Journal of Affective Disorders 311 (2022) 17–30

Contents lists available at ScienceDirect

Journal of Affective Disorders

journal homepage: www.elsevier.com/locate/jad

Review Article

Social anxiety and behavioral assessments of social cognition: A

systematic review
Talha Alvi, Divya Kumar, Benjamin A. Tabak *
Department of Psychology, Southern Methodist University, 6116 N. Central Expressway, Suite 1300, Dallas, TX, USA

A R T I C L E I N F O A B S T R A C T

Keywords: Background: Social anxiety is highly prevalent and has increased in young adults during the COVID-19 pandemic.
Social anxiety Since social anxiety negatively impacts interpersonal functioning, identifying aspects of social cognition that may
Social cognition be impaired can increase our understanding of the development and maintenance of social anxiety disorder.
Empathy
However, to date, studies examining associations between social anxiety and social cognition have resulted in
Theory of mind
mixed findings.
Empathic accuracy
Affect sharing Methods: The aim of this systematic review was to summarize the literature on the association between social
anxiety and social cognition, while also considering several potential moderators and covariates that may in­
fluence findings.
Results: A systematic search identified 52 studies. Results showed mixed evidence for the association between
social anxiety and lower-level social cognitive processes (emotion recognition and affect sharing) and a trend for
a negative association with higher-level social cognitive processes (theory of mind and empathic accuracy). Most
studies examining valence-specific effects found a significant negative association for positive and neutral
stimuli.
Limitations: Not all aspects of social cognition were included (e.g., attributional bias) and we focused on adults
and not children, limiting the scope of the review.
Conclusions: Future studies would benefit from the inclusion of relevant moderators and covariates, multiple
well-validated measures within the same domain of social cognition, and assessments of interpersonal func­
tioning outside of the laboratory. Additional research examining the moderating role of attention or interpre­
tation biases on social cognitive performance, and the potential benefit of social cognitive skills training for social
anxiety, could inform and improve existing cognitive behavioral interventions.

1. Introduction including depression (Aderka et al., 2012). One mechanism through

Social anxiety disorder (SAD) is one of the most common mental
illnesses (Kessler et al., 2005), and levels of social anxiety (SA) have
increased in young adults since the onset of the COVID-19 pandemic
(Hawes et al., 2021). SAD is characterized by a fear of evaluation and
avoidance of social situations, which can negatively affect social func­
tioning (American Psychiatric Association, 2013; Clark and Wells,
1995). Indeed, individuals with SAD report having more interpersonal
problems and difficulty maintaining relationships (Davila and Beck,
2002; Kashdan et al., 2007; Tonge et al., 2020). These issues extend
beyond close relationships and can have a substantial negative impact
on occupational and educational functioning (Schneier et al., 1994;
Wittchen et al., 2000) above and beyond the effects of comorbidities
which SA may negatively impact interpersonal functioning is via alter­
ations in social cognitive processing (Morrison and Heimberg, 2013).
Social cognitive ability (i.e., the ability to accurately understand others'
thoughts and emotions; Brothers, 1990; Frith and Frith, 2007) has been
associated with beneficial outcomes in interpersonal relationships
beginning in childhood and extending into adulthood (Banerjee, 1997;
Gleason et al., 2009; Sened et al., 2017). Given the negative impact of SA
on interpersonal functioning (Alden and Taylor, 2004), a key predictor
of health and well-being (Holt-Lunstad, 2021), identifying specific as­
pects of social cognition that may be impaired can increase our under­
standing of the development and maintenance of SAD.
Although most studies using standardized measures of social cogni­
tion have included individuals with schizophrenia and autism spectrum

* Corresponding author at: Department of Psychology, Southern Methodist University, 6116 N. Central Expressway, Suite 1300, Dallas, TX 75206, USA.
E-mail address: btabak@smu.edu (B.A. Tabak).

https://doi.org/10.1016/j.jad.2022.04.130
Received 6 November 2021; Received in revised form 17 April 2022; Accepted 19 April 2022
Available online 28 April 2022
0165-0327/© 2022 Elsevier B.V. All rights reserved.
T. Alvi et al.
disorder, an increasing amount of research has focused on anxiety
symptomology, including SA. To date, three reviews and meta-analyses
have examined the association between SA and social cognition in
adults, but findings have been inconsistent. A potential reason for the
lack of consistency stems from the complexity of social cognition and the
extent to which different behavioral tasks index both related and distinct
aspects of social cognitive processes. One method of categorizing the
numerous domains of social cognition is to separate them into lower-
level and higher-level processes (Green et al., 2013). Lower-level pro­
cesses involve the recognition of basic social-emotional cues (e.g.,
emotion recognition; Green et al., 2008 and social cue perception; Penn
et al., 2002; Pinkham et al., 2014), as well as affective empathic pro­
cesses (i.e., shared feelings for others; Davis, 1983), such as affect
sharing, which is thought to be automatic and reflexive (Hatfield et al.,
1992; Singer and Lamm, 2009) and may precede cognitive processes
(Fan and Han, 2008). Higher-level social cognitive processes involve the
interpretation of complex social stimuli to make inferences about the
thoughts and intentions of others (e.g., theory of mind; Baron-Cohen
et al., 2001, and empathic accuracy; Ickes et al., 1990).
In the first review and meta-analysis of the association between SA
and social cognition, O’Toole et al. (2013) found a small negative effect
between SA and basic emotion recognition and a moderate negative
effect between SA and complex emotion recognition (higher-level social
cognition). However, Plana et al. (2014) found no association between
SA and emotion recognition, mentalizing ability, or social perception
(no association with lower- or higher-level processes). More recently,
Pittelkow et al. (2021) conducted a meta-analysis examining the asso­
ciation between SA and empathy (a multi-dimensional construct that is
often used synonymously with the term social cognition). They distin­
guished cognitive empathy (i.e., the ability to recognize and identify
others' emotions) from affective empathy (i.e., experiencing an emotion
based on another's emotional experience), both of which can also be
conceptualized as facets of social cognition. Pittelkow et al. (2021)
found a positive association between SA and affective empathy and no
association between SA and cognitive empathy. However, these results
included studies measuring empathic processes based on self-report,
which ask about one's perception of their own empathic tendencies or
abilities, as well as behavioral assessments of social cognition, which
examine the accuracy or congruence of one's responses in relation to
targets or normed reference groups. The distinction between self-report
and behavioral measures of social cognition is important given the weak
correlation between the two, which suggests that self-report measures
are not accurate proxies for ability (Murphy and Lilienfeld, 2019). Thus,
it is important for reviews to distinguish between self-report and
behavioral assessments when interpreting results to avoid conflating
findings for separable constructs. Due in large part to a lack of studies
available, previous reviews (O’Toole et al., 2013; Pittelkow et al., 2021;
Plana et al., 2014) have also not examined the association between SA
and empathic accuracy, a behavioral measure of higher-level social
cognition.
Beyond the type of social cognition assessed, previous reviews of SA
and social cognition have not always examined valence-specific associ­
ations of SA on social cognitive processes or the method of measuring SA
(i.e., dimensional measures, cutoff scores from dimensional measures, or
diagnostic groups). Based on studies showing that greater levels of SA
are associated with lower social cognitive ability for neutral and positive
stimuli (Alvi et al., 2020; Washburn et al., 2016), parsing apart whether
associations are valence-specific (between negative, neutral, and posi­
tive stimuli) may provide further clarity on associations between SA and
social cognition that are not apparent when collapsing across all stimuli.
Indeed, although Pittelkow et al. (2021) considered the potential
moderating role of positive and negative valence, they did not examine
the association between SA and social cognition for neutral stimuli.
Finally, noting whether SA is assessed dimensionally or categorically
may clarify previous findings. For example, when using the same facial
recognition task, some have found no association between SA and
18
Journal of Affective Disorders 311 (2022) 17–30
emotion recognition when SA is assessed dimensionally (Mullins and
Duke, 2004), while others have found a significant, positive association
when SA is categorized based on cut-off scores (Hunter et al., 2009).
Thus, considering the type of SA assessment may help to elucidate
whether associations are specific to individuals with the highest levels of
SA (i.e., those meeting criteria for SAD based on diagnostic interview),
or if associations are evident when assessing SA dimensionally.
In addition to moderators that may help explain mixed findings, the
inclusion of statistical covariates is also important to consider. Depres­
sion (Bora and Berk, 2016), gender (Babchuk et al., 1985; Doherty et al.,
1995; Magen and Konasewich, 2011; Thayer & Johnsen, 2000), age
(Ruffman et al., 2008), alexithymia (Di Tella et al., 2020), and IQ
(Brüne, 2003) have all been associated with social cognitive functioning
and/or SA, and therefore represent potentially competing predictors
that may account for associations between SA and social cognition. For
example, previous work has shown no association between SA and
certain forms of social cognition (i.e., affect sharing) when including
age, gender, depressive symptoms, alexithymia, and target expressivity
as covariates; however, the association becomes statistically significant
when covariates are removed (Alvi et al., in press). As a result, prior
reviews that have not compared findings from studies that accounted for
relevant covariates with those that did not, prevents an understanding of
how much these related factors may explain the associations shown
between SA and social cognitive ability. For example, the most recent
meta-analysis by Pittelkow et al. (2021) did not examine the extent to
which studies accounted for covariates such as depression and
alexithymia.
1.1. Review objectives
The present systematic review assessed whether there is an associ­
ation between SA and behavioral measures of social cognition, while
also considering several potential moderators and covariates that may
influence findings. The primary objectives were to determine trends in
the literature based on the: 1) type of SA assessment, 2) type of social
cognition assessment, 3) domain of social cognition, 4) valence of
stimuli, and 5) inclusion of covariates. By providing a more fine-grained
examination of the different domains of social cognition, and parsing
results based on potential moderators and covariates, our goal was to
help clarify the mixed associations between SA and social cognition that
have been found in prior reviews.
2. Method
The present study was designed in accordance with the Preferred
Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)
statement (Moher et al., 2015).
2.1. Literature search
Articles were identified through searches in PsycInfo, Psych and
Behavioral Sciences Collection, and PsycArticles from the earliest
available dates through September 26th, 2020. In all databases, the
following search string was used: (social anxiety OR social phobia) AND
(emotion recognition OR emotion knowledge OR emotion processing
OR theory of mind OR mentalizing OR mentalising OR social perception
OR empathic accuracy OR cognitive empathy OR social cognition OR
social cognitive ability OR affective empathy OR affect sharing).
Searches were limited to peer-reviewed articles written in English.
Forward and backward searching were used in selected articles to cap­
ture any additional relevant articles.
2.2. Inclusion and exclusion criteria
Inclusion criteria included: a) an adult sample (≥18 years old), b) a
measure of SA and/or a clinical, SA group, and c) at least one behavioral
T. Alvi et al.
measure of social cognition. Based on meta-analytic evidence that self-
report assessments of empathy are only weakly related to behavioral
assessments of social cognition (Murphy and Lilienfeld, 2019), in the
present review, we focus specifically on behavioral assessments. Exclu­
sion criteria included: a) use of a child/adolescent sample, b) no mea­
sure of SA or SA group, c) a sample that included comorbid diagnoses, d)
no behavioral measure of social cognition, and e) imaging studies with
no behavioral outcome. We chose to exclude studies containing in­
dividuals with comorbid disorders who did not have a primary diagnosis
of SA since the focus of our review is on the unique effects of SA on social
cognitive abilities. Importantly, recent evidence has shown that many
interpersonal difficulties experienced by individuals with SAD exist
above and beyond the effects of comorbid depression symptoms (Tonge
et al., 2020).
2.3. Task classification
Social cognition tasks were classified into their respective domains
using the framework proposed by Green et al. (2008). Emotion recogni­
tion tasks include those that prompt participants to identify a discrete
emotion (e.g., happiness, sadness) in response to static or video stimuli
(Heberlein et al., 2004; Joormann and Gotlib, 2006; Nowicki and
Carton, 1993). Tasks measuring social perception include presentation of
social scenes in which participants are asked to assess domains such as
nonverbal cues, status, and intimacy (Conzelmann et al., 2013;
Schroeder, 1995a, 1995b; Veljaca and Rapee, 1998). Measures of theory
of mind include static and video stimuli that prompt participants to infer
the mental states of others (Baron-Cohen, 1995; Baron-Cohen et al.,
2001; Dziobek et al., 2006; Tibi-Elhanany et al., 2011). Studies in the
present review assessed empathic accuracy for strangers using a video
task in which participants continuously rate their perceptions of a tar­
get's emotional state as the target tells an autobiographical story (Kern
et al., 2013). The targets also rate themselves, and the ratings of the
perceiver and target are then correlated to create an empathic accuracy
score. Studies in the present review assessed affect sharing using the same
empathic accuracy video task with altered instructions that ask partic­
ipants to rate their own feelings, as they watch the target individual
(Morrison et al., 2016). All tasks are considered measures of cognitive
empathy, apart from the affect sharing task, which measures affective
empathy.
2.4. Screening and coding procedure
Abstracts were independently screened by two raters using Abstrackr
(http://abstrackr.cebm.brown.edu), an online software tool for
screening and organizing literature searches. After the initial screen,
full-texts were reviewed independently by both raters and assessed for
inclusion criteria. Disagreements were discussed among raters and
resolved through consensus.
Studies were coded for the following: total sample, mean age,
percent female, assessment of SA, assessment of social cognition task,
statistical covariates included, and effect size. Studies were coded for all
covariates included, and then subsequently coded for specific covariates
of interest (depressive symptoms, gender, age, alexithymia, and IQ). For
studies that reported effect sizes, coding included both the metric re­
ported in the original paper and the equivalent using a common metric
(Cohen's d) for ease of interpretation. Coding of all included full-text
articles was completed by the first author. Further, 20% of the articles
were randomly selected and coded by the second author to confirm
agreement. Coders had 95% agreement prior to discussion and resolu­
tion of any disagreements.
19
Journal of Affective Disorders 311 (2022) 17–30
3. Results
3.1. Search results and study selection
An overview of study selection and flow can be seen in Fig. 1.
Database searches from PsycInfo, Psych and Behavioral Sciences
Collection, and PsycArticles resulted in 1196 records. Following dupli­
cate identification and removal (n = 148), 1048 unique records
remained. An additional 16 records were identified through forward and
backward searching, resulting in a total of 1064 records. After the first
round of abstract screening, 935 records were excluded for the following
reasons: no behavioral measure of social cognition (n = 436), no mea­
sure of SA or socially anxious group (n = 425), an imaging study with no
behavioral outcomes (n = 34), child/adolescent samples (n = 33), case
study (n = 7). The remaining 129 full-text articles were assessed for
eligibility, and 81 articles were excluded for various reasons: no
behavioral measure of social cognition (n = 67), no measure of SA or
socially anxious groups (n = 7), samples with comorbid diagnoses (n =
4), child/adolescent samples (n = 2), an imaging study with no behav­
ioral outcomes (n = 1). The final sample included 48 full-text articles.
3.2. Study characteristics
Across 48 full-text articles, 52 studies were identified. Since some of
the articles included multiple studies, the number of social cognitive
assessments and findings are sometimes greater than the total number of
studies. Table 1 shows an overview of study characteristics. Most studies
measured emotion recognition (n = 30), followed by theory of mind (n
= 11), social perception (n = 5), empathic accuracy (n = 3), and affect
sharing (n = 3). Sample size varied across studies (ranging from 20 to
1485 individuals), with an average total sample of 153.67 participants,
and the direction and size of effects varied across studies as well
(Table 2).
3.3. Lower-level social cognition
3.3.1. Emotion recognition
There were no apparent trends in findings from studies examining
the association between SA and emotion recognition. Results based on
SA assessment or emotion recognition assessment were also mixed apart
from video emotion recognition and vocal/prosody tasks. Valence-
specific findings were also mixed, however, there was some consis­
tency with statistically significant associations of SA on neutral stimuli
(i.e., all studies found greater SA to be associated with decreased ac­
curacy for neutral stimuli).
Of the 30 studies examining the association between SA and emotion
recognition, 16 studies found a statistically significant association and
14 studies found a non-significant association. Of the studies that noted
statistical significance, eight found a positive association (i.e., higher
levels of SA were related to greater emotion recognition ability) and 11
studies found a negative association (higher levels of SA were related to
lower emotion recognition ability). In studies that reported an effect
size, values ranged from d = 0–1.96, with an average effect size of d =
0.72.
When examining statistically significant valence-specific associa­
tions, for negative stimuli, six studies found a positive association, with
higher levels of SA associated with greater accuracy in identifying
stimuli as negative (Auyeung and Alden, 2020; Hunter et al., 2009;
Mohlman et al., 2007; Quadflieg et al., 2007; Torro-Alves et al., 2016;
Winton et al., 1995), and four studies found a negative association,
where higher levels of SA were related to less accuracy (Garner et al.,
2009; Gilboa-Schechtman et al., 2008; Oh et al., 2018; Tseng et al.,
2017). For positive stimuli, one study found a positive association
(Hunter et al., 2009) and three studies found a negative association
(Bodner et al., 2012; Oh et al., 2018; Quadflieg et al., 2007). Studies that
examined neutral stimuli had more consistency in findings; all three
T. Alvi et al. Journal of Affective Disorders 311 (2022) 17–30

