Social Anxiety

lable at ScienceDirect

Behaviour Research and Therapy 86 (2016) 50e67

Contents lists avai

Behaviour Research and Therapy

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

The etiology of social anxiety disorder: An evidence-based model

Susan H. Spence a, *, Ronald M. Rapee b

a School of Applied Psychology and Australian Institute for Suicide Prevention and Research, Griffith University, Brisbane, QLD 4121, Australia b Centre for Emotional Health, Department of Psychology, Macquarie University, Sydney, NSW 2109, Australia

a r t i c l e i n f o

Article history: Received 21 January 2016 Received in revised form 7 June 2016 Accepted 29 June 2016 Available online 1 July 2016

* Corresponding author. Australian Institute for Suic Griffith University, Mount Gravatt Campus, Brisbane,

E-mail address: [email protected] (S.H. Spe

http://dx.doi.org/10.1016/j.brat.2016.06.007 0005-7967/© 2016 Elsevier Ltd. All rights reserved.

a b s t r a c t

The current paper presents an update to the model of social anxiety disorder (social phobia) published by Rapee and Spence (2004). It evaluates the research over the intervening 11 years and advances the original model in response to the empirical evidence. We review the recent literature regarding the impact of genetic and biological influences, temperament, cognitive factors, peer relationships, parenting, adverse life events and cultural variables upon the development of SAD. The paper draws together recent literature demonstrating the complex interplay between these variables, and highlights the many etiological pathways. While acknowledging the considerable progress in the empirical litera- ture, the significant gaps in knowledge are noted, particularly the need for further longitudinal research to clarify causal pathways, and moderating and mediating effects. The resulting model will be valuable in informing the design of more effective treatment and preventive interventions for SAD and will provide a useful platform to guide future research directions.

© 2016 Elsevier Ltd. All rights reserved.

1. Introduction

In 2004, we published a paper that presented a model of the etiology of social phobia based on a review of the empirical evi- dence at that time (Rapee & Spence, 2004). The aim of the present paper is to evaluate research over the intervening 11 years and consider its support for the model or whether modifications are needed. We will also examine implications for the prevention and treatment of social phobia, which is now referred to as social anxiety disorder (SAD). Given that the onset of SAD for the great majority of cases occurs during childhood or adolescence (Jin et al., 2005), the focus of this paper is primarily upon evidence relating to young people rather than adults.

1.1. Symptoms and diagnostic criteria for social anxiety disorder

Since 2004, there have been only minor changes to diagnostic classification systems for SAD, with the key elements of both DSM and ICD criteria remaining relatively consistent. According to DSM- 5 (American Psychiatric Association, 2013), the core defining fea- tures of SAD include fear or anxiety in social situations in which the

ide Prevention and Research, QLD 4121, Australia. nce).

individual is exposed to possible scrutiny by others and a fear of acting in a way that will be negatively evaluated by others (either resulting from the individual’s own behavior or from showing anxiety symptoms such as blushing, trembling or sweating). Where possible, the social situations are avoided or are endured with intense anxiety. DSM-5 further specifies that the fear must be out of proportion to the actual threat, that it has persisted for longer than 6-months, is not due to the physiological effects of a substance, another mental disorder, or a medical condition, and that it leads to clinically significant distress or impairment in important areas of social functioning. For a DSM-5 diagnosis in young people, the anxiety must include interactions involving peers, and may be expressed by crying, tantrums, freezing, clinging, shrinking or failing to speak in social situations. Typically, feared social situa- tions for children include speaking in front of the class, asking a teacher a question, performing in front of others, meeting new people, joining in conversations, asking for help in shops or at school, and going to parties or social events where peers are pre- sent (Beidel & Turner, 2007; Rao et al., 2007).

1.2. Epidemiology of SAD in children and adolescents

Recent studies have informed our knowledge of the epidemi- ology of SAD which is relatively common amongst children and adolescents (Burstein et al., 2011; Lawrence et al., 2015). Prevalence rates in community samples are generally found to increase from

S.H. Spence, R.M. Rapee / Behaviour Research and Therapy 86 (2016) 50e67 51

childhood through adolescence (Beesdo et al., 2007; Burstein et al., 2011; Canino et al., 2004; Lawrence et al., 2015), although no such increase was found in a large UK epidemiological study (Ford, Goodman, & Meltzer, 2003). Adult prevalence also appears to differ markedly between countries, with lower reported rates in East Asian countries than elsewhere (Brockveld, Perini, & Rapee, 2014). In our previous paper, we noted that social anxiety is expressed along a severity continuum, with many young people reporting high symptoms of social anxiety without necessarily reaching the threshold for a clinical diagnosis. Recent studies confirm the relatively high prevalence of sub-clinical social anxiety symptoms (eg. Knappe et al., 2011). We also noted that the distinction between high social anxiety and a clinical diagnosis of SAD depends on more than just severity, being influenced by the impact of social functioning and a range of cultural factors (see below). Studies suggest that SAD during childhood and adolescence tends to persist if left untreated (Beesdo-Baum et al., 2012; Burstein et al., 2011; Kessler et al., 2012), with onset prior to 11 years of age increasing the risk of persistence into adulthood (Abidin, 1992; Beesdo et al., 2007; Wittchen & Fehm, 2003). In fact, SAD is one of the more chronic and persistent of the mental disorders across the lifespan although it is more likely to have a waxing and waning course than a stable one (Beesdo-Baum et al., 2012; Bruce et al., 2005).

