Gaze-Based Assessments of Vigilance and Avoidance in Social Anxiety: a Review

Abstract

Purpose of review:
A broad base of research has sought to identify the biases in selective attention which characterize social anxiety, with the emergent use of eye tracking-based methods. This article seeks to provide a review of eye tracking studies examining selective attention biases in social anxiety.

Recent findings:
Across a number of contexts, social anxiety may be associated with a mix of both vigilant and avoidant patterns of attention with respect to the processing of emotional social stimuli. Socially anxious individuals may additionally avoid maintaining eye contact and may exhibit a generalized vigilance via hyperscanning of their environment. The findings highlight the utility of eye tracking methods for increasing understanding of the gaze-based biases which characterize social anxiety disorder, with promising avenues for future research.

Full text

ANXIETY DISORDERS (A PELISSOLO, SECTION EDITOR)
Gaze-Based Assessments of Vigilance and Avoidance in Social
Anxiety: a Review
Nigel T. M. Chen
1,2
&Patrick J. F. Clarke
1,2
#Springer Science+Business Media, LLC 2017
Abstract
Purpose of Review A broad base of research has sought to
identify the biases in selective attention which characterize
social anxiety, with the emergent use of eye tracking-
based methods. This article seeks to provide a review of
eye tracking studies examining selective attention biases in
social anxiety.
Recent Findings Across a number of contexts, social
anxiety may be associated with a mix of both vigilant and
avoidant patterns of attention with respect to the processing
of emotional social stimuli. Socially anxious individuals
may additionally avoid maintaining eye contact and may
exhibit a generalized vigilance via hyperscanning of their
environment.
Summary The findings highlight the utility of eye tracking
methods for increasing understanding of the gaze-based biases
which characterize social anxiety disorder, with promising
avenues for future research.
Keywords Eye tracking .Social anxiety .Attentional bias .
Eye contact .Av oi da nc e .Vigilance
Introduction
Anxiety disorders are widely recognized as the most common
mental health problem with lifetime prevalence rates estimat-
ed at between 26 and 40% in the USA [1]. Of the anxiety
disorders, social anxiety disorder (SAD) is frequently identi-
fied as among the most prevalent [2]. SAD is characterized by
chronic and debilitating fear of social or performance situa-
tions [3] and left untreated, typically runs a chronic course [4]
interrupting the development of social networks and relation-
ships, which form one of the most important protective factors
against other mental health disorders [5]. The desire to under-
stand the patterns of information processing that contribute to
the maintenance of SAD has driven a large body of research
seeking to describe the patterns of selective attention thatchar-
acterize social anxiety. The current review details how the
increasing use of eye tracking in recent years is providing
new insights into the different patterns of selective attention
believed to operate in SAD. The following first provides a
brief outline of influential theoretical models which have
guided research into attentional processes in social anxiety,
in addition to some key findings derived from traditional cog-
nitive reaction-time (RT) tasks, before providing a detailed
overview of recent research emerging from the use of eye
tracking methodologies.
Theoretical Models of Social Anxiety Disorder
Theoretical models consistently implicate information pro-
cessing biases in the development and maintenance of SAD,
and in particular the role of selective attention. Clark and
Wells’[3] influential model suggests that selective attention
plays a role in maintaining social anxiety through its action in
a number of crucial areas. According to this model, the
This article is part of the Topical Collection on Anxiety Disorders
*Nigel T. M. Chen
nigel.chen@uwa.edu.au
1
Centre for the Advancement of Research on Emotion, School of
Psychological Science, University of Western Australia, 35 Stirling
Highway, Crawley, WA 6009, Australia
2
School of Psychology and Speech Pathology, Curtin University,
Bentley, Perth, WA, Australia
Curr Psychiatry Rep (2017) 19:59
DOI 10.1007/s11920-017-0808-4

attentional avoidance of salient social stimuli operates as a
safety behaviour. Safety behaviours are actions designed to
reduce emotional distress and/or prevent a feared negative
event [6]. Thus, during social interactions, a safety behaviour
of avoiding mutual gaze exchanges, or avoiding attending to
emotional social gestures or simply other individuals, serves
the functions of attempting to reduce the likelihood of a feared
confrontation and/or attempting to emotionally self-regulate by
avoiding the registration of real or perceived negative evalua-
tions of others. However, as with many safety behaviours, such
actions typically serve to further perpetuate anxiety [7]. In this
case, such attentional avoidance may create the impression of
being aloof, preoccupied or uninterested during conversation
while also preventing the socially anxious individual from
disconfirming fears of negative evaluations and registering pos-
itive feedback. Therefore, strategic attentional avoidance of so-
cial stimuli is believed to maintain symptoms of SAD and may
be acutely active during social-evaluative situations [3,8].
In addition to general attentional avoidance of socially sa-
lient information, other social anxiety models specifically em-
phasize the role of biased attention for socially threatening
over benign information in SAD. Specifically, Rapee and
Heimberg’s[9] model, updated in Heimberg et al. [10••],
posits that upon entering a feared social-evaluative situation,
socially anxious individuals excessively allocate attention to-
wards external indicators of negative social evaluation. This
pattern of selective attention to socially threatening informa-
tion carries a number of adverse effects that serve to further
perpetuate social anxiety. This includes skewing impressions
of the social environment to be more threatening, and increas-
ing the likelihood that minor signals of negative feedback will
be detected, inflating negative self-evaluations of social per-
formance. As such, selective attention to threat in SAD is
likely to perpetuate the symptoms of social anxiety both gen-
erally and also more acutely during social interactions.
Given that both of the most influential theoretical models of
SAD propose patterns of selective attention in the maintenance
of the disorder, there has now been considerable research devot-
ed to establishing the presence, patterns and conditions under
which selective attention in SAD operates. The following briefly
details some of the experimental origins of this work before
exploring in more detail recent innovations in assessing attention
in SAD that have been permitted through the use of eye tracking.
