Testing the Validity of a Social Anxiety Visual Search
Task Relative to an Implicit Measure of Social Anxiety
Tessa Heinhuis
April 24th 2015
University of Amsterdam
Tessa Heinhuis
Student number: 10319891
Bachelorproject Developmental Psychology: Final Version
Supervisors: Dr. Bram Van Bockstaele, Dr. Kiki Nikolaou & Prof. Dr. Reinout Wiers Extra reader: Mw.Tess den Uijl (MSc)
Word count: 5375
Index Abstract p. 3 Introduction p. 4 Method p. 9 Results p. 14 Discussion p. 16 References p. 21
Abstract
Social Anxiety Disorder (SAD) is the fourth most common psychiatric disorder worldwide
and therefore it is important to know how to detect the disorder correctly. Attentional bias in
relation to SAD is a form of cognitive bias involving preferential attention allocation to
threatening information and therefore neglecting other, more neutral, or positive information.
The current study examined the predictive validity of a Feature Relevant Visual Search Task
(FR-VST) used to assess attentional bias in relation to an Implicit Association Test (IAT),
which is used to assess implicit attitudes towards threat-related stimuli. These attitudes
towards threat-related stimuli have been shown to correlate with the degree of SAD. The
results showed that of the 60 participants included in this study the scores that participants
attained at the FR-VST significantly predicted outcomes on the IAT. There seems to be a
positive linear relation between these two tests. This indicates that the lower implicit anxiety
an individual has, the higher the attentional bias score on the FR-VST is. An explanation for
our findings could be that participants who are implicitly anxious avoid angry faces. Future
research should focus on participants who have clinically been diagnosed with SAD, other
types of anxiety disorders and other types of validity of the FR-VST; like construct validity.
Introduction
Social Anxiety Disorder (SAD) is a type of anxiety disorder which is related to an
uncontrollable fear for various types of social situations, including performing or presenting
oneself in public and interacting in social situations (Staugaard, 2010; Lange, Keijsers,
Becker & Rinck, 2008). SAD is associated with emotional distress and functional impairment
(Acarturk, de Graaf, van Straten, Have, & Cuijpers, 2008; Schneier et al., 1994; Tolman et
al., 2009). SAD is related to persistent fears of situations involving social interaction or social
performance or conditions in which there is the potential for judgment by others and in which
embarrassment can occur (American Psychiatric Association, 1994). Social situations are
preferably avoided or when these circumstances cannot be eluded, social experiences of
people who suffer from the symptoms of SAD involve extreme anxiety and discomfort
(Werner et al., 2012). The DSM-5 states that SAD is a disorder in which the negative
evaluation of others is the core issue. Also, the DSM-5 specifies that SAD occurs at different
age levels and it can be reliably diagnosed in children as young as six years old up until late
adulthood. SAD is the fourth most common psychiatric disorder and therefore it is important
to know how to detect the disorder effectively and correctly (Kessler et al., 2005).
In recent years it has repeatedly been proposed that biases in information processing,
such as attentional biases, memory biases and interpretation biases, of patients who meet the
criteria for SAD (American Psychiatric Association, 2000), have an effect on the symptoms
of the disorder. Attentional bias in relation to SAD is a form of cognitive bias involving
preferential attention allocation to threatening information and consequently neglecting other,
more neutral, or positive information. The tendency to selectively attend to particular cues in
the social environment, which indicates having an attentional bias, becomes a habit for
people with SAD and therefore it creates a vicious cycle in which an ambiguous world is
Rapee and Heimberg (1997) describe a basic model of anxiety in socially evaluative
situations, which states that people who suffer from SAD assume that other people are
fundamentally critical. Also, they indicate that socially anxious people find it important that
other people appraise them positively. They correspondingly suggest that persons with SAD
view the social world in a specific way which emphasizes excessive negative self-judgment;
which is a form of attentional bias. This cognitive model suggests that attentional bias and
fear of negative evaluation are fundamental features of SAD (Rapee & Heimberg, 1997).
Bar-Haim, Lamy, Pergamin, Bakermans-Kranenburg and Van IJzendoorn (2007)
conducted a meta-analysis of 172 studies in which they collected data and results from
articles before 2005 in which the attentional biases of people with either no clinical anxiety
disorder or a General Anxiety Disorder, SAD, specific phobia, obsessive compulsive
syndrome (OCS), post-traumatic stress disorder (PTSD) or panic disorder were checked. The
participants were all tested with either the Emotional Stroop, the dot-probe task or a version
of the emotional spatial cuing task. The researchers wanted to find out if individuals
diagnosed with anxiety disorders (clinical population) and individuals who self-reported high
levels of anxiety on questionnaires (nonclinical population) have a threat-related attentional
bias and if this can be experimentally demonstrated by using different types of attentional
bias tasks under a variety of experimental conditions. In the Emotional Stroop the response
latency of the participants to name the printed color of a word is measured. This reaction time
is compared to when this word is threat-related (e.g., “illness”) relative to when it is neutral
(e.g., “paper”). In the Emotional Spatial Cuing Task, participants get to see a fixation cross in
the middle of their computer screen and they have to detect visual targets presented at either
the left or the right side of the cross (Vogt, De Houwer, Koster, Van Damme & Crombez,
2008). A threat-related attentional bias is discovered when validity effects are larger when the
Another way to measure attentional bias in relation to SAD (or alternate types of anxiety
disorders) is the dot-probe task, which is developed by MacLeod, Mathews, and Tata (1986).
