CoChat
Sign Up
Cited by 215Anxiety, Depression, Psychometrics, Treatment, Cognitive ProcessesNeural and Behavioral Psychology StudiesVisual Attention and Saliency Detection

Attention bias modification (ABM): review of effects of multisession ABM training on anxiety and threat-related attention in high-anxious individuals

Open access
Mogg, Karin, Waters, Allison M., Bradley, Brendan P.·2017·Clinical Psychological Science
DOI 10.1177/2167702617696359PMC PMC5513441
Open this paper in CoChat — free accountOpen PDF

Abstract

Attention bias modification (ABM) aims to reduce anxiety by reducing attention bias (AB) to threat; however, effects on anxiety and AB are variable. This review examines 34 studies assessing effects of multisession-ABM on both anxiety and AB in high-anxious individuals. Methods include ABM-threat-avoidance (promoting attention-orienting away from threat), ABM-positive-search (promoting explicit, goal-directed attention-search for positive/nonthreat targets among negative/threat distractors), and comparison conditions (e.g., control-attention training combining threat-cue exposure and attention-task practice without AB-modification). Findings indicate anxiety reduction often occurs during both ABM-threat-avoidance and control-attention training; anxiety reduction is not consistently accompanied by AB reduction; anxious individuals often show no pretraining AB in orienting toward threat; and ABM-positive-search training appears promising in reducing anxiety. Methodological and theoretical issues are discussed concerning ABM paradigms, comparison conditions, and AB assessment. ABM methods combining explicit goal-directed attention-search for nonthreat/positive information and effortful threat-distractor inhibition (promoting top-down cognitive control during threat-cue exposure) warrant further evaluation.

AI snapshot

Written from the paper’s full text.

TL;DR

This qualitative review of 34 multisession RCTs in high-anxious individuals finds that ABM-threat-avoidance training has inconsistent anxiolytic effects relative to control-attention training, and anxiety reduction often occurs without reduction in preexisting threat-related attention bias. ABM-positive-search training appears promising for reducing anxiety, but evidence is limited to two included studies.

Key findings

  • In about two thirds of studies, ABM-threat-avoidance training was not more effective than comparison conditions in reducing anxiety; 8 studies found it superior to control-attention training, all experimenter-controlled, while 23 studies found no superiority (15 experimenter-controlled, 7 home-based, 1 assessing both settings).
  • Anxiety symptoms often reduced during both ABM-threat-avoidance and control-attention training, sometimes with moderate-to-large effects and often without change in AB to threat.
  • Most studies using standard visual-probe-task scores did not find pretraining AB in orienting toward threat in anxious individuals; anxiety reduction was not consistently accompanied by AB reduction.
  • In two ABM-positive-search studies, anxiety reduction was greater than control or waitlist: 50% of completers versus 8% in control training were free of their principal anxiety disorder in Waters et al. (2013), and 42% versus 8% of waitlist control in Waters et al. (2015).

Methods

  • Qualitative review; searched PsycInfo and Medline and checked reference lists for peer-reviewed English-language studies, identifying 34 studies published between 2008 and end of 2016 (33 publications, one describing two studies).
  • Inclusion criteria: multisession ABM training aimed at modifying threat-related AB and anxiety; threat and nonthreat stimuli; high-anxious subclinical or clinical participants; RCT design; pre-post assessment of anxiety symptoms and AB. Review articles and studies combining ABM with interpretive bias modification were excluded.
  • The review evaluated three predictions: greater anxiety reduction for ABM than comparison conditions, corresponding reduction in AB to threat, and pretraining AB to threat in anxious individuals; it distinguished ABM-threat-avoidance (32 studies) from ABM-positive-search (2 studies) and examined control conditions and AB measures.

Limitations

  • The paper states that a limitation is the small number of ABM-positive-search studies that have assessed effects on both AB and anxiety in anxious individuals.
  • The paper states that null pretraining AB findings must be interpreted cautiously because sample sizes tend to be modest, ABs are inferred indirectly from RT data, and RT-based AB measures have low reliability; it also notes dissemination bias, including publication bias and outcome-reporting bias.
  • The paper states that most studies, but not all, lacked non-attention-training comparison conditions, and that interpretation of AB change is complicated by divergent AB measures and by many studies assessing AB with the same or similar probe-based task used for training, which may reflect task-specific rules rather than generalized change in AB to threat.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1177/2167702617696359

Cites

100 of 100 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 151 cited sources checked · 0 supported · 0 flagged

  • Findingcitation

    Two studies unexpectedly found greater anxiety reduction during CON-attention-training than during ABM-threat-avoidance training in PTSD.

    “In addition, two studies unexpectedly found greater anxiety reduction during CON-attention-training than during ABM-threat-avoidance training in posttraumatic stress disorder (PTSD) (Badura-Brack et al., 2015).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Badura-Brack et al., 2015

      Not checked in the free run.

    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.
  • Findingown data

    In most studies, ABM-threat-avoidance training was not more effective than CON-attention training in reducing anxiety between pre- and posttraining.

