Elsevier

Clinical Psychology Review

Volume 54, June 2017, Pages 107-122
Clinical Psychology Review

Review
Generalized Anxiety Disorder, worry and attention to threat: A systematic review

https://doi.org/10.1016/j.cpr.2017.03.006Get rights and content

Highlights

  • Twenty-nine published articles were included in the final review.

  • Strong evidence of a bias to threat among people with GAD compared to other groups.

  • Few studies had investigated this bias in high trait worriers.

  • Bias to threat in people with GAD is strongest when threat material were words.

Abstract

Among anxious populations, attention has been demonstrated to be preferentially biased to threatening material compared to neutral or other valenced material. Individuals who have high levels of trait worry, such as those with Generalized Anxiety Disorder (GAD), may be biased to threat but research has produced equivocal findings. This review aimed to systematically review the extant experimental literature to establish the current evidence of attentional bias to threat among trait worriers compared to healthy controls and other clinical populations. Twenty-nine published articles were included in the final review. There was strong evidence of a bias to threat among GAD patients compared to other groups and this was found across most experimental paradigms. Few studies had investigated this bias in non-clinical trait worriers. Among GAD patients this bias to threat was most strongly evidenced when visual threat material was in a verbal-linguistic format (i.e., words) rather than when in pictorial form (i.e., images or faces). The bias was also found across several domains of negative material, supporting the general nature of worry. Further research should look to examine the specific components of the threat bias in GAD, as well as investigating the bias to threat in trait worriers.

Introduction

The current review examines the extant literature on attentional bias to threatening stimuli among individuals with a diagnosis of Generalized Anxiety Disorder (GAD) or those with high levels of trait worry who do not have a diagnosis of GAD. The paper begins by highlighting the distinction between “bottom-up” and “top-down” processing before then defining worry itself. The paper also provides a discussion of the different mechanisms associated with attentional biases (i.e., engagement, disengagement, shifting) and then leads into a brief review of neurobiological evidence of threat biases and worry. The paper then discusses Hirsch and Mathews' (2012) cognitive model of pathological worry which focuses on information-processing biases (including attentional bias) and the role of attentional control in promoting uncontrollable worry, before reviewing in detail the literature on attentional bias to threat.

This paper is the first known review that systematically examines the empirical evidence of attentional bias to threat among individuals with GAD and/or pathological levels of trait worry. Although previous papers have examined trait anxiety more broadly, worry has not been specifically targeted for review. However, worry is an integral cognitive component of anxiety, which can interfere with information-processing directly (Hayes, Hirsch, & Mathews, 2008) and has been linked to attentional bias to threat stimuli (Mathews and MacLeod, 2005, Williams et al., 2014). It also represents a cardinal feature of GAD, an anxiety disorder with often difficult to treat symptoms. Therefore, this review will examine the evidence base for attentional bias to threat in individuals with high levels of worry in order to offer important insights into our understanding of attentional bias in those suffering from GAD or pathological worry, to highlight directions of future research and areas for potential treatment innovation. Given the lack of previous reviews targeting this specific characteristic of anxiety, the current paper will aim to focus on the association only between worry and attentional bias to threat, rather than trying to identify the specific direction of the relationship. Indeed, the role of attentional bias in the development of anxiety is rather complex and beyond the remit of this current paper, as Van Bockstaele et al. (2014) eloquently highlighted that “the relation between attentional bias and fear and anxiety is best described as a bidirectional, maintaining, or mutually reinforcing relation.” (page 682).