A typical version of this task shows word pairs which are shortly presented on a computer
screen. One word of a pair is threat related (e.g., “failure”) and the other word is neutral (e.g.,
“pen”). When both words disappear after a small amount of time, a dot appears in the screen
exactly on the same spot as where one of the words was shown before. The reaction time that
is measured can be seen as an indicator of attentional bias: if people respond faster to the dots
that replace threat related words than to dots that replace neutral words, they have an
attentional bias toward threat related stimuli (Bar-Haim et al., 2007). Bar-Haim et al. (2007)
found that a significant threat-related bias was present in anxious participants under different
experimental conditions and when tested with the use of different types of tests, but not in
non-anxious participants. Nevertheless, they mention that there is a need for better
investigation of anxiety disorders and the way it is connected to attentional bias. The main
reason for this is the fact that these tasks do not seem to be valid instruments (Bar-Haim et
al., 2007; Van Damme, Crombez & Notebaert, 2008). The Emotional Stroop has been
widely criticized as an instrument to measure attentional bias because deferred response
latencies with threat-related stimuli may result from late processes that are unrelated to
attention (Bar-Haim et al., 2007).
Even though attentional bias seems to be related to social anxiety, as can be seen in
the meta-analysis of Bar-Haim et al. (2007) as well as in the study of Eldar, Yankelevitch,
Lamy & Bar-Haim (2010), they all used the dot-probe task to measure attentional bias. The
psychometric quality of the task is questionable, since the testing of attentional bias with a
dot-probe task is less reliable for non-clinical individuals who suffer from moderate or low
anxiety (Schmukle, 2005). Schmukle (2005) concluded that the dot-probe task provides a
suggests that the dot-probe task is not internally consistent nor stable over time, and there is
no difference in when words are used as stimuli or when pictures are used as stimuli. He
found this result for two versions of the dot-probe task and the effect was similar for
psychical threat stimuli as for social threat stimuli. The main conclusion of these findings is
that there is clear need for more sound tasks which measure attentional bias. Attentional bias
tasks like the dot-probe are no valid instruments to measure attentional bias, since the
interpretation of results obtained from this paradigm is limited: it is not clear whether effects
are significant because of faster engagement to threat-related stimuli, delayed disengagement
from threat-related stimuli, or a combination of both (Van Damme, Crombez & Notebaert,
2008). This means there is a need for new measures of attentional bias.
The Visual Search Task (VST) is thought to examine attentional bias in a more
ecologically valid manner than does the dot-probe task (Gilboa-Schechtman, Foa & Amir,
1999). Instead of using the dot-probe task or the Emotional Spatial Cuing Task, attentional
bias can be measured with a VST. In this task, participants attempt to find a target object
among non-target objects (Frischen, Eastwood, & Smilek, 2008). An effective way to do this
is by using facial stimuli (Juth, Lundqvist, Karlsson, & Öhman, 2005). Juth, Lundqvist,
Karlsson & Öhman (2005) measured attentional bias in relation to social anxiety by
presenting participants a VST with faces: participants had to look either for a happy face in a
grid of angry faces or for an angry face in a grid of happy faces. Attentional bias was
measured by distracting the reaction time of finding a happy face in a grid of angry faces
from the reaction time of finding an angry face in a grid of happy faces.
We argue that a Feature Relevant Visual Search Task (FR-VST) is a useful candidate
to measure attentional bias in socially anxious stimuli. This task has not been studied
extensively yet and so the psychometrics are not completely acknowledged. Hence, in the
to an implicit measure of attitudes towards certain stimuli, which is measured with an Implicit
Association Test (IAT). The IAT measures strengths of associations between concepts by
comparing response times in two combined discrimination tasks (Gamer, Schmukle,
Krausgrill & Egloff, 2008). The IAT procedure seeks to implicitly measure attitudes by
measuring their underlying automatic evaluation. Participants have to categorize words that
are threat-related and words that are neutral or calm-related. They have to categorize them
either with the label “self” and the label “not-self”. Reaction time is measured and the
discrepancy between the time it takes to categorize words that are threat-related (e.g. “fear”)
with the label “self” (e.g. “me”) and the label “not-self” (e.g. “others”), as well as the latency
of categorizing words that are calm-related (e.g. “ relaxed”) with the label “self” and
“not-self” is measured. The IAT used in this research included pair-combinations of words
associated with the self and words associated with anxiety, as well as pair-combinations of
words associated with the self and words that are associated with non-anxious situations. The
task also included pair-combinations of words not associated with the self and words related
to anxiety, as well as pair-combinations of words not associated with the self and words that
are connected to non-anxious situations. Shorter response latencies are expected when two
congruent concepts share one response key (e.g., for an anxious individual, the concepts
“self” and “anxiety” [as compared to “self” and “calmness”]). The more negative the
individual score on the IAT, the more implicitly socially anxious the participant is.