    “However, in most studies, ABM-threat-avoidance training was not more effective than CON-attention training in reducing anxiety between pre- and posttraining.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.
  • Findingcitation

    Two studies reported that anxiety reduction during ABM-threat-avoidance training was accompanied by reduction in within-session AB variability rather than stable change in standard AB scores between pre- and posttraining.

    “Two other studies, which also used the visual-probe task to both assess and modify AB, reported that anxiety reduction during ABM-threat-avoidance training was accompanied by reduction in within-session AB variability (change between beginning and end of each training session) rather than stable change in standard AB scores between pre- and posttraining”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.
  • Findingreasoning

    Anxiety reduction during ABM training is not consistently accompanied by reduction in attention bias to threat.

    “anxiety reduction is not consistently accompanied by AB reduction;”
    • Also stated in: § Abstract · hedged
      “Second, anxiety reduction may occur during ABM training without AB reduction.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.
  • Findingcitation

    About a third to a half of children with anxiety disorders were free of their principal anxiety disorder after ABM-positive-search training.

    “about a third to a half of children with anxiety disorders were free of their principal anxiety disorder after ABM-positive-search training (50% of those completing ABM versus 8% in control training in Waters et al., 2013, and 42% of those completing ABM versus 8% of waitlist control in Waters et al., 2015).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Waters et al., 2013

      Not checked in the free run.

    • cites Waters et al., 2015

      Not checked in the free run.

    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.
  • May combine two claims. 53
  • The quote only partly states this claim. 1
  • Classified as probably not a substantial claim. 16

  • May combine two claims.

    Findingcitation

    Three studies that found superior anxiolytic effects of ABM-threat-avoidance assessed AB on spatial-cuing tasks and none showed pretraining AB to threat.

    “Three studies, which found superior anxiolytic effects of ABM-threat-avoidance training (relative to CON-attention training), assessed AB on spatial-cuing tasks (Amir, Beard, Taylor et al., 2009; Bar-Haim et al., 2011; Liang & Hsu, 2016). Although none of these studies showed pretraining AB to threat”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Amir, Beard, Taylor et al., 2009

      Not checked in the free run.

    • cites Bar-Haim et al., 2011

      Not checked in the free run.

    • cites Liang & Hsu, 2016

      Not checked in the free run.

    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.

  • May combine two claims.

    Findingown data

    The review identified 34 studies published between 2008 and the end of 2016 that met the inclusion criteria; 32 used ABM-threat-avoidance training and two used ABM-positive-search training.

    “Consequently, we identified 34 studies published between 2008 and the end of 2016 that met the specifications for this review (33 publications, with one describing two studies; indicated in “References”).2 Of these, 32 studies used ABM-threat-avoidance training and two used ABM-positive-search training.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingcitation

    Eight studies indicated that standard ABM-threat-avoidance training was superior to CON-attention training in reducing anxiety, all administered under experimenter-controlled conditions.

    “Specifically, eight studies indicated that standard ABM-threat-avoidance training was superior to CON-attention training in reducing anxiety, all of which administered training under experimenter-controlled conditions”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.

  • May combine two claims.

    Findingcitation

    Nonstandard AB scores on the visual-probe task suggested pretraining AB to threat, which reduced from pre- to posttraining during ABM with fear activation and CON-only training but not during ABM-only training.

    “Nonstandard AB scores on the visual-probe task suggested pretraining AB to threat (slower RTs on threat-present than threat-absent trials), which reduced from pre- to posttraining during ABM with fear activation and CON-only training but not during ABM-only training.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.

  • May combine two claims.

    Findingcitation

    In Eldar et al. (2012), AB reduction was found for threat stimuli used during ABM training and did not generalize to new threat stimuli.

    “AB reduction was found for threat stimuli used during ABM training, which did not generalize to new threat stimuli.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Eldar et al., 2012

      Not checked in the free run.

    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.

  • May combine two claims.

    Findingcitation

    In Amir, Beard, Burns et al. (2009), anxious individuals showed no AB before training and showed threat avoidance after ABM training but not after CON-attention training.

    “In Amir, Beard, Burns et al. (2009), anxious individuals showed no AB before training and showed threat avoidance after ABM training but not after CON-attention training.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Amir, Beard, Burns et al., 2009

      Not checked in the free run.

    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.

  • The quote only partly states this claim.

    Findingown data

    The lack of superiority of ABM-threat-avoidance over CON-attention training in reducing anxiety was found in 23 studies: 15 experimenter-controlled, 7 home-based, and 1 assessing training in both settings.

    “This outcome was found in 23 studies: 15 experimenter-controlled (laboratory-, clinic-, or school-based settings), 7 home-based training, and 1 assessing training in both settings”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.

  • May combine two claims.

    Findingcitation

    Heeren et al. (2012) found that socially anxious adults showed pretraining AB for angry faces on the visual-probe task, which reduced during ABM-threat-avoidance training but did not change during CON or inverse-ABM training.