Identification
Records idenfied through
database searching
(n = 1196)

Records aer duplicates removed

(n = 1048)
Screening

Records excluded
Records idenfied through (n = 935)
forward/backwards searching
(n = 16) No social anxiety (n =425)
No behavioral measure of
social cognion (n=436)
Child/adolescent sample
(n=33)
Records screened
Eligibility

Imaging study (n=34)

(n = 1064)
Case study (n=7)

Full-text arcles assessed Full-text arcles excluded

for eligibility (n = 81)
(n = 129)
No social anxiety (n = 7)
No behavioral measure of
Included

social cognion (n = 67)

Arcles included in Comorbid diagnosis (n = 4)
qualitave synthesis Child/adolescent sample (n = 2)
(n = 48) Imaging study (n = 1)

Fig. 1. PRISMA flow diagram

found a negative association between SA and emotion recognition for
neutral stimuli (Mohlman et al., 2007; Oh et al., 2018; Winton et al.,
1995).
As mentioned previously, most studies used diagnostic groups of
socially anxious individuals and healthy controls. In these studies, nine
found a statistically significant association and nine found a non-
significant association. Studies that created groups based on dimen­
sional assessment also had mixed results, with five finding a statistically
significant association (Bodner et al., 2012; Dickter et al., 2018; Hunter
et al., 2009; Torro-Alves et al., 2016; Winton et al., 1995) and three
finding a non-significant association (Button et al., 2013; Heuer et al.,
2010; Schofield et al., 2007). Finally, in studies using dimensional
assessment of SA, two found a statistically significant association
(Auyeung and Alden, 2020; Lau et al., 2014) and two found a non-
significant association (Alvi et al., 2020; Mullins and Duke, 2004).
Thus, there was no clear trend in findings based on category of SA
assessment.
In considering the type of emotion recognition assessment, in studies
that used a static, facial emotion recognition task, seven studies found a
statistically significant association and eight studies found a non-
significant association. Similarly, in studies that utilized a morphed
faces task, half reported a statistically significant association (Garner
et al., 2009; Gilboa-Schechtman et al., 2008; Montagne et al., 2006;
Torro-Alves et al., 2016) and half reported a non-significant association
(Heuer et al., 2010; Joormann and Gotlib, 2006; Philippot and Douilliez,
2005; Schofield et al., 2007). There was more consistency in results for
studies that used a video emotion recognition task, as three studies
found a statistically significant association (Auyeung and Alden, 2020;
Lau et al., 2014) and one found a non-significant association (Winton
et al., 1995). In addition, all three studies using a vocal/prosody task
found a statistically significant association (Bodner et al., 2012; Quad­
flieg et al., 2007; Tseng et al., 2017). There was only one study that used
an emotional perception of biological motion task (Alvi et al., 2020) and
one study that used an emotion intelligence task (M. Jacobs et al., 2008);
neither found a statistically significant association of SA on emotion
recognition.
3.3.2. Social perception
Overall, most studies found a statistically significant negative asso­
ciation between SA and social perception (i.e., greater SA was associated
with decreased social perception). Although findings with social
perception were generally consistent, there were few studies that
examined this domain of social cognition.
In studies that found a statistically significant association, four
studies found a negative association, with higher levels of SA relating to
reduced social perception (Hampel et al., 2011; Schroeder, 1995b;
Schroeder and Ketrow, 1997; Veljaca and Rapee, 1998) and one study
found a positive association between higher levels of SA and greater
social perception (Veljaca and Rapee, 1998). Only one study found a
non-significant association between SA and social perception
(Schroeder, 1995a). For studies that reported an effect size, values
ranged from d = 0.47–0.75, with an average effect size of d = 0.51.
Only one study examined valence with a moderator, finding a sig­
nificant negative association between SA and social perception for
positive stimuli, such that those with higher levels of SA had lower ac­
curacy in social perception (Veljaca and Rapee, 1998). Of the studies
that used a dimensional measure of SA, the majority found a statistically
significant association (Hampel et al., 2011; Schroeder, 1995b;
Schroeder and Ketrow, 1997) and one study did not (Schroeder, 1995a).
Further, the study that created groups based on dimensional assessment
of SA also found a statistically significant association (Veljaca and
Rapee, 1998). As most studies found a significant association, findings
by measures of social perception were consistent. The studies that uti­
lized a social intelligence task and social cue detection task found

20
T. Alvi et al. Journal of Affective Disorders 311 (2022) 17–30

Table 1
Overview of studies and study characteristics.
Total N Mean age % Sample type Statistical covariates Assessment of Social cognition task
(SA/ SA/HC women social anxiety
HC) SA/HC

Emotion recognition
Alvi et al. (2020) - 1485 25.8 69.0 Student and Age, gender, depressive Composite (LSAS, Emotion Perceptions of Biological Motion
Study 1 Community symptoms, neuroticism, SIAS, SPS) Task (Heberlein et al., 2004)
extraversion, mentalizing,
social anhedonia
Arrais et al. 231 22.3/ 61.5/ Student None Diagnostic groups Facial emotion recognition task using
(2010) (78/ 21.4 65.4 (based on SCID stimuli from Pictures of Facial Affect (
153) for DSM-IV) Ekman and Friesen, 1976)
Auyeung and 134 20.5 79.8 Student Depression, age SIAS-S Video empathy task (with social exclusion
Alden (2020) manipulation)
-Study 1
Auyeung and 126 29.3/ 60.3/ Community Depression, age Diagnostic groups Video empathy task (with social exclusion
Alden (2020) (63/ 29.9 61.9 (based on ADIS manipulation)
-Study 2 63) for DSM-IV)
Bell et al. (2011) 57 37/39 36.7/ Community Gender, medication status Diagnostic groups Morphed faces task using stimuli from
(30/ 40.7 (based on SCID Pictures of Facial Affect (Ekman and
27) for DSM-IV) Friesen, 1976)
Bodner et al. 80 28.6 40.0 Community Depression, age Groups based on Vocal improvisation recognition task (
(2012) (39/ LSAS Bodner and Gilboa, 2006) and vocal prosody
41) recognition task (Berger, 2002)
Button et al. 102 22/23 100/100 Community None Groups based on Facial emotion recognition task using
(2013) (52/ BFNE stimuli from the Karolinska Directed
50) Emotional Faces set (Lundqvist et al., 1998)
Campbell et al. 40 31.9/ 58.3/ Community None Diagnostic groups Facial emotion recognition task using
(2009) (12/ 30.4 35.7 (based on MINI stimuli from the Japanese and Caucasian
18) for DSM-IV) Facial Expressions of Emotion set (
Matsumoto and Ekman, 1988)
Dickter et al. 208 18.9 59.3 Student None Groups based on Facial emotion visual search task using
(2018) - Study (51/ SPAI stimuli from the NimStim Set of Facial
2 63) Expressions (Tottenham et al., 2009)
Garner et al. 33 43.1/ 68.8/ Community None Diagnostic groups Morphed faces task using stimuli from the
(2009) (16/ 39.9 47.1 (based on MINI NimStim Set of Facial Expressions (
17) for DSM-IV) Tottenham et al., 2009)
Gilboa- 202 29.9/ 51.9/ Community None Diagnostic groups Morphed faces task using stimuli from
Schechtman (54/ 27.6 52.3 (based on SCID database of facial expressions (Halberstadt
et al. (2008) - 65) for DSM-IV) and Niedenthal, 1997)
Study 2
Hagemann et al. 42 26.9/ 76.2/ Community None Diagnostic groups Facial emotion recognition task using
(2016) (21/ 27.0 76.2 (based on SCID stimuli from FACES Database (Ebner et al.,
21) for DSM-IV) 2010), Karolinska Directed Emotional Faces
(Lundqvist et al., 1998), Radboud Faces
Database (Langner et al., 2010) and the
NimStim Set of Facial Expressions (
Tottenham et al., 2009)
Heuer et al. 57 20.0/ 100/100 Student None Groups based on Morphed faces task using stimuli from the
(2010) (27/ 20.0 LSAS Karolinska Directed Emotional Faces set (
30) Lundqvist et al., 1998)
Hunter et al. 158 18.71 52.7 Student Gender, depression, trait Groups based on Diagnostic Analysis of Non-verbal Accuracy
(2009) (24/ anxiety SPS (Nowicki and Carton, 1993; Nowicki et al.,
121) 1998)
M. Jacobs et al. 49 32.4/ 46.4/ Community General anxiety, overall Diagnostic groups Mayer-Salovey-Caruso Emotional
(2008) (28/ 36.0 47.6 impairment (based on SCID Intelligence Test- Experiential Emotional
21) for DSM-IV) Intelligence (Mayer et al., 2002)
Joormann and 72 30.2/ 61.5/68 Community Error rate Diagnostic groups Morphed faces task using stimuli from the
Gotlib (2006) (26/ 31.6 (based on SCID Facial Expressions of Emotions-Stimuli and
25) for DSM-IV) Test set (Young et al., 2002)
Lau et al. (2014) 264 College- 69.3 Student Gender, neuroticism SAS-A Video Emotion Recognition Task (Kang and
aged Lau, 2013)
Mohlman et al. 52 21.5/ 62/65 Student Depression, trait anxiety, Diagnostic groups Facial emotion recognition task (with
(2007) (26/ 21.1 stimulus order (based on SCID anxiety manipulation) using stimuli from
26) for DSM-IV) Horstmann (2002)
Montagne et al. 50 36.7/ 58.3/ Community None Diagnostic groups Emotion Recognition Task (Montagne et al.,
(2006) (24/ 37.6 53.8 (based on MINI 2007)
26) for DSM-IV)
Mullins and Duke 70 19.2 100 Student Pre-test anxiety FNES Diagnostic Analysis of Non-verbal Accuracy
(2004) (Nowicki and Carton, 1993) (with anxiety
manipulations)
Oh et al. (2018) 112 27.3/ 46.4/ Community None Diagnostic groups Facial emotion recognition task using
(56/ 25.8 44.6 (based on MINI stimuli from Pictures of Facial Affect (
56) for DSM-IV) Ekman and Friesen, 1976)
Phan et al. (2006) 26.7/ 50.0/ Unknown None Facial emotion recognition task using Penn
26.6 50.0 Emotion Recognition set (Gur et al., 2002)
(continued on next page)

