Systematic Review and Meta-Analysis: Eye-Tracking of Attention to Threat in Child and Adolescent Anxiety
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
REVIEW
Systematic Review and Meta-Analysis: Eye-Tracking of
Attention to Threat in Child and Adolescent Anxiety
Stephen Lisk, MSc, Ayesha Vaswani, BSc, Marian Linetzky, MA,
Yair Bar-Haim, PhD, Jennifer Y.F. Lau, PhD
Objective: Attention biases for threat may reflect an early risk marker for anxiety disorders. Yet questions remain regarding the direction and time-
course of anxiety-linked biased attention patterns in youth. A meta-analysis of eye-tracking studies of biased attention for threat was used to compare the
presence of an initial vigilance toward threat and a subsequent avoidance in anxious and nonanxious youths.
Method: PubMed, PsycARTICLES, Medline, PsychINFO, and Embase were searched using anxiety, children and adolescent, and eye-tracking-
related key terms. Study inclusion criteria were as follows: studies including participants 18 years of age; reported anxiety using standardized mea-
sures; measured attention bias using eye tracking with a free-viewing task; comparison of attention toward threatening and neutral stimuli; and available
data to allow effect size computation for at least one relevant measure. A random effects model estimated between- and within-group effects of first
fixations toward threat and overall dwell time on threat.
Results: Thirteen eligible studies involving 798 participants showed that neither youths with or without anxiety showed significant bias in first fixation
to threat versus neutral stimuli. However anxious youths showed significantly less overall dwell time on threat versus neutral stimuli than nonanxious
controls (g ¼ L0.26).
Conclusion: Contrasting with adult eye-tracking data and child and adolescent data from reaction time indices of attention biases to threat, there was
no vigilance bias toward threat in anxious youths. Instead, anxious youths were more avoidant of threat across the time course of stimulus viewing.
Developmental differences in brain circuits contributing to attention deployment to emotional stimuli and their relationship with anxiety are discussed.
Key words: child and adolescent anxiety, attention bias, threat processing, eye-tracking
J Am Acad Child Adolesc Psychiatry 2020;59(1):88–99.
nxiety disorders are a leading cause of morbidity of daily functioning and manifesting as clinical anxiety.3
A globally.1 As anxiety often starts in youth,2
identifying early risk markers that can be tar-
geted through interventions is a clear public health priority.
Indirect support for these models has emerged in adults
from tasks measuring reaction time (RT) under various
experimental conditions. Most commonly used is the dot-
Biased attention processing of threat may maintain anxiety3 probe task, a measure of biased visual spatial orienting to-
and contribute to its onset,4 making it a potential treatment ward (or away) from threat. Specifically, an attention bias
target. Before the therapeutic potential of targeting atten- toward threat (vigilance) is inferred by shorter RTs to detect
tion biases can be realized, outstanding questions on the a probe replacing a threatening stimulus than one replacing
presence and direction of these biases in relation to youth a nonthreatening stimulus, whereas an attention bias away
anxiety need to be addressed, particularly as most studies from threat (avoidance) is inferred by longer RTs to probe
measuring attention biases and youth anxiety rely on indi- detection following a threatening versus nonthreatening
rect reaction time indices. Here, we present the first meta- stimulus. Using these tasks under short and long presenta-
analysis of newer studies using eye-tracking measures of tion times of the threatening and nonthreatening stimuli,
attention to assess bias for threat and their associations with adult studies demonstrate anxiety-linked attention toward
youth anxiety. threat at early involuntary stages of threat processing,
Effective detection of danger is fundamental to survival. consistent with facilitated threat-orienting (vigilance hypoth-
However, cognitive accounts of anxiety suggest that some esis).5-7 To a lesser extent, strategic avoidance of threat at later
individuals show exaggerated attentional processing of stages of threat processing (vigilance-avoidance hypothesis)
threat-related information, contributing to an interruption has also been reported.8 Another common task, which is
88 www.jaacap.org Journal of the American Academy of Child & Adolescent Psychiatry
Volume 59 / Number 1 / January 2020ATTENTION BIASES AND YOUTH ANXIETY
thought largely to tap the individual’s ability to disengage recording the location of the individual’s first fixation after
attention processing of distracting emotional information stimulus presentation in each trial. Comparison of first
rather than orienting, is the emotional Stroop task. Here, fixations to threat against first fixations to neutral provides a
RTs are compared to color-naming threatening versus probability score to indicate the direction of initial orient-
nonthreatening words. In adults, longer RTs to the former ing. Greater probability of first fixation toward threat in-
versus latter condition have been found; together with data dicates a threat vigilance bias. Alternatively, the latency of
from other variations of the dot-probe task,9,10 these are first fixations to each stimulus type can be compared: faster
interpreted as being consistent with attentional maintenance first fixations to threatening versus neutral stimuli could
on threat in anxiety. Thus, adult RT data suggest the indicate a vigilance bias. To measure maintained attention
contribution of varying components of attentional bias to on threat across stimulus presentation time, dwell time on
anxiety.11 each stimulus type is calculated across the entire trial.
A significant number of studies have applied RT-based Greater mean dwell time on threatening stimuli than on
tasks in children and adolescents with varying anxiety levels. nonthreatening stimuli indicates maintained attention to-
Meta-analyzing 38 studies, Dudeney et al.12 found that ward threat, with the opposite suggesting overall threat
although both anxious and nonanxious youths displayed avoidance. Attentional vigilance and avoidance patterns over
attention biases toward threat, anxious youths demonstrated time can also be derived through dwell time on threatening
a significantly greater bias than nonanxious youths. This and nonthreatening stimuli measured across time windows
between-group difference in bias increased with age across (epochs).
youth; was greater when using the emotional Stroop task Meta-analyses of eye-tracking data from adults15 show
than the dot-probe task; and emerged only in dot-probe that during free viewing and visual search tasks, anxious
studies using a 1,250-millisecond presentation time rather adults demonstrated greater initial vigilance for threat
than shorter presentation times (500 milliseconds). Thus, compared to nonanxious adults, consistent with RT
although these data are clear in suggesting greater atten- studies.5 Total dwell time on threat versus nonthreat stimuli
tional maintenance on threat amongst anxious youths from over the entire stimulus presentation time was not investi-
the Stroop task, data from the dot-probe task are more gated in this meta-analysis. Developmental differences in
mixed over whether there is also “early” vigilance for threat brain circuits contributing to attention deployment in the
as reported in adults.5,11,13,14 presence of emotional stimuli19 and, more particularly, in
One possibility why youth data vary from adult findings how anxiety-linked attention biases change with matura-
is methodological. A longer time to process stimuli before tion12,20 means that adult findings cannot be extrapolated
the button press to maximize accuracy may be needed in to youth populations. Therefore, a growing number of eye-
younger participants.12 Such concerns underscore the distal tracking studies of threat processing in youths have been
relation between attentional processing and behavioral conducted. Mixed findings across studies warrants pooling
response. Indeed, RTs provide a relatively indirect mea- these data to evaluate combined effect sizes of anxiety-linked
surement of attention.15 The resultant RT score does not biases. Most studies have used free-viewing tasks, as these
account for variation in attentional processing after stimulus reflect a more ecologically valid assessment of attention
presentation but before probe appearance, and factors such deployment but also, more importantly, may be suitable for
as preparation and execution of motor response that may children and adolescents, as they are less dependent on task
vary between individuals, particularly children, can performance, which could introduce age-associated con-
confound RT differences.15 Furthermore, although RT founds. Most studies have investigated vigilance to threat by
tasks can potentially separate individual components un- measuring probability of first fixation toward threat, and
derpinning attention bias (eg, facilitated threat orienting maintained attention on threat, through total dwell time on
from attentional maintenance), they require multiple tasks threatening/nonthreatening stimuli.