Gamer et al. (2008) found that prior to treatment, implicit self-anxiety associations
were shown to be stronger in socially anxious participants than in a control group of
non-anxious individuals. Thus, both scores on the IAT and on the attentional bias tasks mentioned
above have shown to be highly associated with the propensity of social anxiety
symptomatology. Understanding and predicting personality-related behavior, like socially
using both direct and indirect measurement procedures (Back, Schmukle, & Egloff, 2009).
We expected that high socially anxious participants will be faster at responding on trials
where the association is between a negative or an anxious word and a “self” association,
based on previous research (De Jong, Pasman, Kindt & Van den Hout, 2001; Gamer et al.,
2008).
We have assessed whether the index of attentional bias that is derived from the VST
has a positive relation with implicit associations as assessed using the IAT. Given that
socially anxious participants show low scores on the IAT and socially anxious participants
show high attentional bias to threatening faces or words, we expected a significant relation
between the IAT and the attentional bias index derived from the VST.
Method
Design / Procedure The experiment was conducted in a laboratorial setting and all the
participants were tested individually. After completing the informed consent, the participants
first filled in two self-report questionnaires to measure their level of social anxiety. Next, the
participants performed the FR-VST. The order of the FR-VST was counter-balanced across
participants and stratified by gender. After this test, all participants completed the IAT. All
the participants filled in questionnaires related to addiction and alcohol use and performed
another version of the VST (the Feature Irrelevant Visual Search Task), but these results will
not be discussed in this study. To measure social anxiety the Social Phobia Scale (SPS;
Mattick & Clark, 1989) and the Fear of Negative Evaluation Scale were administered, but
due to a technical fault the data were not recorded. Therefore these results will not be
discussed further in this research. The testing took 55 minutes per participant and all the tests
Participants Fifty-one undergraduate students (36 females, 25 males; M = 22.2, age
range: 18-29 years) of different tracks of the University of Amsterdam participated in the
study in exchange for research participation credits. Ten participants enrolled in this study in
exchange for money (ten euro’s per person).
Materials
Feature Relevant Visual Search Task (FR-VST). For the FR-VST the face images were
selected from the KDEF face-image database (Lundqvist, Flykt, & Öhman, 1998). The
FR-VST is a type of visual search paradigm in which people have to look for a specific emotional
face in a crowd of distractors. There were 16 pictures of which one is the target stimulus. All
the other pictures of faces were distracters; either you have one happy face in a crowd of
angry faces or you have one angry face in a crowd of happy faces. Each participant
completed two blocks of trials presented in a counterbalanced order. The locations of the
faces were randomized across trials. Every trial terminated when the participant gave a
response. Each trial in both versions of the task began with a fixation cross presented at the
center of the computer screen and this was followed by a centrally presented square grid that
consisted of 16 images. In order to make sure that participants were fixated on the center of
the screen at the beginning of each trial, they had to click on the fixation cross using their
mouse in order for the picture grid for that trial to be presented. In one block they were
instructed to search for the image depicting an angry face within a grid of images of happy
faces (negative block). In the other block they were instructed to search for the image
depicting a happy face within a grid of images of angry faces (positive block). For each trial
in both blocks the latency (time to click on an image location since the start of that specific
trial) has been recorded and whether the response was a correct one (i.e. whether they clicked
on the location where the target appeared). The number of trials for each block is n = 32.
derived from the difference in reaction times between the positive block and the negative
block. Each picture was 80x80 pixels in size.
Implicit Association Test (IAT). The IAT was originally developed by Greenwald, McGhee,
and Schwartz (1998) and it measures the automatic associations between the dimensions
“self” versus “other” and “socially anxious” versus “socially relaxed”. The verbal primes of
the “self”, included the words I (ik), me (mij), mine (mijn), or myself (mijzelf) or own (eigen).
