kaspi et al., 1995, McNally et al., 1994), and in youth with PTSD (Moradi, Taghavi, Heshat Doost, Yule, & Dalgleish, 1999). Conversely, other studies did not find content specificity of attention bias in PTSD using trauma-related words either in adults (e.g., Litz et al., 1996) or children (Ribchester, Yule, & Duncan, 2010). Elsesser, Sartory, and Tackenberg (2004) did not find specificity effect with trauma-related pictures in PTSD patients, even when stimuli were selected idiosyncratically.

There is also accumulating data concerning specificity of threat attention bias in panic disorder, usually tested with words reflecting physical threats. Specificity was found in panic patients using both Stroop and dot-probe tasks (Asmundson et al., 1992, Buckley et al., 2002, respectively). However, other studies did not find this effect neither in stroop (De Cort et al., 2008, Gropalis et al., 2013, Kampman et al., 2002) nor in dot-probe tasks (Beck et al., 1992, Horenstein and Segui, 1997).

Mixed results were also found in social anxiety as well as in obsessive-compulsive disorder (OCD) samples, with some studies reporting evidence for content specificity (Becker et al., 2001, Olatunji et al., 2011), while others failed to find enhanced processing bias when the content of the stimuli was congruent with the anxiety type studied (Foa et al., 1993, Kampman et al., 2002). This inconsistency across studies along with the theoretical and practical implications outlined above articulates the need for a systematic quantitative review of this literature.

In the current meta-analysis we go beyond the primary question of whether content specificity of attention bias to threat in anxiety exists by testing the role of potential moderators of this effect. We tested: a) whether stimulus specificity is evident in specific anxiety disorders and perhaps not (or to a lesser extent) in others; b) whether the type of attention task used to measure threat bias moderated the magnitude of the effects. Specifically, evidence for attention bias specificity in anxiety is derived mainly from studies utilizing the emotional Stroop task (Williams, Mathews, & MacLeod, 1996) and the dot-probe task (MacLeod, Mathews, & Tata, 1986), both allowing computation of a general threat bias index. However, it is also commonly acknowledged that each of these tasks taps into different cognitive processes. The emotional Stroop effect is thought to reflect processes of threat-related interference, and even threat driven general slow-down that is not attentional (Algom, Chajut, & Lev, 2004). In contrast, the dot-probe task is thought to reflect visual-spatial attention. Since the two tasks may reflect different cognitive mechanisms contributing to the diversity in bias specificity effects, we tested task type as a potential moderator; c) whether the nature of the disorder-incongruent comparison stimuli (general threat or threat specifically related to a different anxiety disorder) affect the magnitude of attention bias specificity; and finally, d) we tested whether participants' age (youth under 18 years of age vs. adults) affected the magnitude of the bias specificity effect.

Answers to these questions can inform theoretical conceptualizations and future research as well as clinical decision in the development of attention bias modification protocols.

Section snippets

Literature base

A flow diagram of samples selection is provided in Fig. 1. Studies were collected through a search of the PsycInfo and PubMed databases using the key words attention*, bias*, specific*, content*, relevant* intersected with anxi* (anxiety), phob* (phobia), PTSD, OCD, panic. The references of all the obtained articles were systematically searched for additional relevant studies. This search yielded 47 samples from 40 articles that were potentially eligible for inclusion in the meta-analysis (see

Results

Study characteristics and effect sizes for each of the 37 samples included in the meta-analysis, alongside a forest plot of the effects, are presented in Fig. 2. The overall combined effect size of content-specificity was significant with Cohen's d = 0.28, p < 0.0001. This effect represents a greater threat-related attention bias for stimuli that are disorder-congruent relative to disorder-incongruent. Egger's test indicated no publication bias for this combined effect size, intercept = 0.22, p = 0.89

Discussion

The present meta-analysis provides support for the notion of attention bias specificity in anxiety by showing that among clinically anxious samples attention bias for threat stimuli that are disorder-congruent is larger relative to disorder-incongruent threat stimuli. The 37 samples included in this meta-analysis established a small but significant threat-bias specificity effect, which is not significantly moderated by age, type of anxiety disorder, different visual attention tasks, and the

Role of funding sources

This work was partially supported by the United States–Israel Binational Science Foundation (BSF, Grant # 2009340). BSF had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication.

Contributors

Lee Pergamin-Hight, Reut Naim and Yair Bar- Haim designed the study and wrote the protocol. Lee Pergamin-Hight and Reut Naim conducted literature searches, provided summaries of previous research studies and conducted the statistical analysis. Yair Bar-Haim, Marinus van IJzendoorn, and Marian Backermans-Kranenburg provided feedback on the statistical approach and edited drafts of the manuscript. All authors have approved the final manuscript.