    “By contrast, in Heeren et al. (2012), socially anxious adults showed pretraining AB for angry faces on the visual-probe task, which reduced during ABM-threat-avoidance training (resulting in posttraining AB away from angry relative to happy faces) but did not change during CON or inverse-ABM training”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Heeren et al., 2012

      Not checked in the free run.

    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.

  • May combine two claims.

    Findingcitation

    Khanna et al. (2016) found pretraining AB for combat-related threat words reduced during both ABM and CON training, so this AB was no longer evident posttraining.

    “Khanna et al. (2016) used different attentional tasks to assess and modify AB in PTSD (modified Stroop and visual-probe tasks, respectively) and found pretraining AB for combat-related threat words (indexed by threat-distractor-interference effect on RT) reduced during both ABM and CON training, so this AB was no longer evident posttraining.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Khanna et al., 2016

      Not checked in the free run.

    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.

  • May combine two claims.

    Findingcitation

    Three studies that found superior anxiolytic effects of ABM-threat-avoidance reported RT changes from pre- to posttraining, including faster RTs to probes appearing in a different location to threat cues following ABM training.

    “they reported RT changes from pre- to posttraining, which included faster RTs to probes appearing in a different location to threat cues (i.e., faster RTs on invalid-threat trials) following ABM training.”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Amir, Beard, Taylor et al., 2009

      Not checked in the free run.

    • cites Bar-Haim et al., 2011

      Not checked in the free run.

    • cites Liang & Hsu, 2016

      Not checked in the free run.

    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.

  • May combine two claims.

    Findingcitation

    One study showed moderate to large improvements in clinical outcome measures across training conditions, and anxiety reduction was unexpectedly greater for inverse-ABM training.

    “One of these studies showed moderate to large improvements in clinical outcome measures across training conditions; moreover, anxiety reduction was unexpectedly greater for inverse-ABM training (Boettcher et al., 2013).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Boettcher et al., 2013

      Not checked in the free run.

    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.

  • May combine two claims.

    Findingcitation

    In Enock et al. (2014), Khanna et al. (2016), and Shechner et al. (2014), both anxiety and AB measures reduced during ABM and CON-attention training.

    “Thus, both anxiety and AB measures reduced during ABM and CON-attention training in these studies (Enock et al., 2014; Khanna et al., 2016; Shechner et al., 2014).”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Enock et al., 2014

      Not checked in the free run.

    • cites Khanna et al., 2016

      Not checked in the free run.

    • cites Shechner et al., 2014

      Not checked in the free run.

    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.

  • May combine two claims.

    Findingown data

    Studies that found greater anxiety reduction during ABM than CON commonly reported changes in performance on AB tasks from pre- to posttraining, but the nature of these changes varied across studies.

    “These studies also commonly reported changes in performance on AB tasks from pre- to posttraining. However, the nature of these changes varied across studies.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.

  • May combine two claims.

    Findingcitation

    Although anxiety reduction was greater in ABM with fear activation than CON-only training, these conditions had a similar effect on AB scores.

    “Hence, although anxiety reduction was greater in ABM with fear activation than CON-only training, these conditions had a similar effect on AB scores.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Kuckertz, Gildebrant et al., 2014

      Not checked in the free run.

    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.

  • May combine two claims.

    Findingcitation

    In Heeren et al. (2012), there was similar reduction in self-reported anxiety between pre- and posttraining during both ABM-threat-avoidance and CON-attention training but not during inverse-ABM training, with further anxiety reduction between posttraining and 2-week follow-up for ABM-threat-avoidance training.

    “In Heeren et al. (2012), there was similar reduction in self-reported anxiety between pre- and posttraining during both ABM-threat-avoidance and CON-attention training but not during inverse-ABM training, with further anxiety reduction between posttraining and 2-week follow-up for ABM-threat-avoidance training.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Heeren et al., 2012

      Not checked in the free run.

    • Rests on an assumption: The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.

  • May combine two claims.

    Findingcitation

    Yao et al. (2015) found ABM-threat-avoidance training selectively modified AB to threat but not anxiety.

    “In one study, ABM-threat-avoidance training selectively modified AB to threat but not anxiety (Yao et al., 2015).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Yao et al., 2015

      Not checked in the free run.

  • May combine two claims.

    Findingcitation

    Clerkin et al. (2016) found reductions in AB variability and anxiety symptoms (but not standard AB scores) in socially anxious alcohol-dependent individuals, which occurred during both ABM-threat-avoidance and CON-attention training.

    “Clerkin et al. (2016) found reductions in AB variability and anxiety symptoms (but not standard AB scores) in socially anxious alcohol-dependent individuals, which occurred during both ABM-threat-avoidance and CON-attention training.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Clerkin et al., 2016

      Not checked in the free run.

  • May combine two claims.

    Findingcitation

    In Badura-Brack et al.'s studies, there was greater reduction in PTSD symptoms and AB variability (but not standard AB scores) during CON-attention training compared with ABM-threat-avoidance training.

    “In Badura-Brack et al.’s studies, there was greater reduction in PTSD symptoms and AB variability (but not standard AB scores) during CON-attention training compared with ABM-threat-avoidance training”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Badura-Brack et al., 2015

      Not checked in the free run.