1.3. Comorbidity and association with adverse consequences

SAD in youth is comorbid with a significant number of mental health problems, particularly other anxiety disorders and depres- sion, and with substance use in older adolescents (Beesdo-Baum et al., 2012; Burstein et al., 2011; Wittchen, Stein, & Kessler, 1999). There is insufficient evidence, as yet, to determine whether SAD is a cause or effect of such comorbid problems, or whether these patterns of comorbidity reflect common underlying causal factors. However, longitudinal studies are beginning to examine these complex relationships. For example, Beesdo et al. (2007) demonstrated that SAD during adolescence significantly increased the risk for depression during early adulthood, although this effect was not found by Buckner et al. (2008). Similarly, SAD in adolescence has been shown to increase the risk of future alcohol use (eg. Black et al., 2015) as well as cannabis and alcohol depen- dence in early adulthood (Buckner et al., 2008), although this relationship may be specific to females (Buckner & Turner, 2009). Furthermore, although depression predicted future substance use in the study by Buckner et al. (2008), this effect was explained by its association with SAD, suggesting that SAD in particular may in- crease young people’s risk for future substance abuse. In contrast to these patterns, an early study failed to show that adolescent social phobia predicted any adult disorders other than adult social phobia (Pine, Cohen, Gurley, Brook, & Ma, 1998). Clearly the heterotypic continuity of SAD from adolescence is complex and more longitu- dinal research is needed to cast light on specific patterns, as well as on the mechanisms by which early SAD may continue or may contribute to development of other mental health problems.

2. Summarizing the Rapee and Spence (2004) model

The Rapee and Spence (2004) model was designed to consider factors that might describe how high levels of social anxiety emerge across early development. It began from the assumption that social anxiety exists along a continuum of intensity and severity. Although a diagnosis of SAD is more likely be given toward the upper end of the continuum, the actual clinical diagnosis requires that the symptoms have a marked impact and significantly inter- fere with the individual’s life. Thus the model argued that a clinical

diagnosis of SAD will be affected by a range of risk factors that in- fluence the life interference value of symptoms including age, gender, and cultural “fit”. The model outlined the role of various risk and protective factors that were proposed to influence the shift along the continuum of social anxiety until young people reached the point at which the severity and duration of symptoms, along with their impact are sufficient to warrant a clinical diagnosis of SAD.

We noted that causal and maintaining factors for SAD may not be the same and that different risk factors may be more or less influential at different ages. Risk factors were predicted to interact, with some factors combining to increase the impact upon the development of SAD, and others having a protective effect that may buffer the impact of a particular risk factor. We noted that different pathways and combinations of factors can result in SAD (the principle of equifinality) and that any one risk factor can result in various outcomes, not just SAD (the principle of multi- finality). Furthermore, it was acknowledged that risk factors can be transactional and reciprocal, with young people influencing their own social outcomes which, in turn, impact upon the young person. These points regarding etiology remain relevant today, and we will examine more recent evidence that clarifies these issues.

While acknowledging the considerable limitations in the research up to 2004, we were able to draw together sufficient ev- idence to propose a series of interacting risk and protective factors that influenced the development and maintenance of SAD. There was emerging evidence of a complex interplay between intra- individual factors (genetic, biological processes, cognitive pro- cesses and social skills) and environmental factors (such as parental influences, aversive social experiences, and negative life events) in the etiology of SAD. We proposed that cultural factors can influence both the form and characteristics of social anxiety as well as its life impact. However, there remained a major task for researchers to tease apart these interrelationships and to determine their specific mechanisms of action. A strong need for longitudinal research to clarify cause and effect relationships was noted, to identify what is proximal versus distal in terms of impact on SAD and to inform us about mediators/mechanisms of action, and moderating/interact- ing effects. In the intervening years, many of these issues have received further attention although, as will become evident in this paper, many unanswered questions remain.

The past decade has seen a rapid expansion of research into the etiology and maintenance of SAD in young people. Although there are still significant gaps in the evidence, there have been some exciting studies that inform our understanding of this area. Since 2004 there have been several papers that have reviewed specific literature relating to the development and maintenance of SAD in children and/or adolescents (e.g., Detweiler et al., 2014; Kimbrel, 2008; Knappe, Sasagawa, & Creswell, 2015; Ollendick & Benoit, 2012; Ollendick, Benoit, & Grills-Taquechel, 2014) and we acknowledge their significant contribution to our understanding of this area. There have also been substantial increases in our knowledge of the development and maintenance of SAD in adults (See Wong & Rapee, in press, June 2016 for a review).

3. Developments in the research 2004e2015: intrinsic factors

3.1. Genetics of childhood SAD

There is substantial evidence that SAD has strong familial links - children of parents with SAD have significantly increased risk of experiencing the disorder, and parents of children with SAD are also more likely to experience the problem (See Elizabeth et al., 2006 for a review). Studies suggest that these familial

S.H. Spence, R.M. Rapee / Behaviour Research and Therapy 86 (2016) 50e6752

associations may reflect an increased risk for internalizing prob- lems in general, rather than a specific risk for SAD. Of course family concordance may reflect the influence of either or both heritability and shared environment (and their interaction). A recent meta-analysis by Scaini, Belotti, and Ogliari (2014) exam- ined the distinct influences of these factors on SAD and social anxiety symptoms in children and adults. They found substantial variation in results, with heritability estimates varying from 0.13 to 0.60, with non-shared environmental estimates ranging from 0.31 to 0.78, and shared environment making minimal contribu- tion. It appears that shared environmental factors may explain less variance in symptoms of social anxiety than other forms of anxiety among young people (Eley, Rijsdijk, Perrin, O’Connor, & Bolton, 2008). Hence, unique individual experiences that contribute to non-shared environment, such as school, extra-curricular activ- ities, and peer relationships, may be particularly important in SAD, although we note that estimates of non-shared environmental influence are also affected by measurement error. Interestingly, the Scaini et al. (2014) meta-analysis found that the genetic contribution to SAD in adults was much lower (around half) than that of children. They suggested that this could reflect the longer period of time that adults have been exposed to environmental influences, compared to children. However, it is possible that this finding reflects methodological differences in the studies involving adults versus children, such as differing informants or whether social anxiety rather than diagnosed social phobia was being considered. Scaini et al. (2014) reported markedly lower estimates of heritability when the outcome variable was a clinical diagnosis of SAD (0.27) compared to outcomes assessed via symptoms of social anxiety (0.58). This finding is consistent with our suggestion that social anxiety is a fundamental, personality- like construct whereas the diagnosis of SAD reflects an interac- tion between social anxiety and life impairment. The life impact of social anxiety is likely to be heavily environmentally determined and hence the clinical diagnosis would be expected to be less influenced by heritability than the underlying construct of social anxiety.