The other category included the words they (zij), them ( hun), others (anderen), their (zijzelf)
or you (jullie). The self-related words and the other-related words were combined with either
social-anxious words or social-calm words. The five social-anxious words which are used
are: nervous (nerveus), fearful (bang), criticized (bekritiseerd), ashamed (beschaamd) and
insecure (onzeker). The five words which were used in the social-calm condition are: calm
(kalm), relaxed (ontspannen), accept (aanvaard), careless (onbezorgd) and secure (zeker).
All the participants first practiced the self-other dimension and then the anxious-calm
dimension. The participants got to see one word in the center of the computer screen. They
could respond with either the key “A” or the key “L”. After each answer, a short break of
350ms was included before the next word appeared in the screen. After a wrong answer there
appeared a red ‘X’ on the screen for 450ms. Each block was labelled differently and the
names of the labels were always presented on top of the screen, on the left and the right side
(depending on the instructions). Every block included a ‘waiting trial’ as the first trial, in
which only the labels of the block were shown at the top of the screen.
The first block consisted of practicing with self and not-self related words. There were
21 trials. The labels in this trial were ‘Self’ and ‘Not-self’. The words that were used were the
ten words in the self and the not-self category; each word was used twice in random order.
Block 2 consisted of practicing with anxious and calm words. There were 21 trials. The labels
anxious and calm category; each word was used twice in random order. Block 3 was a
compatible practice block. The words in the self category shared one key on the keyboard
with the words in the calm category and the words in the not-self category shared one key on
the keyboard with the words in the anxious category. There were 21 trials. The labels were
‘Self-Calm’ and ‘Not-Self-Anxious’. Every word from every category was used once in
random order. Block 4 was a compatible test block in which the words in the self category
shared the same key on the keyboard as the words in the calm category, and the words in the
not-self category shared the same key on the keyboard as the words from the anxious
category. There were 61 trials. The labels were ‘Self-Calm’ and ‘Not-self-Anxious’. Every
word of all the categories was used three times in random order. Block 5 was a practicing
block in which the words in the anxious category and the words in the calm category were
shown. There were 21 trials. The labels were ‘Anxious’ and ‘Calm’ and they were switched
in this block: at first the anxious words were on the left side of the screen and the calm words
were on the right side; this was now vice versa. Every word of the anxious and calm category
was used twice in random order. Block 6 was an incompatible practicing block in which
self-related words shared the key with anxious words and the not-self words shared the key with
the words from the calm category. There were 21 trials and the labels were ‘Self-Anxious’
and ‘Not-self-Calm’. Every word from the categories was used once in random order. Block
7 was an incompatible block in which self-related words shared a key with anxious words
and not-self words shared a key with calm words. There were 61 trials; every word from the
categories was used three times in random order.
Half of the participants started with the words associated with not-self at the left side
of the screen and the self-related words on the right side of the screen; which in the first four
blocks lead to selecting the anxious words also on the left side and the calm words on the
screen and not-self related words on the right side of the screen, which lead in the first four
blocks to selecting the words from the calm category on the left side and the anxious words
on the right side. This was counterbalanced. Reaction time was measured for all participants.
Analyses For this study we calculated the errors in each task. We removed all the
errors and calculated the average response latency and the standard deviation for the FR-VST
and the IAT. We removed all the responses that deviated more than three standard deviations
from the means. For each participant in the FR-VST we calculated the individual mean
response latency and the standard deviation. We removed all the responses that deviate more
than three standard deviations from these means. For calculating data for the FR-VST each
participant, we calculated the mean latency in the find-angry block (negative block). For each
participant we subtracted the mean of the find-angry block from the mean of the find-happy
block (positive block), and therefore we calculated an attentional bias score for each
participant.
For calculating data in the IAT, we selected the data from the test block 4 and 7. We
removed the first three trials of each block (because these were “waiting” trials). We removed
all the errors and rescored latencies that were faster than 300ms and changed this into 300ms:
this is our maximum score. We rescored all the reaction times that were slower than 3000ms
and changed it to 3000ms; which is our minimum score. We transformed each reaction time
to its natural logarithm. For each participant and each block, we calculated the mean of these
logarithmic values. For each participant we subtracted the mean of block 4 from the mean of
block 7, which were the scores participants received on the IAT. We wanted to see whether
the attentional bias scores from the FR-VST predicted our outcome variable; the scores on the
Results
In order to ensure quality, all responses were examined in detail after the data
collection was completed. All outliers were excluded, which included all the scores that
deviated more than three standard deviations from the means (see table 1). One participant
did not meet the criteria, due to a technical error, so this person was excluded from the data.
Sixty participants were included in the data analysis.
Table 1
The Average Scores on the FR-VST and the IAT.