  • May combine two claims.

    Findingcitation

    Amir, Beard, Taylor et al. (2009), Bar-Haim et al. (2011), Heeren, Mogoase, McNally et al. (2015), and Liang & Hsu (2016) found no RT difference between invalid-threat and invalid-nonthreat trials before training.

    “or no RT difference between invalid-threat and invalid-nonthreat trials (Amir, Beard, Taylor et al., 2009; Bar-Haim et al., 2011; Heeren, Mogoase, McNally et al., 2015; Liang & Hsu, 2016) before training”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Amir, Beard, Taylor et al., 2009

      Not checked in the free run.

    • cites Bar-Haim et al., 2011

      Not checked in the free run.

    • cites Liang & Hsu, 2016

      Not checked in the free run.

  • May combine two claims.

    Findingnone given

    In about two thirds of studies, ABM-threat-avoidance training was not more effective in reducing anxiety than comparison conditions, which in most studies was control-attention training.

    “in about two thirds of studies, ABM-threat-avoidance training was not more effective in reducing anxiety than comparison conditions, which in most studies was control-attention training”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingcitation

    Other studies reported reduction in within-session AB variability or changes in RT indices on spatial-cuing tasks that may reflect reduction in general threat-related RT slowing and/or faster orienting away from threat.

    “other studies reported reduction in within-session AB variability (Kuckertz, Amir et al., 2014; Li et al., 2008) or changes in RT indices on spatial-cuing tasks that may reflect reduction in general threat-related RT slowing and/or faster orienting away from threat”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Kuckertz, Amir et al., 2014

      Not checked in the free run.

    • cites Li et al., 2008

      Not checked in the free run.

  • May combine two claims.

    Findingcitation

    Kuckertz, Gildebrandt et al. (2014) found ABM combined with fear activation was more effective than CON-attention training without fear activation in reducing anxiety but had a similar effect in reducing nonstandard AB scores.

    “In one study, ABM combined with fear activation was more effective than CON-attention training without fear activation in reducing anxiety (which may have been due to fear activation alone or its combination with ABM) but had a similar effect in reducing nonstandard AB scores (Kuckertz, Gildebrandt et al., 2014).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Kuckertz, Gildebrandt et al., 2014

      Not checked in the free run.

  • May combine two claims.

    Findingcitation

    Price et al. (2016) meta-analysis indicated that 29% of anxious individuals receiving ABM-threat-avoidance training, and 24% of those receiving CON-attention training, showed improvement in anxiety (indexed by a decrease in social anxiety scores of 30% or more at posttraining relative to pretraining).

    “Price et al.’s (2016) meta-analysis of questionnaire data selected from 11 studies, which indicated that 29% of anxious individuals receiving ABM-threat-avoidance training, and 24% of those receiving CON-attention training, showed improvement in anxiety (indexed by a decrease in social anxiety scores of 30% or more at posttraining relative to pretraining)”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Price et al., 2016

      Not checked in the free run.

  • May combine two claims.

    Findingcitation

    Anxious individuals do not consistently show AB in orienting toward threat and sometimes show threat-avoidance or no bias.

    “other evidence also indicates that anxious individuals do not consistently show AB in orienting toward threat and sometimes show threat-avoidance or no bias”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Cisler & Koster, 2010

      Not checked in the free run.

    • cites Dudeney et al., 2015

      Not checked in the free run.

    • cites Van Bockstaele et al., 2014

      Not checked in the free run.

  • May combine two claims.

    Findingcitation

    In three studies, ABM-positive-search training was delivered at home to children with anxiety disorders, and about one-third to a half of them were free of their primary anxiety disorder after ABM training.

    “In the latter three studies, ABM-positive-search training was delivered at home to children with anxiety disorders, and about one-third to a half of them were free of their primary anxiety disorder after ABM training.”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Waters et al., 2013

      Not checked in the free run.

    • cites Waters et al., 2015

      Not checked in the free run.

    • cites Waters et al., 2016

      Not checked in the free run.

  • May combine two claims.

    Findingcitation

    In many studies, anxiety symptoms reduced to a similar degree during both ABM-threat-avoidance and CON-attention training, often without reduction in AB to threat.

    “in many studies, anxiety symptoms reduced to a similar degree during both ABM-threat-avoidance and CON-attention training (often without reduction in AB to threat)”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingcitation

    Only two studies have so far assessed effects of multisession ABM-positive-search training on both anxiety and AB in the same sample of anxious individuals, and these showed anxiolytic effects relative to comparison conditions.

    “To our knowledge, only two studies have so far assessed effects of multisession ABM-positive-search training on both anxiety and AB in the same sample of anxious individuals, and these showed anxiolytic effects relative to comparison conditions (Waters et al., 2013, 2015).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Waters et al., 2013

      Not checked in the free run.

    • cites Waters et al., 2015

      Not checked in the free run.

  • May combine two claims.