and conditions to achieve this, which can cause fatigue and This meta-analysis aimed to address the following
response errors in children. questions: First, do anxious children/adolescents and their
Alternative approaches such as taking eye gaze measures nonanxious counterparts show an absolute bias (signifi-
during the presentation of threatening and nonthreatening cantly different from zero) in probability of first fixation to
stimuli16 have been used to more directly and continuously threatening stimuli (as an index of initial threat-vigilance),
measure attention.17 By measuring fixations (time and and is there a between-group difference on this measure?
location of attentional deployment between saccadic Second, do anxious children/adolescents and their non-
movements18), several components of attention bias can be anxious counterparts show an absolute bias (significantly
assessed. Vigilance toward threat can be indexed by different from zero) in total dwell time on threatening
Journal of the American Academy of Child & Adolescent Psychiatry www.jaacap.org 89
Volume 59 / Number 1 / January 2020LISK et al.
versus neutral stimuli (as an index of maintained attention), the bias measures being investigated (probability of first
and is there a between-group difference on this measure? As fixation to threat; latency of first fixation to threat; total
researchers have addressed these questions using different dwell time on threat versus neutral). These data may be
task parameters and recruiting specific subpopulations, available as mean scores for “anxious” and “nonanxious”
which could affect findings, we investigated the effects of groups, a test statistic for group difference, or a correlation
various procedural and population moderators. (1) Primary between the attention measure and anxiety severity. If data
attention task: as these have varied between a free-viewing are unavailable in the paper, they must be made available by
approach and dot-probe tasks that contain additional the author. (8) The design must allow for the comparison of
active components of probe selection resulting in antici- attention toward threatening and neutral elements of the
patory eye movements during the free-viewing element, this array. Studies pairing threatening stimuli with stimuli of any
may affect first fixation results.21 (2) Stimulus presentation other valence were excluded (such as one that paired fear
time: as brief presentation times are thought to capture and angry faces with a mixture of happy and neutral
involuntary attentional deployment and longer viewing faces27).
times, more strategic processes,13 analyses of total dwell
time from studies using different viewing times may be Information Sources and Search Terms
prone to differential influences of involuntary and strategic In April 2018, PubMed, Psycharticles, Medline, Psychinfo,
processes, affecting the presence and direction of the bias. and Embase databases were searched for eligible studies. We
(3) Age: as attention biases vary with age,12 coinciding with used anxiety-related key terms (anx*, anxiety disorder,
developmental accounts that suggest that all children begin GAD, depress*, fear, phobi*, dysphori*, and panic) that
with an attention bias toward threat, which “corrects” were crossed with key terms for eye-tracking measures (eye*,
during healthy developmental trajectories,22 findings from gaze*, fixation*, dwell time, and saccade) and key terms to
studies using different age ranges (eg, child/adolescent) may identify child and adolescent participants (child*, adol*,
capture distinct attentional response to threat. (4) Clinical pediatric, youth, juvenile, and teen*). Reference lists of
diagnosis: as some studies have found a threat-bias among identified studies were examined further for potentially
clinically anxious youths only,23,24 symptom severity could eligible research, as were any identified relevant review pa-
modify the expression of the attention bias. (5) Anxiety pers. Titles and abstracts were screened for inclusion by the
subtype: as many studies investigating attention biases use authors (S.L., A.V., J.L.) based on eligibility criteria 1
samples containing individuals with mixed anxiety di- through 5. There was 100% agreement across authors.
agnoses or features, this may alter the intensity of the threat Studies that met this eligibility criterion were retained for
stimuli used in the tasks across participants. full-text review to assess whether they met all criteria (S.L.,
A.V.), again with 100% agreement. Where studies met all
METHOD inclusion criteria but further data were required, authors
Eligibility Criteria were contacted to request the necessary data.
We included studies that met the following criteria: (1)
Because of practical considerations relating to translation, Statistical Analysis
and because the majority of biomedical literature is pub- Research Questions and Outcome Measures. Meta-analyses
lished in English-language journals,25 with no clear sys- were carried out to test two questions. First, the vigilance
tematic bias of such language restriction in trials reported in hypothesis was examined, namely, that individuals with an
conventional medicine,26 the study had to be available in anxiety disorder would detect threat more readily, and thus
English. (2) The study must be an original investigation. (3) orient to it more often, than nonanxious controls. The
The study must investigate human participants 18 years vigilance hypothesis was investigated using studies that
of age. (4) The study must use a standardized measure of recorded the direction of initial gaze orienting; specifically,
anxiety (state or trait) for all participants, either clinical measures of probability of first fixation to threat versus
interview or a self/parent-report anxiety questionnaire. (5) neutral stimuli and latency of first fixation toward threat-
The study must use eye tracking to measure attention bia- ening stimuli were used. Studies that did not report first
ses. (6) The study must use a free-viewing task, or a task fixation probability or latency but reported only total fixa-
with a free-viewing element (such as dot-probe), during tion time on threatening stimuli in the first 500þ milli-
which gaze is tracked. As these tasks are less dependent on seconds were excluded from the analysis (k ¼ 2). Second,
task performance, this minimizes age-associated confounds we tested the maintenance hypothesis, namely, that anxiety
in studies of children and adolescents. (7) Appropriate data is characterized by maintained attention on threat; thus,
must be available to compute an effect size for at least one of across the entire trial, individuals with anxiety will more
90 www.jaacap.org Journal of the American Academy of Child & Adolescent Psychiatry
Volume 59 / Number 1 / January 2020ATTENTION BIASES AND YOUTH ANXIETY
often dwell upon threatening than neutral stimuli. This attention task (dot-probe or free-viewing) and the stimulus
hypothesis was investigated using studies that recorded presentation time (2,000 milliseconds or >2,000 milli-
the mean duration of gaze (dwell time) on threat versus seconds). Population variables included the following: age
neutral stimuli, when stimuli were displayed for >1,000 group (adolescent, mean age of 12 years; or child, mean
milliseconds. age .05). In addition, we used the Orwin fail-safe N35
tion measures. Because of a lack of relevant data for within-
to calculate the number of studies with an effect size of zero
group analysis of dwell time, only between-group analysis
that would need to be added to the analysis to produce a
could be carried out for this attention bias measure.
specified “trivial” Hedges’ g value.