21
T. Alvi et al. Journal of Affective Disorders 311 (2022) 17–30

Table 1 (continued )
Total N Mean age % Sample type Statistical covariates Assessment of Social cognition task
(SA/ SA/HC women social anxiety
HC) SA/HC

20 Diagnostic groups
(10/ (based on SCID
10) for DSM-IV)
Philippot and 80 30.0/ 33.3/ Community None Diagnostic groups Morphed faces task using stimuli from Hess
Douilliez (21/ 33.0 51.3 (based on MINI and Blairy (1995)
(2005) 39) for DSM-IV)
Quadflieg et al. 30 23.3/ 53.3/ Student Depressive symptoms Diagnostic groups Emotion prosody recognition task using
(2007) (15/ 23.9 53.3 (based on SCID stimuli from the “Magdeburger Prosodie-
15) for DSM-IV) Korpus” (Wendt and Scheich, 2002)
Schofield et al. 100 18.5/ 75.5/ Student None Groups based on Morphed faces task using stimuli from the
(2007) (49/ 18.9 52.9 BFNE Japanese and Caucasian Facial Expressions
51) of Emotion set (Matsumoto and Ekman,
1988)
Straube et al. 20 25.0/ 60.0/ Student and None Diagnostic groups Facial emotion recognition task using
(2004) (10/ 23.2 60.0 Community (based on SCID stimuli from the NimStim Set of Facial
10) for DSM-IV) Expressions (Tottenham et al., 2009
Toro-Alves et al. 43 20.5/ 40.9/ Student Age Groups based on Morphed faces task using stimuli from the
(2016) (22/ 22.5 42.9 SPIN NimStim Set of Facial Expressions (
21) Tottenham et al., 2009
Tseng et al. 62 29.6/ 45.2/ Community Depression Diagnostic groups Diagnostic Analysis of Non-verbal Accuracy
(2017) (31/ 30.9 45.2 (based on MINI 2- Taiwan version (Chen, 2006)
31) for DSM-IV)
Winton et al. 24 20.6/ 69.2/ Student None Groups based on Facial emotion recognition using stimuli
(1995) (13/ 22.7 54.5 FNES from the Japanese and Caucasian Facial
11) Expressions of Emotion set (Matsumoto and
Ekman, 1988) and the Nonverbal
Discrepancy Test (DePaulo, 1978)
Yoon et al. (2007) 22 27.0/ 54.5/ Unknown None Diagnostic groups Facial emotion recognition task using Penn
(11/ 26.9 54.5 (based on SCID Emotion Recognition set (Gur et al., 2002)
11) for DSM-IV)

Social perception
Hampel et al. 110 31.5 60.0 Community Trait anxiety SBQ, SCQ, SAQ, Magdeburg Test of Social Intelligence-
(2011) and SAM Social Perception (Conzelmann et al., 2013)
Schroeder 84 22.0 59.5 Student None SCS Interpersonal Perception Task (Costanzo
(1995a) and Archer, 1989)
Schroeder 68 18.8 52.9 Student None SCS Interpersonal Perception Task (Costanzo
(1995b) and Archer, 1989)
Schroeder and 161 18–31 55.3 Student None PRCA Interpersonal Perception Task (Costanzo
Ketrow (1997) and Archer, 1989)
Veljaca and 39 22.4/ 89.5/ Student None Groups based on Social cue detection task
Rapee (1998) (19/ 24.2 70.0 APPQ and NOQ
20)

Affect sharing
Alvi et al. (in 202 19.8 68.2 Student Age, gender, depressive Composite (LSAS, Affect sharing video task adapted from Kern
press) - Study 1 symptoms, alexithymia, target SIAS, SPS) et al. (2013)
expressivity
Alvi et al. (in 324 37.1 45.4 Community Age, gender, depressive Composite (LSAS, Affect sharing video task adapted from Kern
press) - Study 2 symptoms, alexithymia, target SIAS, SPS) et al. (2013)
expressivity
Morrison et al. 64 31.9/ 43.7/ Community None Diagnostic groups EC video task adapted from Zaki et al.
(2016) (32/ 31.7 43.7 (based on ADIS (2008)
32) for DSM-IV)

Theory of mind
Alvi et al. (2020) - 1485 25.8 69.0 Student and Age, gender, depressive Composite (LSAS, Reading the Mind in the Eyes Test (Baron-
Study 1 Community symptoms, neuroticism, SIAS, SPS) Cohen et al., 2001)
extraversion, mentalizing,
social anhedonia
Ballespí et al. 113 21.1 85.8 Student None Groups based on Movie for the Assessment of Social
(2018) (33/ SPAI-B and BFNE Cognition task (Dziobek et al., 2006) and
80) induced mentalization in self-referential
paradigm
Buhlmann et al. 140 32.2/ 60.0/ Community Non-social inferencing (control Diagnostic groups Movie for the Assessment of Social
(2015) (35/ 32.7 60.0 items) (based on SCID Cognition task (Dziobek et al., 2006)
35) for DSM-IV)
Hezel and 80 26.5/ 67.5/85 Student and Gender Diagnostic groups Reading the Mind in the Eyes Test (Baron-
McNally (2014) (40/ 20.1 Community (based on MINI Cohen et al., 2001) and Movie for the
40) for DSM-IV) Assessment of Social Cognition task (
Dziobek et al., 2006)
Community
(continued on next page)

22
T. Alvi et al. Journal of Affective Disorders 311 (2022) 17–30

Table 1 (continued )
Total N Mean age % Sample type Statistical covariates Assessment of Social cognition task
(SA/ SA/HC women social anxiety
HC) SA/HC

M. Jacobs et al. 49 32.4/ 46.4/ General anxiety, overall Diagnostic groups Mayer-Salovey-Caruso Emotional
(2008) (28/ 36.0 47.6 impairment (based on SCID Intelligence Test- Strategic Emotional
21) for DSM-IV) Intelligence (Mayer et al., 2002)
Lenton-Brym 113 19.4/ 71.8/ Student None Groups based on Reading the Mind in the Eyes Test (Baron-
et al. (2018) (78/ 20.4 54.3 SPIN Cohen et al., 2001) and Movie for the
35) Assessment of Social Cognition task (
Dziobek et al., 2006)
Lyvers et al. 242 23.2 61.6 Student and Age, gender, depressive SIAS Reading the Mind in the Eyes Test (Baron-
(2019) Community symptoms, alexithymia Cohen et al., 2001)
Maleki et al. 107 27.5/ 54.3/ Community None Diagnostic groups Reading the Mind in the Eyes Test (Baron-
(2020) (35/ 28.4 48.6 (based on SCID Cohen et al., 2001) and faux pas task (Baron-
35) for DSM-IV) Cohen et al., 1999)
Sutterby et al. 56 19.1 59.3/ Student None Groups based on Reading the Mind in the Eyes Test (Baron-
(2012) (27/ 62.1 SPAI Cohen et al., 2001) and the Awareness of
29) Social Inferences Test (McDonald et al.,
2006)
Tibi-Elhanany 43 25.3/25 71.4/ Community Control condition Groups based on Cartoon ToM task (Baron-Cohen, 1995)
and Shamay- (45/ 45.5 LSAS
Tsoory (2011) 22)
Washburn et al. 119 19.8/ 58.3/ Student and Age, gender, ethnicity, Diagnostic groups Reading the Mind in the Eyes Test (Baron-
(2016) (12/ 18.7 65.0 Community education, response latency (based on SCID Cohen et al., 2001) and Movie for the
43) for DSM-IV) Assessment of Social Cognition task (
Dziobek et al., 2006)

Empathic accuracy
Alvi et al. (2020) - 390 19.6 77.2 Student Gender, depressive symptoms, Composite (LSAS, EA video task (Kern et al., 2013)
Study 2 neuroticism, extraversion, SIAS, SPS)
mentalizing, social anhedonia,
target expressivity, video order
Auyeung and 121 20.1 78.7 Student Age SIAS EA video task (with anxiety induction)
Alden (2016)
Morrison et al. 64 31.9/ 43.7/ Community None Diagnostic groups EA video task (Zaki et al., 2008)
(2016) (32/ 31.7 43.7 (based on ADIS
32) for DSM-IV)

Note. LSAS = Liebowitz Social Anxiety Scale, SIAS = Social Interaction Anxiety Scale, SIAS-S = Social Interaction Anxiety Scale-Straight-forward Score, SPS = Social
Phobia Scale, SCID = Structured Clinical Interview for DSM-IV, ADIS = Anxiety Disorders Interview Schedule for DSM-IV, BFNE = Brief Fear of Negative Evaluation
scale, MINI = Mini International Neuropsychiatric Interview for DSM-IV, SPAI = Social Phobia and Anxiety Inventory, SAS-A = Social Anxiety Scale for Adolescents,
FNES = Fear of Negative Evaluation Scale, SPIN = Social Phobia Inventory, SBQ = Social Behavior Questionnaire, SCQ = Social Cognitions Questionnaire, SAQ =
Social Attitudes Questionnaire, SCS = Self-consciousness Scale, PRCA = Personal Report of Communication Apprehension, APPQ = Albany Panic and Phobia
Questionnaire, NOQ = Negative Outcome Questionnaire, ToM = theory of mind, EA = empathic accuracy.