Test Mean Standard Deviation
FR-VST (N = 60) 181.66 673.81
IAT (N = 60) .18 .13
A linear regression analysis was used to test the hypothesis that the scores on the
FR-VST significantly predicted participants' scores on the IAT and that the FR-FR-VST is a valid
way to test social anxiety. First all the assumptions for a linear regression analysis were
checked. The two variables, the attentional bias scores of the FR-VST and the scores of the
IAT, are measured at a continuous level. A scatterplot showed (see figure 1) that there is a
linear relationship between the two variables. The data also shows homoscedasticity and
shows that it is normally distributed. At the α = 0.05 level of significance, there exists enough evidence to conclude that the slope of the population regression line is not zero and, hence,
Figure 1.
The linear relation between the scores on the FR-VST and the IAT.
The results of the regression indicated the predictor explained 7.5% of the variance
(R2 =.075, F(1,59)=4.68, p <.05), which is not a strong indicator. The scores on the FR-VST
significantly predicted IAT scores; b = .186, t(59) = 10.75, p < .05. The positive slope
indicates that the relation between the two tests is positive; the higher the scores on the
FR-VST are per individual, the higher the scores on the IAT for that specific individual will be.
A positive score on the FR-VST means that a person has attentional bias towards
threat-related stimuli and a negative score means that the participant has an avoidance for
threat-related stimuli or has an attentional bias towards happy faces. A score around 0 means
means that the participant has an attitude towards stimuli that have been shown to correlate
with the degree of social anxiety and that that participant more easily associates him- or
herself with anxiety than with calmness. It indicates a high implicit social anxiety. A positive
score on the IAT means that the participant more easily associates him- or herself with
calmness than with anxiety. It indicates a low implicit social anxiety. A positive relation
between the FR-VST and the IAT means all the participants who scored positively on the
FR-VST scored negatively on the IAT; which indicates a high implicit social anxiety and an
attentional bias towards threat-related stimuli.
Discussion
The main aim of this study was to investigate whether the FR-VST is a valid test to
measure attentional bias in relation to SAD. The results showed that the scores that
participants got at the FR-VST significantly predicted outcomes on the IAT. There seems to
be a positive linear relation between these two tests, which shows that the higher the scores
are on the FR-VST, the higher the scores are on the IAT. This indicates that the lower
implicit anxiety an individual has, the higher the attentional bias score on the FR-VST is.
This a slightly unexpected result, so we cannot accept our hypothesis completely. The
FR-VST is a predictive valid test, but in a different direction than we initially assumed. To draw
further conclusions about the FR-VST we have to replicate these findings.
The IAT has already been shown to be a predictive valid test to measure implicit
attitudes that are related to SAD in an earlier study of Egloff and Schmukle (2002). An
alternative explanation for our findings could be that participants who are implicitly anxious
avoid angry faces: this has been found in a previous study in people with low anxiety levels
accompanied by avoidant action tendencies (Mogg & Bradley, 1998). It could be possible
that we did not test participants with severe levels of social anxiety and that we therefore
found positive scores on both variables, and no negative scores on the IAT, which was
initially expected. Also, it could be possible that participants automatically focused on threat
and avoided angry faces at the same time.
We interpret our findings as indicating that the FR-VST can be used to predict scores
on the IAT, but we have to replicate these outcomes to draw stronger conclusions about the
exact direction. Also, the generalization of our findings has been restricted by some
limitations. One important limitation is that it is questionable if the results of the VST can be
generalized to a clinical population due to the use of an analogue sample, since SAD is a type
of disorder in which the disclosure of anxiety in everyday social situations (and not in
laboratory settings) is particularly relevant (Gee, Antony & Koerner, 2013). The DSM-5
states that SAD is a disorder in which the negative evaluation of others is the key symptom,
which can only be measured when an individual is actually in a real-life social situation. The
concern in social situations for people with SAD is that the individual will say or do
something that will result in embarrassment or humiliation (Stein & Stein, 2008), which is
much less likely to occur in laboratorial settings in which all the participants have to do is
complete a computer task in which their results are private and they are sitting alone. Social
anxiety affects attentional bias primarily when social anxiety is activated (Gee, Antony &
Koerner, 2013), which can only be achieved in real-life social situations . Therefore, future
studies should include behavioral observation in everyday situations of the participants.
One more restriction of this study is the fact that participants who enrolled in this
study were free to choose if they wanted to do it; which has an effect on the type of
participants we tested. It could be possible that people who suffer from severe SAD did not
are afraid that someone will discover their disorder or talk about their disorder; which is
likely to occur since the fear of being embarrassed is one of the main symptoms of SAD. The
people who suffer from moderate to extreme levels of SAD could refuse to participate in this
study easily in an effort to protect themselves from possible negative evaluation (Gee,
Anthony & Koerner, 2013). This could have an effect on our results, since it could be
possible that we have only tested people with low or moderate levels of SAD or no SAD at
all. In future studies people who are officially diagnosed with SAD should be asked to
participate and their results should be included in the study. This is important since it will
show the predictive validity of the FR-VST even better, since testing people with low anxiety
could lead to positive scores on the IAT, which indicates that they have low implicit anxiety.
