Journal of Experimental Psychopathology

Avoidance of social situations is a key factor in the maintenance of social anxiety disorder, with overt avoidance an important clinical indicator. This paper examines predictors of the transition from the initial experience of social fears to the development of overt avoidance behaviours. Using the 2007 Australian National Survey of Mental Health and Wellbeing, age, gender, DSM-IV diagnoses of other mental disorders and substance use were examined in discrete time survival models (n = 1,359). A majority of the sample reported overt avoidance within a year of initial social fears. Predictors of faster transitions to overt avoidance included fear of attending parties, entering a room or an older onset age. Predictors of slower transitions to overt avoidance included public performance fears and regular alcohol use. This study provides initial support for factors that may influence the transition from initial fear to overt avoidance in social anxiety disorder.

Social anxiety and paranoia both involve a ‘fear of others’ and often co-occur. Post-event processing (PEP), the mental replaying of social events after their conclusion, may be related to both psychological states, although there has been little test of this premise for paranoia. This study examined PEP after social exclusion as a function of social anxiety and paranoia and the potential moderating role of state anger at three time-points. PEP predicted PEP at later time-points, social anxiety and paranoia predicted greater engagement in PEP, and paranoia amplified levels of PEP at higher levels of social anxiety. State anger moderated the relationship between paranoia and PEP, but not between social anxiety and PEP. These results corroborate the transdiagnostic nature of PEP and underscore the co-occurring relationship of paranoia and social anxiety. Future research is necessary to elucidate shared mechanisms between social anxiety and paranoia to advance models, treatments, and prevention efforts.

Panic attacks (PAs) are experienced by approximately half of all individuals with social anxiety disorder (SAD) and are often situationally triggered. Clinical observation suggests that individuals with SAD experience additional “non-traditional” socially-relevant symptoms during PAs. This study examined whether treatment-seeking individuals with SAD (N = 203) endorsed socially-relevant symptoms during PAs and whether these symptoms might be useful in characterizing PAs among these patients. Based on endorsement rates and factor-analytic results, we developed criteria for socially-relevant PAs, and divided the sample into four groups: no PAs (n = 68), socially-relevant PAs only (n = 44), traditional PAs only (n = 22), and combined socially-relevant/traditional PAs (n = 69). Compared to the no-PA group, the combined and socially-relevant PA groups exhibited greater severity of SAD, whereas the traditional PA group did not. Findings suggest that previously reported associations between PAs and greater severity of SAD may be partially explained by the presence of socially-relevant panic symptoms.

The metacognitive model (Wells & Matthews, 1994) proposes that metacognitions (e.g., positive or negative beliefs about worry and thoughts) are involved in emotional disorders alongside perseverative thinking, such as worry. In social anxiety, worry about forthcoming social situations, termed anticipatory processing (AP), is considered an important maintaining factor (Clark & Wells, 1995), but a role of metacognition is less clear. This study investigated AP and metacognition in 80 high socially anxious individuals asked to engage in either AP or a filler task before delivering a speech. AP and higher uncontrollability/danger metacognitive beliefs were associated with greater state anxiety overall. Individuals with higher positive beliefs about worry experienced less of an increase in anxiety before the speech, but their anxiety persisted until after the speech, compared to individuals with lower beliefs. The results support an effect of metacognitions and are discussed in terms of the social anxiety model and its implications.

This study set out to test metacognitive beliefs and perspective taking in self-imagery as predictors of negative self-evaluation of performance in social anxiety disorder. Forty-seven patients with a primary diagnosis of DSM-IV social anxiety disorder were asked to engage in a speech task. Metacognitive beliefs were assessed before the task, and perspective taking in self-imagery and negative self-evaluations of performance were measured after the task. Positive metacognitive beliefs about worrying and observer perspective imagery were positively correlated with negative self-evaluation. A hierarchical linear regression showed that age, and both positive metacognitive beliefs and the observer perspective, were unique predictors of negative self-evaluation. The results suggest that psychological models, especially those formulating the self-concept, should incorporate metacognitive beliefs.