Meta-analyses were conducted using Comprehensive
Meta-Analysis software (version 3.3.070). All effect sizes
were calculated using Hedges’ g. To interpret effects with RESULTS
this measure, Cohen’s d guidelines were used28 (small effect: Search Results
0.20; moderate effect: 0.50; large effect: 0.80). For the Figure 1 illustrates the literature search and study selection
between-group analysis of both first fixation data and overall process. Initial searches identified 3,871 studies. After
dwell time, effect size direction was calculated so that a removing duplicates, this was reduced to 1,818 studies.
positive effect size indicates that the attentional bias toward After excluding by abstract, this number was reduced to 29
threat is larger in anxious participants than in control par- studies. Full-text screening resulted in exclusion of 16 more
ticipants. In studies that did not use high- and low- studies, resulting in 13 eligible studies.
symptom groups, correlations between symptom severity
and attention bias were used, with a positive effect size Study Characteristics
indicating a greater attention bias toward threat for more Study characteristics are displayed in Table 1.36-48 The
anxious individuals. In the within-group analyses, a positive entire data set was scanned for outliers; these were iden-
effect size indicates that the attentional bias is greater for tified as studies in which 95% CIs did not overlap with the
threat stimuli than for neutral stimuli, with a negative effect 95% CI of the combined effect size. No studies yielded an
size indicating the opposite. A random-effects model was effect size that was an outlier. Therefore, the total sample
chosen to compute combined effect sizes, as heterogeneity included data from 798 participants aged 3 to 18 years,
was expected across studies, and this method allows the from 13 studies. All studies were published in peer-
results to be generalized to similar studies.29 To assess reviewed journals. Although they contained a free-
heterogeneity of overall effect sizes, Cochran’s Q30 was used. viewing element, studies varied on the specific tasks
In addition, the I2 statistic31 was used, indicating the per- used; nine studies used a task that solely involved free
centage of this variation across effect sizes. viewing of the presented stimuli, whereas four studies used
Categorical variables were identified as potential mod- a dot-probe task that required a user action after the period
erators, consisting of procedural and population factors that of free viewing. Nine studies used a clinical sample of
differed across studies. Procedural variables included the anxious participants, and four studies used an unselected
Journal of the American Academy of Child & Adolescent Psychiatry www.jaacap.org 91
Volume 59 / Number 1 / January 2020LISK et al.
sample. Five studies investigated attention bias in relation There was large heterogeneity in the effect sizes for anxious
to social anxiety disorder (SAD) or social phobia (SP), two [Q(5) ¼ 46.32, p < .001, I2 ¼ 89.20%] and nonanxious
used broader overall anxiety scores, one for state anxiety, [Q(5) ¼ 39.48, p < .001, I2 ¼ 87.33%] groups.
and the remaining five included patients with a mixture of
Between-Group Analysis. The meta-analysis examining
anxiety diagnoses (including SAD, SP, generalized anxiety
the between-group differences in first fixation on threat
disorder [GAD], and separation anxiety [SEP]). Ten
(Figure 3) found that anxious individuals did not signifi-
studies used faces as the threatening stimuli, with five of
cantly differ from nonanxious individuals in initial fixation
these using an angry emotion, three using fear, one using
toward threatening versus neutral stimuli (k ¼ 8; g ¼ 0.04,
pain, and one specifying a general “threatening” face.
p ¼ .39, CI ¼ 0.18, 0.26). There was not significant
Two studies used eyes as the threatening stimuli, one as
heterogeneity in the effect sizes [Q(8) ¼ 8.56, p ¼ .29,
part of the face, and the other study only the eyes. The
I2 ¼ 18.25%].
final study used pictures of social scenes, with faces
within the scenes defined as the threatening stimuli.
Effect sizes within each study, and CIs can be seen in Meta-Analysis of Anxiety and Overall Dwell Time
Figures 2 to 4. Between-Group Analysis. The overall effect size for the
meta-analysis examining the association between anxiety
Meta-Analysis of Anxiety and First Fixation Data and dwell time (Figure 4) was significant (k ¼ 12;
Within-Group Analyses. The meta-analyses examining g ¼ 0.26, p ¼ .004, CI ¼ 0.44, 0.08), indicating that
within-group differences in first fixation on threat versus anxious youths avoided threatening stimuli more than
neutral stimuli (Figure 2) show that the combined effect size nonanxious youths across the stimulus-viewing period.
was not significant in anxious participants (k ¼ 6; g ¼ There was not significant heterogeneity in the effect sizes
0.315, p ¼ .21, CI ¼ 0.17, 0.80), or in nonanxious [Q(11) ¼ 15.48, p ¼ .16, I2 ¼ 28.93%]. Of note, because
controls (k ¼ 6; g ¼ 0.27, p ¼ .27, CI ¼ 0.21, 0.75). results from analogue samples can be difficult to interpret,
as nonclinical individuals with high scores on self-reported
anxiety measures do not always show patterns of attention
similar to those of clinical patients, we re-ran the analysis
FIGURE 1 Flowchart of Screening Processes for Study excluding analogue studies. Re-running the between-group
Inclusion analysis on dwell time data without analogue samples still
showed the overall effect [k ¼ 8; g ¼ 0.24, p ¼ .035,
CI ¼ 0.45, 0.02); heterogeneity Q(11) ¼ 9.97, p ¼ .19,
I2 ¼ 29.77%].
Subgroup Moderator analyses
The nonsignificant c2 values in testing for heterogeneity in
variance, and I2 values that are not extremely high, suggest
that the studies in each sample were fairly homogeneous.
However, as the I2 values were approaching 25%, and,
based upon a priori analysis plans, moderator analyses were
conducted.
Subgroup Moderator Results for Between-Group Com-
parisons of First-Fixation Data. There were no significant
moderation effects on first-fixation data by population or
procedural factors identified a priori (Table 2).
Subgroup Moderator Results for Between-Group Com-
parisons of Overall Dwell Time. For anxiety type, signifi-
cantly greater (negative) between-group effect sizes
Note: Criterion 4 did not use standardised measure of anxiety; criterion 6 did not (indicating more avoidance of threat for anxious compared
use appropriate attention task; for Criterion 7, necessary data were unavailable/un- to nonanxious individuals) was found for studies including
obtainable; criterion 8 did not allow for comparison of attention toward threat- participants with a mixture of anxiety types than for studies
ening and neutral stimuli.
using only social anxiety (p ¼ .05) (Table 2).