significant results. The studies that used the interpersonal perception
task had mixed findings; two studies found a statistically significant
association (Schroeder, 1995b; Schroeder and Ketrow, 1997) and one
did not (Schroeder, 1995a). Thus, overall, most studies examining SA
and social perception found a significant, negative association.
3.3.3. Affect sharing
There were no apparent trends in findings in studies examining the
association between SA and behaviorally assessed affect sharing based
on statistically significant associations, type of SA assessment, or valence
of stimuli.
In the three studies that examined the association between SA and
affect sharing, two studies found a non-significant association (Alvi
et al., in press) and one study found a statistically significant negative
association (higher levels of SA were related to lower affect sharing) that
was moderated by valence (Morrison et al., 2016). Specifically, SA was
negatively associated with affect sharing for positive stimuli (Morrison
et al., 2016). Effect sizes ranged from d = 0.02–0.55, with an average
effect size of d = 0.20.
3.4. Higher-level social cognition
3.4.1. Theory of mind
Most studies examining the association between SA and theory of
mind found a statistically significant, negative association (i.e., higher
levels of SA were associated with decreased theory of mind). Valence-
specific findings were also consistent regarding positive and neutral
stimuli (i.e., higher levels of SA were associated with reduced theory of
mind for stimuli with positive and neutral valence). Results based on
social cognitive assessment were also consistent in that nearly all studies
that assessed theory of mind based on the eye region found a statistically
significant association.
Overall, most studies found a statistically significant association (n =
8), with six studies finding a negative association indicating that higher
levels of SA were related to reduced theory of mind ability (Alvi et al.,
2020; Buhlmann et al., 2015; Hezel and McNally, 2014; Lyvers et al.,
2019; Maleki et al., 2020; Washburn et al., 2016), and three studies
finding a positive association, with higher levels of SA relating to better
theory of mind ability (Sutterby et al., 2012; Maleki et al., 2020; Tibi-
Elhanany et al., 2011). Further, four studies found a non-significant
association between SA and theory of mind (Ballespí et al., 2018; M.
Jacobs et al., 2008; Lenton-Brym et al., 2018; Maleki et al., 2020). For
studies that reported an effect size, values ranged from d = 0.06–0.86,
with an average effect size of d = 0.53.
Findings by valence were mixed, although most studies found a
statistically significant negative association (greater SA was related to
reduced theory of mind ability). For negative stimuli, two studies found
a statistically significant negative association (Alvi et al., 2020; Wash­
burn et al., 2016) and one study found a statistically significant positive
association between higher levels of SA and greater theory of mind
ability (Maleki et al., 2020). Three studies, which examined both posi­
tive and neutral stimuli, found that higher levels of SA levels were

23
T. Alvi et al. Journal of Affective Disorders 311 (2022) 17–30

Table 2 Table 2 (continued )

Study findings and effect sizes. Effect size Finding
Effect size Finding
Philippot and d = 0.34 No difference between groups
Emotion recognition Douilliez (2005) on accuracy
Alvi et al. (2020) - For total, d = 0.47; for No association between SA Quadflieg et al. Not reported SA group was less accurate
Study 1 negative, d = 0.44; for and emotion recognition (2007) for happy utterances, but
positive, d = 0.36; for more accurate for fearful and
neutral d = 0.40 sad utterances, than HC
Arrais et al. (2010) Not reported Group X gender interaction- Schofield et al. Not reported No differences between
women in SA group required (2007) groups on accuracy
less emotional intensity to Straube et al. Not reported No differences between
recognize faces displaying (2004) groups on accuracy
fear, happiness, and sadness Torro-Alves et al. Not reported SA group was more accurate
compared to HC (2016) for angry faces with 25%
Auyeung and Not reported Across both conditions, intensity than HC
Alden (2020) - higher SA predicted greater Tseng et al. (2017) d = 0.52 (across stimuli) and HC had greater accuracy than
Study 1 accuracy of target’s negative d = 0.67 (for fear accuracy) SA group across facial and
affect vocal accuracy; SA group was
Auyeung and d = 0.55 Across both conditions, SAD less accurate in fear accuracy
Alden (2020) - group had greater accuracy of for faces
Study 2 target negative affect Winton et al. Not reported SA group was more accurate
compared to HC (1995) at identifying negative facial
Bell et al. (2011) Not reported No differences between expressions, and less accurate
groups on accuracy at neutral expressions, than
Bodner et al. d = 0.94 SA group was less accurate at HC; no difference in groups
(2012) recognition of happy voices in for videos (facial/body and
female voices compared to auditory stimuli)
HC Yoon et al. (2007) Not reported No difference between groups
Button et al. Not reported No difference between groups on accuracy
(2013) on accuracy
Campbell et al. Not reported No difference between groups
Social perception
(2009) on accuracy
Hampel et al. d = 0.52 Negative association between
Dickter et al. d = 0.37 For complex faces, high SA
(2011) SA and social perception
(2018) - Study 2 group was less accurate than
Schroeder (1995a) d = 0.47 No association between SA
low SA group
and interpersonal perception
Garner et al. d = 0.64 SA group was less accurate at
Schroeder (1995b) d = 0.49 Negative association between
(2009) identifying fearful faces than
SA and interpersonal
HC
perception
Gilboa- d = 0.55 SA group less sensitive to
Schroeder and d = 0.54 Negative association between
Schechtman angry expressions than HC
Ketrow (1997) SA and interpersonal
et al. (2008) -
perception
Study 2
Veljaca and Rapee Not reported SA group was more accurate
Hagemann et al. d = 0.55 No difference between groups
(1998) in negative social cue
(2016) on accuracy
detection, and less accurate at
Heuer et al. (2010) d=0 No difference between groups
positive cue detection,
on accuracy
compared to HC
Hunter et al. Not reported SA group was more accurate
(2009) for happy, sad, and fearful
faces compared to HC Affect sharing
M. Jacobs et al. d = 1.62 No difference between groups Alvi et al. (in d = 0.04 No effect of SA on affect
(2008) on accuracy; negative press) - Study 1 sharing
correlation between SA and Alvi et al. (in d = 0.02 No effect of SA on affect
experiential EI press) - Study 2 sharing
Joormann and Not reported No difference between groups Morrison et al. d = 0.55 SA group had less congruence
Gotlib (2006) on accuracy; SA group was (2016) for positive stimuli, compared
able to identify angry faces to HC
more quickly than HC
Lau et al. (2014) d = 0.84 Lower emotion recognition
associated with lower social Theory of mind
anxiety Alvi et al. (2020) - For total, d = 0.86; for Negative association between
Mohlman et al. d = 1.66 (for positive SA group was more accurate Study 1 negative, d = 0.77; for SA and accuracy for total,
(2007) association); d = 1.96 (for at identifying angry faces, positive, d = 0.44; for negative, positive, and
negative association) and less accurate at neutral neutral d = 0.71 neutral stimuli
Ballespí et al. Not reported No difference between groups
faces, than HC, following
(2018) on accuracy in video task;
threat manipulation
Montagne et al. Not reported SA group was less sensitive to hyper-mentalizing in SA
(2006) negative emotions compared group for self-referential
to HC paradigm
Mullins and Duke Not reported No association between SA Buhlmann et al. d = 0.72 SA group was less accurate
(2004) and accuracy, regardless of (2015) than HC in ToM
condition Hezel and McNally d = 0.61 (for videos task) SA group was less accurate on
(2014) and d = 0.70 (for eyes task) both the video and eyes tasks
Oh et al. (2018) Not reported SA group was less accurate
compared to HC
for total, fear, surprise,
neutral, and happy stimuli M. Jacobs et al., d = 0.12 No difference between groups
than HC (2008) on accuracy
Phan et al. (2006) Not reported No difference between groups Lenton-Brym et al. d = 0.06 No difference between groups
on accuracy (2018) on accuracy
d = 0.45
(continued on next page)

24
T. Alvi et al. Journal of Affective Disorders 311 (2022) 17–30

Table 2 (continued ) findings were mixed. There were no apparent trends in findings based on
Effect size Finding statistically significant associations, type of SA assessment, or valence of
stimuli.
Lyvers et al. Negative association between
(2019) SA and ToM
Of the three studies that assessed empathic accuracy, two studies
Maleki et al. Not reported SA group had lower ToM on found a statistically significant association between SA and empathic
(2020) eyes task compared to HC; for accuracy (Alvi et al., 2020; Auyeung and Alden, 2016) and one study
positive and neutral valence, found a non-significant association (Morrison et al., 2016). However,
HC had greater ToM than
the study that found a non-significant association between SA and
SAD group; for negative
valence, HC had lower ToM empathic accuracy had a smaller sample (n = 64) compared to the other
than SAD group; No two (ns = 121–390); thus, these results should be interpreted with
difference between groups on caution. Of the studies that found a statistically significant association,
faux pas accuracy one study found that higher levels of SA were related to reduced
Sutterby et al. d = 0.67 Group X gender interaction
(2012) for tasks; women in SA group
empathic accuracy (Alvi et al., 2020) and the other found the opposite:
were more accurate higher levels of SA were related to increased empathic accuracy
compared to women in HC (Auyeung and Alden, 2016). Findings by valence were also mixed. One
group study found that higher levels of SA were related to reduced empathic
Tibi-Elhanany Not reported SA group had greater
accuracy for positive stimuli (Alvi et al., 2020). In contrast, another
et al. (2011) cognitive accuracy, but less
affective accuracy, than HC study found that higher levels of SA were associated with greater
Washburn et al. d = 0.59 SA group had lower ToM on empathic accuracy for negative stimuli (Auyeung and Alden, 2016).
(2016) eyes task for total, positive, Further, the study that utilized diagnostic groups found a non-significant
and neutral stimuli compared association (Morrison et al., 2016) whereas the two studies that assessed
to HC; no effect for video task
SA dimensionally did find a statistically significant association (Alvi
et al., 2020; Auyeung and Alden, 2016).
Empathic accuracy
Alvi et al. (2020)- Not reported Higher negative association
Study 2 between SA and positive, 3.5. Covariates
compared to negative, stimuli
Auyeung and Not reported Group X condition Twenty-six (50.00%) studies included at least one statistical covar­
Alden (2016) interaction; SA associated iate in their model examining the association between SA and social
greater accuracy for target
negative affect when faced
cognition. Of the studies that did include covariates, 17 (65.38%)
with social threat included two or less covariates. Across these studies, statistical cova­
Morrison et al. d=0 No difference between groups riates included age (11 studies; 42.31%), gender (11 studies; 42.31%),
(2016) on accuracy depression (13 studies; 50.00%), alexithymia (3 studies; 11.54%), and
Note. Effect sizes were converted to Cohen's d to compare effect sizes using a no studies statistically controlled for IQ. Most studies included age and
standard unit; SA = social anxiety; SAD = social anxiety disorder; HC = healthy gender as covariates, and several included depression; yet for all three
controls; ToM = theory of mind. variables, there was no clear trend in the patterns of results that included
nonsignificant and significant associations in both the positive and
associated with reduced theory of mind (Alvi et al., 2020; Maleki et al., negative directions (i.e., with higher levels of SA relating to both
2020; Washburn et al., 2016). Of the studies that used diagnostic groups, increased and decreased social cognitive ability) and effect sizes ranging
four found a statistically significant association (Buhlmann et al., 2015; from small to large. In the only study that controlled for alexithymia
Hezel and McNally, 2014; Maleki et al., 2020; Washburn et al., 2016) when examining SA and theory of mind, a negative association was
and one found a non-significant association (M. Jacobs et al., 2008). found with a moderate effect size wherein individuals with higher levels
Studies that created groups based on dimensional assessment had mixed of SA had reduced theory of mind ability (Lyvers et al., 2019). Inter­
results, with two studies finding a statistically significant association estingly, among the three studies examining SA and affect sharing, a
(Ballespí et al., 2018; Tibi-Elhanany et al., 2011) and two studies finding significant association was only found for the one study that did not
a non-significant association (Lenton-Brym et al., 2018; Sutterby et al., control for alexithymia or depression (Morrison et al., 2016), whereas
2012). The two studies that used dimensional assessment of SA were the other two studies that did control for these variables found no as­
consistent, with both finding a statistically significant association (Alvi sociation (Alvi et al., in press). These results suggest that covariates,
et al., 2020; Lyvers et al., 2019). particularly depression and alexithymia, may be most relevant when
The majority of studies used a theory of mind task based on eye- studying affect sharing; however, future studies are needed to replicate
region, and nearly all found a statistically significant association (Alvi these findings.
et al., 2020; Hezel and McNally, 2014; Lyvers et al., 2019; Maleki et al.,
2020; Sutterby et al., 2012; Washburn et al., 2016), except one (Lenton- 3.6. Trends in results
Brym et al., 2018). Studies that used video-based theory of mind tasks
had mixed results, as three studies found a statistically significant as­ Most studies used diagnostic groups based on clinical interviews for
sociation (Buhlmann et al., 2015; Hezel and McNally, 2014; Sutterby the DSM-IV (45.9%) or cut-off scores through dimensional measures
et al., 2012) and three studies did not (Ballespí et al., 2018; Lenton-Brym (27.9%) in their assessment of SA. Only 26.2% of studies used dimen­
et al., 2018; Washburn et al., 2016). There was only one study that used sional assessment of SA in their analyses. Overall, there was no general
a strategic emotional intelligence task (M. Jacobs et al., 2008) and one trend associated with the type of SA assessment and findings were
study that used a faux pas task (Maleki et al., 2020); neither found a mixed. Most studies examining social perception (80%), theory of mind
statistically significant association of SA. Finally, one study used a (66.6%), and empathic accuracy (66.6%) found a statistically significant
cartoon theory of mind task and found a significant association between association. However, these domains, especially social perception and
SA and theory of mind (Tibi-Elhanany et al., 2011). empathic accuracy, had far fewer studies than others, such as emotion
recognition. Findings related to lower-level processes (emotion recog­
3.4.2. Empathic accuracy nition, social perception, and affect sharing) were mixed. When these
There were only a few studies examining empathic accuracy, and domains were collapsed, no general trend was found, which was largely
driven by the heterogeneity in findings with emotion recognition. For

