Content uploaded by Ann M Kring
Author content
All content in this area was uploaded by Ann M Kring
Content may be subject to copyright.
Interpersonal Consequences of Social Anxiety
Erin A. Heerey
University of Wales, Bangor
Ann M. Kring
University of California, Berkeley
The behavioral manifestations of social anxiety may have implications for social outcomes. Unfortu-
nately, little is known about how anxiety shapes social interaction. The present study examined social
interactions in dyads consisting of either 2 nonsocially anxious (NSA) individuals or 1 socially anxious
(SA) and 1 NSA individual. Behavior, self-reported affect, and perceptions were examined. In compar-
ison with the interactions of NSA pairs, high levels of fidgeting, poor reciprocity of smiling behavior,
more self-talk, and more frequent reassurance seeking and giving characterized interactions between SA
and NSA participants. Both SA participants and their NSA partners rated their interactions as being less
smooth and coordinated than did participants in NSA–NSA dyads. In addition, SA participants’
reassurance seeking and self-talk correlated negatively with partner positive affect and perceptions of
interaction quality. The authors discuss self-focused attention and the interpersonal consequences of
social anxiety.
Keywords: social anxiety, social interaction, emotion, sequential analysis
The anecdotal observation that nonanxious people experience
interactions with socially anxious people as “odd” or “off” (Bar-
low, 2002) has spurred a body of work aimed at understanding the
social consequences of social anxiety. Although this work has
generated valuable knowledge about the nature of social anxiety,
much of it has focused on the acute arousal and anxiety-related
cognitions experienced by people with the disorder. However,
individuals with social anxiety worry about social outcomes, and
these worries manifest in behavior (Leary & Kowalski, 1995a).
Thus, the intrapersonal aspects of social anxiety have interpersonal
consequences, about which much less is known. Insofar as the
experience of interaction affects how relationships develop, be-
havioral manifestations of social anxiety may have far-reaching
implications for social outcomes.
There is general agreement that social anxiety is related to
alterations in behavior during interaction (Wenzel, Graff-Dolezal,
Macho, & Brendle, 2005). Nonetheless, the ways in which behav-
iors change, along with their causes and consequences, are still
debated (Baker & Edelmann, 2002). The concept of self-focused
attention (see Spurr & Stopa, 2002), which manifests in the in-
trapersonal experience of social anxiety (Clark & Wells, 1995),
provides a useful framework for predicting how acute anxiety may
be revealed in behavior.
Evidence suggests that self-focused attention and anxious
arousal are coupled such that increases in anxiety correspond with
increases in self-focus. Specifically, anxious arousal during inter-
action stems from self-presentational concerns, including the idea
that a negative evaluation of one’s self is forthcoming or that signs
of anxiety such as shaking or sweating may be noticeable (e.g.,
Cartwright-Hatton, Tschernitz, & Gomersall, 2005; Clark &
Wells, 1995; Kocovski & Endler, 2000; Wells & Papageorgiou,
1998). These concerns lead to increases in self-focused attention
(Cheek & Briggs, 1990; Kowalski & Leary, 1990; Leary, Kowal-
ski, & Campbell, 1988) that predict participants’ anxiety and affect
across interactions (Woody, 1996). As self-focused attention rises,
the ability to concentrate on social interaction may decline, leading
to disjointed social performances (Clark & McManus, 2002; Per-
owne & Mansell, 2002; Rapee, 1993).
Although individuals with social anxiety wish to avoid self-
disclosure (Clark & Wells, 1995), it may be the case that self-focus
interferes with this desire, particularly in unstructured situations in
which participants’ behavior is not guided by explicit rules. For
example, studies have shown that anxious individuals disclose less
in structured situations designed to elicit disclosure (DePaullo,
Epstein, & LeMay, 1990; Meleshko & Alden, 1993), although
naturalistic studies have failed to replicate the finding (Reno &
Kenny, 1992; Thompson & Rapee, 2002). One reason for the
discrepancy may be that shy or anxious individuals are less likely
to direct conversation during naturalistic social interactions (Pilko-
nis, 1977). Reticence to do so may have paradoxical effects: Those
who fail to direct conversation may find themselves answering
more questions and revealing more personal information.
Another consequence of self-focused attention may be increased
awareness of anxious arousal. To allay this anxiety, socially anx-
ious individuals report engaging in actions designed to forestall the
occurrence of feared social outcomes (Clark & Wells, 1995; Wells,
Clark, Salkovskis, & Ludgate, 1995). One such anxiety regulation
strategy may be to engage in “checking” behavior to ascertain the
likelihood of a feared outcome (Mennin, Heimberg, Turk, &
Fresco, 2002). Excessive worry, along with the perception that
outcomes are controllable, both elements of social anxiety (e.g.,
Erin A. Heerey, School of Psychology, University of Wales, Bangor,
United Kingdom; Ann M. Kring, Psychology Department, University of
California, Berkeley.
We wish to thank Steve Bistricky, Wei-Bing Chen, Kathy Derangi,
Jennifer Lacy, Janell Lynch, Natalya Maisel, and Christine Wong for their
invaluable assistance with coding videotapes. We also wish to thank James
Gold for his helpful comments on drafts of this article.
Correspondence concerning this article should be addressed to Erin A.
Heerey, School of Psychology, University of Wales, Bangor, Adeliad
Brigantia, Gwynedd LL57 2AS, United Kingdom. E-mail:
e.heerey@bangor.ac.uk
Journal of Abnormal Psychology Copyright 2007 by the American Psychological Association
2007, Vol. 116, No. 1, 125–134 0021-843X/07/$12.00 DOI: 10.1037/0021-843X.116.1.125
125
Leary & Kowalski, 1995b), has been related to compulsive check-
ing in generalized anxiety and obsessive-compulsive disorders
(Mennin et al., 2002). Similar behavior may occur in social anx-
iety, although this idea has not yet been examined. If so, one might
expect individuals with social anxiety to seek affirmation or reas-
surance at higher rates than nonanxious individuals.