92 www.jaacap.org Journal of the American Academy of Child & Adolescent Psychiatry
Volume 59 / Number 1 / January 2020Volume 59 / Number 1 / January 2020
Journal of the American Academy of Child & Adolescent Psychiatry
TABLE 1 Study Characteristics
% Primary
n n Age, Age, Sex Sample Anxiety Attention Threat Threat Number of Display
Study N (Clinical) (Control) Range (y) mean (y) (Female) Type Problem Task Stimulus Emotion Stimuli Time (ms)
Capriola-Hall et al. 41 41 N/A Adolescents 14.54 68% Clinical SAD Free- Face Angry 2 3,000
(2018)36 (12L16) viewing
Dodd et al. (2015)37 83 37 46 Children 3.99 59% Clinical SAD, GAD, Free- Face Angry 2 1,250
(3L4) SEP, SP viewing
Haller et al. (2017)38 51 N/A N/A Adolescents 16.73 100% Analogue SAD Free- Scene Social Varying 5,000
(14L18) viewing
Heathcote et al. 37 N/A N/A Adolescents 12.1 64% Analogue State anxiety Free- Face Pain 2 3,500
(2016)39 (8L17) viewing
Kleberg et al. 25 25 N/A — 15.2 84% Clinical SAD Free- Eyes Eyes 4 2,000
(2017)40 Adolescents viewing
Michalska et al. 82 N/A N/A Children 11.81 60 % Analogue Overall Free- Face Eyes 1 7,000L8,000
(2017)41 (9L13) anxiety viewing
score
Price et al. (2013)42 94 74 20 Children 10.57 52% Clinical GAD, SEP, SP Dot- Face Fear 2 2,000
— probe
Price et al. (2016)43 67 67 N/A Children 11.1 53.7% Clinical GAD, SEP, SP Dot- Face Fear 2 2,000
(9L14) probe
Schmidtendorf 79 37 42 Children 11.45 61% Clinical SAD Free- Face Angry 2 5,000
et al. (2018)44 — viewing
Seefeldt et al. 73 30 43 Children 9.9 44% Clinical SP Dot- Face Angry 2 3,000
(2014)45 (8L12) probe
Shechner et al. 33 18 15 Adolescents 13.19 58% Clinical GAD, SAD, SP Free- Face Angry 2 10,000
(2013)46 (8L17) viewing
Shechner et al. 45 19 26 Adolescents 12.63 44% Clinical GAD, SAD, SP Free- Face Threat 3 5,000
(2017)47 (8L17) viewing
Tsypes et al. 88 N/A N/A Children 9.26 44% Analogue Overall Dot- Face Fear 2 1,000
ATTENTION BIASES AND YOUTH ANXIETY
(2017)48 — anxiety probe
score
www.jaacap.org
Note: GAD ¼ generalized anxiety disorder; SAD ¼ social anxiety disorder; SEP ¼ separation anxiety disorder; SP ¼ social phobia (spider).
93LISK et al.
FIGURE 2 Forest Plot of Within-Group First Fixation Bias for Threatening Stimuli
Note: 95% CIs and study weights illustrate contribution to overall effect size. Diamond represents estimate of combined effect size.
Publication Bias 0 added to the analysis to increase the p value to >.05, that is,
Funnel plots were inspected (see Figures S1 and S2, available to produce a statistically nonsignificant cumulative effect. In
online), and no evidence of asymmetry was observed. The addition to this, using the Orwin fail-safe N, to bring our
Egger test33 and rank correlation test32 results were all criterion down to a Hedges’ g value of 0.1, it would take 21
nonsignificant (all p values >.49). Furthermore, using the extra studies with an effect size of 0.
Duval and Tweedie trim and fill procedure,34 no evidence of
publication bias was found for any of the measures. For the DISCUSSION
dwell time meta-analysis, the fail-safe N49 was 25, meaning This first meta-analysis of eye-tracking measures of atten-
that there would need to be 25 studies with an effect size of tion bias in child and adolescent anxiety included data from
FIGURE 3 Forest Plot of Between-Group First Fixation Bias for Threatening Stimuli
Note: 95% CIs and study weights illustrate contribution to overall effect size. Diamond represents estimate of combined effect size.
94 www.jaacap.org Journal of the American Academy of Child & Adolescent Psychiatry
Volume 59 / Number 1 / January 2020ATTENTION BIASES AND YOUTH ANXIETY
FIGURE 4 Forest Plot of Between-Group Dwell Time Bias for Threatening Stimuli
Note: 95% CIs and study weights illustrate contribution to overall effect size. Diamond represents estimate of combined effect size.
798 participants aged 3 to 18 years across 13 studies. A threat. As our total dwell time score is unable to infer
significantly greater tendency to direct first fixations on specific patterns of attention bias over time, it is less clear
threatening over neutral stimuli did not characterize or whether the direction of these biases indeed fluctuate across
differentiate anxious and nonanxious children or adoles- time in anxious versus nonanxious youths. In-Albon et al.
cents. Instead, anxious youths showed a greater tendency to found a pattern of vigilance-avoidance in anxious youth in
avoid maintaining their gaze on threat compared to non- two studies,50,51 although a third study52 failed to report
anxious youths, a difference that emerged only in studies in similar patterns. These conflicting data underscore a need
which samples comprised mixed anxiety diagnoses. for more studies using time windows with fixations and
At first glance, our findings that biased orienting toward dwell time across different epochs. There are alternative
threat did not differentiate anxious and nonanxious children measures derived from eye tracking that can be used to
and adolescents but that over the course of stimulus assess anxiety-linked differences across the entire viewing
viewing, anxious youths avoided threatening over period of complex stimuli,53 such as assessing the visual scan
nonthreatening stimuli more than nonanxious controls path, a sequence of fixations and saccades thought to reflect
seems incompatible with the meta-analyses of RT data in the manner in which information is attended to, reap-
children and adolescents.12 However, it is possible that first praised, and integrated.
fixation data may not be equivalent to attention capture/ Second, our meta-analytic findings appear inconsistent
engagement in RT-based paradigms. There is therefore still with adult RT5 and eye-tracking data15 which suggest that
work to do in mapping how RT-based indices relate to eye- anxiety is characterized by facilitated detection and orient-
tracking indices. Closer inspection of the moderator analysis ing of initial attention toward threat and greater maintained
in the meta-analyses also shows that anxiety group differ- attention on threat in anxious individuals.54-58 These dif-
ences for attention bias emerged only only when stimuli in ferences may instead underscore the importance of devel-
dot-probe tasks were presented at 1,250 milliseconds (rather opmental accounts of anxiety. Such accounts need to
than those >500 milliseconds) and only when they were recognize greater variability in attention bias expression
greater than dot-probe when considering emotional Stroop among young people compared to adults, manifesting be-
task results. These data could be interpreted to imply that tween initial vigilance, rapid avoidance, sustained threat
any bias in attentional deployment for threat in youths is monitoring, and vigilance-avoidance patterns, and may be
likely to occur beyond initial fixation and is driven by dis- attributable to the influences of multiple cognitive and
turbances in voluntary top-down processes. Our findings learning processes unique to the developing individual.59
extend these interpretations by suggesting that these later- However, that developmental factors may moderate atten-
stage biases result in an eventual strategic avoidance of tion bias expression across youth was not supported by our
Journal of the American Academy of Child & Adolescent Psychiatry www.jaacap.org 95
Volume 59 / Number 1 / January 2020LISK et al.