Visual attention can be captured by salient or distinctive information in everyday environments, such as a smiling face, a growling dog, or a speeding car. At a basic level, selective attention can be defined as “any cognitive operation that results in the selection of some information over other information” (Weierich, Treat, & Hollingworth, 2008, p. 988). This selection can be stimulus-driven, such as changes in perceptual events which may capture attention automatically (Franconeri & Simons, 2003), or can be more strategically controlled, such as avoiding certain stimuli in order to regulate emotion (Calvo & Avero, 2005). The former is often regarded as being mediated by sub-cortical “bottom-up” pathways designed to rapidly detect salient stimuli in the environment (Davis & Whalen, 2001), whilst the latter is believed to be regulated by “top-down” pathways located in more prefrontal cortex regions, associated with attentional control, working memory, and goal-driven behaviour (Miller & Cohen, 2001). One factor that may influence the selection of attention is the level of threat attached to the stimulus, which may bias individuals to attend to it over neutrally valenced stimuli in the environment (e.g., MacLeod, Mathews, & Tata, 1986). This preferential processing of threat is regarded as being evolutionarily adaptive (to monitor danger in the environment) (Ohman, 1986) and is thus applicable to most individuals, but it is more pronounced in anxious individuals compared to non-anxious populations (Bar-Haim et al., 2007, Beck et al., 1985, Mathews and MacLeod, 1994). This ‘attentional bias’ to threat among anxious populations is well established and may be implicated in the maintenance of anxiety symptoms (Yiend, 2010). However, the attentional system comprises several components and is modulated by multiple mechanisms and so understanding the distinct processes involved within attentional bias to threat among anxious individuals is warranted to inform clinical treatments (Bar-Haim et al., 2007, Cisler et al., 2009, Cisler and Koster, 2010).

Worry is a feature of most anxiety disorders (Beck & Clark, 1997), but in particular is the core criterion of Generalized Anxiety Disorder (GAD; Diagnostic and Statistical Manual 5th Edition, DSM-V; American Psychiatric Association, 2013). Worry is often associated with elevated feelings of anxiety, but is conceptually distinct, as anxiety is more broadly defined as including feelings of tension and autonomic arousal (Borkovec, Robinson, Pruzinsky, & DePree, 1983). Thus high worriers represent a subset of anxious individuals, for whom repetitive negative thoughts (typically in quasi-verbal form) are particularly prominent. Those with high levels of trait worry may experience negative health outcomes, regardless of whether or not they currently qualify for a GAD diagnosis (Brosschot & van der Doef, 2006). Consequently, it is important to identify factors that cause and maintain excessive worry, with attentional biases providing a possible avenue of research (e.g., Oathes, Siegle, & Ray, 2011). Although studies have found attentional bias to threat in GAD patients (e.g., Mogg & Bradley, 2005) and in high trait anxious groups (see Bar-Haim et al., 2007), less research has examined threat bias in non-clinical worriers, who represent an at-risk group for the development of GAD. Importantly, as will be discussed briefly later, investigations of emotional processing have revealed certain neural characteristics that seem to distinguish high worriers from non-worrying high trait anxious individuals (e.g. Engels et al., 2007, Paulesu et al., 2010).

Posner (1980) postulated three components of attention: engagement, disengagement, and shifting. Engagement refers to the orienting of the attentional resources on a particular stimulus, whilst shifting refers to the process of switching from one stimulus to another (Clarke, MacLeod, & Guastella, 2013). In order for shifting to occur though, the individual has to first disengage their attention from the current attended to stimulus. Clarke and colleagues defined biased engagement as “the rapid orientation of attention to a threat stimulus due to its enhanced ability to “capture” or “draw” attention” (2013, p. 3), whilst they defined biased disengagement as the “delayed withdrawal of attention from a threat stimulus due to its ability to “hold” attention” (2013, p. 3). Different methods have been used to assess attentional bias to threat in the anxiety literature, with each having advantages and disadvantages. Posner (1980) developed the spatial cueing task, which involves participants attending to a cue which is located in the same location as a to-be-identified target in the majority of trials and then in the remaining trials the target is in the opposite location (opposite to an original fixation cross). This task was modified by Yiend and Mathews (2001) and Fox, Russo, Bowles, and Dutton (2001) who used different emotional cues (threat/neutral) to identify preferential processing of different emotional stimuli. This task is thought to detect biased engagement and delayed disengagement as inferred by speeded reactions to targets in valid trials (emotional cue and target in same location) and by delayed reactions to invalid trials (target in opposite location to emotional cue), respectively. Fox et al. (2001), Fox, Russo, and Dutton (2002) and Yiend and Mathews (2001) concluded from their use of this paradigm that attentional bias is primarily due to delayed disengagement from threat rather than facilitated engagement to threat. However, some believe that the task measures disengagement better than engagement (Cisler and Koster, 2010, Clarke et al., 2013) and has been criticised for not distinguishing between disengagement and a general behavioural slowing that occurs in the presence of threat (Mogg, Holmes, Garner, & Bradley, 2008; although see Yiend, 2010 for a critique of this). The affective Stroop task (Mathews & MacLeod, 1985) and the attentional probe task (MacLeod et al., 1986) have been commonly used, although other paradigms have also included the visual search task, which typically involves participants having to decide if a target stimulus is present or absent in the presence of distractor stimuli (Müller & Krummenacher, 2006); or the attentional blink task, where a stream of stimuli are displayed and respondents are required to identify a target presented shortly after the first target has been presented (Raymond, Shapiro, & Arnell, 1992).