The FR-VST has to be studied when testing people with higher levels of implicit anxiety in
order to check the validity further.
Also, in this research, we only considered the predictive validity. To actually state that
the FR-VST is a completely valid test, we should also check this test on construct validity
and concurrent validity. In future research this has to be done. Furthermore, this study is
conducted with little consideration of other aspects of cognitive processes besides attentional
bias, such as interpretation bias, memory bias and the processes which are responsible for the
verbal content of worries, self-knowledge and beliefs. Next to that, we did not measure
different types of anxiety disorders (e.g. general anxiety disorder or a specific phobia) with
the VST, which has to be done in further research to make the implications of a predictive
valid test even stronger.
Next, it is worth to mention that explicit measurements and implicit measurements are
hard to compare with each other, since they both measure completely different things. They
may not simply be two indices of the same constructs (Tanner, Stopa & De Houwer, 2006).
evaluations of the self may reflect two qualitatively different kinds of self-evaluation. They
state that people have a natural tendency to evaluate themselves more positive when
measured implicitly, because this is part of a primitive self-enhancement mechanism, in
which people always try to feel better by thinking more positive about themselves. Explicit
evaluations of the self include more realistic cognitive thoughts, which therefore will be less
positive. This idea is not new: it is called the self-discrepancy theory which is explained by
Higgins (1987). This theory states that people tend to have an ideal self and a real self.
Amico, Bruch, Haase, and Sturmer (2004) argued that self-discrepancies were useful in
predicting discomfort related to social interaction, since discomfort means that there is a
discrepancy between the ideal self and the real self. In future studies implicit measurements
can be correlated with each other and explicit measures can be correlated with each other to
make it less likely that two different constructs are compared.
Finally, it is possible that the attentional bias of people who suffer from SAD is not
merely a negative bias, but that it is due to the fact that these people lack a positive
attentional bias that is characteristic for non-anxious people. The absence of a positive bias
may act as a maintaining factor in social anxiety along with more explicit strategic biases in
self-appraisal (Tanner, Stopa & De Houwer, 2006). This means that it could be that our
results from the FR-VST task should be interpreted even more nuanced than we first thought
we could do. In future studies we should also consider testing the fact that people with
attentional bias and SAD might lack a positive bias. This can be tested in forthcoming studies
by presenting only positive social cues (e.g. seeing a visual image of people smiling in a
group) in a special VST and check what the latency is of the participants on this type of VST
in comparison with a VST with merely negative social cues (e.g. seeing a visual image of
There still is a lot to investigate when it comes to attentional bias. We can assume that
the FR-VST is a predictive valid test to measure this type of cognitive bias, which is a
symptom of SAD, but we have to be careful with drawing our conclusion here, since the
direction of the relation between the measured attentional bias and the measured implicit
attitudes characteristic for social anxiety is that the higher the measured attentional bias in the
FR-VST is, the lower the implicit attitudes toward social anxiety seem to be. Nevertheless,
this information has considerably advanced the knowledge we have on the visual search task
References
Acarturk, C., de Graaf, R., Van Straten, A., Ten Have, M., & Cuijpers, P. (2008). Social phobia and number of social fears, and their association with comorbidity, health-related quality of life and help seeking. Social Psychiatry and Psychiatric
Epidemiology, 43, 273 – 279. doi: 10.1007/s00127-008-0309-1
American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental
Disorders (4th edition). Washington DC: APA.
http://dx.doi.org/10.1176/appi.books.9780890425596
American Psychiatric Association (2000). Diagnostic and Statistical Manual of Mental
Disorders (revised 4th edition). Washington, DC: American Psychiatric Association.
American Psychiatric Association. (2013). Diagnostic and statistical manual of mental
disorders (5th edition). Arlington, VA: American Psychiatric Publishing.
Amico, K. R., Bruch, M. A., Haase, R. F., & Sturmer, P. J. (2004). Trait shyness, actual-ought self-discrepancy and discomfort in social interaction. Personality and
Individual Differences, 36, 1597 – 1610. doi:10.1016/j.paid.2003.06.008
Back, M. D., Schmukle, S. C., & Egloff, B. (2009). Predicting actual behavior from the explicit and implicit self-concept of personality. Journal of Personality and Social
Psychology, 97, 533 – 548. doi: 10.1037/a0016229
Bar-Haim, Y., Lamy, D., Pergamin, L., Bakermans-Kranenburg, M. J., & Van Ijzendoorn, M. H., (2007). Threat-related attentional bias in anxious and nonanxious individuals: a meta-analytic study. Psychological Bulletin, 133, 1 – 24. doi:
10.1037/0033-2909.133.1.1
Beard, C., & Amir, N. (2010). Negative interpretation bias mediates the effect of social anxiety on state anxiety. Cognitive Therapy and Research, 34, 292 – 296. doi: 10.1007/s10608-009-9258-6
Birchwood, M., Smith, J. O., Cochrane, R., Wetton, S., & Copestake, S. (1990). The Social Functioning Scale. The development and validation of a new scale of social
adjustment for use in family intervention programmes with schizophrenic patients.