25
T. Alvi et al.
studies examining higher-level processes (theory of mind and empathic
accuracy), the majority (66.6%) found a statistically significant associ­
ation. Further, most of these studies (63.7%) found a negative associa­
tion between SA and higher-level social cognition, with those with
higher levels of SA exhibiting reduced abilities. Among studies that
found a valence-specific effect, most found negative associations for
positive (90%) and neutral (100%) stimuli, whereas findings for nega­
tive stimuli were mixed. In the two studies that examined gender
moderation, both found a positive association between SA and social
cognition (emotion recognition and theory of mind) for females only.
Among the four studies that included experimental anxiety inductions
(Table 1), results were mixed and no trend in findings could be identified
based on the type of study design.
4. Discussion
This systematic literature review synthesized studies in adults that
investigated the association between SA and behaviorally assessed social
cognition. Further, it considered trends across studies based on the type
of SA and social cognition assessment, the domain of social cognition,
relevant covariates, and the valence of stimuli. The main finding of this
review was that most studies examining the association between SA and
higher-level social cognition found a statistically significant effect, with
the majority finding a negative association (higher levels of SA relating
to less social cognitive ability). These findings are consistent with
O’Toole et al. (2013) who found a negative association between SA and
complex emotions (higher-level social cognition) and suggest that
greater SA is associated with impairment in higher-level social cogni­
tion. Findings related to lower-level social cognition were mixed,
although the majority of studies examining the domain of social
perception found a statistically significant negative effect of SA. Most
studies did not examine valence-specific effects, but among those that
did, findings suggest SA is negatively associated with the ability to
accurately perceive positive and neutral stimuli.
The mixed findings associated with lower-level social cognitive
processes may be explained by several factors. First, the assessments
used to measure lower-level processes (e.g., emotion recognition) have
been more varied than the assessments used to measure other domains
of social cognition; this increased variability may explain the differences
in findings. Indeed, the majority of studies assessing higher-level pro­
cesses used the same measures of social cognition (81% of studies
examining theory of mind utilized one or both of the same tasks, and all
of the studies examining empathic accuracy utilized a version of the
same task). In contrast, emotion recognition tasks varied greatly in terms
of the stimuli presented (i.e., the assessment of facial, vocal, or body
cues), as well as the sensory modality (i.e., videos, pictures, sounds).
Further, the greater number of studies that examined emotion recogni­
tion relative to other forms of social cognition, in conjunction with the
different forms of assessment, contributes to this variability. The
discrepancy in the number of studies between domains of social cogni­
tion may be due in part to feasibility. Most studies examining emotion
recognition used a static emotion recognition task, which are easily
administered, scored, and analyzed. In comparison, behavioral mea­
sures of empathic accuracy and affect sharing often require specific
computer software to run and analyze data, greater time for adminis­
tration, and advanced statistical techniques (e.g., multilevel modeling).
Finally, the mixed findings with lower-level social cognition may also be
explained by the infrequent inclusion of covariates. The majority of
studies examining these domains either included no covariates, or only a
few (only 17.3% of all studies included more than two covariates), and
these were generally limited to age, gender, and depressive symptoms.
Future studies are needed to systematically examine the extent to which
the factors we have identified contribute to increased replicability (e.g.,
do findings replicate using the same sample and covariates but with
different measures of emotion recognition).
Evidence for a negative association between SA and positive or
26
Journal of Affective Disorders 311 (2022) 17–30
neutral (but not negative) stimuli may be explained by socially anxious
individuals' attentional biases away from positive information (Taylor
et al., 2010), and the tendency to suppress positive emotions (Farmer
and Kashdan, 2012), which can result in the misinterpretation of these
social cues. Further, socially anxious individuals tend to interpret
ambiguous stimuli more negatively (Morrison and Heimberg, 2013),
which may explain the negative association between SA and accuracy
for neutral cues. Future studies should include measures of cognitive
biases to test whether associations between SA and social cognition are
moderated by such attentional and/or interpretation biases. Evidence of
such moderation would suggest that social cognition in socially anxious
individuals may be improved by incorporating attention training toward
positive stimuli (Heeren et al., 2012; Li et al., 2008), or positive affect
treatment (Craske et al., 2019), within cognitive behavioral therapy.
Results from our review also found no apparent trend in findings of
studies that included relevant covariates (age, gender, depressive
symptoms, alexithymia, and IQ). Studies that included these covariates
had mixed findings with effect sizes ranging from small to large. These
results were comparable to studies that did not include covariates,
suggesting that the variability in findings to date cannot be explained by
the inclusion of covariates alone. Studies that included gender as a
moderator found similar results (i.e., a positive association between SA
and social cognition for females), which is supported by previous work
demonstrating increased social cognitive ability in women (Babchuk
et al., 1985; Thayer & Johnsen, 2000). However, only two studies
examined this moderation, so future studies are needed to replicate this
across domains of social cognition. Importantly, when examining spe­
cific domains of social cognition, alexithymia and depression seem to be
relevant to consider when assessing affect sharing. Indeed, alexithymia
is highly correlated with SA (Alvi et al., in press) and may mediate the
association between SA and affect sharing (Morrison et al., 2016). Given
the instructions in the affect sharing task that were used in the studies
reviewed (i.e., assessing how you as a perceiver feel when watching a
target recount an autobiographical story), it is likely that difficulty
identifying emotions may impact performance on this task. Similarly,
depression is highly comorbid with SA (Kessler et al., 2005) and shares
an affective profile with SA of high negative and low positive affect
(Kashdan, 2002), highlighting the importance of controlling for one
when examining the other. As with studies that examined gender
moderation, future work is needed to replicate the few findings of SA
with affect sharing with a particular emphasis on these variables.
Although findings based on the type of social cognitive assessment
within each domain were mixed, there were some noticeable trends. For
example, although only two studies included emotion recognition tasks
involving vocal/prosody stimuli, both found a negative association with
SA (Bodner et al., 2012; Quadflieg et al., 2007). Studies measuring
theory of mind with the Reading the Mind in the Eyes Test (RMET;
Baron-Cohen et al., 2001) were also generally consistent in their asso­
ciation with SA (71.4% found a statistically significant negative asso­
ciation). This trend suggests that the construct measured by this task,
which has been conceptualized as theory of mind (Baron-Cohen et al.,
2001), but also emotion recognition (Oakley et al., 2016), is reliably
associated with symptoms of SA. Although this task may be more closely
associated with emotion recognition, it differs from most lower-level
emotion recognition tasks that rely on basic emotions (e.g., happy,
sad, angry) and instead assesses complex emotions (e.g., jealous,
cautious, grateful) with a restricted amount of information from facial
expressions (i.e., only the eye region is presented). Since the RMET likely
lies within a grey area between lower (i.e., emotion recognition) and
higher-level social cognition (i.e., theory of mind), it may help to
characterize this task as an intermediate-level social cognitive process
moving forward. Alternatively, it could be characterized as theory of
mind but distinguished by its assessment of decoding ability, rather than
reasoning, due to the lack of context or additional information provided
with the images (Sabbagh, 2004).
Importantly, when examining studies based on the use of a decoding
T. Alvi et al.
task (i.e., the RMET) vs. a reasoning task (i.e., MASC), results to date are
still comparable. Most studies utilizing the MASC also found a signifi­
cant negative association (i.e., 60% of studies), suggesting separation of
findings based on decoding and reasoning tasks may not change the
pattern of results. However, since far more studies of SA and social
cognition have included the RMET compared to the MASC, this pattern
of results should be viewed as preliminary. In sum, although most
studies examining what we initially categorized as higher-level social
cognitive processes found an association with SA, this finding is largely
driven by studies that utilized the Reading the Mind in the Eyes Test.
Thus, future within-person studies are needed to examine differential
effects of SA on theory of mind that involves decoding vs. reasoning.
4.1. Limitations and future directions
Our systematic review has several limitations that are important to
point out. First, all aspects of social cognition were not included. This
review focused on processes and tasks that are based on accuracy (or the
correlation between perceiver and target responses as in the affect
sharing task), however previous research has included domains such as
attributional and attentional biases within social cognition (Green et al.,
2012; Pinkham et al., 2016). Future reviews should examine findings in
these additional areas to further understand the effect of SA on social
cognition. Additionally, this review did not include unpublished studies,
which may bias findings. Although unpublished studies may be of lesser
quality due to the absence of the peer review process (Egger et al.,
2003), publication bias may result from the exclusion of important
findings in unpublished work. Indeed, unpublished studies often have
smaller effects than published work (Hopewell et al., 2007); thus, the
exclusion of grey literature may artificially inflate the overall signifi­
cance and the effects of findings in the current review. Further, although
this review focused on adult samples, SA may be related to social
cognitive impairments in children and adolescents as well (Pittelkow
et al., 2021). Future reviews should include studies in children and
adolescents to determine if the trends in the present findings based on
social cognitive domains, assessments, stimuli valence, and the inclusion
of relevant covariates mirror those in studies with adult samples.
Our review also identified that research on SA and social cognition is
far less common in some domains (e.g., affect sharing) compared to
others (e.g., emotion recognition). Future studies should focus on more
understudied areas to examine whether effects from previous studies are
replicated. In addition, given the heterogeneity among measures within
each social cognitive domain, future studies examining SA and social
cognition should include large samples with well-powered designs
involving multiple measures of the same social cognitive domain to
confirm that associations are domain rather than task-specific. Since
behavioral tasks assessing social cognition are known to have poor
psychometric properties (Pinkham et al., 2014), future research would
also benefit from the inclusion of more well-validated tasks, such as the
Geneva Emotion Recognition Test (Schlegel et al., 2019). Although our
review did not identify differences based on type of SA assessment, it is
difficult to interpret these findings based on only a few studies that have
measured SA dimensionally. Symptoms of psychological distress are
thought to be more accurately operationalized using dimensional mea­
sures (Kessler, 2002), and creating groups based on cutoffs and relying
on diagnostic groups may restrict the range of symptoms. Thus, future
studies may further elucidate the association of SA and social cognition
if SA is measured dimensionally.
Finally, future research examining the association between SA and
social cognition would also benefit from the assessment of real-world
social functioning. Although it is typically assumed that laboratory-
based assessments of social cognitive performance correlate with real
world functioning, there is limited work in this area as it pertains to SA.
Studies employing daily diary, or ecological momentary assessment
(Hur et al., 2019), would be particularly beneficial in this regard as this
would allow researchers to extend their findings beyond the laboratory
27
Journal of Affective Disorders 311 (2022) 17–30
to examine whether differences in social cognition among socially
anxious individuals translate into difficulties outside of the laboratory.
5. Conclusion and implications
In sum, this systematic review highlights several important findings.
First, although there is mixed evidence for the association between SA
and lower-level social cognitive processes, there is a trend in studies
showing negative associations between SA and intermediate to higher-
level social cognitive processes (theory of mind and empathic accu­
racy). Second, findings across studies suggest that stimuli valence is
important to consider as a moderator in these associations. Of the studies
that examined valence-specific associations, most found a significant
negative association between SA and the ability to accurately perceive
positive and neutral stimuli. Third, this review highlights the lack of
inclusion of statistical covariates across studies, which may be most
relevant when examining SA and affect sharing. To further clarify the
relation between SA and social cognition, future studies would benefit
from the inclusion of relevant moderators and covariates, multiple
measures within the same domain of social cognition, and dimensional
assessment of SA.
Findings related to impairment in higher-level social cognitive pro­
cesses among those with SA symptoms may be explained cognitive
models of SA, which propose that social cognitive misinterpretations
result in and maintain symptoms (Beck, 2011; Clark and Wells, 1995).
For example, individuals with SA may misinterpret social cues in
everyday life resulting in automatic negative thoughts. These negative
interpretations impact social behaviour (e.g., through avoidance of so­
cial situations) and interpersonal functioning. Thus, impairment in
higher-level social cognitive processes, which includes mental state
inference of complex social stimuli, is consistent with this theoretical
understanding of SA. Likewise, this aligns with the cognitive model of
SA which posits that emotions and behaviors are influenced by in­
dividuals' thoughts (Beck, 2011). Thus, these findings highlight the
utility of focusing on cognitive, as opposed to emotional, processes in
psychotherapy for SAD. In addition, based on the present findings, it
may be more beneficial for clinicians to focus on higher-level, rather
than lower-level social cognitive processes. Indeed, our findings suggest
that trials examining the efficacy of social skills training in combination
with cognitive behavioral therapy (Beidel et al., 2014) may benefit from
even more targeted social cognitive skills training (Kurtz et al., 2016).
Future research using social cognitive interventions in individuals with
SA is needed to examine these potential clinical implications.
CRediT authorship contribution statement
Talha Alvi: Conceptualization, Data curation, Formal analysis,
Investigation, Methodology, Project Administration, Writing-original
draft, Writing-reviewing and editing. Divya Kumar: Formal analysis,
Project administration, Writing-review & editing. Benjamin A. Tabak:
Conceptualization, Supervision, Writing-review & editing.
Role of the funding source
This research did not receive any specific grant from funding
agencies in the public, commercial, or not-for-profit sectors.
Conflict of interest
All authors declare no conflicts of interest.
Acknowledgements
We thank Dr. Austin Baldwin for providing helpful feedback on an
earlier draft of this manuscript.
T. Alvi et al. Journal of Affective Disorders 311 (2022) 17–30