The results of studies using these different tasks points to attentional bias to threat among anxious individuals in general, but it is unclear whether the bias to threat is a result of facilitated engagement, delayed disengagement, or impaired or biased shifting. This uncertainty is due to a lack of studies that have specifically distinguished the components of attentional bias (Bar-Haim et al., 2007) and the lack of reliability in the methodological designs to confirm the contribution of each component on attentional bias (Clarke et al., 2013). Further, research looking at the neural mechanisms underpinning attentional bias point to different neural networks and locations involved in the bias, as described below.

The attentional system and the regulation of emotion are regarded as operating through an interaction of the amygdala and cortical regions (Bishop, 2007, Blair and Blair, 2012, Cisler and Koster, 2010), which has also been reported in the context of individual differences in anxiety (Mathews, Yiend, & Lawrence, 2004). The initial rapid orienting of attention to threat is regarded as being relatively automatic and has been shown to be coordinated by sub-cortical structures, such as the amygdala (Davis & Whalen, 2001). However, most of the research cited above has not investigated the specific role of worry, as opposed to elevated state or trait anxiety, and current evidence indicates that elevated worry is distinguished by involvement of the so-called “extended Amygdala”; and specifically, the Base Nucleus of the Stria Terminalis or BNST, which is particularly active under conditions of uncertain threat (Paulesu et al., 2010: Yassa, Hazlett, Stark, & Hoehn-Saric, 2012).

Biased engagement of attention with threat cues is often shown at short stimulus exposures in most experimental paradigms (Sagliano, Trojano, Amoriello, Migliozzi, & D'Olimpio, 2014) suggesting a degree of automaticity in the initial capture of attention by threat cues. When a stimulus is exposed for longer durations then it falls within conscious awareness (i.e. is ‘supraliminal’) and it is generally assumed that at these longer stimulus exposures there are more top-down strategic (or controlled) processes contributing towards the allocation of resources (Cisler & Koster, 2010). These top-down processes are believed to be governed by frontal brain structures, such as the prefrontal cortex (Blair et al., 2012), which are involved in disengaging and selectively shifting attention (Miller & Cohen, 2001). As a result, there may be more variation in experimental findings when using supraliminal exposures, as individuals may have different attentional goals. For example, several studies have found a bias towards threat at later exposure durations and concluded that it is due to delayed disengagement from threat, which may be due to an impaired ability to disengage attention from threatening material due to poor top-down attentional control (Derryberry & Reed, 2002) and/or it may represent an underlying emotion regulation strategy to remain focused on threatening information (Wells, 1995). Equally, though, other researchers have found biases away from threat among anxious individuals under supraliminal conditions, and have inferred that this is due to an attentional avoidance of threat (Mogg & Bradley, 1998), which is also governed by top-down processes.