The British Journal of Psychiatry, 157, 853 – 859. doi: 10.1192/bjp.157.6.853
Collins, K. A., Westra, H. A., Dozois, D. J., & Stewart, S. H. (2005). The validity of the brief version of the Fear of Negative Evaluation Scale. Journal of Anxiety Disorders, 19, 345 – 359. doi:10.1016/j.janxdis.2004.02.003
Cronbach, L. J., & Meehl, P. E. (1955). Construct validity in psychological tests.
Psychological Bulletin, 52, 281 – 302.
De Jong, P. J., Pasman, W., Kindt, M., & Van den Hout, M. A. (2001). A reaction time paradigm to assess (implicit) complaint-specific dysfunctional beliefs. Behaviour
Research and Therapy, 39, 101 – 113. doi:10.1016/S0005-7967(99)00180-1
Derakshana, N. & Koster, E. (2010). Processing efficiency in anxiety: Evidence from eye-movements during visual search. Behaviour Research and Therapy, 48, 1180 – 1185. doi:10.1016/j.brat.2010.08.009Egloff, B., & Schmukle, S. C. (2002). Predictive
validity of an Implicit Association Test for assessing anxiety. Journal of Personality
and Social Psychology, 83, 1441 – 1455. doi: 10.1037/0022-3514.83.6.1441
Eldar, S., Yankelevitch, R., Lamy, D., & Bar-Haim, Y. (2010). Enhanced neural reactivity and selective attention to threat in anxiety. Biological Psychology, 85, 252 – 257. doi:10.1016/j.biopsycho.2010.07.010
Fletcher, R.B., Nielsen, B., Marshall, D., & Hattie, J. (2004). Validity of test score
interpretations for three self-concept measures based on differing theoretical models.
Perceptual And Motor Skills, 99, 565 – 574. doi: 10.2466/pms.99.2.565-574
Frischen, A., Eastwood, J. D., & Smilek, D. (2008). Visual search for faces with emotional expressions. Psychological Bulletin, 134, 662 – 676. doi:
10.1037/0033-2909.134.5.662
Gamer, J., Schmukle, S. C., Luka-Krausgrill, U., & Egloff, B. (2008). Examining the
dynamics of the implicit and the explicit self-concept in social anxiety: changes in the implicit association test–anxiety and the social phobia anxiety inventory following treatment. Journal of Personality Assessment, 90, 476 – 480.
doi:10.1080/00223890802248786
Gee, B. A., Antony, M. M., & Koerner, N. (2013). Disclosure of anxiety in everyday life: Effects of social anxiety. Personality and Individual Differences, 54, 438 – 441. doi:10.1016/j.paid.2012.10.016
Gilboa-Schechtman, E., Foa, E. B., & Amir, N. (1999). Attentional biases for facial
expressions in social phobia: The face-in-the-crowd paradigm. Cognition & Emotion,
13, 305 – 318. doi: 10.1080/026999399379294
Greenwald, A. G., & Farnham, S. D. (2000). Using the implicit association test to measure self-esteem and self-concept. Journal of Personality and Social Psychology, 79, 1022 – 1038.
Greenwald, A. G., McGhee, D. E., & Schwartz, J. L. K. (1998). Measuring individual
differences in implicit cognition: The implicit association task. Journal of Personality
and Social Psychology, 74, 1464 – 1480.
Heimberg, R. G., Hofmann, S. G., Liebowitz, M. R., Schneier, F. R., Smits, J. A., Stein, M. B., & Craske, M. G. (2014). SOCIAL ANXIETY DISORDER IN DSM‐5. Depression and anxiety, 31, 472 – 479.
Juth, P., Lundqvist, D., Karlsson, A., & Öhman, A. (2005). Looking for foes and friends: perceptual and emotional factors when finding a face in the crowd. Emotion, 5, 379 – 395. doi: 10.1037/1528-3542.5.4.379
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. Archives of General Psychiatry, 62, 593 – 602. doi:10.1001/archpsyc.62.6.593.
Koole, S.L., Dijksterhuis A., Van Knippenberg A. (2001). What's in a name: Implicit self-esteem and the automatic self. Journal of Personality and Social Psychology, 80, 669 – 685. doi: I0.1037//0022-3514.80.4.669
Lange, W. G., Keijsers, G., Becker, E. S., & Rinck, M. (2008). Social anxiety and evaluation of social crowds: Explicit and implicit measures. Behaviour Research and Therapy,
46, 932 – 943. doi:10.1016/j.brat.2008.04.008
Lundqvist, D., Bruce, N., & Öhman, A. (2014). Finding an emotional face in a crowd: Emotional and perceptual sti-mulus factors influence visual search efficiency.