Awareness of one’s high negative and low positive affect, an
aspect of self-focused attention (Flory, Raikkonen, Matthews, &
Owens, 2000), may impact nonverbal behavior. In particular,
individuals with social anxiety may inadvertently communicate the
distress of which they are so keenly aware (Keltner & Haidt,
1999). Fidgeting, for example, signals psychomotor agitation
(Okazaki, Liu, Longworth, & Minn, 2002) and may be elevated
among individuals with social anxiety. Frowns also indicate neg-
ative affect (Ekman, 1992) and may, therefore, be elevated. More-
over, general anxiety is associated with reductions in smiling
(Field et al., 2005; Yovetich, Dale, & Hudak, 1990). Although
studies have found that smiling is “adequate” among individuals
with social anxiety (Baker & Edelmann, 2002), mixed results have
been found among shy individuals (Pilkonis, 1977). Acute aware-
ness of one’s reduced positive and elevated negative affect
(Hirsch, Meynen, & Clark, 2004; Kashdan & Roberts, 2004)
suggests reduced smiling, particularly with respect to smiles of
pleasure (Ekman, 1992) rather than the polite smiles associated
with sociability (Keltner & Haidt, 2001). These types of smiles and
their relationship to social anxiety have not been carefully disam-
biguated in the literature.
Taken together, the behaviors individuals communicate may
play an important role in shaping an interaction as well as more
distal social outcomes. Studies of negative feedback solicitation
hint at how this process may work. Individuals who solicit more
negative feedback from their social partners also receive more of
it (Casbon, Burns, Bradbury, & Joiner, 2005), along with more
negative social evaluations (Joiner & Metalsky, 1995; Katz &
Joiner, 2001). The effects of self-focus in social anxiety may be
similar, leading to subtle reinforcement of anxiety-related cogni-
tions. Studies suggest that willingness to engage in a future inter-
action with someone is related to that individual’s interpersonal
behavior, including smiles (LaFrance & Hecht, 1999; Palmer &
Simmons, 1995), communication of positive emotion (Berry &
Hansen, 1996), and the degree to which interaction is perceived as
being smooth and coordinated (Cappella, 1997). If the presence of
social anxiety disrupts these aspects of social behavior, individuals
with social anxiety may inadvertently provoke the negative per-
ceptions they seek to avoid.
In the present study, we sought to address the interpersonal
consequences of social anxiety by asking nonsocially anxious
(NSA) individuals to interact with either an NSA or a socially
anxious (SA) conversation partner. We expected SA participants to
ask fewer questions and to engage in more frequent self-focused
talk. Furthermore, we expected that self-talk would be related to
partner questions for SA but not NSA participants. We also pre-
dicted that SA participants would engage in more frequent
reassurance-seeking or support solicitations, which their conver-
sation partners were expected to provide, and would display less
positive affect (e.g., smiles) and more negative affect (e.g., fidg-
eting and frowns).
Dyads that included SA participants were expected to rate their
interactions as lower in quality than dyads that did not. We
predicted that self-reported affect would depend on partner anxiety
status. Because most people enjoy social interaction (Berry &
Hansen, 1996), we hypothesized that participants in NSA dyads
would report an increase in positive affect from pre- to postinter-
action, with little change in negative affect. Participants in mixed-
anxiety dyads were expected to report a rise in negative affect from
pre- to postinteraction, with little change in positive affect. Finally,
we expected that affect and ratings of interaction quality would be
related to partner behavior.
Method
Participants
Participants (n ⫽ 120) were recruited from a large undergrad-
uate sample (N ⫽ 2,754) on the basis of their scores on the
Interaction Anxiousness Scale (Leary & Kowalski, 1993). Individ-
uals scoring in the top and bottom 20th percentile of the sample
made up the SA (n ⫽ 30) and NSA (n ⫽ 90) groups, respectively.
Study participants additionally completed the Social Phobia and
Anxiety Inventory (SPAI; Turner, Beidel, Dancu, & Stanley,
1989). Relative to NSA participants, SA participants had signifi-
cantly higher SPAI scores, F(1, 118) ⫽ 217.06, p ⬍ .001 (see
Table 1), that fell within the SPAI clinical range (Turner et al.,
1989).
Thirty NSA women were randomly paired with either an SA
(n ⫽ 15) or NSA (n ⫽ 15) female partner, likewise for men.
Together, there were 30 SA–NSA dyads and 30 NSA–NSA dyads.
Participant characteristics appear in Table 1.
Procedure
Potential participants were recruited by phone for a study of
“how people feel when they get to know one another.” They
anticipated interacting twice with the same partner. On arrival, the
experimenter introduced participants and ascertained that they
were strangers. Participants completed a short assessment of cur-
rent affect and were escorted to a quiet reading lounge for the
interaction. A hidden camera recorded behavior. With the permis-
sion of the university’s institutional review board, we did not
inform participants of the videotaping procedures until study com-
pletion. Participants were instructed to “get to know” one another
and were left alone for the 5-min interaction period. When the time
had elapsed, the experimenter ended the interaction and distributed
questionnaires. To ensure independence of ratings, the experi-
menter had participants complete questionnaires in separate rooms.
Participants were then debriefed, informed about the videotaping
procedure, and given the opportunity to provide fully informed
consent. No participant declined consent.
Questionnaires
Self-report data consisted of the Positive and Negative Affect
Schedule (PANAS) and the Quality of Interaction (QI) scale. The
PANAS includes 10 positive affect words (e.g., attentive, inter-
ested) and 10 negative affect words (e.g., jittery, upset), rated on a
5-point Likert scale (1 ⫽ not at all;5⫽ very much) for the degree
to which participants “currently feel” each emotion (Watson,
Clark, & Tellegen, 1988). It has well-established psychometric
properties (Crawford & Henry, 2004). This measure was collected
126
HEEREY AND KRING
both immediately before and after the interaction. The QI, adapted
from Berry & Hansen (1996), measures perceptions of the smooth-
ness and coordination of interactions and satisfaction therewith
(e.g., “To what degree did you find the interaction enjoyable?” “To
what degree did you feel the interaction was awkward or
strained?”). The QI is rated on an 8-point scale (1 ⫽ not at all;8⫽
very much), and reliability analyses in the present sample showed
that the measure had good internal consistency (␣⫽.82).
Coders
Eight research assistants independently coded either verbal
(three coders) or nonverbal behavior (five coders) from videotapes
of the study sessions. All were naive to study hypotheses and
participants’ anxiety status. Prior to coding study data, verbal and
nonverbal behavior coders were trained to 95% agreement on a set
of sham interactions. As a check on posttraining reliability, each
coder independently coded the same three pilot interactions. Ver-
bal behavior coders achieved 94% agreement, and nonverbal be-
havior coders achieved 91% agreement. We examined agreement
by having two verbal behavior coders overlap their coding for half
the sessions (30 sessions, both participants). Agreement was as-
sessed among nonverbal behavior coders in similar fashion.