    Findingnone given

    Anxiety symptoms often reduced during both ABM-threat-avoidance and control-attention training (sometimes with moderate-to-large effect sizes, and often without change in AB to threat).

    “Anxiety symptoms often reduced during both ABM-threat-avoidance and control-attention training (sometimes with moderate-to-large effect sizes, and often without change in AB to threat).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingcitation

    Inclusion of additional ABM studies in this review does not alter conclusions from previous reviews that ABM-threat-avoidance training has an inconsistent anxiolytic effect, relative to control-attention training, and that any additional benefit of this ABM method is largely restricted to experimenter-controlled settings.

    “Inclusion of additional ABM studies in this review does not alter conclusions from previous reviews that ABM-threat-avoidance training has an inconsistent anxiolytic effect, relative to control-attention training, and that any additional benefit of this ABM method is largely restricted to experimenter-controlled settings”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Cristea et al., 2015

      Not checked in the free run.

    • cites MacLeod & Clarke, 2015

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Findingnone given

    The clinical effectiveness of multisession ABM-positive-search training remains to be established.

    “However, the clinical effectiveness of other ABM methods, such as multisession ABM-positive-search training, remains to be established.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingnone given

    Such comparisons do not control for expectancy effects or provide information about specific features of ABM training that may convey therapeutic benefit.

    “However, such comparisons do not control for expectancy effects or provide information about specific features of ABM training that may convey therapeutic benefit”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Outcome measures of anxiety vary considerably across studies, which poses a problem for meta-analyses.

    “Outcome measures of anxiety vary considerably across studies, which poses a problem for meta-analyses (Cristea et al., 2015)”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Cristea et al., 2015

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    Hazen et al. (2009) noted that conclusions were limited by small sample sizes at posttraining (n = 9 and n = 8 in ABM and CON groups, respectively).

    “but noted that conclusions were limited by small sample sizes at posttraining (n = 9 and n = 8 in ABM and CON groups, respectively).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Hazen et al., 2009

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    The control-attention-training condition is designed to control for the effects of multisession practice on the visual-probe task and threat-cue exposure but without modifying AB to threat.

    “This control-attention-training (CON-attention-training) condition is designed to control for the effects of multisession practice on the visual-probe task and threat-cue exposure but without modifying AB to threat.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    Top-down attention control processes may also be modified by control training conditions, thereby contributing to evidence of anxiety reductions in both active ABM and control conditions.

    “Top-down attention control processes may also be modified by control training conditions, thereby contributing to evidence of anxiety reductions in both active ABM as well as control conditions”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    ABM-threat-avoidance training tends to be more effective in reducing anxiety relative to control conditions when delivered in laboratory settings but not when delivered at home.

    “Moreover, the efficacy of ABM-threat-avoidance training depends on its mode of delivery; that is, it tends to be more effective in reducing anxiety relative to control conditions when delivered in laboratory settings but not when delivered at home”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Cristea et al., 2015

      Not checked in the free run.

    • cites MacLeod & Clarke, 2015

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Replication failures and meta-analyses indicate inconsistent effects of ABM-threat-avoidance training on anxiety and question its clinical utility.

    “Although early studies using ABM-threat-avoidance training with clinically anxious individuals were promising (e.g., Amir, Beard, Burns et al., 2009; Amir, Beard, Taylor et al., 2009), replication failures and meta-analyses indicate inconsistent effects on anxiety and question its clinical utility”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Beard, Sawyer, & Hofmann, 2012

      Not checked in the free run.

    • cites Cristea, Kok, & Cuijpers, 2015

      Not checked in the free run.

    • cites Emmelkamp, 2012

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Of 22 multisession studies in MacLeod and Clarke (2015), 45% indicated change in both AB and anxiety during ABM training relative to control conditions, whereas the others showed no difference in change in either measure.

    “Of 22 multisession studies, 45% indicated change in both AB and anxiety during ABM training relative to control conditions, whereas the others showed no difference in change in either measure.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites MacLeod and Clarke, 2015

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    AB measures in orienting to threat and in threat-distractor interference are typically uncorrelated, consistent with different underlying processes.

    “These AB measures are typically uncorrelated, consistent with different underlying processes (Cisler, Bacon, & Williams, 2009).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Cisler, Bacon, & Williams, 2009

      Not checked in the free run.

  • May combine two claims.

    Premisenone given

    Meta-analyses commonly treat AB as a unitary construct and do not discriminate between differing ABM training methods and AB indices.

    “For example, meta-analyses commonly treat AB as a unitary construct and do not discriminate between differing ABM training methods and AB indices.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    AB reduction may be a consequence of improved attention control and/or reduced anxiety rather than being a cause of anxiety reduction.

    “If so, AB reduction may be a consequence of improved attention control and/or reduced anxiety (rather than being a cause of anxiety reduction).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    The task-specific rule applies to both visual-probe and spatial-cuing tasks, so switching between these tasks for assessment and training would not eliminate this problem.