Cognition and Emotion, (ahead-of-print), 1 -13. doi:10.1080/02699931.2014.927352
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.
MacLeod, C., & Mathews, A. (2012). Cognitive bias modification approaches to anxiety.
Annual Review of Clinical Psychology, 8, 189 – 217. doi:
10.1146/annurev-clinpsy-032511-143052
Mogg, K., & Bradley, B. P. (1998). A cognitive-motivational analysis of anxiety. Behaviour
Research and Therapy, 36, 809 – 848. doi:10.1016/S0005-7967(98)00063-1
Mogg, K., Bradley, B. P., Dixon, C., Fisher, S., Twelftree, H., & McWilliams, A. (2000). Trait anxiety, defensiveness and selective processing of threat: An investigation using two measures of attentional bias. Personality and Individual Differences, 28, 1063 – 1077. doi:10.1016/S0191-8869(99)00157-9
Rapee, R. M., & Heimberg, R. G. (1997). A cognitive-behavioral model of anxiety in social phobia. Behaviour Research and Therapy, 35, 741 – 756.
doi:10.1016/S0005-7967(97)00022-3
Rossignol, M., Campanella, S., Bissot, C., & Philippot, P. (2013). Fear of negative evaluation and attentional bias for facial expressions: an event-related study. Brain and
Cognition, 82, 344 – 352. doi:10.1016/j.bandc.2013.05.008
Schmukle, S. C. (2005). Unreliability of the dot probe task. European Journal of Personality,
19, 595 – 605. doi: 10.1002/per.554
Schmidt, N. B., Richey, J. A., Buckner, J. D., & Timpano, K. R. (2009). Attention training for generalized social anxiety disorder. Journal of Abnormal Psychology, 118, 5 – 14. DOI: 10.1037/a0013643
Schneier, F. R., Heckelman, L. R., Garfinkel, R., Campeas, R., Fallon, B. A., Gitow, A. & Liebowitz, M. R. (1994). Functional impairment in social phobia. Journal of Clinical
Psychiatry.
Staugaard, S. R. (2010). Threatening faces and social anxiety: a literature review. Clinical psychology review, 30, 669 –690. doi:10.1016/j.cpr.2010.05.001
Stein, M. B., & Stein, D. J. (2008). Social anxiety disorder. The Lancet, 371, 1115 – 1125. doi:10.1016/S0140-6736(08)60488-2
Stroop, J. R. (1938). Factors affecting speed in serial verbal reactions. Psychological
Tanner, R. J., Stopa, L., & De Houwer, J. (2006). Implicit views of the self in social anxiety. Behaviour Research and Therapy, 44(10), 1397-1409.
doi:10.1016/j.brat.2005.10.007
Tolman, R. M., Himle, J., Bybee, D., Abelson, J. L., Hoffman, J., & Van Etten-Lee, M. (2009). Impact of social anxiety disorder on employment among women receiving welfare benefits. Psychiatric Services, 60, 61 – 66. doi: 10.1176/appi.ps.60.1.61 Van, Bockstaele, B. Verschuere, B., Koster, E. H., Tibboel, H., De Houwer, J., & Crombez,
G. (2011). Differential predictive power of self-report and implicit measures on behavioural and physiological fear responses to spiders. International Journal of
Psychophysiology, 79, 166 – 174. doi:10.1016/j.ijpsycho.2010.10.003
Van, Bockstaele, B., Verschuere, B., Tibboel, H., De Houwer, J., Crombez, G., & Koster, E. H. (2014). A review of current evidence for the causal impact of attentional bias on fear and anxiety. Psychological Bulletin, 140, 682 – 721. doi: 10.1037/a0034834 Van Damme, S., Crombez, G., & Notebaert, L. (2008). Attentional bias to threat: a perceptual
accuracy approach. Emotion, 8, 820 – 827. doi: 10.1037/a0014149
Vogt, J., De Houwer, J., Koster, E. H., Van Damme, S., & Crombez, G. (2008). Allocation of spatial attention to emotional stimuli depends upon arousal and not valence. Emotion,
8, 880 – 885. doi: 10.1037/a0013981
Werner, K. H., Jazaieri, H., Goldin, P. R., Ziv, M., Heimberg, R. G., & Gross, J. J. (2012). Self-compassion and social anxiety disorder. Anxiety, Stress & Coping, 25, 543 – 558. doi:10.1080/10615806.2011.608842
Williams, J. M. G., Watts, F. N., MacLeod, C., & Mathews, A. (1997). Cognitive psychology