Individuals with Social Anxiety Disorder have difficulty disengaging from self-processing in social situations. Metacognitive therapy interventions for enhancing attentional control were administered to a convenience sample of 24 with a Social Anxiety Disorder diagnosis. Using a cross-over design, 11 participants were given four weekly sessions of Attention Training Technique (ATT), followed by four weekly sessions of Situational Attentional Refocusing (SAR). For the other 13 participants the two treatment components were given in the reverse order. All participants made significant reductions on interview rated and self-reported measures of social and general levels of anxiety by the end of the first intervention (either ATT or SAR). Following completion of the second treatment components, further reductions were observed and 46% (n = 11) of the total sample no longer met DSM-IV criteria for Social Anxiety Disorder diagnosis. Two large order effects were found favoring patients receiving SAR interventions first. Overall these brief techniques aimed at increasing attentional flexibility were associated with large and clinically significant changes in Social Anxiety Disorder symptoms.

We present a study designed to investigate fear of positive vs. negative evaluation within the context of a laboratory-based paradigm designed to evoke social threat. Eighty-nine undergraduates with high (n = 43) or low (n = 46) levels of trait social anxiety took part in a “getting acquainted” task. Participants rated their anxiety about receiving prospective positive vs. negative evaluation in anticipation of receiving public feedback on a filmed introduction of themselves that they had made for an unknown social partner whom they expected they would later meet. Results demonstrated, in contrast to extant theories of fear of positive evaluation in social anxiety, that all participants, including those with high levels of social anxiety, rated the prospect of positive evaluation as anxiety reducing. This finding raises important questions about the construct of fear of positive evaluation and how to measure it “in vivo” in an ecologically valid manner.

Fear of evaluation in general is important in social anxiety, including fear of positive evaluation (FPE) and fear of negative evaluation (FNE). The present study examined various FPE- and FNE-associated state responses (i.e., affective, cardiovascular, and neuroendocrine) to an impromptu speech task which integrated simultaneous and systematic delivery of positive and negative social threat cues (n = 100 [unselected]). Both FPE and FNE related positively to state anxiety and heart rate changes from anticipation of the speech to during the speech itself, and these effects were partly conjoint and partly unique. Furthermore, high FPE alone was associated with dampened cortisol in response to the speech task in contrast to a more normative, robust response to social threat. Last, consistent with hypothesis and prior findings, state anxiety during the speech mediated the relationship between trait FPE and state disqualification of positive social outcomes (a mental safety behavior for FPE-related state anxiety). These results further inform upon the commonalities and distinctions between these two socio-evaluative fears. Implications for the theoretical conceptualization and treatment of social anxiety are discussed.

Process-outcome research in psychotherapy has mainly focused on between-person data (e.g., how differences in psychological process among patients are related to differences in outcome among the patients). However, this level of analysis is in danger of missing its target because psychotherapy models and therapists focus primarily on within-person relationships (e.g., whether change in a patient's cognitive process during the course of therapy may lead to a reduction of symptoms in that client). The study of within-person processes requires collection of repeated data and a disaggregation of the between- and within-person components of time-varying process predictors. The purpose of this study was to examine whether the results of a previously published longitudinal process-outcome study of cognitive and interpersonal therapy for social anxiety disorder (SAD) (Hoffart, Borge, Sexton, & Clark, 2009) were maintained when the process predictors were disaggregated. Eighty social phobic patients were randomized to 10-week residential cognitive or interpersonal psychotherapy. In the present reanalysis, time-varying predictors were disaggregated by use of person-mean centering. For the cognitive process predictors (self-focus, estimated probability and estimated cost of negative social events, safety behaviors), the within-person relationships between predictors and subsequent social anxiety remained significant when disaggregating the predictors. On the other hand, the previously significant within-person relationship between the interpersonal variable of perceived acceptance by others and subsequent social anxiety disappeared with disaggregation. Disaggregated social anxiety also predicted fluctuations in self-focus, estimated probability and estimated cost of negative social events, but not in safety behaviors and perceived acceptance.