We used onset times from a time code stamped on the tape to
link dyad members’ behavior. The codes for each verbal behavior
(described below) were defined as mutually exclusive—that is, no
verbal code could occur simultaneously with another verbal code.
Nonverbal facial behaviors were also considered mutually exclu-
sive, as were hand movements (described below). Behaviors were
coded in 1-s time intervals. Coders identified which behavior was
most prominent during each second, including episodes of silence,
rest, and “other” behaviors not of theoretical interest to the present
investigation. Thus, each second of each participant’s interaction
included a code for one verbal behavior, one nonverbal facial
behavior, and one hand behavior. As there were no gaps in record-
ing, the offset time of one behavior was considered to have
happened during the second prior to the onset of the new behavior
(for a complete description of this method of coding and recording
data, see Bakeman & Quera, 1995).
Verbal Coding
Among the many verbal behaviors participants generated, we
focused on four a priori hypotheses: reassurance seeking (RS);
empathy and support (ES); questions asked (Q); and information
sharing, which we subdivided into talk about the self (ST) or
general/non–self-talk (GT). Verbal coders additionally rated epi-
sodes of silence. RS included complaints (e.g., “These research
studies make me kind of nervous”); apologies (e.g., “I’m sorry for
interrupting”; “I didn’t really mean that”); and direct requests for
advice (e.g., “What would you do?”), support (e.g., “Isn’t that
awful?”), or agreement (e.g., “Don’t you think so too?”). ES
included behaviors such as empathetic comments (e.g., “I feel the
same way”), defending (e.g., “I would have done the same thing”),
and advice (e.g., “You should go talk to the professor about that”).
Q involved information solicitation (e.g., “Where are you from?”
“Did you see the game last night?”). ST included information
sharing with the self as the primary focus (e.g., “I’m living in the
dorms this semester”; “I have three brothers”), whereas GT was
general information not related to self (e.g., “Professor Covington
is the teacher for Psych 1 this semester”; “I heard that Psych 192
is one of the best classes on campus”). Verbal behavior was absent
during episodes of silence.
Tests of interrater agreement were conducted on the 30 sessions
that had been coded by two raters. Cohen’s kappa coefficients
ranged from .72 to .96. Coders demonstrated acceptable reliability
as evidenced by kappas of at least .70 (Bakeman & Gottman,
1997).
Nonverbal Coding
Coders examined participants’ faces for polite smiles (smiles
lacking involvement of the eye region), pleasurable smiles (smiles
Table 1
Participant Demographics, Anxiety Ratings, and Personality Characteristics
Variable
Nonsocially anxious (NSA) Socially anxious (SA)
nM SDnM SD
Participant 90 30
Men 45 15
Women 45 15
Age (years) 18.75 0.82 18.97 1.05
Ethnicity
Caucasian 48 14
Asian 26 12
Middle Eastern 4 2
Latino/Latina 5 1
Indian 3 1
African American 4 0
IAS 13.81 4.64 48.09 4.29
SPAI (Social Phobia subscale)
a
25.91 18.96 86.61 20.50
Note. Mean differences across groups on the Interaction Anxiousness Scale (IAS) and Social Phobia and
Anxiety Inventory (SPAI) were statistically significant as expected ( p ⬍ .001). No other differences existed.
a
Means from the SPAI normative samples as reported in Turner et al. (1989) for clinically identified students
⫽ 72.2 and for adults with social phobia ⫽ 94.0.
127
SOCIAL ANXIETY IN DYADIC INTERACTION
involving changes in the apparent shape of the eye), and frowns
(brow furrowing with or without involvement of the mouth). They
also coded participants’ hands for fidgeting. Other hand and face
behaviors along with episodes of rest were not considered relevant
to the present study and are not discussed. As with verbal behavior,
two raters coded half the sessions to check agreement. Kappas
ranged from .85 to .94.
Data Analysis
The design included three participant types: NSA participants
who interacted with other NSA participants (NSA with NSA),
NSA participants who interacted with SA participants (NSA with
SA), and SA participants (SA). Participant types were nested
within dyads that were of two dyad types: dyads consisting of two
NSA participants (NSA–NSA) or dyads including one SA and one
NSA participant (SA–NSA). Finally, because individuals inter-
acted within dyads, conversation partners’ experiences during in-
teraction were not statistically independent, creating a third source
of variance. Although this variance was not of theoretical interest
in the present investigation, it was necessary to partition it out of
the statistical model so that participant type differences could be
tested, independent of the random effects associated with dyadic
pairing. These analyses, mathematically equivalent to the intra-
class correlation (see Kenny, 1996; Kenny & la Voie, 1985), are
not reported here.
Hypotheses predicting differences in behavior and self-report
measures were tested using the analysis of variance model dis-
cussed in Kenny and la Voie (1985) to control dyad-level inter-
dependence. We further examined significant omnibus tests of
participant type for differences between the three levels of this
variable by using Scheffe´’s correction for Type I error. To more
closely examine predictions about social reciprocity, we examined
sequences of verbal and nonverbal behaviors at the dyad level for
the presence of reciprocal interaction patterns (see Bakeman &
Gottman, 1997), such as one participant smiling in response to the
other’s smiles. Chi-square analyses measured the degree to which
nonverbal sequences across dyad types fit an expected model.
Results
We predicted that SA participants would ask fewer questions
and produce more self-focused talk, which was in turn expected to
relate to questions asked by their partners. In addition, we pre-
dicted that SA participants would seek, and their partners would
provide, more reassurance. Except as noted, descriptive statistics
and analyses (results and p values) are reported in Table 2.