Having established that genetics do play a role, albeit only moderately, researchers have attempted to identify the exact genes and biological pathways involved. Several investigators have looked for candidate genes, with a strong focus on those associated with the serotonergic (5-HT) and dopaminergic pathways. To date, however, there have been no consistent findings to demonstrate conclusively that there are specific genes associated with SAD over and above those linked to a general predisposition to anxiety or to predisposing temperament such as BI discussed below (Knappe et al., 2015).

In summary, although there is now a substantial body of research, it is still difficult to draw firm conclusions about the role of genetic factors in the development of SAD. Although there ap- pears to be a modest genetic influence during childhood, devel- opmental changes are evident in the relative contribution of genes versus non-shared environment, with genetic factors possibly becoming less influential and non-shared environment conse- quently increasing in impact. Interestingly, shared environment factors appear to make minimal contribution to the development of SAD, although this does not exclude the possibility that shared environmental influences may interact with genetic factors in the expression of social anxiousness. Although research has not yet identified the specific genes associated with SAD, it is important that these investigations continue and that we are able to clarify the specific biological mechanisms through which genes have their action. As noted below, in the section relating to behavioral inhi- bition, an important area for future research is to identify envi- ronmental variables that influence the impact of genetic risk factors

upon the development and maintenance of SAD.

3.2. Biological influences upon SAD

A detailed review of the literature relating to biological contri- butions to SAD is beyond the scope of this paper. Nevertheless, there are some important points that warrant a mention. The research in this area is developing rapidly but we are still a long way from being able to draw firm conclusions. The search continues for brain structures, neural pathways, neurotransmitters, biochemistry, and psychophysiological processes with specific as- sociations to SAD.

In terms of brain structures and neural pathways, the most strongly researched brain pathways relate to the amygdala and pre-frontal cortex (Fox & Kalin, 2014), with increasing evidence from neuroimaging studies implicating these areas in social anx- iety and associated precursors such as behavioral inhibition (Detweiler et al., 2014). Very basically, the brain areas implicated in social anxiety are typically those associated with emotional processing of stimuli and activation of the higher cognitive pro- cesses involved in emotion regulation and behavioral responding to the stimuli (Detweiler et al., 2014). Neuroimaging studies also implicate these areas in emotional responsiveness and attention regulation in the processing of social threat, with a recent meta- analysis confirming that adult SAD patients tend to show increased bilateral amygdala activation in response to socially emotive stimuli (Hattingh et al., 2012). With respect to pre-frontal cortex activity, reflecting executive processing and emotional regulation, the findings from imaging studies are unclear. Different patterns of results have been found for the ventromedial versus the ventrolateral regions of the prefrontal cortex, and in some instances results appear to indicate increased activation whereas others suggest reduced activation (Yokoyama et al., 2015). Thus, we cannot yet draw conclusions about the role of prefrontal cortex in SAD.

Caouette and Guyer (2014) reviewed brain imaging studies with children and adolescents with SAD or behaviorally inhibited temperaments and concluded that, compared to non-anxious controls, these young people tend to show amygdala hyper- activation in response to attention to threatening facial stimuli. However, similar effects are evident for children with generalized anxiety disorder (GAD) who also show greater right amygdala activation to social threat stimuli compared to non-anxious controls (Monk et al., 2008). This effect could not be attributed to comorbidity between GAD and SAD as there was no significant difference in amygdala activation between GAD youth with and without SAD. Consistent with this finding, Hamm et al. (2014) found that although youth with anxiety disorders showed aber- rant amygdala connectivity compared to non-anxious controls, there was no difference between those with a primary diagnosis of GAD versus those with SAD. Together these findings suggest that, while young people with SAD show increased amygdala activation indicative of increased vigilance and emotional responding to social threat, this biological factor may not be specific to SAD and may be a characteristic of heightened anxiety in general.

Other brain structures, pathways and processes have also been implicated in SAD, with recent research suggesting that adolescents with SAD show elevated activation in the neural pathways that coordinate motivated behavior and reward processing (see Caouette & Guyer, 2014 for a review). These networks include the basal ganglia and their projections from frontal cortical regions (e.g., vlPFC), the hippocampus, amygdala, and anterior cingulate cortex. Caouette and Guyer (2014), drawing on evidence from a series of imaging studies conducted by Guyer and her colleagues,

S.H. Spence, R.M. Rapee / Behaviour Research and Therapy 86 (2016) 50e67 53

outline a model proposing that social anxiety and BI in youngers is characterized by aberrant information processing of anticipated social reward incentive cues. Although the research is in its early stages, Caouette and Guyer (2014) cite evidence suggesting that young people with SAD or a BI temperament show a pattern of brain activity of heightened striatal reactivity, in the caudate nu- cleus and the putamen region, suggestive of increased sensitivity to rewards, heightened vigilance toward evaluating their perfor- mance, and excessive valuing of positive outcomes. Importantly, Guyer et al. (2012) did not find this effect for adolescents with GAD and suggest that it may explain the heightened valence placed by youth with SAD upon positive evaluation by others, and the increased desire for positive social outcomes and yet the greater likelihood of avoidance.