Reaction-Time and Gaze-Based Assessments
of Selective Attention
Of the traditional cognitive-experimental assessments of bi-
ased attention, one of the most commonly employed has been
the attentional probe task. The attentional probe is a RT-based
task in which two stimuli that differ on a critical dimension
(e.g. threatening vs neutral information) are presented,
followed by a target probe appearing in the location vacated
by either stimulus. Faster RTs to probes appearing in the lo-
cation of one stimulus type over another are taken to indicate
an attentional bias to this stimulus.
Using RT tasks such as this, research has supported the
predictions of the two theoretical models outlined above [3,
10••], though with mixed findings across studies. Socially
anxious individuals have shown attentional avoidance of emo-
tional stimuli [11,12]. While in other studies, socially anxious
individuals have exhibited a bias towards threatening stimuli
[13,14], with further evidence suggesting that this threat-bias
may be evident during early attentional processing [15,16].
Such mixed findings have prompted the development of a
‘vigilance-avoidance’model of selective attention [17]. This
account proposes that among individuals with anxiety disor-
ders, attention may be automatically drawn to emotionally
negative information initially, with subsequent strategic
avoidance employed as an attempt to regulate anxious mood
provoked through the registration of threat.
While RT tasks have been invaluable in enabling the identi-
fication of ‘early’and ‘later’patterns of selective attention which
characterize social anxiety, such assessments may be limited as
they only provide a ‘snapshot’of attention at specific points in
time [18]. When a stimulus pair is presented for a typical dura-
tion of 500 ms, it is theoretically possible for multiple shifts of
attention to occur during this interval [19]. As such, RT mea-
sures may be non-optimal for detecting these more dynamic
aspects of attention. For instance, if a threat-neutral stimulus pair
is presented, a participant in principle could initially attend to the
neutral face for 400 ms and then shift attention to the threat face
until the 500-ms stimulus offset. Although the majority of time
was spent attending to the neutral face, the probe RT data will
reflect an attentional bias towards the threat stimulus.
As eye gaze can provide a continuous and direct measure
of visual attention in real-time [20], a growing base of research
has incorporated gaze-based measures of selective attention.
Evidence suggests that eye movements may provide crucial
insights in a number of neurological, psychiatric and psycho-
logical domains [21,22]. Given the emergent interest in eye
tracking methodologies, it is imperative that we begin to iden-
tify convergent themes, as well as any inconsistencies in find-
ings across the literature. The following provides an overview
of existing eye tracking research into the biases in selective
attention which characterize SAD. We consider recent studies
which have assessed social anxiety-linked biases in selective
attention across clinical SAD samples, non-clinical socially
anxious individuals or through the induction of social-
evaluative stress. In line with the predominant eye tracking
approaches evident, we have broadly categorized these studies
into tasks which have examined gaze-relevant selective atten-
tion biases in response to competing stimuli, selective atten-
tion biases when viewing single faces and selective attention
biases during social-evaluative contexts.
59 Page 2 of 9 Curr Psychiatry Rep (2017) 19:59

A Note on Eye Tracking Terminology
Eye tracking studies have used a variety of differing terminol-
ogy to describe specific aspects of attention assessed. As such,
we have sought to first clarify the intended meaning of these
terms, as used in this review.
Attentional Maintenance The degree to which attention is
aligned with a given stimulus over time, providing an overall
measure of attentional preference. Most commonly, this is
measured by calculating the total dwell time or total fixation
count towards a stimulus, pooling together the fixations that
may occur across multiple instances of gaze orienting and
disengagement. Greater dwell time is taken to reflect greater
attentional maintenance.
Initial Orienting The propensity to initially saccade towards,
thus engaging visual attention with, a given stimulus. This is
commonly examined in response to the onset of a single stim-
ulus, or multiple competing stimuli, and measured with either
the latency of initial orienting, or the relative proportion of
trials in which initial orienting to a given stimulus occurred.
Initial orienting has been alternatively referred to as attentional
engagement, first fixation or early processing.
Disengagement The definition of attentional disengagement
for the present review has been based on similar conceptual-
izations of disengagement by Posner et al. [23,24].
Specifically, we use of this term to indicate either the latency
to saccade away from a given stimulus, following fixation at
this stimulus, or by the relative proportion of trials in which
this critical disengagement saccade occurs. However, it is not-
ed that some studies have also used the term ‘disengagement’
as a non-specific reference to attentional maintenance during
later (vs earlier) time intervals, e.g. [25].
Hyperscanning While the aforementioned gaze processes in-
volve a certain vigilance towards, or avoidance of a specific
type of stimulus, it has further been suggested that a stimulus
non-specific vigilance may be further relevant for the under-
standing of SAD [26]. This generalized vigilance may involve
the excessive monitoring and scanning of the surrounding en-
vironment and has been previously operationalized in terms of
an increased length of scanpath over a given duration [27,28•].
Selective Attention to Competing Stimuli
Onelineofgaze-basedresearchinsocialanxietyhasbeento
examine patterns of biased attention when multiple stimuli
simultaneously compete for attention. Studies have typically
presented either pairs of emotional and neutral stimuli (e.g. an
angry and neutral face) or larger arrays of such stimuli for free
viewing, while eye movements are continuously recorded.
When initial orienting is examined in such studies, there is
some evidence to suggest that social anxiety is associated with
greater orienting towards threat [29–31]withsomefindings
suggesting greater orienting to emotional stimuli in general
(i.e. both threatening and positive facial expressions) [32,
33], consistent with the notion of an early hypervigilance in
response to the onset of such stimuli. One study further sug-
gests greater initial orienting to threat may be correlated with
social anxiety symptoms in childhood SAD [29]. In contrast
to hypervigilance findings, it is noted that other studies have
observed reduced initial orienting to emotional stimuli in so-
cial anxiety [34–36], with others observing no orienting dif-
ferences in socially anxious individuals [37–39]. Interestingly,
significant findings of a bias in initial orienting in social anx-
iety appear to be confined to tasks that employ stimulus pair
presentations, while no social anxiety-linked effects have been
observed using visual array viewing tasks [40–43]. This is
consistent with recent suggestions that initial orienting mea-
sures may be significantly impacted by the number of com-
peting stimuli presented, such that greater perceptual load or
greater spatial eccentricity of stimuli may eliminate initial
orienting biases which may otherwise be observed using stim-
ulus pairs [40,44].