The relative contribution of bottom-up and top-down pathways to attentional bias to threat is still not fully understood. Egloff and Hock (2003) concluded from their assessment of both the affective Stroop and attentional probe tasks that stimulus-driven (bottom-up) biases occurring rapidly in the attentional process (short stimulus exposures outside of conscious awareness) are distinct from biases that occur when stimuli are presented supraliminally, within conscious awareness (where there is opportunity for top-down control). Examination of the research literature on neural mechanisms underpinning attentional bias certainly indicates a more complex picture than simply top-down versus bottom-up processes, as briefly highlighted below.

Neural correlates of worry include activation of medial prefrontal and anterior cingulate regions as well as the BNST (Paulesu et al., 2010), consistent with the phenomenological pattern of thinking reported by high worriers. Similarly, Bishop (2007) concluded in her review of neurocognitive mechanisms of anxiety that amygdala-prefrontal circuitry is likely responsible for biases to threat in anxious individuals. However, the exact location and/or pathways are unclear and it remains uncertain whether the same pathways are similarly involved in the processing of other emotional material (e.g. positive stimuli). For example, dorsolateral prefrontal cortex regions have been shown to mediate processing of positive, as well as threatening, material (Bishop, 2007, Herrington et al., 2005), whereas Miscovic and Schmidt (2010) concluded that variations in attention to threat was jointly accounted for by right frontal EEG asymmetry and low cardiac vagal tone. Furthermore, Clarke, Browning, Hammond, Notebaert and MacLeod (2014) demonstrated the role of the left dorsolateral prefrontal cortex in modulating attentional bias to threat. This left-right difference between studies is likely due to the type of stimuli used, since Miscovic and Schmidt (2010) used facial stimuli and Clarke, Browning et al. (2014) used words, that presumably require more processing in the left-dominated language centres (Zatorre, Meyer, Gjedde, & Evans, 1996). This distinction is further supported by Avram, Baltes, Miclea, and Miu's (2010) study of anxious individuals using the Stroop task and EEG data. These authors found attentional bias to threat using facial stimuli that corresponded with greater right frontal activation.

The left versus right hemisphere processing of emotional material was specifically examined by Engels et al. (2007). These authors also distinguished between anxious apprehensive individuals (i.e. worriers) and others characterised more by high anxious arousal (with corresponding low worry) in their processing of threat information, and concluded that biased processing of threat words in worriers involved left frontal regions whereas in the latter group, bias involved right inferior temporal regions. Interestingly, Engels et al. (2007) also examined the role of left dorsolateral prefrontal cortex regions in processing positive versus negative information and concluded that among worriers there are distinct regions responsible for processing threat (inferior frontal gyrus) versus positive information (dorsolateral prefrontal cortex).

The above evidence of frontal involvement in attentional bias to threat is consistent with other findings of impaired control in anxious individuals (for example, when trying to disengage from threat). However, Eldar, Yankelevitch, Lamy, and Bar-Haim's (2010) ERP data demonstrated higher C1 amplitude (at around 80 ms stimulus onset asynchrony; SOA) in the anxious versus non-anxious group when responding to threat stimuli, suggestive of a biased early engagement with threat. This biased engagement is more likely mediated by areas other than just the prefrontal cortex. However, the two processes are probably interlinked and operate alongside each other in contributing to threat-related attentional bias. Indeed, Pessoa (2005) has argued that there is considerable interdependence between stimulus-driven attention to threat and top-down control, because demonstrating attention to threat actually depends upon the availability of top-down attentional resources (i.e. high levels of perceptual load can block the supposedly ‘automatic’ attention to threat distracters).