TABLE 2 Moderator Results for Between-Group Comparisons of Attentional Vigilance and Attentional Maintenance
Effect size Heterogeneity Moderation
Moderators for Between-
Group Comparisons of
Attentional Vigilance k g 95% CI I2 Q p
Age group
Adolescent 3.00 0.15 e0.62, 0.93 70.27 0.20 .66
Child 5.00 e0.03 e0.25, 0.2 0.00
Presentation Time, ms
2,000 4.00 0.20 e0.18, 0.58 32.90
Task
Dot-probe 4.00 0.01 e0.26, 0.28 0.00 0.06 .81
Free-viewing 4.00 0.08 e0.44, 0.6 63.17
Anxiety Type
Mixed 4.00 0.21 e0.11, 0.54 26.71 2.61 .11
SAD/SP 4.00 e0.14 e0.43, 0.15 0.00
Moderators for Between- Effect size Heterogeneity Moderation
Group Comparisons of
Attentional Maintenance k g 95% CI I2 Q P
Age group
Adolescent 5 e0.19 e0.61, 0.22 49.82 0.20 .653
Child 7 e0.30* e0.48, e0.11 14.22
Presentation Time, ms
2,000 7 e0.16 e0.45, 0.13 50.47
Task
Dot-probe 4 e0.24 e0.57, 0.09 49.04 0.02 .881
Free-viewing 8 e0.27* e0.5, e0.05 26.70
Sample Type
Analogue 4 e0.30 e0.63, 0.04 41.35 0.07 .791
Clinical 8 e0.24* e0.46, e0.02 29.77
Anxiety Type
Mixed 6 e0.43* e0.63, e0.24 0 3.83* .050
SAD/SP 5 e0.08 e0.37, 0.21 14.07
Note: The number of studies using an analogue group (k ¼ 0) was not enough to test moderation of “sample type.” Mixed ¼ studies including
patients with a range of anxiety diagnoses; SAD ¼ Social Anxiety Disorder; SP ¼ Social Phobia.
*
p < .05.
data. We found no within-group vigilance effect in anxious whereas the adolescent group studies did not. This is sur-
or nonanxious children and adolescents, and no moderating prising, as the literature proposes avoidance as a maladaptive
effect of age on between-group differences in vigilance within emotion regulation strategy, driven largely by executive
this age range. These data therefore seem to speak against control processes that mature in youth.60,61 Instead, putting
developmental accounts that all children may begin with an our findings with those from adults suggests that although
attention bias toward threat, which then “corrects” during attentional avoidance of threat characterizes anxious children
healthy developmental trajectories.22 However caution is and difficulty disengaging from threat characterizes anxious
needed before drawing firm conclusions. Albeit not reaching adults, there are no clear attentional strategies in anxious
significance as a moderator, when categorizing the studies by adolescents, possibly as brain circuits underlying attentional
age, there were differences in the strength of the association deployment are still undergoing reorganization during
between dwell time on threat and anxiety across age groups: adolescence. However, there was a relatively high heteroge-
the child group studies showed a significant avoidance, neity of variance between effect sizes in the adolescent group,
96 www.jaacap.org Journal of the American Academy of Child & Adolescent Psychiatry
Volume 59 / Number 1 / January 2020ATTENTION BIASES AND YOUTH ANXIETY
indicating that other factors may be affecting the associations first fixations to be around 175 milliseconds,52 whereas eye-
between anxiety and avoidance. Furthermore, as many tracking studies from anxious individuals generally show
studies used wide age ranges, we could not investigate the first-fixation latency to be longer (250400 ms16,65). There
influence of age through a meta-regression. Instead, we relied are also suggestions first fixation measures of threat pro-
on subgroup analysis, which crudely used mean age of the cessing aren’t as reliable as expected,54,66,67 and may be
sample to dichotomously categorize studies into children and affected by participants favoring fixation to the top or left
adolescents. Further research assessing the association be- image regardless of its emotional valence. Relatedly, the
tween anxiety-linked attention patterns across specific ages free-viewing approach across long stimulus duration times
(or developmental stages) within youth would help to (>1,000 milliseconds) used by many eye-tracking studies
elucidate the role of maturation and/or experience on the may have an impact on identifying anxiety group differences
expression of these biases. in first fixations. As this task measures only spontaneous
Finally, the only factor that significantly moderated viewing behavior, and not attentional behavior related to
maintained attention was “anxiety type”; only studies using task demands, it may be less powerful in tapping attentional
participants with a mixture of anxiety types found a sig- engagement/disengagement, as neither is necessary for task
nificant between-group difference in avoidance. It should be completion. Indeed, group differences in attention bias are
noted, however that studies using mixed anxiety groups all more readily identified when a task action is required, such
included patients with social anxiety within their samples; as a visual search task.68-70 Another way of potentially
plus, given high level of homotypic comorbidity in anxiety informing the attentional components contributing to
disorders,62 several of the “only” social anxiety studies may anxiety is to simultaneously collect pupillary dilation data.
have included co-occurring anxiety disorders, making it Future studies could try to associate these measures and to
difficult to disentangle biases in maintained attention per gain information on the online interplay between the
disorder. However, as a whole, the results imply that specific temporal dynamics of gaze behavior and brain-mediated
diagnostic subgroups other than social anxiety disorder are emotional responsiveness.
driving this avoidance effect. Using more specific disorder Finally, many studies may not always accurately identify
and symptom boundaries in future study designs may be biases in relation to threat due to differences in threat
more informative as attentional components increasingly evaluations. For instance, all facial stimuli may be consid-
show disorder and symptom specificity.63,64 ered somewhat threatening in socially anxious individuals,
There are several limitations to our study. Compared to and as such avoidance of all faces may occur.71 Avoidance of
other meta-analyses of attention bias to threat5,12,15, there all perceived social threat may mask any group differences in
were fewer studies in this meta-analysis, with a handful of attention bias picked up with current measures, as only
published studies that were excluded because of inadequate between-face differences are generally calculated. It could
and unavailable data to compute an effect size. Null results also be possible that no threat evaluation occurs because
may have emerged from low power. A low number of such threat stimuli lack personal relevance to young people,
studies also prevented some moderator analyses from being and that this also masks within or between-group anxiety
carried out, and others being considered (such as a meta- differences in attention bias.