Comparably fewer studies have examined social anxiety-
linked bias in attentional disengagement, following initial
fixational processing, although available evidence is mixed.
Chen et al. [37] found slower disengagement from threat,
relative to positive stimuli in clinical SAD compared to con-
trols, which is similar to recent findings that socially anxious
individuals may be slower to disengage from threat, when
compared to low levels of social anxiety [41,45]. In contrast,
Garner et al. [33] observed social anxiety-linked speed disen-
gagement
1
from both positive and threat stimuli, and Lange
et al. [42] found socially anxious individuals to be faster to
disengage from initially fixated threat. Thus, at this stage,
while there is evidence for initial orienting and disengagement
biases in SAD, there lacks consistency as to the specific di-
rection of these biases.
Studies which have examined social anxiety-linked bias in
the maintenance of attention have similarly yielded mixed
results. Some studies have found evidence supporting greater
attention to threat in social anxiety [25,40–42,46•], with one
study observing greater attention for emotional stimuli [45].
However, across several other studies, socially anxious indi-
viduals have also been found to avoid maintaining attention to
either threat [39,43,47] or emotional stimuli [32,34,35,37,
1
In this study, the latency to saccade away from a given stimulus following
initial orienting was examined. While this was termed attentional maintenance
by the authors, it is consistent with the present manuscript’s definition of
attentional disengagement.
Curr Psychiatry Rep (2017) 19:59 Page 3 of 9 59

38], while a further few studies have found no social anxiety-
linked effect on attentional maintenance when examined in
socially anxious youth [29,30,36]. One study further ob-
served reduced attentional maintenance to positive stimuli to
be correlated with greater state anxiety in clinical SAD [37].
We considered the possibility that these contrasting effects
of vigilance and avoidance in attentional maintenance may be
accounted for by time-course changes in bias. Typically, stud-
ies have explored this by dividing the duration of stimulus
presentation into a number of time bins, before examining
for changes in bias over time. While there is some evidence
of earlier vigilance (0–1000 ms) followed by subsequent
avoidance (1000–1500 ms) of emotional stimuli in socially
anxious individuals [32], such vigilance-avoidance patterns
have not been replicated in subsequent time-course analysis
studies [35,38]. Similarly, when existing studies are consid-
ered together, both social anxiety-linked vigilance and avoid-
ance biases have been found across a wide variety of stimulus
presentation durations, with vigilance being evident from
1500 ms [45]to10s[41] exposures, and avoidance from
1500 ms [35]to5s[43], thus precluding any clear time-
course modulations of attentional maintenance bias in SAD.
We further considered the possibility that differential pat-
terns of vigilance and avoidance may be observable when
contrasting clinical versus non-clinical levels of social anxiety.
However, the available evidence does not suggest any clear
distinctions. That is, when initial orienting is considered, clin-
ical studies have shown both SAD-linked vigilance [29–31]
and avoidance [35,36] effects, which is consistent with the
mixed vigilant [32,33] and avoidant [34] initial orienting
findings observed in non-clinical social anxiety. When the
maintenance of attention is considered, the pattern of results
is similar: both vigilance [40] and avoidance [35,37,38,43]
effects have been found in clinical SAD, and both vigilance
[25,41,42,45,46•] and avoidance [32,34,39,47]effects
have been found in non-clinical social anxiety. Finally, when
attentional disengagement is considered, the one available
clinical SAD study suggests that these individuals may be
faster to disengage from positive, compared to threat stimuli
[37], while the existing studies in non-clinical social anxiety
suggest a mix of both vigilant [41,45] and avoidant [33,42]
disengagement processes.
Taken together, available evidence suggests that when
competing stimuli are considered, SAD may be characterized
by a combination of both attentional vigilance and avoidance
of emotional social stimuli. Interestingly, both of these vigi-
lance and avoidance biases appear to be evident across all
initial orienting, disengagement and maintenance components
of selective attention considered, and do not appear to be
differentiated based on the time-course of stimulus exposure,
or the severity of social anxiety. The findings do not appear to
be consistent with the notion of an earlier vigilance followed
by a later avoidance in SAD, as posited by vigilance-
avoidance models [17]. Rather the findings are suggestive of
the need for further inquiry into potential moderators, which
may determine whether vigilance or avoidance is observed in
SAD.
Existing evidence further implicates the biased processing
of both positive and threatening social stimuli in SAD, in
contrast to longstanding view of social anxiety being charac-
terized by an attentional bias to threat [9,10••]. Consistent
with this, contemporary theories have proposed that SAD
may be characterized by a bivalent fear of evaluation [48],
that is, a fear of positive social evaluation [49,50], in addition
to the traditional fear of negative social evaluation. This theory
suggests that positive social regard may be aversively per-
ceived in SAD, as it may, by consequence, signify direct social
comparison to others, which in turn, may cause the anxious
individual to feel highly conspicuous and place them in great-
er danger of negative social evaluation [48,49].
Selective Attention While Viewing Faces
Research has also investigated the manner in which socially
anxious individuals scan the faces of others. These studies
typically present a single face for free viewing, while eye
movements are continuously recorded. When gaze to the core
facial features, notably the eye and mouth regions, are
analysed, the majority of research suggests that social anxiety
is associated with the avoidance of eye contact, observed
across both clinical SAD and non-clinical socially anxious
individuals. Several studies have shown this effect through
the observation of a comparative reduction in the maintenance
of attention towards eye regions of a presented face in socially
anxious individuals [51–55,56•]. However, two studies have
observed the opposite pattern, with social anxiety being asso-
ciated with increased attention to eye regions [57,58]. In
clinical SAD, the magnitude of eye contact avoidance has
further been found to correlate with social anxiety symptoms
[52–54]. Interestingly, there is also evidence that social
anxiety-linked eye contact avoidance may be modulated by
the emotional expression of the presented face. That is, while
socially anxious individuals may avoid eye contact across a
range of facial expressions, this effect may be most pro-
nounced in response to angry faces [52,54,59,60].
These findings are consistent with the notion that eye con-
tact may be perceived as aversive and highly arousing in SAD.