Conversation
Overall, participants in SA–NSA dyads were silent for a greater
proportion of the interaction, and NSA–NSA dyads engaged in
more frequent episodes of general talk. Consistent with prediction,
SA participants’ verbal behavior consisted of more frequent epi-
sodes of self-talk (ST), relative to either of the other participant
groups. SA participants also asked fewer questions than did their
partners, although SA participants did not differ from participants
in NSA–NSA dyads in number of questions asked. It is possible
that questions elicited less information in SA–NSA dyads, as
participants in NSA–NSA dyads tended to talk longer after each
Table 2
Verbal and Nonverbal Behavior
Variable
NSA–NSA
dyads NSA–SA dyads
Dyad type
difference
Participant type
difference
NSA with
NSA
(n ⫽ 60)
NSA with
SA (n ⫽ 30) SA (n ⫽ 30)
MSDMSDMSDFpF p
Total silence .40 .10 .42 .09 .46 .10 4.30 .04 1.98 .16
Verbal behavior
Self-talk 15.82 4.84 15.97 5.91 19.46 7.37 2.59 .11 5.75 .02
a,b
General talk 10.74 4.74 9.52 4.59 7.79 5.22 4.50 .04 1.64 .21
Questions asked 11.74 4.86 14.76 6.74 11.46 5.43 2.11 .15 5.00 .03
c
Reassurance seeking 2.18 1.66 2.72 2.20 4.64 2.92 2.34 .13 5.88 .02
a
Empathy and support 0.92 0.76 3.14 1.39 0.82 0.61 4.15 .05 9.96 < .01
c,d
Facial behavior
Polite smiles 10.66 3.51 10.98 4.32 13.93 4.09 3.12 .08 8.93 < .01
a,b
Pleasant smiles 6.08 3.76 5.65 3.14 7.15 3.32 0.86 .36 0.49 .48
Frowns 2.50 3.69 1.96 2.78 1.22 1.69 1.80 .19 0.96 .33
Hand behavior
Fidgeting .24 .18 .31 .21 .34 .24 4.80 .03 0.18 .67
Note. Bold type indicates significant differences. Total silence and fidgeting are reported as proportions of total
interaction time because the frequencies of these behaviors were highly related to episodes of partner behavior
(silence with partner speech acts and fidgeting episodes with partner fidgeting episodes, ps ⬍ .01). All other
behaviors are reported as average frequencies. NSA ⫽ nonsocially anxious; SA ⫽ socially anxious.
a
SA participants differ from NSA with NSA participants ( p ⬍ .05).
b
SA participants differ from NSA with
SA participants ( p ⬍ .05).
c
NSA with SA participants differ from SA participants ( p ⬍ .05).
d
NSA with
SA participants differ from NSA with NSA participants ( p ⬍ .05).
128
HEEREY AND KRING
question than did those in SA–NSA dyads (M
NSA–NSA
⫽ 3.60,
SD ⫽ 2.45; M
SA–NSA
⫽ 2.46, SD ⫽ 1.56), F(1, 54) ⫽ 3.26, p ⫽
.08. To test the idea that SA participants generated self-talk in
response to partner questions, we tallied instances of self-talk that
were preceded by partner questions and those preceded by any
other partner behavior. Contrary to prediction, SA participants
produced more unsolicited self-talk than did their NSA partners,
likelihood
2
(1, N ⫽ 60) ⫽ 3.90, p ⫽ .04 (see Figure 1
). NSA with
NSA participants tended to engage in more unsolicited versus
solicited self-talk, although this difference did not reach traditional
levels of statistical significance,
L
2
(1, N ⫽ 60) ⫽ 2.84, p ⫽ .09.
With respect to RS, SA participants produced more than NSA
with NSA participants and tended to produce more than their own
partners ( p ⫽ .07; see Figure 2). More ES was provided in
SA–NSA dyads, with the NSA members of those dyads producing
more than both SA participants and participants in NSA–NSA
dyads. Sequential analyses indicated that RS was more likely than
any other type of behavior to precede ES,
goodness of fit
2
(2, N ⫽
120) ⫽ 6.73, p ⬍ .01, and did so 92% of the time. Despite the fact
that more RS and ES occurred among participants in SA–NSA
dyads, the pattern of behavior, RS followed by ES, did not differ
across the dyads,
L
2
(1, N ⫽ 60) ⫽ 1.06, p ⫽ .89.
Nonverbal Behavior
We had expected SA participants to display fewer pleasant and
more polite smiles. Surprisingly, pleasant smiles were displayed
with similar frequency across both participant types and dyad
types. Consistent with predictions, frequency of polite smiles
tended to differ for dyad types, and SA participants produced more
polite smiles than did either of the NSA participant groups. To
examine smiling reciprocity, we tallied the polite and pleasurable
smile onsets of each participant, given the smiling behavior of his
or her partner, at a lag of 1 s. In both dyad types, participants
frequently smiled back at their partners rather than engaging in any
other facial behavior,
GFI
2
(4, N ⫽ 120) ⫽ 30.18, p ⬍ .01. In
SA–NSA dyads, SA participants were likely to smile politely in
response to both types of partner smiles,
L
2
(1, N ⫽ 60) ⫽ 1.54,
p ⫽ .25, whereas their NSA partners were more likely to respond
with a matching smile,
L
2
(1, N ⫽ 60) ⫽ 7.02, p ⫽ .01. Polite and
pleasurable smile responses typically matched the partner’s smile
type in NSA–NSA dyads,
L
2
(1, N ⫽ 60) ⫽ 12.17, p ⬍ .01 (see
Figure 3). Thus, although SA participants returned their partners’
smiles, they were less likely to reciprocate with a smile of the same
type.
We predicted that SA participants would display more negative
affect, including frowns and fidgeting, than NSA participants.
Although the frequency of frowns did not differ across dyad or
participant types, SA–NSA dyads fidgeted more than NSA–NSA
dyads. Interestingly, fidgeting did not discriminate participants in
SA–NSA dyads. Sequential analysis was used to examine the
hypothesis that fidgeting may be “contagious.” It was often true
that partners in both dyad types fidgeted simultaneously, rather
than when a partner was still (time spent fidgeting jointly:
M
NSA–NSA
⫽ 62.61 s, SD ⫽ 30.01; M
SA–NSA
⫽ 86.42 s, SD ⫽
44.18; time spent fidgeting alone: M
NSA–NSA
⫽ 9.47 s, SD ⫽
4.17; M
SA–NSA
⫽ 11.17 s, SD ⫽ 4.61),
L
2
(1, N ⫽ 60) ⫽ 0.05,
p ⫽ .82. Further analysis confirmed that SA participants initi-
ated bouts of joint fidgeting more frequently than their NSA
partners,
L
2
(1, N ⫽ 60) ⫽ 6.78, p ⬍ .01. Thus, fidgeting
appeared to be transmitted across interaction partners and was
more frequently initiated by SA participants.
Interaction Outcomes
Prior to beginning the interaction, participants differed on
reports of both negative and positive affect (descriptive statis-
tics and all analyses are shown in Table 3). Relative to NSA
participants, SA participants reported less positive affect (PA)
and more negative affect (NA). We expected participants in
SA–NSA dyads to report increased NA from pre- to postinter-
action, yet all participants reported a drop in NA. At the dyad
level, analyses of change scores showed a tendency for SA–
NSA dyads to decrease in NA more than NSA–NSA dyads.
SA participants, who showed the largest drop in NA, likely
powered this change. Overall levels of postinteraction NA were
not significantly different among either dyad or participant
types.