There is also evidence that children with SAD show differences in physiological responses associated with fear pathways. For example, compared to non-anxious children they show indicators of higher baseline sympathetic activity (heart rate, electrodermal activity) and lower parasympathetic activation (respiratory sinus arrhythmia) (Kramer et al., 2012; Schmitz, Kramer, Tuschen-Caffier, Heinrichs, & Blechert, 2011). In response to a specific social threat, one study showed that children with SAD had similar elevation in heart rate to controls, but showed lower parasympathetic responding and slower heart rate recovery, suggesting that it is the recovery from social distress that may be especially impaired (Schmitz et al., 2011).

Although the findings described above are interesting, it is not yet clear whether the results are specific to SAD, reflect anxiety (or internalizing) in general, or reflect the temperamental un- derpinnings (of SAD specifically or anxiety more generally) such as BI where similar findings have been demonstrated (Schmitz et al., 2011). A further limitation in biological investigations of SAD to date is that the majority of the research has been cross-sectional in design and has tended to focus on participants who already exhibit SAD making it unclear whether the particular neural structures or processes represent manifestations of social anxiety or play a part in its causation. It is only recently that longitudinal studies have started to examine brain and biological factors in the development of SAD in cohorts that do not already meet diagnostic criteria for SAD, and these have primarily emerged from the childhood temperament literature.

A further complicating factor in interpreting the research investigating the neurobiology of SAD is that we need to take into account the potential impact of developmental level of the child. Throughout childhood and adolescence, young people proceed through developmental stages in which specific changes take place within the brain, and indeed throughout the body. Furthermore, with increasing age, children are required to deal with increasingly complex social challenges. Thus, the search for neurobiological structures and processes associated with SAD needs to take into account both biological and environmental differences across the age span. For example, during early childhood children show cognitive developmental shifts in the acquisition of effortful control strategies that influence emotion regulation and reactivity capac- ities, which influence approach and avoidance behavior (Rothbart, Sheese, Rueda, & Posner, 2011). Similarly, in adolescence there are major hormonal and structural changes within the brain, particu- larly in the pre-frontal cortex, amygdala and their inter- connections, that need to be considered, as they have a signifi- cant impact upon the development of the appraisal, expression, and regulation of emotion (Del Piero, Saxbe, & Margolin, 2016). In trying to identify neurobiological factors associated with the development and maintenance of SAD, researchers need to recog- nise that these different factors are likely to be important at different age levels.

3.3. Temperament

3.3.1. Behavioral inhibition Temperament is one pathway through which genetic factors

might influence the development of mental health problems. Behavioral inhibition is one particular dimension of temperament that was strongly implicated in the development of SAD in the Rapee and Spence (2004) paper, although the degree to which BI could be shown to be a specific precursor of SAD rather than to anxiety in general was unclear at that stage. Behavioral inhibition represents a tendency to respond with heightened sensitivity to novel auditory and visual stimuli, and avoid unfamiliar situations and people (Kagan, Reznick, Clarke, Snidman, & Garcia-Coll, 1984). Such patterns can be identified from infancy onwards, and are also associated with specific physiological response patterns. Although there have been some inconsistent findings in the literature, there is a good deal of evidence to suggest that, relative to uninhibited children, inhibited children are more likely to show physiological signs of arousal at rest, including higher cortisol levels, and greater reactivity to novelty reflected by higher heart rate and heart rate acceleration, decreased heart rate habituation, pupil dilation, and laryngeal muscle tension (Fox, Henderson et al., 2005; Hirshfeld- Becker, Micco, Wang, & Henin, 2014). Recent brain imaging studies have demonstrated hyper-responsivity and structural dif- ferences in the amygdala and associated circuitry in adults with a history of BI, compared to non-BI (Hirshfeld-Becker et al., 2014; Roy et al., 2014), that closely resemble those associated with SAD. This suggests that BI and SAD may have shared biological un- derpinnings, although as noted earlier, heightened amygdala reactivity is more likely a reflection of greater general anxiousness.

Arguably, the critical component of BI is behavioral withdrawal and avoidance (Rapee, 2002). Children high in BI show several characteristic behaviors that are typically used to define the construct, especially during observational assessment. Key differ- ences between children high and low in BI include limited eye contact, maintained proximity to attachment (safety) figures, lack of verbal utterances, and avoidance of threat stimuli (Kagan et al. 1984; Rapee, 2002). Of critical importance to BI’s relationship with social anxiety, several of these behaviors are core features of social anxiety and have been highlighted in models of the main- tenance of SAD (Clark & Wells, 1995; Rapee & Heimberg, 1997). Within the context of SAD, these behaviors are variously concep- tualized as safety behaviors and social skills deficits and they are often seen as central to the etiology and/or maintenance of the disorder. Given that some of these behaviors are fundamental to the construct of BI and can be identified by two years of age or earlier, it is possible that they are critical to the development of social anxiety and may even be hard-wired or endophenotypes.