This is also in line with previous research which has observed
greater subjective and physiological arousal in socially anx-
ious individuals while viewing faces with direct, as compared
to averted gaze [61,62]. It is likely that socially anxious indi-
viduals may avoid maintaining eye contact as a safety-seeking
strategy in order to emotionally regulate heightened negative
affect [7,8]. It has also been suggested that eye contact avoid-
ance in SAD may represent a submissive gesture, from an
59 Page 4 of 9 Curr Psychiatry Rep (2017) 19:59

evolutionary standpoint, used in order to minimize the possi-
bility of any conflict or challenge from a potentially dangerous
and more socially dominant individual [52,60,63].
A few studies have additionally examined specific saccadic
biases in response to the onset of the face stimulus. Studies
which have assessed biases in initial orienting have yielded
mixed results. One study observed that individuals with SAD,
compared to controls, had a greater tendency to initial orient to
eye regions [57], while similar examinations of initial
orienting have not found social anxiety-linked differences
[53,54]. Studies which have pre-cued gaze to the eye region
and measured the subsequent latency to disengage have also
yielded mixed findings. One study found that socially anxious
individuals were faster to disengage eye contact from angry,
compared to happy, facial expressions, consistent with the
notion of social threat avoidance [60], while another study
found socially anxious individuals to be slower to disengage
eye contact, in comparison to low anxious individuals [64].
Thus, while the avoidance ofeye contact in SAD appears to be
a consistent finding when the maintenance of attention is con-
sidered, there is less evidence to support the notion that such
avoidance may pertain to either the initial orienting or disen-
gagement components of selective attention.
Further to the gaze processing of the eye and mouth re-
gions, research additionally suggests that SAD may be asso-
ciated with the hyperscanning of emotional faces, indicated by
alonger,andpresumablymoreerratic,lengthofscanpath[51,
52]. It has been suggested that this increased scanpath may be
reflective of a general hypervigilant strategy, which may en-
compass the extensive monitoring of visual information [26,
27,28•,51,52]. Considered together, existing evidence sug-
gests that during the viewing of faces, SAD may be charac-
terized by both the avoidance of eye contact and a concurrent
generalized hypervigilant processing strategy.
Selective Attention During Social-Evaluative
Contexts
While the research reviewed thus far has shown the potential
utility of gazed-based measures in informing specific aspects
of biased selective attention in social anxiety, a limitation of
these approaches is that they potentially lack ecological valid-
ity. Specifically, because models of SAD [8,10••] suggest that
information processing biases will be most active, and most
acutely problematic, during social-evaluative situations, stud-
ies delivered in the absence of such a social context may not
capture the problematic patterns of bias that characterize SAD.
In addition, the social stimuli presented during typical exper-
imental contexts may lack direct practical relevance to the fear
of social evaluation experienced in SAD. A further potential
advantage of eye tracking is its capacity to be implemented
unobtrusively during naturalistic contexts. Given this, a small
number of studies have sought to examine whether the atten-
tional biases which have been thought to characterize SAD
may indeed extend to social settings with greater ecological
validity.
A recent study by Howell et al. [56•]examinedgazepro-
cesses during a dyadic interaction, whereby participants en-
gaged in a live, 4-min conversation with a confederate via a
webcam, while eye gaze was continually recorded. Participant
trait social anxiety was found to be associated with reduced
attentional maintenance to confederate eye regions, suggest-
ing an avoidance of eye contact during social interaction. A
similar avoidance of eye contact has been found in individuals
with SAD when viewing videos of confederates expressing
positive and negative social feedback [55]. Interestingly, eye
contact avoidance was observed in response to both positively
and negatively valenced social feedback, consistent with the
notion that SADmay be characterize by a bivalent fear ofboth
positive and negative social evaluation [65].
Two studies have further examined biased gaze processes
in SAD during a public speaking simulation [28•,66•]. In
these studies, clinically socially anxious participants and
healthy controls were required to perform an impromptu
speech in front of a pre-recorded audience of confederates
intermittently displaying positive and negative social gestures.
Across both studies, socially anxious individuals avoided
maintaining attention, by way of either avoiding emotional
social stimuli in general [66•] or by preferentially allocating
gaze to the non-social regions in-between and around the con-
federates [28•]. Chen et al. [28•] further observed SAD-linked
hyperscanning of the audience display. Chen et al. [66•]addi-
tionally integrated a gaze-cueing procedure (cf. [37]) through-
out the speech in order to examine biased orienting of atten-
tion following the onset of emotional confederate gestures.
While control participants were faster to initially orient to
positive, compared to negative gestures, this positivity bias
was absent in SAD, suggesting that the socially anxious indi-
viduals may neglect positive stimuli more than their non-
anxious counterparts.
Taken together, the available work examining the gaze-
based biases which operate during social-evaluative situations
has produced findings largely in line with theoretical models
of SAD. During a social-evaluative situation, socially anxious
individuals may exhibit vigilance of their surrounding envi-
ronment [9,10••], by way of a hypervigilant scanning strategy,
but may also seek to avoid maintaining their attention towards
relevant emotional social stimuli, by either avoiding eye con-
tact or by opting to hold their gaze at non-social spaces.
Consistent with theoretical models, such avoidance is likely
employed as a safety-seeking strategy, but may consequently
hinder the opportunity for accurate reappraisals of their social
environment [3,7,8]. In addition, when positive social ges-
tures are presented, socially anxious individuals may be less
likely to orient towards them. It is fitting that this absence of a
Curr Psychiatry Rep (2017) 19:59 Page 5 of 9 59

positivity bias in early processing may resultantly skew indi-
viduals’mental representation of themselves as seen in their
social environment, which in turn may exacerbate negative
beliefs and other symptoms of social anxiety [9,10••].
Future Directions
While we reviewed the evidence for the selective attention
biases which characterize SAD, what remains to be seen is
the functional impact of these biases, as posited by theoretical
models. For instance, Çek et al. [46•]recentlyidentifiedan
interesting indirect path whereby the effect of social anx-
iety on negative post-event processing following a speech
stressor may be partially account for by mediating effects
of attentional bias to threat and emotional reactivity to the
stressor. Future research into these potential direct and indirect
paths of influence can provide a more detailed picture of
how these processes exert their influence on emotional
vulnerability in SAD.