Examination of change in PA over the interaction revealed no
differences at the level of dyad type but did reveal significant
participant type differences. These did not occur as predicted. NSA
with SA participants were the only participants who did not report
an increase in PA. They differed significantly from both other
participant groups in the degree to which positive emotion
changed. Although SA participants experienced the greatest rise in
positive emotion, their postinteraction levels of PA remained sig-
nificantly lower than those of NSA with NSA participants, al-
though SA participants did not differ from their own partners ( p ⫽
.99).
Figure 1. Average frequency of self-talk episodes preceded by questions
and those preceded by other behaviors. Data from nonsocially anxious
(NSA) with NSA participants reflect both members of the NSA–NSA
dyads, as these individuals are not distinguishable from one another on the
basis of a priori characteristics. The columns representing NSA with
socially anxious (SA) participants and SA participants together reflect the
data from SA–NSA dyads. In this case, participants’ data are displayed
independently as the participant types are distinguishable.
129
SOCIAL ANXIETY IN DYADIC INTERACTION
As predicted, NSA–NSA dyads reported higher QI than did
SA–NSA dyads, although participant types did not differ. This
finding suggests that NSA with SA participants experienced their
interactions differently from participants in NSA–NSA dyads.
Consequences of Affect and Behavior for Partner
Outcomes
Several behaviors distinguished SA participants from NSA
participants. These included more self-relevant talk, altered
smiling behavior, and more RS, which resulted in increased ES
from partners. In addition, participants differed on self-reported
affect and interaction quality. To understand how participants’
behavior and affect related to their partners’ affect and percep-
tions of interaction quality, we computed partial correlations
between an interactant’s behavior and affect and partner’s af-
fect and QI ratings, controlling for the partner’s preinteraction
affect. With respect to behavior, among NSA with NSA partic-
ipants and among SA participants, increases in RS predicted
decreases in partner QI (NSA with NSA: r
p
⫽ –.30, p ⫽ .02;
SA: r
p
⫽ –.36, p ⫽ .03). Among SA participants, increases in
self-talk corresponded to decreases in partner PA (r
p
⫽ –.35,
p ⫽ .03) and QI (r
p
⫽ –.34, p ⫽ .04). In addition, the more
pleasant smiling SA participants produced, the more PA their
partners reported (r
p
⫽ .38, p ⫽ .02). Finally, the more ES that
NSA with SA participants produced, the more PA their SA
partners reported (r
p
⫽ .65, p ⬍ .001). Relationships between
participants’ affect and partners’ affect and QI ratings showed
somewhat different patterns. In NSA–NSA dyads, dyad mem-
bers’ PA was correlated (r
p
⫽ .34, p ⫽ .01), and participant PA
also related to partner QI (r
p
⫽ .29, p ⫽ .02). Similarly, NSA
with SA participants’ PA related to partner PA (r
p
⫽ .33, p ⫽
.04). SA participants’ affect was unrelated to partner affect or
QI. Notably, NA was uncorrelated with partner QI, behavior, or
affect. These findings suggest that alterations in PA and social
Figure 2. A: Reassurance-seeking behaviors across participant types. B: Empathy and support provided across
participant types. Data from nonsocially anxious (NSA) with NSA participants reflect both members of the
NSA–NSA dyads, as these individuals are not distinguishable from one another on the basis of a priori
characteristics. The columns representing NSA with socially anxious (SA) participants and SA participants
together reflect the data from SA–NSA dyads. In this case, participants’ data are displayed independently as the
participant types are distinguishable.
Figure 3. Percentage of responses to polite and pleasurable stimulus smiles
across participant type. These do not sum to 100 because participants did not
respond to smiles with other smiles 100% of the time. Both types of nonso-
cially anxious (NSA) participants reciprocated their partners’ smile types
( ps ⬍ .05). Data from NSA participants reflect both members of the NSA–
NSA dyads, as these individuals are not distinguishable from one another on
the basis of a priori characteristics. The columns representing NSA with
socially anxious (SA) participants and SA participants together reflect the data
from SA–NSA dyads. In this case, participants’ data are displayed indepen-
dently as the participant types are distinguishable.
130
HEEREY AND KRING
behavior have consequences for a partner’s experience of PA
and perception of interaction quality.
Discussion
The presence of social anxiety during interaction had effects on
behavior, affect, and perception that were salient to both anxious
and nonanxious interactants alike. Interactions between SA and
NSA participants were characterized by less reciprocity in smiling
and more fidgeting. The NSA members of those dyads offered
more support and empathy compared with individuals in NSA–
NSA dyads, viewed their interaction quality as being lower, and
failed to experience the increase in positive affect reported by
other participants. SA participants, despite a desire to execute
smooth social performances, asked fewer questions, engaged in
more self-focused talk, and sought more reassurance. Both self-
talk and reassurance seeking decreased partner perceptions of
interaction quality. In “getting to know you” interactions, a fre-
quent occurrence on college campuses, these signals of anxiety
may seem out of place and uncomfortable, thereby serving to
decrease the likelihood of future interaction.
These findings support the idea that self-focused attention fig-
ures prominently into the characterization of social anxiety, al-
though they suggest that it is not necessarily related to decreased
willingness to direct conversation (Pilkonis, 1977). According to
models of self-focus in social anxiety (e.g., Clark & Wells, 1995),
when anxious individuals enter a social interaction, they focus
attention on the self for the purposes of monitoring self-
presentation. Thus, attention is shifted to internally generated
information, including physiological changes, thoughts, feelings,
and beliefs. Worries about social performance are then used to
generate negative feelings that serve as feedback, thereby increas-
ing anxiety (Spurr & Stopa, 2002). Insofar as being self-focused
leads to self-focused communication, SA participants in the
present study demonstrated self-awareness both verbally and non-
verbally. Their partners appeared to notice. Research has shown
that individuals who converse with a distressed partner report
lower feelings of engagement and more social restraint (Furr &
Funder, 1998; Joiner & Metalsky, 1995). Judging by the degree to
which their partners inhibited spontaneous self-talk and showed
diminished perceptions of interaction quality, it is likely that the
NSA partners of SA participants perceived the unbalanced con-
versational focus and found these interactions less rewarding than
usual.
Excessive reassurance seeking in conversation may prove a
good target for behavioral intervention in social anxiety. The
reassurance seeking in which SA participants engaged is consistent
with the idea of compulsive checking behavior as an emotion
regulation strategy (Casbon et al., 2005; Mennin et al., 2002). This
study provides initial evidence that such behavior is related to the
presence of social anxiety and may be one of a number of “safety-
seeking behaviors” used by anxious individuals (Clark & Wells,
1995). Although the present design did not allow a clear test of
this, it may have been the case that fluctuations in negative
emotion during the task fueled instances of reassurance seeking.