Conflict of interest

All authors declare that they have no conflicts of interest.

Acknowledgment

This work was partially supported by the United States–Israel Binational Science Foundation (BSF, Grant # 2009340).

References (72)

  • Y. Hakamata et al.

    Attention bias modification treatment: A meta-analysis toward the establishment of novel treatment for anxiety

    Biological Psychiatry

    (2010)
  • C. Hunt et al.

    Anxiety sensitivity, conscious awareness and selective attentional biases in children

    Behaviour Research and Therapy

    (2007)
  • M. Kampman et al.

    The emotional Stroop: A comparison of panic disorder patients, obsessive-compulsive patients, and normal controls, in two experiments

    Journal of Anxiety Disorders

    (2002)
  • A. Kaur et al.

    Health threat increases attentional bias for negative stimuli

    Journal of Behavior Therapy and Experimental Psychiatry

    (2013)
  • E. Keogh et al.

    Selective attentional bias for pain-related stimuli amongst pain fearful individuals

    Pain

    (2001)
  • M. Kyrios et al.

    Automatic and strategic processing in obsessive-compulsive disorder: Attentional bias, cognitive avoidance or more complex phenomena?

    Journal of Anxiety Disorders

    (1998)
  • E. Maidenberg et al.

    Specificity of attentional bias in panic disorder and social phobia

    Journal of Anxiety Disorders

    (1996)
  • R.J. McNally et al.

    Cognitive processing of idiographic emotional information in panic disorder

    Behaviour Research and Therapy

    (1994)
  • R.J. McNally et al.

    Selective processing of threat cues in panic disorder

    Behaviour Research and Therapy

    (1990)
  • K. Mogg et al.

    A cognitive-motivational analysis of anxiety

    Behaviour Research and Therapy

    (1998)
  • M.G. Newman et al.

    Diagnostic comorbidity in adults with generalized anxiety disorder: Impact of comorbidity on psychotherapy outcome and impact of psychotherapy on comorbid diagnoses

    Behavior Therapy

    (2010)
  • B.O. Olatunji et al.

    A selective impairment in attentional disengagement from erotica in obsessive–compulsive disorder

    Progress in Neuro-Psychopharmacology & Biological Psychiatry

    (2011)
  • A.J. Ouimet et al.

    Cognitive vulnerability to anxiety: A review and an integrative model

    Clinical Psychology Review

    (2009)
  • S.L. Pineles et al.

    Attentional biases in PTSD: More evidence for interference

    Behaviour Research and Therapy

    (2009)
  • I. Spector et al.

    Selective attentional bias related to the noticeability aspect of anxiety symptoms in generalized social phobia

    Journal of Anxiety Disorders

    (2003)
  • J. Wikström et al.

    Preattentive bias for snake words in snake phobia?

    Behaviour Research and Therapy

    (2004)
  • D. Algom et al.

    A rational look at the emotional Stroop phenomenon: A generic slowdown, not a Stroop effect

    Journal of Experimental Psychology: General

    (2004)
  • V. Ashley et al.

    Attentional bias for trauma-related words: Exaggerated emotional Stroop effect in Afghanistan and Iraq war veterans with PTSD

    BMC Psychiatry

    (2013)
  • Y. Bar-Haim

    Research review: Attention bias modification (ABM): A novel treatment for anxiety disorders

    Journal of Child Psychology and Psychiatry

    (2010)
  • Y. Bar-Haim et al.

    Threat-related attentional bias in anxious and nonanxious individuals: A meta-analytic study

    Psychological Bulletin

    (2007)
  • S.J. Bishop et al.

    Prefrontal cortical function and anxiety: Controlling attention to threat-related stimuli

    Nature Neuroscience

    (2004)
  • T.C. Buckley et al.

    Automatic and strategic processing of threat stimuli: A comparison between PTSD, panic disorder, and nonanxiety controls

    Cognitive Therapy and Research

    (2002)
  • K. De Cort et al.

    A specific attentional bias in panic disorder?

    Depression and Anxiety

    (2008)
  • S. Duval et al.

    Trim and fill: A simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis

    Biometrics

    (2000)
  • A. Ehlers et al.

    Selective processing of threat cues in subjects with panic attacks

    Cognition and Emotion

    (1988)
  • K. Elsesser et al.

    Attention, heart rate, and startle response during exposure to trauma-relevant pictures: A comparison of recent trauma victims and patients with posttraumatic stress disorder

    Journal of Abnormal Psychology

    (2004)
  • Cited by (135)

    View all citing articles on Scopus
    1

    This work reflects equal contribution of the first two authors (LPH and RN).

    View full text