Future research may additionally seek to identify important
moderators of social anxiety-linked attentional bias.
Specifically, it is evident that several studies observe vigi-
lance, while several others observe avoidance. It will be im-
portant therefore to determine the conditions under which so-
cial anxiety-linked vigilance or avoidance will emerge. For
instance, recent findings by Byrow et al. [34] suggest that
the avoidance of emotional stimuli may be more likely when
social anxiety co-occurs with an avoidant attachment style.
Moreover, it is possible that contextual relevance may simi-
larly be important for understanding social anxiety-linked at-
tentional bias. For instance, while a mix of vigilant and
avoidant attentional maintenance have been observed from
experimental tasks, when emotional stimuli are presented in
a more naturalistic social-evaluative context, attentional
avoidance has been consistently observed [28,55,56•,66•].
While further research is required, the examination of in vivo
attentional processes during social-evaluative contexts repre-
sents a promising avenue for eye tracking research.
Clinical Implications
While the research reviewed above has made significant prog-
ress towards identifying the precise patterns of attention that
characterize SAD, the translational potential of much of these
findings remains to be seen. With this in mind, it is relevant to
consider how the current findings may inform and enhance
clinical outcomes in SAD. An obvious area for future research
in this regard is to integrate eye tracking research designs into
therapeutic protocols for the treatment of SAD. Incorporating
eye tracking assessments of selective attention biases operat-
ing both before and following therapy will enable researchers
to address a number of applied questions. This includes iden-
tifying the biases that serve to predict treatment response to
established SAD interventions, the biases which may index
reductions in anxiety symptomatology and the biases which
may predict maintenance/relapse following treatment. Such
research will provide a significant step towards establishing
the patterns of selective attention that functionally mediate
SAD symptomatology.
Another avenue for future research is to directly target the
patterns of biased attention thought to maintain SAD, through
gaze-contingent modification procedures. A significant body of
research over the last decade has implemented probe-based
experimental tasks to re-train selective attention away from
threatening information. While this procedure, attentional bias
modification(ABM),hasbeenshowntoleadtoreductionsin
SAD symptoms, when a successful change in attentional bias is
achieved [67–69], recent null findings suggest that the overall
therapeutic effect of ABM may not be consistent [70,71], thus
prompting the need for more effective methods of bias modifi-
cation [72]. Just as eye tracking has provided a means to assess
dimensions of selective attention not otherwise accessible via
traditional cognitive tasks, similarly, it could provide novel
ways of re-training maladaptive patterns of attention in SAD.
Indeed some initial work is beginning to emerge in this area. An
initial study by Price et al. [73•] used a gaze-contingent operant
conditioning task to direct attention towards happy faces, with
results suggesting that it was effective in reducing attention to
negative social information and that it attenuated emotional
reactivity to a subsequent stressor, in a non-clinical sample.
Furthermore, in a recent randomized controlled trial in
clinical SAD, Lazarov et al. [74•] trained attention away
from negative facial expressions, by playing music select-
ed by the participant contingent on their gaze direction.
Specifically, music played only while attention was aligned
with neutral faces, within an array of emotional faces, and
was immediately muted when participants attended to dis-
gust faces. Eight sessions of this training across 4 weeks
was associated with significant reductions in self-report
and clinician-rated social anxiety symptoms, with therapeu-
tic effects being maintained at 3 months follow-up. These early
results therefore provide encouragement for the potential clin-
ical utility of gaze-contingent attention training paradigms.
Interestingly however, we are not aware of any studies that
have sought to specifically re-train patterns of attentional
avoidance, either with respect to maintaining eye contact or
maintaining gaze to emotional social stimuli, as put forward in
theoretical models [3,8], and evidenced in the reviewed re-
search. It is possible that a training task which targets such
avoidance (particularly during social-evaluative stress) could
in principle deliver the greatest benefits in terms of reducing
social anxiety and thus provide an interesting avenue for fu-
ture research seeking to translate existing knowledge of atten-
tional processes in SAD into clinical benefit.
59 Page 6 of 9 Curr Psychiatry Rep (2017) 19:59

Conclusions
The research reviewed suggests that SAD may be characterized
by a number of biases in selective attention. When competing
social stimuli are considered, a mix of both vigilance and
avoidance may be evident in SAD across all initial orienting,
disengagement and maintenance components of attention.
Although, when such biases are examined during social-
evaluative contexts, avoidance in the maintenance of attention
is consistently observed, with some evidence of hyperscanning
and the absence of a positivity orienting bias. When viewing
faces, SAD may be associated with the avoidance of eye con-
tact, which may be more pronounced for threatening expres-
sions, and hyperscanning. The findings highlight the utility of
eye tracking methods in examining the mechanisms of biased
selective attention, as well as enabling the assessment of in vivo
attentional processing during social-evaluative contexts.
Compliance with Ethical Standards
Conflict of Interest The authors declare that they have no conflict of
interest.
Human and Animal Rights and Informed Consent This article does
not contain any studies with human or animal subjects performed by any
of the authors.
References
Papers of particular interest, published recently, have been
highlighted as:
•Of importance
•• Of major importance
1. Kessler RC, Petukhova M, Sampson NA, Zaslavsky AM, Wittchen
HU. Twelve-month and lifetime prevalence and lifetime morbid
risk of anxiety and mood disorders in the United States. Int J
Methods Psychiatr Res. 2012;21(3):169–84. doi:10.1002/mpr.
1359.
2. Ruscio AM, Brown TA, Chiu WT, Sareen J, Stein MB, Kessler RC.
Social fears and social phobia in the USA: results from the National
Comorbidity Survey Replication. Psychol Med. 2008;38(1):15–28.
doi:10.1017/s0033291707001699.
3. Clark DM, Wells A. A cognitive model of social phobia. In:
Heimberg R, Liebowitz M, Hope DA, Schneier FR, editors.
Social phobia: diagnosis, assessment and treatment. New York:
Guilford Press; 1995. p. 69–93.