References Conzelmann, K., Weis, S., Süß, H.-M., 2013. New findings about social intelligence.
J. Individ. Differ. 34 (3), 119–137. https://doi.org/10.1027/1614-0001/a000106.
Costanzo, M., Archer, D., 1989. Interpreting the expressive behavior of others: the
Aderka, I.M., Hofmann, S.G., Nickerson, A., Hermesh, H., Gilboa-Schechtman, E.,
interpersonal perception task. J. Nonverbal Behav. 13, 225–245. https://doi.org/
Marom, S., 2012. Functional impairment in social anxiety disorder. J. Anxiety
10.1007/BF00990295.
Disord. 26 (3), 393–400. https://doi.org/10.1016/J.JANXDIS.2012.01.003.
Craske, M.G., Meuret, A.E., Ritz, T., Treanor, M., Dour, H., Rosenfield, D., 2019. Positive
Arrais, K.C., Machado-de-Sousa, J.P., Trzesniak, C., Filho, A.S., Ferrari, M.C.F., Osório, F.
affect treatment for depression and anxiety: a randomized clinical trial for a core
L., Loureiro, S.R., Nardi, A.E., Hetem, L.A.B., Zuardi, A.W., Hallak, J.E.C., Crippa, J.
feature of anhedonia. J. Consult. Clin. Psychol. 87 (5), 457–471. https://doi.org/
A.S., 2010. Social anxiety disorder women easily recognize fearfull, sad and happy
10.1037/ccp0000396.
faces: the influence of gender. J. Psychiatr. Res. 44 (8), 535–540. https://doi.org/
Davila, J., Beck, J.G., 2002. Is social anxiety associated with impairment in close
10.1016/j.jpsychires.2009.11.003.
relationships?A preliminary investigation. Behavior Therapy 33 (3), 427–446.
Alden, L.E., Taylor, C.T., 2004. Interpersonal processes in social phobia. Clin. Psychol.
https://doi.org/10.1016/S0005-7894(02)80037-5.
Rev. 24 (7), 857–882. https://doi.org/10.1016/j.cpr.2004.07.006.
Davis, M.H., 1983. Measuring individual differences in empathy: evidence for a
Alvi, T., Kouros, C.D., Lee, J., Fulford, D., Tabak, B.A., 2020. Social anxiety is negatively
multidimensional approach. J. Pers. Soc. Psychol. https://doi.org/10.1037/0022-
associated with theory of mind and empathic accuracy. J. Abnorm. Psychol. 129 (1)
3514.44.1.113.
https://doi.org/10.1037/abn0000493.
DePaulo, B.M., 1978. Decoding discrepant nonverbal cues. J. Pers. Soc. Psychol. 36 (3),
Alvi, T., Rosenfield, D., Sunahara, C.S., Wallmark, Z., Lee, J., Tabak, B.A., 2022.
313–323. https://doi.org/10.1037/0022-3514.36.3.313.
Examining unique associations of social anxiety and depression on behaviorally
Di Tella, Marialaura, Adenzato, Mauro, Catmur, Caroline, Miti, Francesca, Castelli, Lorys,
assessed affective empathy. Clin. Psychol. Sci. in press.
Ardito, Rita, 2020. The role of alexithymia in social cognition: Evidence from a non-
American Psychiatric Association, 2013. DSM V. In: Diagnostic and Statistical Manual of
clinical population. Journal of Affective Disorders 273, 482–492. https://doi.org/
Mental Disorders, Fifth edition. https://doi.org/10.1109/mwc.2007.4300985.
10.1016/j.jad.2020.05.012.
Auyeung, K., Alden, L.E., 2020. Accurate empathy, social rejection, and social anxiety
Dickter, C.L., Burk, J.A., Fleckenstein, K., Kozikowski, C.T., 2018. Autistic traits and
disorder. Clin. Psychol. Sci. 8 (2), 266–279. https://doi.org/10.1177/
social anxiety predict differential performance on social cognitive tasks in typically
2167702619885410.
developing young adults. PLoS ONE 13 (3). https://doi.org/10.1371/journal.
Auyeung, K., Alden, L.E., 2016. Social anxiety and empathy for social pain. Cogn. Ther.
pone.0195239.
Res. https://doi.org/10.1007/s10608-015-9718-0.
Doherty, R.W., Orimoto, L., Singelis, T.M., Hatfield, E., Hebb, J., 1995. Emotional
Babchuk, W.A., Hames, R.B., Thompson, R.A., 1985. Sex differences in the recognition of
contagion: gender and occupational differences. Psychol. Women Q. 19, 355–371.
infant facial expressions of emotion: the primary caretaker hypothesis. Ethol.
https://doi.org/10.1111/j.1471-6402.1995.tb00080.x.
Sociobiol. 6 (2), 89–101. https://doi.org/10.1016/0162-3095(85)90002-0.
Dziobek, I., Fleck, S., Kalbe, E., Rogers, K., Hassenstab, J., Brand, M., Kessler, J.,
Ballespí, S., Pérez-Domingo, A., Vives, J., Sharp, C., Barrantes-Vidal, N., 2018. Childhood
Woike, J.K., Wolf, O.T., Convit, A., 2006. Introducing MASC: a movie for the
behavioral inhibition is associated with impaired mentalizing in adolescence. PLoS
assessment of social cognition. J. Autism Dev. Disord. https://doi.org/10.1007/
ONE 13 (3). https://doi.org/10.1371/journal.pone.0195303.
s10803-006-0107-0.
Banerjee, M., 1997. Hidden Emotions: Preschoolers’ Knowledge of Appearance-reality
Ebner, N.C., Riediger, M., Lindenberger, U., 2010. FACES–a database of facial
and Emotion Display Rules. Soc. Cogn. 15 (2), 107–132. https://doi.org/10.1521/
expressions in young, middle-aged, and older women and men: development and
SOCO.1997.15.2.107. Doi:10.1521/Soco.1997.15.2.107.
validation. Behav. Res. Methods 42 (1), 351–362. https://doi.org/10.3758/
Baron-Cohen, S., 1995. Mindblindness: An essay on autism and theory of mind. In:
BRM.42.1.351.
Mindblindness: An Essay on Autism and Theory of Mind. The MIT Press.
Egger, M., Juni, P., Bartlett, C., Holenstein, F., Sterne, J., 2003. How important are
Baron-Cohen, S., O'Riordan, M., Stone, V., Jones, R., Plaisted, K., 1999. Recognition of
comprehensive literature searches and the assessment of trial quality in systematic
faux pas by normally developing children and children with asperger syndrome or
reviews? Empirical study. Health Technol. Assess. (Winch. Eng.) 7 (1), 1–76.
high-functioning autism. J. Autism Dev. Disord. 29 (5), 407–418. https://doi.org/
Ekman, P., Friesen, W.V., 1976. Pictures of Facial Affect. Consulting Psychologists Press,
10.1023/a:1023035012436.
Palo Alto, CA.
Baron-Cohen, S., Wheelwright, S., Hill, J., Raste, Y., Plumb, I., 2001. The “Reading the
Fan, Y., Han, S., 2008. Temporal dynamic of neural mechanisms involved in empathy for
mind in the eyes” test revised version : a study with normal adults, and adults with
pain: an event-related brain potential study. Neuropsychologia 46 (1), 160–173.
asperger syndrome or high-functioning autism. J. Child Psychol. Psychiatry 42 (2),
https://doi.org/10.1016/j.neuropsychologia.2007.07.023.
241–251. https://doi.org/10.1111/1469-7610.00715.
Farmer, A.S., Kashdan, T.B., 2012. Social anxiety and emotion regulation in daily life:
Beck, J., 2011. Cognitive Behavior Therapy: Basics and Beyond. The Guilford Press.
spillover effects on positive and negative social events. Cogn. Behav. Ther. https://
https://doi.org/10.1017/CBO9781107415324.004.
doi.org/10.1080/16506073.2012.666561.
Bell, C., Bourke, C., Colhoun, H., Carter, F., Frampton, C., Porter, R., 2011. The
Frith, C.D., Frith, U., 2007. Social cognition in humans. Current Biology. https://doi.org/
misclassification of facial expressions in generalised social phobia. J. Anxiety Disord.
10.1016/j.cub.2007.05.068.
25 (2), 278–283. https://doi.org/10.1016/j.janxdis.2010.10.001.
Garner, M., Baldwin, D.S., Bradley, B.P., Mogg, K., 2009. Impaired identification of
Beidel, D.C., Alfano, C.A., Kofler, M.J., Rao, P.A., Scharfstein, L., Wong Sarver, N., 2014.
fearful faces in generalised social phobia. J. Affect. Disord. 115 (3), 460–465.
The impact of social skills training for social anxiety disorder: a randomized
https://doi.org/10.1016/j.jad.2008.10.020.
controlled trial. J. Anxiety Disord. 28 (8), 908–918. https://doi.org/10.1016/J.
Gilboa-Schechtman, E., Foa, E., Vaknin, Y., Marom, S., Hermesh, H., 2008. Interpersonal
JANXDIS.2014.09.016.
sensitivity and response bias in social phobia and depression: labeling emotional
Berger, Y., 2002. Ability to Express and Perceive Emotional Vocal Expressions, by Verbal
expressions. Cogn. Ther. Res. 32 (5), 605–618. https://doi.org/10.1007/s10608-
and Non-verbal Learning Disabled Children and by Children Without Learning
008-9208-8.
Disabilities. Ilan University, Ramat Gan. MA Thesis.
Gleason, K.A., Jensen-Campbell, L.A., Ickes, W., 2009. The role of empathic accuracy in
Bodner, E., Aharoni, R., Iancu, I., 2012. The effect of training with music on happiness
adolescents’ peer relations and adjustment. Personal. Soc. Psychol. Bull. 35 (8),
recognition in social anxiety disorder. J. Psychopathol. Behav. Assess. 34 (4),
997–1011. https://doi.org/10.1177/0146167209336605.
458–466. https://doi.org/10.1007/s10862-012-9304-7.
Green, M.F., Bearden, C.E., Cannon, T.D., Fiske, A.P., Hellemann, G.S., Horan, W.P.,
Bodner, E., Gilboa, A., 2006. Emotional communicability in music therapy: different
Kee, K., Kern, R.S., Lee, J., Sergi, M.J., Subotnik, K.L., Sugar, C.A., Ventura, J.,
instruments for different emotions? Nord. J. Music. Ther. 15 (1), 3–16. https://doi.
Yee, C.M., Nuechterlein, K.H., 2012. Social cognition in schizophrenia, part 1:
org/10.1080/08098130609478147.
performance across phase of illness. Schizophr. Bull. https://doi.org/10.1093/
Bora, E., Berk, M., 2016. Theory of mind in major depressive disorder: a meta-analysis.
schbul/sbq171.
J. Affect. Disord. 191, 49–55. https://doi.org/10.1016/j.jad.2015.11.023.
Green, M.F., Lee, J., Ochsner, K.N., 2013. Adapting social neuroscience measures for
Brothers, L., 1990. The neural basis of primate social communication. Motiv. Emot. 14
schizophrenia clinical trials, part 1: ferrying paradigms across perilous waters.
(2), 81–91. https://doi.org/10.1007/BF00991637.
Schizophr. Bull. https://doi.org/10.1093/schbul/sbt131.
Brüne, M., 2003. Theory of mind and the role of IQ in chronic disorganized
Green, M.F., Penn, D.L., Bentall, R., Carpenter, W.T., Gaebel, W., Gur, R.C., Kring, A.M.,
schizophrenia. Schizophr. Res. 60 (1), 57–64. https://doi.org/10.1016/S0920-9964
Park, S., Silverstein, S.M., Heinssen, R., 2008. Social cognition in schizophrenia: an
(02)00162-7.
NIMH workshop on definitions, assessment, and research opportunities. Schizophr.
Buhlmann, U., Wacker, R., Dziobek, I., 2015. Inferring other people’s states of mind:
Bull. https://doi.org/10.1093/schbul/sbm145.
comparison across social anxiety, body dysmorphic, and obsessive-compulsive
Gur, R.C., Sara, R., Hagendoorn, M., Marom, O., Hughett, P., Macy, L., Turner, T.,
disorders. J. Anxiety Disord. 34, 107–113. https://doi.org/10.1016/j.
Bajcsy, R., Posner, A., Gur, R.E., 2002. A method for obtaining 3-dimensional facial
janxdis.2015.06.003.
expressions and its standardization for use in neurocognitive studies. J. Neurosci.
Button, K., Lewis, G., Penton-Voak, I., Munafò, M., 2013. Social anxiety is associated
Methods 115 (2), 137–143. https://doi.org/10.1016/s0165-0270(02)00006-7.
with general but not specific biases in emotion recognition. Psychiatry Res. 210 (1),
Hagemann, J., Straube, T., Schulz, C., 2016. Too bad: bias for angry faces in social
199–207. https://doi.org/10.1016/j.psychres.2013.06.005.
anxiety interferes with identity processing. Neuropsychologia 84, 136–149. https://
Campbell, D.W., Sareen, J., Stein, M.B., Kravetsky, L.B., Paulus, M.P., Hassard, S.T.,
doi.org/10.1016/j.neuropsychologia.2016.02.005.
Reiss, J.P., 2009. Happy but not so approachable: the social judgments of individuals
Halberstadt, J.B., Niedenthal, P.M., 1997. Emotional state and the use of stimulus
with generalized social phobia. Depress. Anxiety 26 (5), 419–424. https://doi.org/
dimensions in judgment. J. Pers. Soc. Psychol. 72 (5), 1017–1033. https://doi.org/
10.1002/da.20474.
10.1037//0022-3514.72.5.1017.
Chen, S.H., 2006. Emotion Recognition and Social Rehabilitation in Patients with
Hampel, S., Weis, S., Hiller, W., Witthöft, M., 2011. The relations between social anxiety
Schizophrenia: Development and Revision of Measure for Nonverbal Emotion
and social intelligence: a latent variable analysis. J. Anxiety Disord. 25 (4), 545–553.
Recognition (DOH94-TD-M-113-020; DOH95-TD-M-113-006). Department of
https://doi.org/10.1016/j.janxdis.2011.01.001.
Health, Science and Technology Unit.
Hatfield, E., Cacioppo, J.T., Rapson, R.L., 1992. Primitive emotional contagion. In:
Clark, D.A., Wells, A., 1995. A cognitive model of social phobia. In: Social Phobia:
Emotion and Social Behavior.
Diagnosis, Assessment, and Treatment.