Given the many common biological, behavioral and information processing features between BI and SAD, the question arises as to whether BI is merely a manifestation of SAD or represents a related, but distinct construct. In a detailed review of the literature, Rapee and Coplan (2010) concluded that there was sufficient evidence to suggest that temperamental styles such as BI could be concep- tually distinguished from clinical syndromes such as SAD, although they noted the strong overlap and lack of clear evidence. Consistent with this conceptual commonality, Rapee and Spence (2004) noted that certain temperament styles may represent early behavioral and emotional manifestations of the same construct that, when present at a more severe level and associated with impairment or disruption to daily living, are later labeled a disorder. Cultural fac- tors may be especially critical in determining impairment and there is growing evidence that temperamentally-related behaviors such as withdrawal and social avoidance are more socially accepted in some cultures than in others (Chen & Tse, 2008; Chen, DeSouza,

S.H. Spence, R.M. Rapee / Behaviour Research and Therapy 86 (2016) 50e6754

Chen, & Wang, 2006; Rapee et al., 2011). There is now substantial evidence to demonstrate that a stable

childhood temperament of behavioral inhibition significantly in- creases the risk of developing SAD. In a meta-analysis of longitu- dinal studies, Clauss and Blackford (2012) found that a BI classification during toddlerhood or early childhood increased the risk of developing SAD during childhood to adolescence sevenfold. The level of risk was increased irrespective of whether laboratory tasks or parent report were used to identify BI and the ages at which BI was assessed, although the degree of risk decreased with the length of follow-up interval. Furthermore, there is good empirical support for the view that BI is a stronger risk factor for SAD, rather than for other anxiety disorders or depression (Chronis- Tuscano et al., 2009; Hirshfeld-Becker et al., 2007; Muris, van Brakel, Arntz, & Schouten, 2011; Rapee, 2014). For example, Rapee (2014) followed through to adolescence a sample of children who were categorized as BI or non-BI at age 4. Almost 37% of those who were categorized as BI at age 4 demonstrated SAD at age 15, compared with 15% of uninhibited children, and preschool BI was not a significant predictor of other forms of anxiety disorder.

What remains unclear is why not all behaviorally inhibited children remain inhibited or develop SAD. For example, the Clauss and Blackford (2012) meta-analysis found that 43% of the behav- iorally inhibited children across the combined studies developed SAD at a future follow-up, which indicates that 57% did not. Thus we need to identify those factors either intrinsic or extrinsic to the child that influence the stability of BI and its impact on SAD. Similarly, it is important to recognise that a small, but significant, proportion of young children who are not identified as having a behaviorally inhibited temperament subsequently proceed to develop SAD (15% in the Clauss and Blackford (2012) meta- analysis). Thus, models of the etiology and maintenance of SAD need to recognise that there are pathways to social anxiety that are not dependent upon BI.

3.3.2. BI in interaction with other intrinsic factors A recent review by Henderson, Pine, and Fox (2015) suggested

that the association between BI and future anxiety is not straight- forward and that we need to consider how BI interacts with other factors. They suggest that BI is more likely to increase the risk for SAD if the child also shows an information processing pattern reflecting heightened automatic processing, low attention shifting or relatively high inhibitory control. These authors draw on Roth- bart and colleagues’ (eg. Rothbart et al., 2011) model to differentiate reactive/automatic cognitive processes (eg. automatic processing of novel stimuli) and self-regulatory processes (voluntary, effortful control e e.g. inhibitory control, attention shifting, conflict moni- toring, and response monitoring). Normally, important changes in these cognitive processes occur with maturation during childhood, with the gradual development of effortful control processes (such as voluntary attentional control and response inhibition) that begin to modulate the reactive, automatic, stimulus-driven information processing present from infancy. Children gradually begin to develop more effortful control processes that enable planning, reasoning and goal-directed behaviors and modulate reactive/ automatic processes. Henderson et al. proposed that the BI phenotype is characterized by a pattern of heightened reactive/ automatic processing in toddlerhood, which may even precede the behavioral expression of BI, and poor effortful control later in childhood. Furthermore, they proposed that the behavioral mani- festation of BI will significantly increase the risk of future social anxiety in children who also show the information processing styles of greater reactivity and weaker effortful control.

Henderson et al. described several imaging studies demon- strating that early BI is associated with structural and functional

properties in subcortical and cortical networks related to salience detection, threat sensitivity, and attention orienting particularly involving exaggerated amygdala responses (indicative of height- ened reactivity). There is also evidence that BI is a stronger pre- dictor of future anxiety if the child also shows a reactive/automatic pattern of information processing. Barker et al. (2015) assessed BI at ages 2 and 3, startle responses on a laboratory task at age 7, and parent reported social anxiety and internalizing problems in gen- eral at age 9. Early BI predicted later internalizing problems and social anxiety only among those children who showed higher startle responses at age 7. This suggests that a more reactive in- formation processing style moderates the impact of BI upon later social anxiety. Clearly considerably more empirical evidence is needed to evaluate this very interesting suggestion.

There is also some evidence that effortful control processes may moderate the impact of BI upon future anxiety. White, McDermott, Degnan, Henderson, and Fox (2011) found that high levels of BI in toddlerhood predicted high levels of anxiety symptoms during the preschool years only for children exhibiting low levels of attention shifting (an indicator of low effortful control). The ability to flexibly shift attention is proposed to play an important role in the regu- lation of negative temperamental reactivity, possibly by enabling the shift of attention away from negative stimuli or cognitions. White et al. also found that high levels of inhibitory control increased the risk for anxiety problems in BI children. Although the explanation for this finding is unclear, it may be that these children have difficulty in engaging their control processes in a flexible way, resulting in a pattern of rigid, inflexible, and over-controlled be- haviors, including withdrawal and avoidance. Once again, consid- erably more evaluation of these potentially critical temperamental interactions is required.

These studies illustrate some of the exciting research in cogni- tive neuropsychology that is informing us about the ways in which BI may impact upon future SAD and provides indications about ways in which early preventive programs could be developed. Once reliable and significant moderators have been identified, it may be possible to select children with a stable pattern of BI who also show these particular information processing patterns, and to re-train these information processing styles to reduce the risk of SAD.

3.3.3. BI in interaction with environmental factors There are now several longitudinal studies demonstrating that

environmental factors influence the impact of BI on mental health. For example, Frenkel et al. (2015) showed that stable BI in early childhood significantly increased the risk for adult anxiety disor- ders, but only for young people who had a low level of social involvement during adolescence. Insecure attachment is another variable that has been shown to moderate the impact of BI upon future anxiety. For example, Lewis-Morrarty et al. (2015) found that a consistent pattern of BI assessed at 14 and 24 months was only associated with social anxiety during adolescence for those par- ticipants with a history of insecure attachment during infancy, and only in males. Research with primary school children has also found that those who show both a behaviorally inhibited temper- ament and an insecure attachment style show greater increases in anxiety disorders symptoms in the future (Muris et al., 2011).