4. Wittchen HU, Stein MB, Kessler RC. Social fears and social phobia
in a community sample of adolescents and young adults: preva-
lence, risk factors and co-morbidity. Psychol Med. 1999;29(2):
309–23.
5. Deater-Deckard K. Annotation: recent research examining the role
of peer relationships in the development of psychopathology. J
Child Psychol Psychiatry. 2001;42(5):565–79.
6. Salkovskis PM. The importance of behaviour in the maintenance of
anxiety and panic: a cognitive account. Behav Psychother.
2009;19(1):6–19. doi:10.1017/S0141347300011472.
7. Hofmann SG. Cognitive factors that maintain social anxiety disorder:
a comprehensive model and its treatment implications. Cogn Behav
Ther. 2007;36(4):193–209. doi:10.1080/16506070701421313.
8. Clark DM. A cognitive perspective on social phobia. In: Crozier
WR, Alden LE, editors. International handbook of social anxiety:
concepts, research and interventions relatingto the self and shyness.
New York, NY: John Wiley; 2001. p. 405–30.
9. Rapee RM, Heimberg RG. A cognitive-behavioral model of anxi-
ety in social phobia. Behav Res Ther. 1997;35(8):741–56. doi:10.
1016/S0005-7967(97)00022-3.
10.•• Heimberg RG, Brozovich FA, Rapee RM. A cognitive-behavioral
model of social anxiety disorder: update and extension. In:
Hofmann SG, PM DB, editors. Social anxiety (third edition). San
Diego: Academic Press; 2014. p. 705–28. An important theoret-
ical model of social anxiety.
11. Mansell W, Clark DM, Ehlers A, Chen YP. Social anxiety and
attention away from emotional faces. Cogn Emot. 1999;13(6):
673–90. doi:10.1080/026999399379032.
12. Chen YP, Ehlers A, Clark DM, Mansell W. Patients with general-
ized social phobia direct their attention away from faces. Behav Res
Ther. 2002;40(6):677–87.
13. Asmundson GJG, Stein MB. Selective processing of social threat in
patients with generalized social phobia: evaluation using a dot-
probe paradigm. J Anxiety Disord. 1994;8(2):107–17. doi:10.
1016/0887-6185(94)90009-4.
14. Gilboa-Schechtman E, Foa EB, Amir N. Attentional biases for fa-
cial expressions in social phobia: the face-in-the-crowd paradigm.
Cogn Emot. 1999;13(3):305–18. doi:10.1080/026999399379294.
15. Mogg K, Philippot P, Bradley BP. Selective attention to angry faces
in clinical social phobia. J Abnorm Psychol. 2004;113(1):160–5.
doi:10.1037/0021-843x.113.1.160.
16. Mogg K, Bradley BP. Selective orienting of attention to masked
threat faces in social anxiety. Behav Res Ther. 2002;40(12):1403–14.
17. Mogg K, Bradley BP, Miles F, Dixon R. Time course of attentional
bias for threat scenes: testing the vigilance-avoidance hypothesis.
Cogn Emot. 2004;18(5):689–700. doi:10.1080/02699930341000158.
18. Bradley BP, Mogg K, Millar NH. Covert and overt orienting of
attention to emotional faces in anxiety. Cogn Emot. 2000;14(6):
789–808. doi:10.1080/02699930050156636.
19. Rayner K. Eye movements in reading and information processing:
20 years of research. Psychol Bull. 1998;124(3):372–422. doi:10.
1037/0033-2909.124.3.372.
20. Duchowski AT. A breadth-first survey of eye-tracking applications.
Behav Res Methods Instr Comput. 2002;34(4):455–70. doi:10.
3758/bf03195475.
21. Carpenter RHS. Frontal cortex: choosing where to look. Curr Biol.
1994;4(4):341–3. doi:10.1016/S0960-9822(00)00074-9.
22. Hutton SB, Ettinger U. The antisaccade task as a research tool in
psychopathology: a critical review. Psychophysiology. 2006;43(3):
302–13. doi:10.1111/j.1469-8986.2006.00403.x.
23. Posner MI, Inhoff AW, Friedrich FJ, Cohen A. Isolating attentional
systems—a cognitive-anatomical analysis. Psychobiology.
1987;15(2):107–21. doi:10.3758/BF03333099.
24. Posner MI, Petersen SE. The attention system of the human brain.
Annu Rev Neurosci. 1990;13:25–42. doi:10.1146/annurev.ne.13.
030190.000325.
25. Buckner JD, Maner JK, Schmidt NB. Difficulty disengaging atten-
tion from social threat in social anxiety. Cogn Ther Res. 2010;34(1):
99–105. doi:10.1007/s10608-008-9205-y.
26. Richards HJ, Benson V, Donnelly N, Hadwin JA. Exploring the
function of selective attention and hypervigilance for threat in anx-
iety. Clin Psychol Rev. 2014;34(1):1–13. doi:10.1016/j.cpr.2013.
10.006.
27. Green MJ, Williams LM, Davidson D. In the face of danger: spe-
cific viewing strategies for facial expressions of threat? Cogn Emot.
2003;17(5):779–86. doi:10.1080/02699930302282.
Curr Psychiatry Rep (2017) 19:59 Page 7 of 9 59

28.•Chen NTM, Thomas LM, Clarke PJF, Hickie IB, Guastella AJ.
Hyperscanning and avoidance in social anxiety disorder: the visual
scanpath during public speaking. Psychiatry Res. 2015;225(3):
667–72. doi:10.1016/j.psychres.2014.11.025.This study
examines the selective attention biases which operate in SAD
during a social-evaluative speech task
29. Seefeldt WL, Kramer M, Tuschen-Caffier B, Heinrichs N.
Hypervigilance and avoidance in visual attention in children with
social phobia. J Behav Ther Exp Psychiatry. 2014;45(1):105–12.
doi:10.1016/j.jbtep.2013.09.004.
30. Shechner T, Jarcho JM, Britton JC, Leibenluft E, Pine DS, Nelson
EE. Attention bias of anxious youth during extended exposure of
emotional face pairs: an eye-tracking study. Depress Anxiety.