regression of the ratio of female-to-male participants in Notwithstanding these limitations there are some
the sample), as well as the presence of comorbidity with clinical implications of these findings. Based on relatively
nonanxiety disorders in some samples (eg, autism spectrum robust findings from anxious adults of an association with
disorders or depression). For one moderator, presentation attention bias for threat,5 attention bias modification
time, the rationale for selecting a cut-off (2,000 millisec- (ABM) tasks have been used as anxiety-reducing in-
onds) was somewhat arbitrary, driven by the distribution of terventions in adults72,73 and young people,74 mainly
the parameters used in individual studies, and the need to using the dot-probe task but also eye-tracking tasks.75,76
achieve a largely even split of studies into long and short These paradigms train attention away from threat
durations (five versus seven). Where moderation was mostly toward nonthreatening stimuli. Results using
examined, differential effects across levels of some variables ABM in anxious youth have been mixed, with meta-
may have reached significance with larger samples. analyses finding that ABM did not lead to a signifi-
Second, although first-fixation data via eye tracking cantly greater symptom reduction than a control condi-
provide a more precise indication of where overt attention is tion.77 The current meta-analysis results suggest that
first directed, it is unclear whether these measures in fact rather than modify an initial orienting bias for threat it
reflect a mixture of stimulus-driven and strategic processes. may be valuable to modify strategic processes that reduce
Previous research has found typical latency of exogenous threat avoidance. Some studies have already suggested
Journal of the American Academy of Child & Adolescent Psychiatry www.jaacap.org 97
Volume 59 / Number 1 / January 2020LISK et al.
that ABM reduces anxiety by improving strategic atten- Accepted June 26, 2019.
tion control processes,78 and, within this, some theorists Mr. Lisk, Ms. Vaswani, and Dr. Lau are with Institute of Psychiatry, Psychology
suggest that visual search tasks may be more appropriate and Neuroscience, King’s College London. Ms. Linetzky and Dr. Bar-Haim are
with the School of Psychological Sciences, Tel Aviv University, Israel. Dr. Bar-
for modifying these voluntary aspects of attention.13 Haim is also with the Sagol School of Neuroscience, Tel Aviv University, Israel.
Indeed, in youth, implementations of visual search Mr. Lisk is supported by a UK Medical Research Council studentship (MR/
tasks, where participants search for a benign target K50130X/1) and the European Commission FP7 Braintrain grant (602186).
Dr. Lau has received funding from the UK Medical Research Council (MR/
(smiling face) from among negative distractors (negative N006194/1).
faces) has resulted in consistent symptom reduction,79,80 This article is part of a special series devoted to the subject of anxiety and
although it remains to be seen whether this reduction can OCD. The series covers current topics in anxiety and OCD, including epide-
miology, translational neuroscience, and clinical care. The series was edited by
be explained by threat avoidance specifically, rather than Guest Editor Daniel A. Geller, MBBS, FRACP.
exposure to threatening faces per se. Disclosure: Dr. Bar-Haim has active competitive grants from the U.S. Depart-
In summary, the current meta-analyses suggest that ment of Defense, the US-Israel Binational Science Foundation, the Israel Sci-
ence Foundation, and JOY Ventures. Dr. Lau has active competitive grants
anxious and non-anxious youth do not differ in overt initial from the British Academy and Mental Health Research UK. Mr. Lisk, Ms. Vas-
orienting to threat, as measured by eye movements; how- wani, and Ms. Linetzky have reported no biomedical financial interests or po-
tential conflicts of interest.
ever, our results demonstrate a small effect of anxious youth
Correspondence to Jennifer Y.F. Lau, PhD, Department of Psychology, Insti-
avoiding threat. Future research with large sample sizes is tute of Psychiatry, Psychology and Neuroscience, King’s College London, De
required to investigate the avoidant pattern of strategic Crespigny Park, London SE5 8AF, UK; e-mail: jennifer.lau@kcl.ac.uk
attention across time more discretely, and to delineate the 0890-8567/$36.00/ª2019 American Academy of Child and Adolescent
Psychiatry
factors contributing to the individual differences found in
https://doi.org/10.1016/j.jaac.2019.06.006
attention bias expression amongst anxious youth.
REFERENCES
1. World Health Organization. Depression and Other Common Mental Disorders: Global 19. Scerif G. Attention trajectories, mechanisms and outcomes: at the interface between
Health Estimates. Geneva: World Health Organization; 2017. developing cognition and environment. Dev Sci. 2010;13:805-812.
2. Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Walters EE. Lifetime 20. Cohen Kadosh K, Heathcote LC, Lau JY. Age-related changes in attentional control
prevalence and age-of-onset distributions of DSM-IV disorders in the National Co- across adolescence: how does this impact emotion regulation capacities? Front Psychol.
morbidity Survey Replication. Arch Gen Psychiatry. 2005;62:593-602. 2014;5:111.
3. MacLeod C, Mathews A, Tata P. Attentional bias in emotional disorders. J Abnorm 21. Rommelse NN, Van der Stigchel S, Sergeant JA. A review on eye movement studies in
Psychol. 1986;95:15-20. childhood and adolescent psychiatry. Brain Cogn. 2008;68:391-414.
4. Lau JY, Waters AM. Annual research review: An expanded account of information- 22. Field AP, Lester KJ. Is there room for ’development’ in developmental models of in-
processing mechanisms in risk for child and adolescent anxiety and depression. J Child formation processing biases to threat in children and adolescents? Clin Child Fam Psy-
Psychol Psychiatry. 2017;58:387-407. chol Rev. 2010;13:315-332.
5. Bar-Haim Y, Lamy D, Pergamin L, Bakermans-Kranenburg MJ, van IJzendoorn MH. 23. Waters AM, Henry J, Mogg K, Bradley BP, Pine DS. Attentional bias towards angry
Threat-related attentional bias in anxious and nonanxious individuals: a meta-analytic faces in childhood anxiety disorders. J Behav Ther Exp Psychiatry. 2010;41:
study. Psychol Bull. 2007;133:1-24. 158-164.
6. Beck AT, Clark DA. An information processing model of anxiety: automatic and strategic 24. Waters AM, Mogg K, Bradley BP, Pine DS. Attention bias for angry faces in children
processes. Behav Res Ther. 1997;35:49-58. with social phobia. J Exp Psychopathol. 2011;2:jep-018111.