Parenting style has been examined as a potential moderator of the impact of a BI temperament upon social anxiety. There is some evidence that an over-controlling parenting style influences the link between a stable BI temperament and subsequent social anx- iety. Lewis-Morrarty et al. (2012) found that a consistently high pattern of behavioral inhibition during infancy and childhood predicted high social anxiety during adolescence (as reported by the mother, but not the child) only for those children whose mothers showed high over-control at child-age 7. A similar pattern

S.H. Spence, R.M. Rapee / Behaviour Research and Therapy 86 (2016) 50e67 55

of high BI combined with parental intrusion has been shown to predict shyness among preschool-aged children (Rubin, Burgess, & Hastings, 2002). However Muris et al. (2011), in a prospective assessment of children aged from 5 to 8 years, did not find either overprotective or anxious parental styles to moderate the rela- tionship between BI and development of social anxiety.

In an interesting study, Guyer et al. (2015) investigated whether the experience of authoritarian parenting (characterized by lack of emotional warmth and support for the child, non-transparent declaration of rules, and high levels of control) would exacerbate the impact of a BI temperament upon a pattern of brain functioning indicative of anxious-emotional responding to a peer social stressor task. The study examined three brain regions associated with both BI and social anxiety, namely the amygdala, striatum and ventro- lateral prefrontal cortex (vlPFC) using functional magnetic reso- nance imaging (fMRI). Adolescents who had shown a pattern of BI in early childhood and experienced authoritarian parenting at age 7, showed a decreased vlPFC response to a social rejection task. The vlPFC response to rejection did not vary as a function of authori- tarian parenting for those without a BI history. In terms of amygdala activity, BI youth showed greater reactivity than non-BI youth, but parenting style did not affect this relationship. Guyer et al. noted that vlPFC activity reflects processes that include modulating af- fective response to threat or reward stimuli and updating goals and guides for action. They suggested that a BI temperament in com- bination with severe and harsh parenting may result in disruption of the regulatory function of vlPFC and impaired modulation of affective response and social learning. Although the results of the Guyer et al. study are interesting, we note that Williams et al. (2009) did not find authoritarian parenting style to moderate the impact of BI upon change in internalizing problems over time.

Researchers are also starting to examine in more detail the ways in which genetic influences thought to be associated with the development of BI may be moderated by environmental factors such as parenting style and family functioning. For example, Fox, Nichols, et al. (2005) found that a polymorphism in the promoter region of the gene for the serotonin transporter (5-HTT) (linked to serotonin activity) interacted with maternal social support in pre- dicting inhibited behavior with unfamiliar peers in middle child- hood. Children with the combination of a short 5-HTT allele (either two short or one long, one short) and low social support were at increased risk for behavioral inhibition in middle childhood. Similar results were reported in a study showing that higher levels of maternal overprotection, as reported by the child, were associated with elevations in BI among children carrying two copies of the lower expressing 5-HTTLPR alleles (S or LG), but not among those carrying only one copy or those homozygous for the LA allele (Burkhouse, Gibb, Coles, Knopik, & McGeary, 2011). In addition, this interaction was specific for the social component of BI, not the nonsocial component and was maintained even after statistically controlling for children’s and mothers’ psychopathologies.

A prospective adoption study by Natsuaki et al. (2013) also found that parenting behavior could influence genetic predisposi- tion to BI. This study examined whether an enriched, supportive environment involving a high level of parental responsiveness at 18 months would moderate the impact of high genetic risk upon the development of behavioral inhibition and social anxiety at 27 months. Having a birth mother with a history of a diagnosis of SAD was taken as the indicator of genetic risk for BI and social anxiety. For children whose birth mothers had no history of SAD, the responsiveness of the adoptive mother did not influence the tra- jectory of child BI or social anxiety. In contrast, adoptive mothers’ responsiveness when the child was 18 months old predicted decreasing levels of BI and social anxiety at 27 months in children whose birth mothers had experienced social phobia. Such evidence

suggests that a highly positive environment may buffer the impact of a genetic predisposition to BI and social anxiety and indicates protective factors that could be a focus for preventive interventions with children at risk for SAD.

It is interesting to speculate about the possible three-way interaction between parenting, BI, and cultural acceptance. In a series of studies, Chen and colleagues have shown that parents from East Asian countries such as China, express greater positivity towards expressions of inhibition and reticence in their child (Chen et al., 1998). In turn, parental control and protection may be more anxiety-generating among children from Western than from Eastern countries (Oh, Shin, Moon, Hudson, & Rapee, 2002).

Taken together, current research provides convincing evidence of a link between BI and SAD, and suggests that this association is stronger than for other anxiety or related disorders. Furthermore the link is complex, with both intrinsic and environmental factors having the capability of modifying the BI in terms of its own sta- bility and its influence on SAD. Ultimately, it may be possible to identify those children most at risk of developing SAD who possess both a BI temperament and exposure to modifiable environmental or intrinsic factors that could be targeted in programs to prevent the development of SAD (Kennedy, Rapee, & Edwards, 2009).

3.4. Cognitive factors and safety behaviors

3.4.1. Cognitive factors At the time of our earlier paper, preliminary evidence had

emerged suggesting that socially anxious young people show similar cognitive biases and distortions to adults, in line with those proposed by cognitive models such as Clark and Wells (1995) and Rapee and Heimberg (1997) to maintain socially phobic affect and behaviors. We noted, however, that there was very little evidence to determine how young children come to develop such maladaptive cognitive processes. In the absence of sufficient evidence at that time, we could only speculate about the potential influence of factors such as parenting style and adverse social experiences upon cognitive processes.