    “This task-specific rule applies to both visual-probe and spatial-cuing tasks (i.e., learning this rule would have a similar effect on RTs on threat-incongruent trials on the visual-probe task and invalid-threat trials on the spatial-cuing task), so switching between these tasks for assessment and training would not eliminate this problem.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    The combination of attention training and threat exposure may have an anxiolytic effect, whereas trying to modify the direction of AB in orienting away from threat conveyed no additional anxiolytic benefit in many studies.

    “Thus, the combination of attention training and threat exposure may have an anxiolytic effect (which is common to ABM-threat-avoidance and CON-attention-training), whereas trying to modify the direction of AB in orienting away from threat conveyed no additional anxiolytic benefit in many studies.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductcitation

    Null findings have to be interpreted with caution because sample sizes tend to be modest, ABs are inferred indirectly from RT data, and RT-based AB measures have low reliability.

    “Null findings have to be interpreted with caution (e.g., sample sizes tend to be modest; ABs are inferred indirectly from RT data; and RT-based AB measures have low reliability, which poses a challenge for ABM research; MacLeod & Clarke, 2015).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites MacLeod & Clarke, 2015

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    ABM-positive-search training seems promising in treating anxiety disorders in home settings, and its anxiolytic effect does not seem to depend on reducing AB in orienting to threat.

    “Thus, this ABM method seems promising in treating anxiety disorders in home settings, and its anxiolytic effect does not seem to depend on reducing AB in orienting to threat.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    Participants may implicitly or explicitly learn a task-specific rule during ABM-threat-avoidance training-that probes are more likely to appear in the opposite location to threat cues-which would speed RTs on threat-incongruent or invalid-threat trials and result in apparent AB reduction.

    “participants may implicitly or explicitly learn a task-specific rule during ABM-threat-avoidance training—that is, probes are more likely to appear in the opposite location to threat cues—which would speed RTs on threat-incongruent or invalid-threat trials and result in apparent AB reduction”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Most cognitive models of anxiety propose that both bottom-up and top-down processes play an important role in anxiety and attention responses to threat.

    “However, most cognitive models of anxiety propose that both bottom-up (automatic) and top-down (controlled) processes play an important role in anxiety and attention responses to threat”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Bar-Haim et al., 2007

      Not checked in the free run.

    • cites Beck & Clark, 1997

      Not checked in the free run.

    • cites Cisler & Koster, 2010

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    Multisession ABM-positive-search training seems promising in reducing anxiety, including when used in home settings, and its anxiolytic effect does not depend on reducing AB in orienting to threat.

    “Multisession ABM-positive-search training seems promising in reducing anxiety, including when used in home settings, and its anxiolytic effect does not depend on reducing AB in orienting to threat.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Bottom-up functions include automatic stimulus-salience evaluation, alerting, automatic interrupt/inhibition of ongoing goal-directed activity, and automatic attention orienting to threat.

    “Bottom-up functions include automatic stimulus-salience evaluation, alerting, automatic interrupt/inhibition of ongoing goal-directed activity, and automatic attention orienting to threat”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Corbetta & Shulman, 2002

      Not checked in the free run.

    • cites Pessoa & Adolphs, 2010

      Not checked in the free run.

    • cites Petersen & Posner, 2012

      Not checked in the free run.

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    A cognitive view of anxiety proposes that anxious individuals are characterized by an enhanced automatic AB toward threat, which plays a causal role in anxiety.

    “ABM-threat-avoidance training is supported by a cognitive view of anxiety, which proposes that anxious individuals are characterized by an enhanced automatic AB toward threat, which plays a causal role in anxiety (e.g., Williams, Watts, MacLeod, & Mathews, 1988).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Williams et al., 1988

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Top-down functions include goal-directed inhibitory control, cognitive flexibility, maintaining and updating goals in working memory, reason-based appraisal, and controlled attention orienting.

    “top-down functions include goal-directed inhibitory-control (which is a core component underlying a wide range of top-down control functions), cognitive flexibility, maintaining and updating goals in working memory, reason-based appraisal, and controlled attention orienting”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Diamond, 2013

      Not checked in the free run.

    • cites Miyake & Friedman, 2012

      Not checked in the free run.

    • cites Petersen & Posner, 2012

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisereasoning

    The cognitive view predicts that an intervention that reduces preexisting AB toward threat in anxious individuals and instead makes them threat avoidant should reduce anxiety.

    “Hence, it predicts that an intervention that reduces preexisting AB toward threat in anxious individuals and instead makes them threat avoidant should reduce anxiety.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The cognitive view assumes that low-anxious individuals are characterized by attentional avoidance of threat.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Top-down functions typically operate in a unified manner to support goal-directed activity.

    “These top-down functions typically operate in a unified manner to support goal-directed activity (Duncan, 2010).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Duncan, 2010

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisenone given

    Both excessive anxiety and ABs to threat reflect imbalance between bottom-up and top-down systems.

    “According to this framework, both excessive anxiety and ABs to threat reflect imbalance between these bottom-up and top-down systems.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisereasoning

    AB to threat is not a unitary construct, and its manifestations reflect differing combinations of processes.