7. Eysenck MW, Derakshan N, Santos R, Calvo MG. Anxiety and cognitive performance: 25. Bown MJ, Sutton AJ. Quality control in systematic reviews and meta-analyses. Eur J
attentional control theory. Emotion. 2007;7:336-353. Vasc Endovasc Surg. 2010;40:669-677.
8. Mogg K, Bradley BP, Miles F, Dixon R. Brief report time course of attentional bias 26. Morrison A, Polisena J, Husereau D, et al. The effect of English-language restriction on
for threat scenes: testing the vigilance-avoidance hypothesis. Cogn Emot. 2004;18: systematic review-based meta-analyses: a systematic review of empirical studies. Int J
689-700. Technol Assess Health Care. 2012;28:138-144.
9. Fox E, Russo R, Bowles R, Dutton K. Do threatening stimuli draw or hold visual 27. Lagattuta KH, Kramer HJ. Try to look on the bright side: children and adults can
attention in subclinical anxiety? J Exp Psychol Gen. 2001;130:681-700. (sometimes) override their tendency to prioritize negative faces. J Exp Psychol Gen.
10. Fox E, Russo R, Dutton K. Attentional bias for threat: evidence for delayed disengage- 2017;146:89-101.
ment from emotional faces. Cogn Emot. 2002;16:355-379. 28. Cohen J. Statistical Power Analysis for the Behavioral Sciences. Second ed. Hillsdale, NJ:
11. Cisler JM, Koster EH. Mechanisms of attentional biases towards threat in anxiety dis- Erlbaum; 1988.
orders: an integrative review. Clin Psychol Rev. 2010;30:203-216. 29. Field AP. Meta-analysis of correlation coefficients: a Monte Carlo comparison of fixed-
12. Dudeney J, Sharpe L, Hunt C. Attentional bias towards threatening stimuli in children and random-effects methods. Psychol Methods. 2001;6:161-180.
with anxiety: a meta-analysis. Clin Psychol Rev. 2015;40:66-75. 30. Hedges L, Olkin I. Statistical Methods for Meta-analysis. San Diego, CA: Academic
13. Mogg K, Bradley BP. Anxiety and attention to threat: cognitive mechanisms and Press; 1985.
treatment with attention bias modification. Behav Res Ther. 2016;87:76-108. 31. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med.
14. Williams JMG, Watts FN, MacLeod C, Matthews A. Cognitive Psychology and 2002;21:1539-1558.
Emotional Disorders. Oxford, UK: John Wiley & Sons; 1988. 32. Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for publi-
15. Armstrong T, Olatunji BO. Eye tracking of attention in the affective disorders: a meta- cation bias. Biometrics. 1994;50:1088-1101.
analytic review and synthesis. Clin Psychol Rev. 2012;32:704-723. 33. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a
16. Garner M, Mogg K, Bradley BP. Orienting and maintenance of gaze to facial expressions simple, graphical test. BMJ. 1997;315:629-634.
in social anxiety. J Abnorm Psychol. 2006;115:760-770. 34. Duval S, Tweedie R. Trim and fill: a simple funnel-plot-based method of testing and
17. Duchowski AT. Eye tracking methodology: theory and practice. London: Springer- adjusting for publication bias in meta-analysis. Biometrics. 2000;56:455-463.
Verlag London; 2007. 35. Orwin RG. A fail-safe N for effect size in meta-analysis. J Educ Stat. 1983;8:157-159.
18. Salvucci DDG, J. H. Identifying fixations and saccades in eye-tracking protocols. Pro- 36. Capriola-Hall NN, Wieckowski AT, Ollendick TH, White SW. The influence of social
ceedings of the 2000 Symposium on Eye Tracking Research & Applications. ACM. communication impairments on gaze in adolescents with social anxiety disorder. Child
2000;71-78. Psychiatry Hum Dev. 2018;49:672-679.
98 www.jaacap.org Journal of the American Academy of Child & Adolescent Psychiatry
Volume 59 / Number 1 / January 2020ATTENTION BIASES AND YOUTH ANXIETY
37. Dodd HF, Hudson JL, Williams T, Morris T, Lazarus RS, Byrow Y. Anxiety and 60. Amso D, Scerif G. The attentive brain: insights from developmental cognitive neuro-
attentional bias in preschool-aged children: an eyetracking study. J Abnorm Child Psy- science. Nat Rev Neurosci. 2015;16:606-619.
chol. 2015;43:1055-1065. 61. Paus T. Mapping brain maturation and cognitive development during adolescence.
38. Haller SPW, Doherty BR, Duta M, Kadosh KC, Lau JYF, Scerif G. Attention allocation Trends Cogn Sci. 2005;9:60-68.
and social worries predict interpretations of peer-related social cues in adolescents. Dev 62. Beesdo K, Knappe S, Pine DS. Anxiety and anxiety disorders in children and adolescents:
Cogn Neurosci. 2017;25:105-112. developmental issues and implications for DSM-V. Psychiatr Clin North Am. 2009;32:
39. Heathcote LC, Lau JY, Mueller SC, et al. Child attention to pain and pain tolerance are 483-524.
dependent upon anxiety and attention control: an eye-tracking study. Eur J Pain. 2017; 63. Grafton B, Southworth F, Watkins E, MacLeod C. Stuck in a sad place: biased atten-
21:250-263. tional disengagement in rumination. Emotion. 2016;16:63-72.
40. Kleberg JL, Hogstrom J, Nord M, Bolte S, Serlachius E, Falck-Ytter T. Autistic traits and 64. Pergamin-Hight L, Naim R, Bakermans-Kranenburg MJ, van IJzendoorn MH, Bar-
symptoms of social anxiety are differentially related to attention to others’ eyes in social Haim Y. Content specificity of attention bias to threat in anxiety disorders: a meta-
anxiety disorder. J Autism Dev Disord. 2017;47:3814-3821. analysis. Clin Psychol Rev. 2015;35:10-18.
41. Michalska KJ, Machlin L, Moroney E, et al. Anxiety symptoms and children’s eye gaze 65. Mogg K, Millar N, Bradley BP. Biases in eye movements to threatening facial expressions
during fear learning. J Child Psychol Psychiatry. 2017;58:1276-1286. in generalized anxiety disorder and depressive disorder. J Abnorm Psychol. 2000;109:
42. Price RB, Siegle GJ, Silk JS, et al. Sustained neural alterations in anxious youth per- 695-704.
forming an attentional bias task: a pupilometry study. Depress Anxiety. 2013;30:22-30. 66. Wermes R, Lincoln TM, Helbig-Lang S. How well can we measure visual attention?
43. Price RB, Rosen D, Siegle GJ, et al. From anxious youth to depressed adolescents: Psychometric properties of manual response times and first fixation latencies in a visual
prospective prediction of 2-year depression symptoms via attentional bias measures. search paradigm. Cogn Ther Res. 2018;41:588-599.