Cognitive models at that time focussed primarily on factors that were proposed to maintain SAD rather than its initial development (Clark & Wells, 1995; Rapee & Heimberg, 1997). For example, Clark and Wells (1995) aimed to explain why social anxiety persists despite repeated exposure to feared social situations in the absence of aversive outcomes and why extinction of fear does not occur. In brief, they proposed that socially phobic individuals hold an excessively strong desire to convey a favourable impression of themselves to others and hold beliefs that, when they enter a social situation, they are at serious risk of behaving in an inappropriate or unskilled way, leading to disastrous social consequences. Thus so- cial situations are viewed as highly threatening, triggering a mal- adaptive cycle of events in which i) the somatic and behavioral symptoms of anxiety are interpreted by the individual as a further source of threat (that others will observe the symptoms and appraise them negatively), ii) the individual becomes preoccupied with their somatic responses and negative thoughts, which in turn interferes with their ability to focus their attention on the social task, and iii) the anxious appearance and behaviors of the indi- vidual may actually lead others to respond less positively towards them.

Rapee and Heimberg (1997) added to the conceptualization in their model of the generation and maintenance of anxiety in social/ evaluative situations by emphasizing that social phobia can be regarded as a position along a continuum ranging from mild shyness to social phobia and avoidant personality disorder, with the cognitive processes associated with social anxiety also pro- posed to lie on a continuum. They proposed that individuals, upon

S.H. Spence, R.M. Rapee / Behaviour Research and Therapy 86 (2016) 50e6756

entering a social situation form a mental representation of them- selves as they believe others (the audience) perceive them. This representation was suggested to be influenced by memories of prior experiences, internal cues, and external cues (audience feedback). Typically, the socially phobic person assumes that the audience holds unrealistically high standards and thus un- derestimates the quality of their own performance, and anticipates adverse outcomes that, in turn, trigger cognitive, behavioral and somatic symptoms of anxiety. Importantly, this model notes that prior experiences and learning history can contribute to the development of social anxiety through influence upon the in- dividual’s mental representation of self. In their discussion of etiological factors, Rapee and Heimberg noted the role of genetic influences upon the tendency to preferentially allocate attentional resources to threat. They suggested, however, that this “tendency” interacts with family environmental factors to steer the allocation of attentional resources towards social situations that are thus perceived as dangerous and a source of threat. Rapee and Heimberg placed particular importance on the role of parents in communi- cating information that influences their child’s expectations of negative evaluation by others and in promoting avoidance of social situations.

Both models proposed that socially anxious people engage in safety behaviors that, while being intended to prevent adverse outcomes, generally lead the individual to attribute any social successes to the safety behaviors, rather than to their own skill. Hence these behaviors help to maintain threat expectancies even in the absence of genuine negative experiences. Perhaps more importantly, such safety behaviors often constitute poor social performance and in this way may reduce the likelihood of suc- cessful social outcomes.

At the time of our review, there was relative little research to determine whether socially anxious children showed similar cognitive biases and associated behavioral responses, such as safety behaviors, to those found in adults. One of the few studies in the area at that time had shown that social phobic children tended to anticipate negative outcomes from social-evaluative situations, evaluate their own performance more negatively, and reported a higher level of negative cognitions in relation to social-evaluative tasks in comparison to a matched sample of non-anxious peers (Spence et al. (1999). Subsequently, a good deal of evidence has emerged indicating that the cognitive biases and processes pro- posed in the cognitive models of SAD in adults (e.g. Clark & Wells, 1995) are also evident in socially anxious children and adolescents. In comparison to non-socially anxious peers, children and/or ado- lescents with high levels of social anxiety show:-

a. More negative anticipated outcomes from social tasks (Alfano, Beidel, & Turner, 2006) and more negative expectations regarding the quality of one’s own performance (Blote, Miers, Heyne, Clark, & Westenberg, 2014)

b. A greater level of self-focused attention (Blote et al., 2014; Hodson, McManus, Clark, & Doll, 2008; Kley, Tuschen-Caffier, & Heinrichs, 2012; Schreiber, Hofling, Stangier, Bohn, & Steil, 2012)

c. More negative cognitions/self-talk during social performance tasks (Alfano et al., 2006; Kley et al., 2012; Ranta, Tuomisto, Kaltiala-Heino, Rantanen, & Marttunen, 2014)

d. More negative evaluation of one’s performance after the event (Alfano et al., 2006; Inderbitzen-Nolan, Anderson, & Johnson, 2007; Miers, Blote, Bokhorst, & Westenberg, 2009), although this may be a reflection of real performance deficits (see section on social skills)

e. More negative pre-event processing (Hodson et al., 2008)

f. Greater post-event processing and rumination (Hodson et al., 2008; Schmitz, Kramer, Blechert, & Tuschen-Caffier, 2010)

g. Greater likelihood of negative interpretation of social informa- tion or situations, including negative perception of audience responses (Blote et al., 2014; Miers, Blote, Bogels, & Westenberg, 2008) and negative observer-perspective images (Ranta et al., 2014), and

h. Increased frequency and vividness of negative distorted self- images, associated with greater distress relating to social situ- ations (Schreiber & Steil, 2013; Schreiber et al., 2012),