An alternative form of interaction between bottom-up and top-down processes in attentional bias was suggested by Hirsch and Mathews (2012) in their cognitive model of worry. Hirsch and Mathews (2012) reviewed evidence that trait worriers have an increased tendency to engage with threatening information than non-worriers and also have greater difficulty disengaging from it, either due to impaired attentional control ability or a goal-driven focus on threat. Such a goal-driven focus could arise from mistaken positive meta-cognitive beliefs about the benefits of worrying (Wells, 1995), for example, that worry is helpful in avoiding threats or solving problems. Alternatively, the difficulty in disengaging could reflect a reduced ability to redirect attention away to neutral or positive topics (Derryberry & Reed, 2002). It is possible that this poor attentional control is actually a product of the worry process itself (Hayes et al., 2008, Stefanopoulou et al., 2014), which takes up attentional control resources in high worriers and those with a diagnosis of GAD, thereby reducing the attentional resources available to switch to non-worry topics (Klein & Boals, 2001). Leigh and Hirsch (2011) found substantially reduced attentional control resources in high worriers when they were worrying in their usual quasi-verbal form rather than thinking in the form of mental imagery. Thus the verbal nature of worry itself (Hirsch, Hayes, Mathews, Perman, & Borkovec, 2012) may be partly responsible for impaired control and to the special difficulty experienced when high worriers try to switch focus to other topics.

It is important to add that attentional bias to threat is just one potential information-processing bias associated with worry. In addition, it has been previously demonstrated that worry is maintained in individuals with a diagnosis of GAD and among high worriers by their interpretation bias to threat (Hayes et al., 2010, Hirsch et al., 2009). In fact, this interpretation bias is another key component of Hirsch and Mathews' (2012) cognitive model of worry described above. In their model, they highlight the fact that worriers will interpret ambiguous information in a threatening manner and as such the product of this interpretation will be threatening in nature and these thoughts will then form a focus for attentional bias. Therefore, although attentional bias may be influenced by potential attentional control deficits and/or positive beliefs about worrying (see above), it may also simply be that worriers and those with GAD are interpreting ambiguous stimuli as threatening more of the time than non-worriers and are subsequently directing their focus on this ‘threat’ material (Hirsch & Mathews, 2012). Although information-processing biases do not exist in isolation but will interact in important ways (Hirsch, Clark, & Mathews, 2006), for the purpose of the current review, bias of attention to threat will be the point of focus, in order to establish the hypothesised association with worry.

Attentional biases to threat have been found across a range of anxious populations, both clinical and non-clinical and have been implicated in the maintenance of anxiety disorders. Importantly, GAD, with its hallmark feature of worry, has been shown to often temporally precede other affective disorders and influence the subsequent course and outcomes of these secondary conditions (Kessler, Keller, & Wittchen, 2001). Therefore, despite information processing biases being suggested to be important across anxiety disorders, it would be helpful to systematically review the evidence of attentional bias to threat in individuals with GAD and those with high levels of worry (without a diagnosis of GAD), since this may help inform our theoretical understanding of worry and GAD, as well as helping to guide future treatment innovation.

This paper aims to systematically review experimental studies of attentional bias in individuals with high levels of trait worry only and as such, it excludes studies that recruited participants using measures of trait anxiety more broadly. Further, given that GAD has ‘excessive worry’ as a core criterion (American Psychiatric Association, 2013) this review will also include studies that have investigated samples of individuals with a diagnosis of GAD. It is widely believed that most organisms have evolved preferential processing of threatening versus neutral information so as to promote survival (Ohman, 1986). Hence simply identifying a bias favouring threat versus neutral information within GAD or high worry (GAD/worry) samples would not add to the current knowledge base. Rather, this review aims to determine if high trait worry samples have an exaggerated attentional bias to threat compared to healthy controls and/or other clinical populations. This bias will be identified by experimental paradigms rather than through associations between reported individual differences, as experimental paradigms provide a more objective measure of attentional bias than self-report measures and have established attentional biases in other anxious populations (Bar-Haim et al., 2007). Further aims are to then establish which aspects of empirical evidence moderate the GAD/worry attentional bias effect, asking the following specific questions:

  • Do adults with GAD/worry (high trait worry) demonstrate increased attentional bias to threat compared to healthy controls?

  • Does an attentional bias to threat among GAD/worry groups differ from those seen in other clinical groups?

  • Is a similar attentional bias to threat among GAD/worry groups found across all experimental paradigms?

  • Does any bias in GAD/worry groups vary at different stimuli exposure times, indicating predominantly bottom-up or top-down processing?