Nonverbal behaviors provide a significant amount of informa-
tion during the course of interaction (Chovil, 1991; Keltner &
Haidt, 1999). The pleasurable smiles associated with positive
emotion have been shown to be a form of social reward
(O’Doherty et al., 2003). Receiving these types of smiles is related
to positive emotion (Cappella, 1997) and reciprocal smiling be-
Table 3
Self-Reported Interaction Quality and Affect
Measure
NSA–NSA SA–NSA
Dyad type
difference
Participant type
difference
NSA with
NSA (n ⫽
60)
NSA with
SA (n ⫽ 30) SA (n ⫽ 30)
MSDMSDMSDFp F p
Interaction quality 41.97 7.12 35.79 8.18 36.21 8.45 9.84 .01 0.82 .43
Preinteraction affect
a
Negative 19.92 6.11 21.03 6.23 26.09 7.46 19.34 < .01
Positive 28.65 7.28 28.59 5.02 24.94 6.39 7.68 < .01
Postinteraction affect
b
Negative 17.26 3.54 18.56 5.75 20.41 6.73 0.44 .51 0.11 .75
Positive 30.61 8.45 28.06 5.58 27.94 7.15 1.07 .50 5.04 .03
c
Difference score
Negative ⫺2.67 4.95 ⫺2.47 4.78 ⫺5.69 5.76 2.77 .10 5.07 .02
d
Positive 1.95 5.78 ⫺0.53 3.97 3.00 5.59 0.47 .50 9.35 < .01
e
Note. Bold type indicates significant differences. Scores refer to group means. Difference scores reflect
postinteraction-preinteraction values. NSA ⫽ nonsocially anxious; SA ⫽ socially anxious.
a
Dyad type was not tested because no interaction had occurred at the time of rating; both types of NSA
participants were combined into one group. SA participants differed from all NSA participants.
b
Pre- and
postinteraction negative affect (NA) were correlated; likewise for positive affect (PA). Therefore, variance
associated with preinteraction affect (either NA or PA) was used as a covariate in these analyses.
c
SA
participants differ from NSA with NSA participants ( p ⬍ .05).
d
SA participants differ from both NSA with
NSA and NSA with SA participants ( p ⬍ .05).
e
NSA with SA participants differ from both NSA with NSA
and SA participants ( p ⬍ .05).
131
SOCIAL ANXIETY IN DYADIC INTERACTION
havior (Fridlund, 1991). Although SA participants smiled in re-
sponse to their partners’ smiles, they failed to match polite and
pleasant smile types. Although it may be socially appropriate to
return a smile, important social rewards may be lost when plea-
surable smiles are frequently reciprocated with polite ones. NSA
participants with SA partners were the only group that did not
experience a rise in positive emotion. By failing to match their
partners’ smile types, SA participants may have failed to provide
appropriate social rewards, leading their partners to experience less
positive emotion.
1
Moreover, SA participants elicited fidgeting, a
signal of anxiety (Okazaki et al., 2002), in their partners. It is not
inconceivable that the NSA partners of SA participants experi-
enced some degree of agitation in response to increased fidgeting.
Surprisingly, participants’ self-reported affect did not change as
predicted over the course of interaction. Although we had antici-
pated that participants in SA–NSA dyads would show increased
NA during the interaction, in fact they did not. All participants
reported decreases in NA from pre- to postinteraction. It has been
suggested that social interaction leads to decreases in negative
emotion (Berry & Hansen, 1996), which may have been the case
even for the NSA partners of SA participants. It may also have
been true that all participants, and particularly those in SA–NSA
dyads, were relieved to have finished the interaction portion of the
study.
With respect to positive affect, participants who interacted with
NSA participants reported increases in positive affect during the
interaction, whereas those who interacted with SA participants
reported essentially baseline levels of positive affect. In contrast to
the present study, previous findings that socially anxious individ-
uals experience more negative affect and less positive affect during
interaction have typically involved confederates (e.g., Kashdan &
Roberts, 2004). Because confederates are carefully trained to be-
have consistently across study interactions, they may have failed to
respond to some of the SA participants’ behavior (e.g., not pro-
viding reassurance), leading SA participants to experience and
report more negative affect and less positive affect.
Reciprocity is an important aspect of social interaction. Behav-
iors, such as smiling, fidgeting, and providing support in response
to distress, are typically reciprocated. However, all of these be-
haviors were altered in SA participants. NSA partners of SA
participants changed their behavior in response to these alterations.
Previous research has suggested that people who interact with
partners who reciprocate social cues in atypical fashion report
decreased interaction quality and liking for their social partners
(Cappella, 1985). The increased self-talk, persistent fidgeting, and
inconsistency in smiling reciprocity on the part of the SA partic-
ipants may have interfered with the expected patterns of these
behaviors, leading to NSA with SA participants’ less positive
affect and perceptions of interaction quality.
Although this is one of the first studies to address sequential
social behavior in social anxiety, it is important to acknowledge
some limitations. In particular, the target participants were socially
anxious college students. Although these participants experienced
significant social distress, indicated by SPAI scores near the clin-
ical mean (Turner et al., 1989), anxious college students likely
have more opportunity for interaction than many individuals with
clinical diagnoses. However, it has been argued that the differ-
ences between socially anxious and socially phobic individuals are
more a matter of distress intensity than of differences in symptom
quality (Beidel & Turner, 1999; Leary & Kowalski, 1995b). It may
be true, then, that symptoms vary by degree but are not qualita-
tively different between individuals with social anxiety and those
with social phobia. This idea is related to a second limitation, that
several of the reported effects were small in size. Small effect sizes
may be due to having studied a nonclinical sample. It may also be
the case that the 5-min interaction was not especially powerful, and
a longer interaction period may have strengthened the results. As
with any study, replication of these findings, particularly in a
clinically diagnosed sample, will be important. Even so, the
present study confirms the idea that social anxiety is manifested in
social behavior and is, therefore, important to understanding both
the nature and maintenance of social anxiety.