28
T. Alvi et al.
Hawes, M.T., Szenczy, A.K., Klein, D.N., Hajcak, G., Nelson, B.D., 2021. Increases in
depression and anxiety symptoms in adolescents and young adults during the
COVID-19 pandemic. Psychol. Med. https://doi.org/10.1017/S0033291720005358.
Heberlein, A.S., Adolphs, R., Tranel, D., Damasio, H., 2004. Cortical regions for
judgments of emotions and personality traits from point-light walkers. J. Cogn.
Neurosci. 16 (7), 1143–1158. https://doi.org/10.1162/0898929041920423.
Heeren, A., Reese, H.E., McNally, R.J., Philippot, P., 2012. Attention training toward and
away from threat in social phobia: effects on subjective, behavioral, and
physiological measures of anxiety. Behav. Res. Ther. 50 (1), 30–39. https://doi.org/
10.1016/j.brat.2011.10.005.
Hess, U., Blairy, S., 1995. Set of Emotional Facial Stimuli. Department of Psychology,
University of Quebec at Montreal, Montreal, Canada.
Heuer, K., Lange, W.-G., Isaac, L., Rinck, M., Becker, E.S., 2010. Morphed emotional
faces: emotion detection and misinterpretation in social anxiety. J. Behav. Ther. Exp.
Psychiatry 41 (4), 418–425. https://doi.org/10.1016/j.jbtep.2010.04.005.
Hezel, D.M., McNally, R.J., 2014. Theory of mind impairments in social anxiety disorder.
Behav. Ther. 45 (4), 530–540. https://doi.org/10.1016/j.beth.2014.02.010.
Holt-Lunstad, J., 2021. The major health implications of social connection. Current
Directions in Psychology Science 30 (3). https://doi.org/10.1177/
2F0963721421999630.
Hopewell, S., McDonald, S., Clarke, M., Egger, M., 2007. Grey literature in meta-analyses
of randomized trials of health care interventions. Cochrane Database Syst. Rev. 2,
MR000010 https://doi.org/10.1002/14651858.MR000010.pub3.
Horstmann, G., 2002. Facial expressions of emotion: does the prototype represent central
tendency, frequency of instantiation, or an ideal? Emotion 2 (3), 297–305. https://
doi.org/10.1037/1528-3542.2.3.297.
Hunter, L.R., Buckner, J.D., Schmidt, N.B., 2009. Interpreting facial expressions: the
influence of social anxiety, emotional valence, and race. J. Anxiety Disord. 23 (4),
482–488. https://doi.org/10.1016/j.janxdis.2008.10.004.
Hur, J., DeYoung, K.A., Islam, S., Anderson, A.S., Barstead, M.G., Shackman, A.J., 2019.
Social context and the real-world consequences of social anxiety. Psychol. Med. 50
(12), 1989–2000. https://doi.org/10.1017/S0033291719002022.
Ickes, W., Stinson, L., Bissonnette, V., Garcia, S., 1990. Naturalistic social cognition:
empathic accuracy in mixed-sex dyads. J. Pers. Soc. Psychol. https://doi.org/
10.1037/0022-3514.59.4.730.
Jacobs, M., Snow, J., Geraci, M., Vythilingam, M., Blair, R.J.R., Charney, D.S., Pine, D.S.,
Blair, K.S., 2008. Association between level of emotional intelligence and severity of
anxiety in generalized social phobia. J. Anxiety Disord. 22 (8), 1487–1495. https://
doi.org/10.1016/j.janxdis.2008.03.003.
Joormann, J., Gotlib, I.H., 2006. Is this happiness i see? Biases in the identification of
emotional facial expressions in depression and social phobia. J. Abnorm. Psychol.
115 (4), 705–714. https://doi.org/10.1037/0021-843X.115.4.705.
Kang, S.M., Lau, A.S., 2013. Revisiting the out-group advantage in emotion recognition
in a multicultural society: further evidence for the in-group advantage. Emotion 13
(2), 203–215. https://doi.org/10.1037/a0030013.
Kashdan, T.B., 2002. Social anxiety dimensions, neuroticism, and the contours of positive
psychological functioning. Cogn. Ther. Res. https://doi.org/10.1023/A:
1021293501345.
Kashdan, T.B., Volkmann, J.R., Breen, W.E., Han, S., 2007. Social anxiety and romantic
relationships: the costs and benefits of negative emotion expression are context-
dependent. J. Anxiety Disord. 21 (4), 475–492. https://doi.org/10.1016/J.
JANXDIS.2006.08.007.
Kern, R.S., Penn, D.L., Lee, J., Horan, W.P., Reise, S.P., Ochsner, K.N., Marder, S.R.,
Green, M.F., 2013. Adapting social neuroscience measures for schizophrenia clinical
trials, part 2: trolling the depths of psychometric properties. Schizophr. Bull. 39 (6),
1201–1210. https://doi.org/10.1093/schbul/sbt127.
Kessler, R.C., 2002. The categorical versus dimensional assessment controversy in the
sociology of mental illness. J. Health Soc. Behav. 43 (2), 171. https://doi.org/
10.2307/3090195.
Kessler, R.C., Berglund, P., Demler, O., Jin, R., Merikangas, K.R., Walters, E.E., 2005.
Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the
national comorbidity survey replication. Arch. Gen. Psychiatry. https://doi.org/
10.1001/archpsyc.62.6.593.
Kurtz, M.M., Gagen, E., Rocha, N.B.F., Machado, S., Penn, D.L., 2016. Comprehensive
treatments for social cognitive deficits in schizophrenia: a critical review and effect-
size analysis of controlled studies. Clin. Psychol. Rev. 43, 80–89. https://doi.org/
10.1016/J.CPR.2015.09.003.
Langner, O., Dotsch, R., Bijlstra, G., Wigboldus, D.H.J., Hawk, S.T., van Knippenberg, A.,
2010. Presentation and validation of the radboud faces database. Cognit. Emot. 24
(8), 1377–1388. https://doi.org/10.1080/02699930903485076.
Lau, A.S., Wang, S., Fung, J.J., Namikoshi, M., 2014. What happens when you “can’t read
the air”? Cultural fit and aptitude by values interactions on social anxiety. J. Soc.
Clin. Psychol. 33 (10), 853–866. https://doi.org/10.1521/jscp.2014.33.10.853.
Lenton-Brym, A.P., Moscovitch, D.A., Vidovic, V., Nilsen, E., Friedman, O., 2018. Theory
of mind ability in high socially anxious individuals. Anxiety Stress Coping 31 (5),
487–499. http://10.0.4.56/10615806.2018.1483021.
Li, S., Tan, J., Qian, M., Liu, X., 2008. Continual training of attentional bias in social
anxiety. Behav. Res. Ther. 46 (8), 905–912. https://doi.org/10.1016/j.
brat.2008.04.005.
Lundqvist, D., Flykt, A., Öhman, A., 1998. The Karolinska Directed Emotional Faces -
KDEF, CD ROM from Department of Clinical Neuroscience. Psychology section,
Karolinska Institutet, ISBN 91-630-7164-9.
Lyvers, M., Scott, K., Thorberg, F.A., 2019. Social anxiety and alexithymia in relation to
problematic drinking and theory of mind. Am. J. Psychol. 132 (3), 325–341. https://
doi.org/10.5406/amerjpsyc.132.3.0325.
29
Journal of Affective Disorders 311 (2022) 17–30
Magen, E., Konasewich, P.A., 2011. Women support providers are more susceptible than
men to emotional contagion following brief supportive interactions. Psychol. Women
Q. 35 (4), 611–616. https://doi.org/10.1177/0361684311423912.
Maleki, G., Zabihzadeh, A., Richman, M.J., Demetrovics, Z., Mohammadnejad, F., 2020.
Decoding and reasoning mental states in major depression and social anxiety
disorder. BMC Psychiatry 20. https://doi.org/10.1186/s12888-020-02873-w.
Matsumoto, D., Ekman, P., 1988. Japanese and Caucasian Facial Expressions of Emotion
(JACFEE) [Slides]. Intercultural and Emotion Research Laboratory, Department of
Psychology, San Francisco State University, San Francisco, CA.
Mayer, J.D., Salovey, P., Caruso, D., 2002. Mayer-salovey-caruso emotional intelligence
test (MSCEIT) users manual. Multi-Health Systems, Toronto, Ontario.
McDonald, S., Bornhofen, C., Shum, D., Long, E., Saunders, C., Neulinger, K., 2006.
Reliability and validity of the awareness of social inference test (TASIT): a clinical
test of social perception. Disabil. Rehabil. 28 (24), 1529–1542. https://doi.org/
10.1080/09638280600646185.
Mohlman, J., Carmin, C.N., Price, R.B., 2007. Jumping to interpretations: social anxiety
disorder and the identification of emotional facial expressions. Behav. Res. Ther. 45
(3), 591–599. https://doi.org/10.1016/j.brat.2006.03.007.
Moher, D., Shamseer, L., Clarke, M., Ghersi, D., Liberati, A., Petticrew, M., et al., 2015.
Preferred reporting items for systematic review and meta-analysis protocols
(PRISMA-P) 2015 statement. Syst. Rev. 4 (1).
Montagne, B., Kessels, R.P.C., De Haan, E.H.F., Perrett, D.I., 2007. The emotion
recognition task: a paradigm to measure the perception of facial emotional
expressions at different intensities. Percept. Mot. Skills 104 (2), 589–598. https://
doi.org/10.2466/pms.104.2.589-598.
Montagne, B., Schutters, S., Westenberg, H.G.M., van Honk, J., Kessels, R.P.C., de
Haan, E.H.F., 2006. Reduced sensitivity in the recognition of anger and disgust in
social anxiety disorder. Cognitive Neuropsychiatry 11 (4), 389–401. https://doi.org/
10.1080/13546800444000254.
Morrison, A.S., Heimberg, R.G., 2013. Social anxiety and social anxiety disorder. Annu.
Rev. Clin. Psychol. 9 (1), 249–274. https://doi.org/10.1146/annurev-clinpsy-
050212-185631.
Morrison, A.S., Mateen, M.A., Brozovich, F.A., Zaki, J., Goldin, P.R., Heimberg, R.G.,
Gross, J.J., 2016. Empathy for positive and negative emotions in social anxiety
disorder. Behav. Res. Ther. 87, 232–242. https://doi.org/10.1016/j.
brat.2016.10.005.
Mullins, D.T., Duke, M.P., 2004. Effects of social anxiety on nonverbal accuracy and
response time I: facial expressions. J. Nonverbal Behav. 28 (1), 3–33. https://doi.
org/10.1023/B:JONB.0000017865.24656.98.
Murphy, B.A., Lilienfeld, S.O., 2019. Are self-report cognitive empathy ratings valid
proxies for cognitive empathy ability? Negligible meta-analytic relations with
behavioral task performance. Psychol. Assess. 31 (8), 1062–1072. https://doi.org/
10.1037/pas0000732.