  • Is bias in GAD/worry groups found with all types of threat stimuli (verbal-linguistic vs imagery; threat vs other emotional content valence; general threat vs specific threat)?

Section snippets

Search strategy

The current systematic review conducted literature searches on PsycINFO and Medline in week 4 of October 2014, using the following search terms: attention; attentional bias; worry; worrie*; generalized anxiety disorder; anxiety disorders; anxiety. Mapped terms and auto exploding of search results were used where available. The terms were then combined and studies limited to studies written in English and studying human samples only. Articles were retained if they fulfilled the following

Results

The search process can be seen in the PRISM Flowchart in Fig. 1. The searches resulted in 1933 articles in PsychINFO and 2471 articles in Medline (Medline search terms differed due to different abilities to ‘explode’ search terms between the two search databases). Therefore, the total number of articles returned was 4404, which was reduced to 3610 after duplicates were removed. Titles were then screened for relevance and 303 papers were retained for further inspection. Papers were screened

Discussion

The current systematic review examined the empirical evidence of attentional bias to threat in those with a diagnosis of GAD and high trait worriers (without GAD), compared to healthy controls and other clinical populations. The key conclusion is that there is evidence within the studies reviewed here to suggest that individuals with a clinical diagnosis of GAD show attentional bias favouring threatening information relative to non-clinical populations, as this evidence was found in over

Conclusions

In conclusion, this review updated the evidence of attentional processing biases to threatening stimuli among people with a clinical diagnosis of GAD and individuals with high levels of trait worry. Although there was a dearth of studies investigating non-diagnosed high worriers, the majority of studies found a significant attentional bias to threat among both GAD and high worry groups compared to healthy controls or other clinical/anxious samples. This positive finding was established across

Acknowledgments

The last author received salary support from the National Institute for Health Research (NIHR), Mental Health Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and Kings College London.

References (105)

  • S. Eldar et al.

    Enhanced neural reactivity and selective attention to threat in anxiety

    Biological Psychiatry

    (2010)
  • C.R. Hirsch et al.

    Imagery and interpretations in social phobia: Support for the combined cognitive biases hypothesis

    Behavior Therapy

    (2006)
  • C.R. Hirsch et al.

    A cognitive model of pathological worry

    Behaviour Research and Therapy

    (2012)
  • A. Jarde et al.

    Q-Coh: A tool to screen the methodological quality of cohort studies in systematic reviews and meta-analyses

    International Journal of Clinical and Health Psychology

    (2013)
  • R.C. Kessler et al.

    The epidemiology of generalized anxiety disorder

    Psychiatric Clinics of North America

    (2001)
  • E. Leigh et al.

    Worry in imagery and verbal form: Effect on residual working memory capacity

    Behaviour Research and Therapy

    (2011)
  • C. MacLeod et al.

    Biased cognitive operations in anxiety: Accessibility of information or assignment processing priorities?

    Behaviour Research and Therapy

    (1991)
  • M. Martin et al.

    Does anxiety lead to selective processing of threat-related information?

    Behaviour Research and Therapy

    (1991)
  • A. Mathews et al.

    Selective processing of threat cues in anxiety

    Behaviour Research and Therapy

    (1985)
  • A. Mathews et al.

    Effect of psychological treatment on cognitive bias in generalized anxiety disorder

    Behaviour Research and Therapy

    (1995)
  • T.J. Meyer et al.

    Development and validation of the Penn State worry questionnaire

    Behaviour Research and Therapy

    (1990)
  • K. Mogg et al.

    A cognitive-motivational analysis of anxiety

    Behaviour Research and Therapy

    (1998)
  • K. Mogg et al.

    A follow-up study of cognitive bias in generalized anxiety disorder

    Behaviour Research and Therapy

    (1995)
  • K. Mogg et al.

    Effects of threat cues on attentional shifting, disengagement and response slowing in anxious individuals

    Behaviour Research and Therapy

    (2008)
  • K. Mogg et al.

    Biased cognitive operations in anxiety: Artefact, processing priorities or attentional search?