2013;30(1):14–21. doi:10.1002/da.21986.
31. Stevens S, Rist F, Gerlach AL. Eye movement assessment in indi-
viduals with social phobia: differential usefulness for varying pre-
sentation times? J Behav Ther Exp Psychiatry. 2011;42(2):219–24.
doi:10.1016/j.jbtep.2010.11.001.
32. Wieser MJ, Pauli P, Weyers P, Alpers GW, Muhlberger A. Fear of
negative evaluation and the hypervigilance-avoidance hypothesis:
an eye-tracking study. J Neural Transm. 2009;116(6):717–23. doi:
10.1007/s00702-008-0101-0.
33. Garner M, Mogg K, Bradley BP. Orienting and maintenance of
gaze to facial expressions in social anxiety. J Abnorm Psychol.
2006;115(4):760–70. doi:10.1037/0021-843x.115.4.760.
34. Byrow Y, Broeren S, de Lissa P, Peters L. Anxiety, attachment &
attention: the influence of adult attachment style on attentional
biases of anxious individuals. J Exp Psychopathol. 2016;7(1):
110–28. doi:10.5127/jep.046714.
35. Byrow Y, Chen NTM, Peters L. Time courseof attention in socially
anxious individuals: investigating the effects of adult attachment
style. Behav Ther. 2016;47(4):560–71. doi:10.1016/j.beth.2016.
04.005.
36. Gamble AL, Rapee RM. The time-course of attentional bias in
anxious children and adolescents. J Anxiety Disord. 2009;23(7):
841–7. doi:10.1016/j.janxdis.2009.04.001.
37. Chen NTM, Clarke PJF, MacLeod C, Guastella AJ. Biased atten-
tional processing of positive stimuli in social anxiety disorder: an
eye movement study. Cogn Behav Ther. 2012;41(2):96–107. doi:
10.1080/16506073.2012.666562.
38. Schofield CA, Inhoff AW, Coles ME. Time-course of attention
biases in social phobia. J Anxiety Disord. 2013;27(7):661–9. doi:
10.1016/j.janxdis.2013.07.006.
39. Singh JS, Capozzoli MC, Dodd MD, Hope DA. The effects of
social anxiety and state anxiety on visual attention: testing the
vigilance-avoidance hypothesis. Cogn Behav Ther. 2015;44(5):
377–88. doi:10.1080/16506073.2015.1016447.
40. Lazarov A, Abend R, Bar-Haim Y. Social anxiety is related to
increased dwell time on socially threatening faces. J Affect
Disord. 2016;193:282–8. doi:10.1016/j.jad.2016.01.007.
41. Liang CW, Tsai JL, Hsu WY. Sustained visual attention for com-
peting emotional stimuli in social anxiety: an eye tracking study. J
Behav Ther Exp Psychiatry. 2017;54:178–85. doi:10.1016/j.jbtep.
2016.08.009.
42. Lange WG, Heuer K, Langner O, Keijsers GPJ, Becker ES, Rinck
M. Face value: eye movements and the evaluation of facial crowds
in social anxiety. J Behav Ther Exp Psychiatry. 2011;42(3):355–63.
doi:10.1016/j.jbtep.2011.02.007.
43. Shechner T,Jarcho JM, Wong S, Leibenluft E, Pine DS, Nelson EE.
Threats, rewards, and attention deployment in anxious youth and
adults: an eye tracking study. Biol Psychol. 2017;122:121–9. doi:
10.1016/j.biopsycho.2015.10.004.
44. Armstrong T, Olatunji BO. Eye tracking of attention in the affective
disorders: a meta-analytic review and synthesis. Clin Psychol Rev.
2012;32(8):704–23. doi:10.1016/j.cpr.2012.09.004.
45. Schofield CA, Johnson AL, Inhoff AW, Coles ME. Social anxiety
and difficulty disengaging threat: evidence from eye-tracking. Cogn
Emot. 2012;26(2):300–11. doi:10.1080/02699931.2011.602050.
46.•Çek D, Sánchez A, Timpano KR. Social anxiety-linked attention
bias to threat is indirectly related to post-event processing via sub-
jective emotional reactivity to social stress. Behav Ther.
2016;47(3):377–87. doi:10.1016/j.beth.2016.02.007.This study
provides an interesting investigation into how attentional bias
may influence other cognitive and emotional aspects of social
anxiety
47. Macatee RJ, Albanese BJ, Schmidt NB, Cougle JR. Attention bias
towards negative emotional information and its relationship with
daily worry in the context of acute stress: an eye-tracking study.
Behav Res Ther. 2017;90:96–110. doi:10.1016/j.brat.2016.12.013.
48. Weeks JW, Howell AN. The bivalent fear of evaluation model of
social anxiety: further integrating findings on fears of positive and
negative evaluation. Cogn Behav Ther. 2012;41(2):83–95. doi:10.
1080/16506073.2012.661452.
49. Weeks JW, Heimberg RG, Rodebaugh TL. The fear of positive
evaluation scale: assessing a proposed cognitive component of so-
cial anxiety. J Anxiety Disord. 2008;22(1):44–55. doi:10.1016/j.
janxdis.2007.08.002.
50. Weeks JW, Heimberg RG, Rodebaugh TL, Norton PJ. Exploring
the relationship between fear of positive evaluation and social anx-
iety. J Anxiety Disord. 2008;22(3):386–400. doi:10.1016/j.janxdis.
2007.04.009.
51. Horley K, Williams LM, Gonsalvez C, Gordon E. Social phobics
do not see eye to eye: a visual scanpath study of emotional expres-
sion processing. J Anxiety Disord. 2003;17(1):33–44. doi:10.1016/
S0887-6185(02)00180-9.
52. Horley K, Williams LM, Gonsalvez C, Gordon E. Face to face:
visual scanpath evidence for abnormal processing of facial expres-
sions in social phobia. Psychiatry Res. 2004;127(1–2):43–53. doi:
10.1016/j.psychres.2004.02.016.
53. Moukheiber A, Rautureau G, Perez-Diaz F, Jouvent R, Pelissolo A.
Gaze behaviour in social blushers. Psychiatry Res. 2012;200(2–3):
614–9. doi:10.1016/j.psychres.2012.07.017.