    “Thus, AB to threat is not a unitary construct, and its manifestations reflect differing combinations of processes (Fig. 1).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S4 (based).

    Interpretationnone given

    Excessive anxiety may arise from a combination of inadequate top-down goal-directed inhibitory control and hyperreactivity to threat cues of the bottom-up threat-salience-evaluation mechanism.

    “Excessive anxiety may arise from a combination of inadequate top-down goal-directed inhibitory control and hyperreactivity to threat cues of the bottom-up threat-salience-evaluation mechanism”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Interpretationreasoning

    Such findings may be explained by a differing theoretical view, which proposes that dysfunction in multiple processes (including top-down attention control and threat-distractor inhibition) contributes to anxiety and differing manifestations of ABs and may be modified by both active ABM and control-attention training.

    “such findings may be explained by a differing theoretical view, which proposes that dysfunction in multiple processes (including top-down attention control and threat-distractor inhibition) contributes to anxiety and differing manifestations of ABs and may be modified by both active ABM and control-attention training”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationcitation

    Preliminary studies indicate that ABM-positive-search training is effective in reducing nonclinical and clinical anxiety in both experimenter-controlled and home-based settings.

    “However, preliminary studies indicate that this method is effective in reducing nonclinical and clinical anxiety in both experimenter-controlled and home-based settings and thus merits further evaluation (Dandeneau et al., 2007; De Voogd et al., 2014; Waters et al., 2013, 2015, 2016).”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Dandeneau et al., 2007

      Not checked in the free run.

    • cites De Voogd et al., 2014

      Not checked in the free run.

    • cites Waters et al., 2013, 2015, 2016

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Potential sources of bias in ABM research include publication bias and outcome-reporting bias.

    “Potential sources of bias include publication bias (“file-drawer” problem) and outcome-reporting bias (“cherry-picking”), which may influence the ratio of significant to null findings in published reports.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Cristea et al., 2015

      Not checked in the free run.

    • cites Song, Hooper, & Loke, 2013

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisecitation

    Recent models of anxiety predict that ABM methods that recruit multiple top-down goal-directed cognitive-control processes to oppose threat processing will be more effective in reducing anxiety than methods targeting one specific bias in direction of orienting toward threat.

    “recent models of anxiety predict that ABM methods that recruit multiple top-down goal-directed cognitive-control processes to oppose threat processing will be more effective in reducing anxiety than methods targeting one specific bias in direction of orienting toward threat”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Mogg & Bradley, 2016

      Not checked in the free run.

    • cites Waters & Craske, 2016

      Not checked in the free run.

  • May combine two claims.

    Interpretationcitation

    Attention biases may have both state- and trait-like characteristics and reflect an individual's current motivational/goal priorities that are dominant in a particular situation.

    “ABs may have both state- and trait-like characteristics and reflect an individual’s current motivational/goal priorities that are dominant in a particular situation (e.g., to automatically detect unattended threat; to maintain attention on threat to allow more detailed evaluation; to avoid attending to threat cues to minimize subjective discomfort; or to suppress threat-related AB to support task-”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Mogg & Bradley, 2016

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S4 (based).

    Interpretationnone given

    Anxiolytic effects of ABM may relate to improvements in attention control and threat-distractor inhibition, which are distinct from modifying the direction of AB in orienting away from threat.

    “Anxiolytic effects of ABM may relate to improvements in attention control and threat-distractor inhibition, which are distinct from modifying the direction of AB in orienting away from threat.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable).

    Premisecitation

    Establishing whether an intervention is effective in treating anxiety requires a systematic approach comparing it with a series of different conditions to examine specific issues.

    “Establishing whether an intervention is effective in treating anxiety requires a systematic approach comparing it with a series of different conditions to examine specific issues (Lohr, Lilienfeld, & Rosen, 2012).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Lohr, Lilienfeld, & Rosen, 2012

      Not checked in the free run.

  • May combine two claims.

    Interpretationcitation

    Multiple top-down and bottom-up mechanisms may contribute to anxiolytic effects of ABM, such as goal-directed attention control and threat-distractor inhibition.

    “it highlights the role of multiple top-down and bottom-up mechanisms, which are proposed by cognitive models to underlie anxiety and which may contribute to anxiolytic effects of ABM, such as goal-directed attention control and threat-distractor inhibition (Mogg & Bradley, 2016);”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Mogg & Bradley, 2016

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Interpretationcitation

    These priorities may depend on multiple task-related and individual-difference variables.

    “and these priorities may in turn depend on multiple task-related and individual-difference variables”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Mogg & Bradley, 2016

      Not checked in the free run.

65 of 65

  • The paper relies on this without support.

    The included studies' measures of anxiety and attention bias are sufficiently valid and comparable to permit a qualitative synthesis of their findings.

    “In common with MacLeod and Clarke (2015), we conducted a qualitative review to allow detailed examination of variations in methodology and findings across studies that may be obscured in a quantitative review (Cristea et al., 2015).”

  • The authors state this as an assumption.