    Behaviour Research and Therapy

    (1991)
  • K. Mogg et al.

    Selective processing of threat cues in anxiety states: A replication

    Behaviour Research and Therapy

    (1989)
  • M.G. Newman et al.

    Preliminary reliability and validity of the generalized anxiety disorder questionnaire-IV: A revised self-report diagnostic measure of generalized anxiety disorder

    Behavior Therapy

    (2002)
  • L. Pessoa

    To what extent are emotional visual stimuli processed without attention and awareness?

    Current Opinion in Neurobiology

    (2005)
  • A.M. Ruscio

    Delimiting the boundaries of generalized anxiety disorder: Differentiating high worriers with and without GAD

    Journal of Anxiety Disorders

    (2002)
  • D. Sheehan et al.

    The validity of the Mini International Neuropsychiatric Interview (MINI) according to the SCID-P and its reliability

    European Psychiatry

    (1997)
  • C. Stokes et al.

    Engaging in imagery versus verbal processing of worry: Impact on negative intrusions in high worriers

    Behaviour Research and Therapy

    (2010)
  • M. Williams et al.

    Verbal worry facilitates attention to threat in high-worriers

    Journal of Behavior Therapy and Experimental Psychiatry

    (2014)
  • M. Albu

    Automatic and intentional inhibition in patients with generalized anxiety disorder

    Cognition, Brain, & Behavior

    (2008)
  • American Psychiatric Association

    Diagnostic and statistical manual of mental disorders: DSM-5

    (2013)
  • J. Avram et al.

    Frontal EEG activation asymmetry reflects cognitive biases in anxiety: Evidence from an emotional face Stroop task

    Applied Psychophysiological Biofeedback

    (2010)
  • I.A. Bargh

    Conditional automaticity: Varieties of automatic influence in social perception and cognition

  • Y. Bar-Haim et al.

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

    Psychological Bulletin

    (2007)
  • A.T. Beck et al.

    Anxiety disorders and phobias: A cognitive perspective

    (1985)
  • K.S. Blair et al.

    A cognitive neuroscience approach to generalized anxiety disorder and social phobia

    Emotion Review

    (2012)
  • B.P. Bradley et al.

    Selective processing of negative information: Effects of clinical anxiety, concurrent depression, and awareness

    Journal of Abnormal Psychology

    (1995)
  • B.P. Bradley et al.

    Attentional bias for emotional faces in generalized anxiety disorder

    British Journal of Clinical Psychology

    (1999)
  • D. Broadbent et al.

    Anxiety and attentional bias: State and trait

    Cognition & Emotion

    (1988)
  • J.F. Brosschot et al.

    Daily worrying and somatic health complaints: Testing the effectiveness of a simple worry reduction intervention

    Psychology & Health

    (2006)
  • M.G. Calvo et al.

    Time course of attentional bias to emotional scenes in anxiety: Gaze direction and duration

    Cognition & Emotion

    (2005)
  • J. Chen et al.

    Differential attentional bias in generalized anxiety disorder and panic disorder

    Neuropsychiatric Disease and Treatment

    (2013)
  • J.M. Cisler et al.

    Phenomenological characteristics of attentional biases towards threat: A critical review

    Cognitive Therapy and Research

    (2009)
  • P.J.F. Clarke et al.

    Is selective attention in anxiety characterised by biased attentional engagement with or disengagement from threat: Evidence from a colour-naming paradigm

    Journal of Experimental Psychopathology

    (2014)
  • P.J.F. Clarke et al.

    Assessing the role of spatial engagement and disengagement of attention in anxiety-linked attentional bias: A critique of current paradigms and suggestions for future research directions

    Anxiety, Stress, and Coping

    (2013)
  • M. Davis et al.

    The amygdala: Vigilance and emotion

    Molecular Psychiatry

    (2001)
  • D. Derryberry et al.

    Anxiety-related attentional biases and their regulation by attentional control

    Journal of Abnormal Psychology

    (2002)
  • Cited by (84)

    View all citing articles on Scopus
    View full text