Conclusion
During interaction, SA participants were less likely to match the
smile types of their partners, initiated more joint fidgeting, en-
gaged in more self-talk, and asked fewer questions. They also
engaged in more reassurance seeking, to which their partners
responded. These patterns may have led their partners to experi-
ence the interactions as subtly one-sided—focused on their SA
partners. One implication of these findings is that internal states
are revealed in external behavior and, therefore, have conse-
quences for social outcomes. Thus, the self-focus exhibited by SA
participants may affect both their social interactions and social
relationships. For NSA participants, who likely carry on numerous
pleasurable interactions with other NSA individuals over the
course of a day, interactions with SA participants did not produce
increases in positive affect. Failure to experience an interaction as
rewarding may decrease the likelihood of future interaction,
thereby serving to reinforce anxiety-related cognitions in individ-
uals with social anxiety.
The degree to which one is able to enjoy and benefit from
interaction impacts the development of personal relationships. By
remaining fixed on their internal experience, evident in the trans-
mission of anxious arousal, poor reciprocation of positive affect,
and repeated reassurance seeking, SA individuals may inadver-
tently hinder the development of personal relationships and social
support networks. As with any study, these findings require rep-
lication. Nonetheless, it is clear that the presence of social anxiety
and self-focused attention have important consequences within the
context of social interaction that are apparent to conversation
partners and have the potential to affect social outcomes.
1
As a caveat to the finding of differences in polite and pleasurable
smiles, it is important to note that, although every effort was made to
accurately code the tapes, a precise facial coding system, such as the Facial
Action Coding System (Ekman & Friesen, 1978), was not used. Because of
the design of the study, it was not possible to code facial behavior at close
range, making the use of the Facial Action Coding System impossible.
Despite the fact that coders achieved high levels of reliability and were
blind to participants’ anxiety status, it may have been true that some smiles
were coded incorrectly simply because coders were unable to detect facial
changes. For this reason, differences in smiling behavior must be inter-
preted with caution.
132
HEEREY AND KRING
References
Bakeman, R., & Gottman, J. M. (1997). Observing interaction: An intro-
duction to sequential analysis (2nd ed.). New York: Cambridge Univer-
sity Press.
Bakeman, R., & Quera, V. (1995). Analyzing interaction: Sequential
analysis with SDIS & GSEQ. New York: Cambridge University Press.
Baker, S. R., & Edelmann, R. J. (2002). Is social phobia related to lack of
social skills? Duration of skill-related behaviours and ratings of behav-
ioural adequacy. British Journal of Clinical Psychology, 41, 243–257.
Barlow, D. H. (2002). Anxiety and its disorders: The nature and treatment
of anxiety and panic (2nd ed.). New York: Guilford Press.
Beidel, D. C., & Turner, S. M. (1999). The natural course of shyness and
related syndromes. In L. A. Schmidt & J. Schulkin (Eds.), Extreme fear,
shyness, and social phobia: Origins, biological mechanisms, and clini-
cal outcomes (pp. 203–223). New York: Oxford University Press.
Berry, D. S., & Hansen, J. S. (1996). Positive affect, negative affect, and
social interaction. Journal of Personality and Social Psychology, 71,
796 –809.
Cappella, J. N. (1985). Production principles for turn-taking rules in social
interaction: Socially anxious vs. socially secure persons. Journal of
Language and Social Psychology, 4, 193–212.
Cappella, J. N. (1997). Behavioral and judged coordination in adult infor-
mal social interactions: Vocal and kinesic indicators. Journal of Per-
sonality and Social Psychology, 72, 119 –131.
Cartwright-Hatton, S., Tschernitz, N., & Gomersall, H. (2005). Social
anxiety in children: Social skills deficit, or cognitive distortion? Behav-
iour Research and Therapy, 43, 131–141.
Casbon, T. S., Burns, A. B., Bradbury, T. N., & Joiner, T. E. (2005).
Receipt of negative feedback is related to increased negative feedback
seeking among individuals with depressive symptoms. Behaviour Re-
search and Therapy, 43, 485–504.
Cheek, J. M., & Briggs, S. R. (1990). Shyness as a personality trait. In
W. R. Crozier (Ed.), Shyness and embarrassment: Perspectives from
social psychology (pp. 315–337). New York: Cambridge University
Press.
Chovil, N. (1991). Social determinants of facial displays. Journal of
Nonverbal Behavior, 15, 141–154.
Clark, D. M., & McManus, F. (2002). Information processing in social
phobia. Biological Psychiatry, 51, 92–100.
Clark, D. M., & Wells, A. (1995). A cognitive model of social phobia. In
R. G. Heimberg, M. R. Liebowitz, D. A. Hope, & F. R. Schneier (Eds.),
Social phobia: Diagnosis, assessment, and treatment (pp. 69 –93). New
York: Guilford Press.
Crawford, J. R., & Henry, J. D. (2004). The Positive and Negative Affect
Schedule (PANAS): Construct validity, measurement properties and
normative data in a large non-clinical sample. British Journal of Clinical
Psychology, 43, 245–265.
DePaullo, B. M., Epstein, J. A., & LeMay, C. S. (1990). Responses of the
socially anxious to the prospect of interpersonal evaluation. Journal of
Personality, 58, 623–640.
Ekman, P. (1992). An argument for basic emotions. Cognition & Emotion,
6, 169 –200.
Ekman, P., & Friesen, W. V. (1978). Facial Action Coding System. Palo
Alto, CA: Consulting Psychologists Press.
Field, T., Hernandez-Reif, M., Vera, Y., Gil, K., Diego, M., Bendell, D., et
al. (2005). Anxiety and anger effects on depressed mother–infant spon-
taneous and imitative interactions. Infant Behavior & Development, 28,
1–9.
Flory, J. D., Raikkonen, K., Matthews, K. A., & Owens, J. F. (2000).
Self-focused attention and mood during everyday social interactions.
Personality and Social Psychology Bulletin, 26, 875– 883.
Fridlund, A. J. (1991). Sociality of solitary smiling: Potentiation by an
implicit audience. Journal of Personality and Social Psychology, 60,
229 –240.
Furr, R. M., & Funder, D. C. (1998). A multimodal analysis of personal
negativity. Journal of Personality and Social Psychology, 74, 1580–
1591.
Hirsch, C. R., Meynen, T., & Clark, D. M. (2004). Negative self-imagery
in social anxiety contaminates social interactions. Memory, 12, 496 –
506.
Joiner, T. E., & Metalsky, G. I. (1995). A prospective test of an integrative
interpersonal theory of depression: A naturalistic study of college room-
mates. Journal of Personality and Social Psychology, 69, 778 –788.
Kashdan, T. B., & Roberts, J. E. (2004). Social anxiety’s impact on affect,
curiosity, and social self-efficacy during a high self-focus social threat
situation. Cognitive Therapy and Research, 28, 119 –141.