J Abnorm Psychol. 2016;125:267-278. 67. Waechter SN AL, Wright C, Hyatt A, Oakman J. Measuring attentional bias to
44. Schmidtendorf S, Wiedau S, Asbrand J, Tuschen-Caffier B, Heinrichs N. Attentional threat: reliability of dot probe and eye movement indices. Cogn Ther Res. 2014;38:
bias in children with social anxiety disorder. Cogn Res Ther. 2018;42:273-288. 313-333.
45. Seefeldt WL, Kramer M, Tuschen-Caffier B, Heinrichs N. Hypervigilance and avoidance in 68. Rinck M, Reinecke A, Ellwart T, Heuer K, Becker ES. Speeded detection and increased
visual attention in children with social phobia. J Behav Ther Exp Psychiatry. 2014;45:105-112. distraction in fear of spiders: evidence from eye movements. J Abnorm Psychol. 2005;
46. Shechner T, Jarcho JM, Britton JC, Leibenluft E, Pine DS, Nelson EE. Attention bias of 114:235-248.
anxious youth during extended exposure of emotional face pairs: an eye-tracking study. 69. Huijding J, Mayer B, Koster EH, Muris P. To look or not to look: an eye movement
Depress Anxiety. 2013;30:14-21. study of hypervigilance during change detection in high and low spider fearful students.
47. Shechner T, Jarcho JM, Wong S, Leibenluft E, Pine DS, Nelson EE. Threats, rewards, Emotion. 2011;11:666-674.
and attention deployment in anxious youth and adults: an eye tracking study. Biol 70. Dodd HF, Vogt J, Turkileri N, Notebaert L. Task relevance of emotional information
Psychol. 2017;122:121-129. affects anxiety-linked attention bias in visual search. Biol Psychol. 2017;122:13-20.
48. Tsypes A, Owens M, Gibb BE. Suicidal ideation and attentional biases in children: an 71. Kuckertz JM, Strege MV, Amir N. Intolerance for approach of ambiguity in social
eye-tracking study. J Affect Disord. 2017;222:133-137. anxiety disorder. Cogn Emot. 2017;31:747-754.
49. Rosenthal R. The file drawer problem and tolerance for null results. Psychol Bull. 1979; 72. Liu H, Li X, Han B, Liu X. Effects of cognitive bias modification on social anxiety: a
86:638. meta-analysis. PLoS One. 2017;12:e0175107.
50. In-Albon T, Kossowsky J, Schneider S. Vigilance and avoidance of threat in the eye movements 73. Mogg K, Waters AM, Bradley BP. Attention Bias Modification (ABM): Review of Effects
of children with separation anxiety disorder. J Abnorm Child Psychol. 2010;38:225-235. of Multisession ABM Training on Anxiety and Threat-Related Attention in High-
51. In-Albon T, Schneider S. Does the vigilance-avoidance gazing behavior of children with Anxious Individuals. Clin Psychol Sci. 2017;5(4):698-717.
separation anxiety disorder change after cognitive-behavioral therapy? J Abnorm Child 74. Bar-Haim Y. Research review: attention bias modification (ABM): a novel treatment for
Psychol. 2012;40:1149-1156. anxiety disorders. J Child Psychol Psychiatry. 2010;51:859-870.
52. Gamble AL, Rapee RM. The time-course of attentional bias in anxious children and 75. Lazarov A, Pine DS, Bar-Haim Y. Gaze-contingent music reward therapy for social
adolescents. J Anxiety Disord. 2009;23:841-847. anxiety disorder: a randomized controlled trial. Am J Psychiatry. 2017;174:
53. Chen NT, Thomas LM, Clarke PJ, Hickie IB, Guastella AJ. Hyperscanning and 649-656.
avoidance in social anxiety disorder: the visual scanpath during public speaking. Psy- 76. Linetzky M, Kahn M, Lazarov A, Pine DS, Bar-Haim Y. Contingent music reward
chiatry Res. 2015;225:667-672. therapy for clinically anxious 7- to 10-year-olds: an open multiple baseline feasibility
54. Lazarov A, Abend R, Bar-Haim Y. Social anxiety is related to increased dwell time on study. J Clin Child Adolesc Psychology. 2019 Mar 25;1-8. https://doi.org/10.1080/
socially threatening faces. J Affect Disord. 2016;193:282-288. 15374416.2019.1573685 [Epub ahead of print].
55. Buckner JD, Maner JK, Schmidt NB. Difficulty disengaging attention from social threat 77. Cristea IA, Mogoase C, David D, Cuijpers P. Practitioner review: Cognitive bias
in social anxiety. Cogn Ther Res. 2010;34:99-105. modification for mental health problems in children and adolescents: a meta-analysis.
56. Liang CW, Tsai JL, Hsu WY. Sustained visual attention for competing emotional stimuli J Child Psychol Psychiatry. 2015;56:723-734.
in social anxiety: an eye tracking study. J Behav Ther Exp Psychiatry. 2017;54:178-185. 78. Heeren A, Mogoase C, McNally RJ, Schmitz A, Philippot P. Does attention bias
57. Schofield CA, Johnson AL, Inhoff AW, Coles ME. Social anxiety and difficulty disen- modification improve attentional control? A double-blind randomized experiment with
gaging threat: evidence from eye-tracking. Cogn Emot. 2012;26:300-311. individuals with social anxiety disorder. J Anxiety Disord. 2015;29:35-42.
58. Wieser MJ, Pauli P, Weyers P, Alpers GW, Muhlberger A. Fear of negative evaluation 79. De Voogd EL, Wiers RW, Prins PJ, Salemink E. Visual search attentional bias modi-
and the hypervigilance-avoidance hypothesis: an eye-tracking study. J Neural Transm fication reduced social phobia in adolescents. J Behav Ther Exp Psychiatry. 2014;45:
(Vienna). 2009;116:717-723. 252-259.
59. Waters AM, Craske MG. Towards a cognitive-learning formulation of youth anxiety: a 80. Waters AM, Zimmer-Gembeck MJ, Craske MG, Pine DS, Bradley BP, Mogg K. Look
narrative review of theory and evidence and implications for treatment. Clin Psychol Rev. for good and never give up: a novel attention training treatment for childhood anxiety
2016;50:50-66. disorders. Behav Res Ther. 2015;73:111-123.
Journal of the American Academy of Child & Adolescent Psychiatry www.jaacap.org 99
Volume 59 / Number 1 / January 2020LISK et al.
FIGURE S1 Funnel Plot for Between-Group First-Fixation Analysis
FIGURE S2 Funnel Plot for Between-Group Dwell-Time Analysis
99.e1 www.jaacap.org Journal of the American Academy of Child & Adolescent Psychiatry
Volume 59 / Number 1 / January 2020You can also read