    ABM-threat-avoidance training assumes that anxious individuals have a preexisting AB to orient attention toward threat cues relative to nonthreat cues.

    “ABM-threat-avoidance training assumes that anxious individuals have a preexisting AB to orient attention toward threat cues relative to nonthreat cues”

  • The authors state this as an assumption.

    The cognitive view assumes that low-anxious individuals are characterized by attentional avoidance of threat.

    “This view further assumes that low-anxious individuals are characterized by attentional avoidance of threat.”

  • The authors state this as an assumption.

    The anxiety-related AB toward threat is assumed to operate automatically and outside awareness.

    “the anxiety-related AB toward threat is assumed to operate automatically and outside awareness”
4 of 4

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Many ABM studies have shown that redirecting attention from threat to non-threat cues yields greater reduction of self-reported and clinically rated fear symptoms than attending directly to phobic stimuli, though findings are inconsistent when compared to more rigorous control conditions.

    “Many ABM studies have shown that this redirection of attention yields greater reduction of self-reported and clinically rated fear symptoms than attending directly to phobic stimuli (Hakamata, Lissek, Bar-Haim, Britton, et al. 2010; Heeren, Reese, McNally & Philippot, 2012; however, see Mogg, Waters & Bradley (2017) for inconsistent findings when ABM is compared to more rigorous control conditions”

    From Advancing the treatment of anxiety disorders in transition‐age youth: a review of the therapeutic effects of unconscious exposure. (2025) · cites this paper as Mogg, Waters & Bradley (2017)

  • ABM threat-avoidance training studies often use similar tasks to both modify AB and assess AB change.

    “In addition, ABM threat-avoidance training studies often use similar tasks to both modify AB and assess AB change (e.g., probe-based spatial attention tasks are used in both ABM threatavoidance training and AB assessment), and it is therefore unclear whether effects of training transfer to other AB measures or related cognitive functions [7,82].”

    From Anxiety and Threat-Related Attention: Cognitive-Motivational Framework and Treatment (2018) · cites this paper as [7]

  • The co-occurrence of reductions in anxiety and AB measures during ABM training is not consistently found.

    “However, this cooccurrence is not consistently found [7], and conclusions are complicated by methodological issues (e.g., differing outcome measures, confounding variables, low-quality studies, publication bias) [9,79].”

    From Anxiety and Threat-Related Attention: Cognitive-Motivational Framework and Treatment (2018) · cites this paper as [7]

  • It is unclear whether effects of ABM training transfer to other AB measures or related cognitive functions.

    “In addition, ABM threat-avoidance training studies often use similar tasks to both modify AB and assess AB change (e.g., probe-based spatial attention tasks are used in both ABM threatavoidance training and AB assessment), and it is therefore unclear whether effects of training transfer to other AB measures or related cognitive functions [7,82].”

    From Anxiety and Threat-Related Attention: Cognitive-Motivational Framework and Treatment (2018) · cites this paper as [7]

  • AB assessment methods include visual probe, spatial cueing, modified Stroop, flanker, and visual search tasks.

    “AB assessment methods include visual probe, spatial cueing, modified Stroop, flanker, and visual search tasks [6,7,12,19,35,53-55,86,98-100].”

    From Anxiety and Threat-Related Attention: Cognitive-Motivational Framework and Treatment (2018) · cites this paper as [7]

Cites

Flagged first, then by citations

Research works printed but not linked: 2

  1. *Maoz K., 2013, Frontiers of Human Neuroscience, 7, 389
  2. Van Bockstaele B., 2015, Cognition and Emotion, 23, 1

Cited by

Sorted by citations

Related

Authors
Mogg, K., Waters, A. M., & Bradley, B. P.
Affiliations
  • University of Southampton (GB)
  • Griffith University (AU)
Corresponding author
Mogg, Karin
ORCID
Venue
Clinical Psychological Science
Publisher
SAGE Publications
Published
2017-04-26
Volume
5
Issue
4
Pages
698-717
ISSN
  • 2167-7026 (ISSN-L)
  • 2167-7034
DOI
10.1177/2167702617696359
PMC
PMC5513441
OpenAlex
W2607627610
Semantic Scholar
33dc86184688b3d716fef304156afd0a429a078d
Cited by
215
References
100
Citations by year
2026: 7 · 2025: 16 · 2024: 16 · 2023: 17 · 2022: 18 · 2021: 23 · 2020: 65 · 2019: 24 · 2018: 22 · 2017: 7
FWCI
5.10
Type
journal article
Language
en
Open access
Licence
Locations
  • Clinical Psychological Science (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • ePrints Soton (University of Southampton) (acceptedVersion)
  • Griffith Research Online (submittedVersion, cc-by)
Topics
  • Social Sciences › Psychology › Experimental and Cognitive Psychology › Anxiety, Depression, Psychometrics, Treatment, Cognitive Processes
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Neural and Behavioral Psychology Studies
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Visual Attention and Saliency Detection
Keywords
attentional bias modification, attention bias to threat, anxiety reduction, threat avoidance, top-down cognitive control, attentional orienting

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-09.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-09.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review