Katz, J., & Joiner, T. E. (2001). The aversive interpersonal context of
depression: Emerging perspectives on depressotypic behavior. In R. M.
Kowalski (Ed.), Behaving badly: Aversive behaviors in interpersonal
relationships (pp. 117–147). Washington, DC: American Psychological
Association.
Keltner, D., & Haidt, J. (1999). Social functions of emotions at four levels
of analysis. Cognition & Emotion, 13, 505–521.
Keltner, D., & Haidt, J. (2001). Social functions of emotions. In T. J.
Mayne & G. A. Bonanno (Eds.), Emotions: Current issues and future
directions (pp. 192–213). New York: Guilford Press.
Kenny, D. A. (1996). The design and analysis of social-interaction re-
search. Annual Review of Psychology, 47, 59 – 86.
Kenny, D. A., & la Voie, L. (1985). Separating individual and group
effects. Journal of Personality and Social Psychology, 48, 339 –348.
Kocovski, N. L., & Endler, N. S. (2000). Social anxiety, self-regulation,
and fear of negative evaluation. European Journal of Personality, 14,
347–358.
Kowalski, R. M., & Leary, M. R. (1990). Strategic self-presentation and
the avoidance of aversive events: Antecedents and consequences of
self-enhancement and self-depreciation. Journal of Experimental Social
Psychology, 26, 322–336.
LaFrance, M., & Hecht, M. A. (1999). Option or obligation to smile: The
effects of power and gender on facial expression. In P. Philippot & R. S.
Feldman (Eds.), Social context of nonverbal behavior (pp. 45–70). New
York: Cambridge University Press.
Leary, M. R., & Kowalski, R. M. (1993). The Interaction Anxiousness
Scale: Construct and criterion-related validity. Journal of Personality
Assessment, 61, 136–146.
Leary, M. R., & Kowalski, R. M. (1995a). The self-presentation model of
social phobia. In R. G. Heimberg, M. R. Liebowitz, D. A. Hope, & F. R.
Schneier (Eds.), Social phobia: Diagnosis, assessment, and treatment
(pp. 94 –112). New York: Guilford Press.
Leary, M. R., & Kowalski, R. M. (1995b). Social anxiety. New York:
Guilford Press.
Leary, M. R., Kowalski, R. M., & Campbell, C. D. (1988). Self-
presentational concerns and social anxiety: The role of generalized
impression expectancies. Journal of Research in Personality, 22, 308 –
321.
Meleshko, K. G., & Alden, L. E. (1993). Anxiety and self-disclosure:
Toward a motivational model. Journal of Personality and Social Psy-
chology, 64, 1000–1009.
Mennin, D. S., Heimberg, R. G., Turk, C. L., & Fresco, D. M. (2002).
Applying an emotion regulation framework to integrative approaches to
generalized anxiety disorder. Clinical Psychology: Science and Practice,
9, 85–90.
O’Doherty, J., Winston, J., Critchley, H., Perrett, D., Burt, D. M., & Dolan,
R. J. (2003). Beauty in a smile: The role of medial orbitofrontal cortex
in facial attractiveness. Neuropsychologia Special Issue: The Cognitive
Neuroscience of Social Behavior, 42, 147–155.
Okazaki, S., Liu, J. F., Longworth, S. L., & Minn, J. Y. (2002). Asian
American–White American differences in expressions of social anxiety:
133
SOCIAL ANXIETY IN DYADIC INTERACTION
A replication and extension. Cultural Diversity and Ethnic Minority
Psychology, 8, 234–247.
Palmer, M. T., & Simmons, K. B. (1995). Communicating intentions
through nonverbal behaviors: Conscious and nonconscious encoding of
liking. Human Communication Research, 22, 128 –160.
Perowne, S., & Mansell, W. (2002). Social anxiety, self-focused attention,
and the discrimination of negative, neutral and positive audience mem-
bers by their non-verbal behaviours. Behavioural and Cognitive Psycho-
therapy, 30, 11–23.
Pilkonis, P. A. (1977). The behavioral consequences of shyness. Journal of
Personality, 45, 596– 611.
Rapee, R. M. (1993). The utilisation of working memory by worry.
Behaviour Research and Therapy, 31, 617– 620.
Reno, R. R., & Kenny, D. A. (1992). Effects of self-consciousness and
social anxiety on self-disclosure among unacquainted individuals: An
application of the Social Relations Model. Journal of Personality, 60,
79 –94.
Spurr, J. M., & Stopa, L. (2002). Self-focused attention in social phobia
and social anxiety. Clinical Psychology Review, 22, 947–975.
Thompson, S., & Rapee, R. M. (2002). The effect of situational structure
on the social performance of socially anxious and non-anxious partici-
pants. Journal of Behavior Therapy and Experimental Psychiatry, 33,
91–102.
Turner, S. M., Beidel, D. C., Dancu, C. V., & Stanley, M. A. (1989). An
empirically derived inventory to measure social fears and anxiety: The
Social Phobia and Anxiety Inventory. Psychological Assessment, 1,
35– 40.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and vali-
dation of brief measures of positive and negative affect: The PANAS
scales. Journal of Personality and Social Psychology, 54, 1063–1070.
Wells, A., Clark, D. M., Salkovskis, P., & Ludgate, J. (1995). Social
phobia: The role of in-situation safety behaviors in maintaining anxiety
and negative beliefs. Behavior Therapy, 26, 153–161.
Wells, A., & Papageorgiou, C. (1998). Social phobia: Effects of external
attention on anxiety, negative beliefs, and perspective taking. Behavior
Therapy, 29, 357–370.
Wenzel, A., Graff-Dolezal, J., Macho, M., & Brendle, J. (2005). Commu-
nication and social skills in socially anxious and nonanxious individuals
in the context of romantic relationships. Behaviour Research and Ther-
apy, 43, 505–519.
Woody, S. R. (1996). Effects of focus of attention on anxiety levels and
social performance of individuals with social phobia. Journal of Abnor-
mal Psychology, 105, 61– 69.
Yovetich, N. A., Dale, J. A., & Hudak, M. A. (1990). Benefits of humor in
reduction of threat-induced anxiety. Psychological Reports, 66, 51–58.
Received April 13, 2005
Revision received October 23, 2006
Accepted October 24, 2006 䡲
134
HEEREY AND KRING
A preview of this full-text is provided by American Psychological Association.
Content available from Journal of Abnormal Psychology
This content is subject to copyright. Terms and conditions apply.