Nowicki, S., Carton, J., 1993. The measurement of emotional intensity from facial
expressions. J. Soc. Psychol. https://doi.org/10.1080/00224545.1993.9713934.
Nowicki Jr., S., Glanville, D., Demertzis, A., 1998. A test of the ability to recognize
emotion in the facial expressions of african american adults. J. Black Psychol. 24 (3),
335–350. https://doi.org/10.1177/00957984980243005.
Oh, K.-S., Lee, W.H., Kim, S., Shin, D.-W., Shin, Y.-C., Lim, S.-W., 2018. Impaired facial
expression recognition in patients with social anxiety disorder: a case-control study.
Cogn. Neuropsychiatry 23 (4), 218–228. https://doi.org/10.1080/
13546805.2018.1462695.
O’Toole, M.S., Hougaard, E., Mennin, D.S., 2013. Social anxiety and emotion knowledge:
a meta-analysis. J. Anxiety Disord. https://doi.org/10.1016/j.janxdis.2012.09.005.
Oakley, B.F.M., Brewer, R., Bird, G., Catmur, C., 2016. Theory of mind is not theory of
emotion: a cautionary note on the reading the mind in the eyes test. J. Abnorm.
Psychol. 125 (6), 818–823. https://doi.org/10.1037/abn0000182.
Penn, D.L., Ritchie, M., Francis, J., Combs, D., Martin, J., 2002. Social perception in
schizophrenia: the role of context. Psychiatry Res. 109 (2), 149–159. https://doi.
org/10.1016/S0165-1781(02)00004-5.
Phan, K.L., Fitzgerald, D.A., Nathan, P.J., Tancer, M.E., 2006. Association between
amygdala hyperactivity to harsh faces and severity of social anxiety in generalized
social phobia. Biol. Psychiatry 59 (5), 424–429. https://doi.org/10.1016/j.
biopsych.2005.08.012.
Philippot, P., Douilliez, C., 2005. Social phobics do not misinterpret facial expression of
emotion. Behav. Res. Ther. 43 (5), 639–652. https://doi.org/10.1016/j.
brat.2004.05.005.
Pinkham, A.E., Penn, D.L., Green, M.F., Buck, B., Healey, K., Harvey, P.D., 2014. The
social cognition psychometric evaluation study: results of the expert survey and
RAND panel. Schizophr. Bull. 40 (4), 813–823. https://doi.org/10.1093/schbul/
sbt081.
Pinkham, A.E., Penn, D.L., Green, M.F., Harvey, P.D., 2016. Social cognition
psychometric evaluation: results of the initial psychometric study. Schizophr. Bull.
https://doi.org/10.1093/schbul/sbv056.
Pittelkow, M.-M., Aan Het Rot, M., Seidel, L.J., Feyel, N., Roest, A.M., 2021. Social
anxiety and empathy: a systematic review and meta-analysis. Journal of Anxiety
Disorders 78, 102357. https://doi.org/10.1016/j.janxdis.2021.102357.
Plana, I., Lavoie, M.-A., Battaglia, M., Achim, A.M., 2014. A meta-analysis and scoping
review of social cognition performance in social phobia, posttraumatic stress
disorder and other anxiety disorders. J. Anxiety Disord. 28 (2), 169–177. https://doi.
org/10.1016/j.janxdis.2013.09.005.
Quadflieg, S., Wendt, B., Mohr, A., Miltner, W.H.R., Straube, T., 2007. Recognition and
evaluation of emotional prosody in individuals with generalized social phobia: a
pilot study. Behav. Res. Ther. 45 (12), 3096–3103. https://doi.org/10.1016/j.
brat.2007.08.003.
T. Alvi et al.
Ruffman, T., Henry, J.D., Livingstone, V., Phillips, L.H., 2008. A meta-analytic review of
emotion recognition and aging: implications for neuropsychological models of aging.
Neurosci. Biobehav. Rev. 32 (4), 863–881. https://doi.org/10.1016/j.
neubiorev.2008.01.001.
Sabbagh, M.A., 2004. Understanding orbitofrontal contributions to theory-of-mind
reasoning: implications for autism. Brain Cogn. 55 (1), 209–219. https://doi.org/
10.1016/j.bandc.2003.04.002.
Schlegel, K., Fontaine, J.R.J., Scherer, K.R., 2019. The nomological network of emotion
recognition ability. Eur. J. Psychol. Assess. 35 (3), 352–363. https://doi.org/
10.1027/1015-5759/a000396.
Schneier, F.R., Heckelman, L.R., Garfinkel, R., Campeas, R., Fallon, B.A., Gitow, A.,
Street, L., Del Bene, D., Liebowitz, M.R., 1994. Functional impairment in social
phobia. J. Clin. Psychiatry 55 (8), 322–331.
Schofield, C.A., Coles, M.E., Gibb, B.E., 2007. Social anxiety and interpretation biases for
facial displays of emotion: Emotion detection and ratings of social cost. Behav. Res.
Ther. 45 (12), 2950–2963. https://doi.org/10.1016/j.brat.2007.08.006.
Schroeder, J.E., 1995a. Interpersonal perception skills: self-concept correlates. Percept.
Mot. Skills 80 (1), 51–56. https://doi.org/10.2466/pms.1995.80.1.51.
Schroeder, J.E., 1995b. Self-concept, social anxiety, and interpersonal perception skills.
Personal. Individ. Differ. 19 (6), 955–958. https://doi.org/10.1016/S0191-8869(95)
00108-5.
Schroeder, J.E., Ketrow, S.M., 1997. Social anxiety and performance in an interpersonal
perception task. Psychological Reports 81 (3, Pt 1), 991–996. https://doi.org/
10.2466/pr0.1997.81.3.991.
Sened, H., Lavidor, M., Lazarus, G., Bar-Kalifa, E., Rafaeli, E., Ickes, W., 2017. Empathic
accuracy and relationship satisfaction: a meta-analytic review. J. Fam. Psychol. 31
(6), 742–752. https://doi.org/10.1037/FAM0000320.
Singer, T., Lamm, C., 2009. The social neuroscience of empathy. Ann. N. Y. Acad. Sci.
https://doi.org/10.1111/j.1749-6632.2009.04418.x.
Straube, T., Kolassa, I.-T., Glauer, M., Mentzel, H.-J., Miltner, W.H.R., 2004. Effect of task
conditions on brain responses to threatening faces in social phobics: an event-related
functional magnetic resonance imaging study. Biol. Psychiatry 56 (12), 921–930.
https://doi.org/10.1016/j.biopsych.2004.09.024.
Sutterby, S.R., Bedwell, J.S., Passler, J.S., Deptula, A.E., Mesa, F., 2012. Social anxiety
and social cognition: the influence of sex. Psychiatry Res. 197 (3), 242–245. https://
doi.org/10.1016/j.psychres.2012.02.014.
Taylor, C.T., Bomyea, J., Amir, N., 2010. Attentional bias away from positive social
information mediates the link between social anxiety and anxiety vulnerability to a
social stressor. J. Anxiety Disord. https://doi.org/10.1016/j.janxdis.2010.02.004.
Thayer, J.F., Johnsen, B.H., 2000. Sex differences in judgement of facial affect: a
multivariate analysis of recognition errors. Scand. J. Psychol. 41 (3), 243–246.
https://doi.org/10.1111/1467-9450.00193.
Tibi-Elhanany, Y., Shamay-Tsoory, S.G., 2011. Social cognition in social anxiety: first
evidence for increased empathic abilities. Isr. J. Psychiatry Relat. Sci. 48 (2),
98–106.
30
Journal of Affective Disorders 311 (2022) 17–30
Tibi-Elhanany, Y., Shamay-Tsoory, S.G., Shamay-Tsoory, S., 2011. Social cognition in
social anxiety: first evidence for increased empathic abilities. Isr. J. Psychiatry Relat.
Sci. 48 (2).
Tonge, N.A., Lim, M.H., Piccirillo, M.L., Fernandez, K.C., Langer, J.K., Rodebaugh, T.L.,
2020. Interpersonal problems in social anxiety disorder across different relational
contexts. J. Anxiety Disord. 75, 102275 https://doi.org/10.1016/J.
JANXDIS.2020.102275.
Torro-Alves, N., Claudino, R.G.E., Rodrigues, M.R., Machado-de-Sousa, J.P., Crippa, J.A.,
Bezerra, I.A.de O., Osório, F.de L., 2016. Facial emotion recognition in social
anxiety: the influence of dynamic information. Psychology & Neuroscience 9 (1),
1–11. https://doi.org/10.1037/pne0000042.
Tottenham, N., Tanaka, J.W., Leon, A.C., McCarry, T., Nurse, M., Hare, T.A., Marcus, D.
J., Westerlund, A., Casey, B.J., Nelson, C., 2009. The NimStim set of facial
expressions: judgments from untrained research participants. Psychiatry Res. 168
(3), 242–249. https://doi.org/10.1016/j.psychres.2008.05.006.
Tseng, H.-H., Huang, Y.-L., Chen, J.-T., Liang, K.-Y., Lin, C.-C., Chen, S.-H., 2017. Facial
and prosodic emotion recognition in social anxiety disorder. Cogn. Neuropsychiatry
22 (4), 331–345. https://doi.org/10.1080/13546805.2017.1330190.
Veljaca, K.-A., Rapee, R.M., 1998. Detection of negative and positive audience
behaviours by socially anxious subjects. Behav. Res. Ther. 36 (3), 311–321. https://
doi.org/10.1016/S0005-7967(98)00016-3.
Washburn, D., Wilson, G., Roes, M., Rnic, K., Harkness, K.L., 2016. Theory of mind in
social anxiety disorder, depression, and comorbid conditions. J. Anxiety Disord. 37,
71–77. https://doi.org/10.1016/j.janxdis.2015.11.004.
Wendt, B., Scheich, H., 2002. The ‘Magdeburger Prosodie-Korpus’. In: Bel, B., Martien, I.
(Eds.), Proceedings of the Speech Prosody 2002 Conference, 11– 13 April 2002.
Laboratoire Parole et Langage, Aix-en-Provence, pp. 699–701 in press.
Winton, E.C., Clark, D.M., Edelmann, R.J., 1995. Social anxiety, fear of negative
evaluation and the detection of negative emotion in others. Behav. Res. Ther. 33 (2),
193–196. https://doi.org/10.1016/0005-7967(94)E0019-F.
Wittchen, H.U., Fuetsch, M., Sonntag, H., Müller, N., Liebowitz, M., 2000. Disability and
quality of life in pure and comorbid social phobia. Findings from a controlled study.
Eur. Psychiatry 15 (1), 46–58. https://doi.org/10.1016/S0924-9338(00)00211-X.
Yoon, K.L., Fitzgerald, D.A., Angstadt, M., McCarron, R.A., Phan, K.L., 2007. Amygdala
reactivity to emotional faces at high and low intensity in generalized social phobia: a
4-Tesla functional MRI study. Psychiatry Res. 154 (1), 93–98. https://doi.org/
10.1016/j.pscychresns.2006.05.004.
Young, A., Perrett, D., Calder, A., Sprengelmeyer, R., Ekman, P., 2002. Facial Expressions
of Emotion – Stimuli and Tests (FEEST). Thames Valley Test, Bury St. Edmunds,
England.
Zaki, J., Bolger, N., Ochsner, K., 2008. It takes two: the interpersonal nature of empathic
accuracy. Psychol. Sci. 19, 399–404. https://doi.org/10.1111/j.1467-
9280.2008.02099.x.