54. Moukheiber A, Rautureau G, Perez-Diaz F, Soussignan R, Dubal S,
Jouvent R, et al. Gaze avoidance in social phobia: objective mea-
sure and correlates. Behav Res Ther. 2010;48(2):147–51. doi:10.
1016/j.brat.2009.09.012.
55. Weeks JW, Howell AN, Goldin PR. Gaze avoidance in social anx-
iety disorder. Depress Anxiety. 2013;30(8):749–56. doi:10.1002/
da.22146.
56.•Howell AN, Zibulsky DA, Srivastav A, Weeks JW. Relations
among social anxiety, eye contact avoidance, state anxiety, and
perception of interaction performance during a live conversation.
Cogn Behav Ther. 2016;45(2):111–22. doi:10.1080/16506073.
2015.1111932.This study provides an interesting use of eye
tracking in the assessment of eye contact during a live social
interaction
57. Boll S, Bartholomaeus M, Peter U, Lupke U, Gamer M. Attentional
mechanisms of social perception are biased in social phobia. J
Anxiety Disord. 2016;40:83–93. doi:10.1016/j.janxdis.2016.04.
004.
58. Brunet PM, Heisz JJ, Mondloch CJ, Shore DI, Schmidt LA.
Shyness and face scanning in children. J Anxiety Disord.
2009;23(7):909–14. doi:10.1016/j.janxdis.2009.05.009.
59. Van Dillen LF, Enter D, Peters LPM, van Dijk WW, Rotteveel M.
Moral fixations: the role of moral integrity and social anxiety in the
selective avoidance of social threat. Biol Psychol. 2017;122:51–8.
doi:10.1016/j.biopsycho.2016.01.016.
60. Terburg D, Syal S, Rosenberger LA,Heany SJ, Stein DJ, van Honk
J. Testosterone abolishes implicit subordination in social anxiety.
Psychoneuroendocrinology. 2016;72:205–11. doi:10.1016/j.
psyneuen.2016.07.203.
59 Page 8 of 9 Curr Psychiatry Rep (2017) 19:59

61. Myllyneva A, Ranta K, Hietanen JK. Psychophysiological re-
sponses to eye contact in adolescents with social anxiety disorder.
Biol Psychol. 2015;109:151–8. doi:10.1016/j.biopsycho.2015.05.
005.
62. Schneier FR, Rodebaugh TL, Blanco C, Lewin H, Liebowitz MR.
Fear and avoidance of eye contact in social anxiety disorder. Compr
Psychiatry. 2011;52(1):81–7. doi:10.1016/j.comppsych.2010.04.
006.
63. Gilbert P. Evolution and social anxiety—the role of attraction, so-
cial competition, and social hierarchies. Psychiatr Clin North Am.
2001;24(4):723–51.
64. Wieser MJ, Pauli P, Alpers GW, Muehlberger A. Is eye to eye
contact really threatening and avoided in social anxiety?—an eye-
tracking and psychophysiology study. J Anxiety Disord.
2009;23(1):93–103. doi:10.1016/j.janxdis.2008.04.004.
65. Weeks JW, Howell AN. The bivalent fear of evaluation model of
social anxiety: further integrating findings on fears of positive and
negative evaluation. Cogn Behav Ther. 2012;41(2):83–95. doi:10.
1080/16506073.2012.661452.
66.•Chen NTM, Clarke PJF, MacLeod C, Hickie IB, Guastella AJ.
Aberrant gaze patterns in social anxiety disorder: an eye movement
assessment during public speaking. J Exp Psychopathol.2016;7(1):
1–17. doi:10.5127/jep.040313.This study examines the selective
attention biases which operate in SAD during a social-
evaluative speech task
67. Clarke PJF, Notebaert L, MacLeod C. Absence of evidence or ev-
idence of absence: reflecting on therapeutic implementations of
attentional bias modification. BMC Psychiatry. 2014;14(1):1–6.
doi:10.1186/1471-244X-14-8.
68. MacLeod C, Clarke PJF. The attentional bias modification approach
to anxiety intervention. Clin Psychol Sci. 2015;3(1):58–78. doi:10.
1177/2167702614560749.
69. Price RB, Wallace M, Kuckertz JM, Amir N, Graur S, Cummings L,
et al. Pooled patient-level meta-analysis of children and adults com-
pleting a computer-based anxiety intervention targeting attentional bi-
as. Clin Psychol Rev. 2016;50:37–49. doi:10.1016/j.cpr.2016.09.009.
70. Boettcher J, Berger T, Renneberg B. Internet-based attention train-
ing for social anxiety: a randomized controlled trial. Cogn Ther Res.
2012;36(5):522–36. doi:10.1007/s10608-011-9374-y.
71. Neubauer K, von Auer M, Murray E, Petermann F, Helbig-Lang S,
Gerlach AL. Internet-delivered attention modification training as a
treatment for social phobia: a randomized controlled trial. Behav
Res Ther. 2013;51(2):87–97. doi:10.1016/j.brat.2012.10.006.
72. Kuckertz JM, Amir N. Attention bias modification for anxiety and
phobias: current status and future directions. Curr Psychiatry Rep.
2015;17(2):9. doi:10.1007/s11920-014-0545-x.
73.•Price RB, Greven IM, Siegle GJ, Koster EH, De Raedt R. A novel
attention training paradigm based on operant conditioning of eye
gaze: preliminary findings. Emotion. 2016;16(1):110–6. doi:10.
1037/emo0000093.This study provides an innovative gaze-
contingent approach to training the selective attention biases
which maintain SAD
74.•Lazarov A, Pine DS, Bar-Haim Y. Gaze-contingent music reward
therapy for social anxiety disorder: a randomized controlled trial.
Am J Psychiatry. 2017:appi.ajp.2016.16080894. doi:10.1176/appi.
ajp.2016.16080894.This study provides an innovative gaze-
contingent approach to training the selective attention biases
which maintain SAD.
Curr Psychiatry Rep (2017) 19:59 Page 9 of 9 59