Importantly, these effects are still evident after controlling for depression (Hodson et al., 2008; Miers et al., 2008). Thus they cannot be attributed solely to a general cognitive bias associated with negative affect. Taken together, the research provides strong evidence that socially anxious adolescents and children as young as 8 years show similar biases and distortions in cognitive processes that may explain the maintenance of social anxiety. Researchers have also started to examine the inter-relationships between these cognitive processes in children, consistent with cognitive models of the maintenance of SAD. For example, Blote et al. (2014) explored the mediating role of negative self-evaluation cognitions (measured by negative performance expectations) and self- focussed attention in the link between social anxiety and chil- dren’s perceptions of their audience’s reactions to a pre-prepared speech. The audience reaction had been carefully controlled to be neutral. The results showed that children with high levels of pre- existing social anxiety were significantly more likely to rate their audience reaction negatively. This relationship was mediated by negative expectations of performance, which in turn influenced self-focussed attention, which was then associated with the child’s perceptions about the audience. Thus, socially anxious children expected that they would perform badly, which increased the focus of their attention upon their own behavior, negative thoughts, feelings, and physical symptoms. In turn, this self-focus was sug- gested to detract from accurate and valid monitoring of audience behavior, resulting in a perception of the audience that is aligned more with their own negative thoughts and interpretations of their performance (Blote et al., 2014). Furthermore, the lack of attention to the audience’s actual reactions may then fail to provide evidence to disconfirm the child’s negative perceptions about their audience, resulting in a perpetuation of their negative beliefs. The focus on the self rather than on others in the interaction may also result in failure to notice important social cues, meaning that the young person does not modify their behavior appropriately to increase their chance of success on the social task. It is encouraging to see studies with young people examining the inter-relationships be- tween the cognitive factors that are associated with social anxiety. Clarifying the way in which various cognitive processes can influ- ence each other may inform the development of more effective treatments for SAD in children and adolescents.

As noted above, cognitive models of SAD have focussed pri- marily on processes that maintain anxiety, and it is unclear whether they also play a role in its initial development. It is important to examine how such processes come to be established and whether they interact with other factors to increase the risk of developing SAD. It is also feasible that they play a role as mediating factors in explaining how distal causal factors (such as genetic factors, childhood temperament and parenting practices) operate to increase the risk of developing SAD. A major weakness in the literature to date, however, has been the reliance upon cross- sectional and correlational designs involving youth who already manifest high social anxiety, making it difficult to determine whether such biases are simply a cognitive reflection of social anxiety, or whether they play a role in its development.

S.H. Spence, R.M. Rapee / Behaviour Research and Therapy 86 (2016) 50e67 57

There has been limited longitudinal research investigating the causal role of cognitive factors in the development of SAD. Miers, Blote, de Rooij, Bokhorst, and Westenberg (2013) reported a community-based longitudinal study involving a cohort-sequential design in which young people aged 9e17 years were followed up at 4 time points over 3e5 years. Three main trajectories of social anxiety were identified i) initially higher and changing (increasing from age 10e13 years, and then decreasing); ii) moderate and decreasing; and iii) low and slightly decreasing. Higher levels of negative interpretations of ambiguous situations and self-focussed attention, but not performance evaluation predicted the high and changing rather than the moderate trajectory. However, this study did not identify a trajectory of increasing social anxiety over this period thus the results do not clarify whether the cognitive pro- cesses were associated with initial etiology rather than mainte- nance of existing high levels of anxiety.

In a further paper, Miers, Blote, Heyne, and Westenberg (2014) examined predictors of social avoidance behavior in the community-based cohort study. Two distinct trajectories were found: one that began with low levels of social avoidance that increased between 9 and 17 years, followed by a slight decrease, and a second involving relatively stable, low avoidance. A high level of post-event rumination, but not anticipatory processing or self- focussed attention, significantly predicted the increasing- avoidance trajectory after controlling for initial depression and social anxiety. Miers et al. (2014) noted that this finding is consis- tent with the updated version of the Rapee and Heimberg model (Heimberg, Brozovich, & Rapee, 2010) that emphasized the role of post-event rumination in the maintenance of SAD and associated avoidance. The revised model suggested that the process of persistently recalling social interactions and interpreting them in a negative way leads to an increasingly negative view of the social world. When future social events are anticipated, recollections of previous social experiences are brought to mind and are then linked to thoughts and images about the forthcoming situation, resulting in anticipatory anxiety and avoidance. The findings of Miers et al. (2014) suggest that the tendency to engage in post- event rumination may be particularly important in both the development and maintenance of social avoidance.

Information about causal processes can also be obtained through the manipulation of variables thought to be involved in the etiology or maintenance of SAD and observing the impact on social anxiety. For example, Kley, Tuschen-Caffier, and Heinrichs (2011) experimentally manipulated self-focussed attention in children and found a detrimental impact upon anxiety subsequently expe- rienced during a social task in both socially anxious and non- anxious children. Irrespective of SAD status, children who were instructed to use internally self-focussed attention strategies, compared to those instructed to use externally focussed strategies, reported more anxiety, lower expectations regarding evaluation of their performance by others, and more frequent negative cogni- tions during the task. This finding is consistent with suggestions from models of maintenance that a shift of attention towards in- ternal aspects of oneself, such as one’s behaviors, thoughts, emo- tions, appearance and physiology, may enhance awareness of negative mental representations of the self and thus increase negative cognitions and feelings of anxiety.

3.4.2. Safety behaviors Consistent with the adult literature, several studies have

confirmed that children and adolescents with high social anxiety also engage in maladaptive safety behaviors in an attempt to reduce the probability of an aversive outcome from a social situation (Kley et al., 2012; Ranta et al., 2014; Schreiber et al., 2012). Some commonly used safety behaviors reported by children with SAD

include avoidance of eye contact, trying to behave in a way that does not attract attention, saying little, and avoiding talking about themselves (Kley et al., 2012). Socially anxious children do not seem to use some of the more complex strategies reported by adults with SAD, such as covering the face or attempting to disguise symptoms of sweating. It is likely that the safety behaviors selected by socially anxious children may lead to adverse perceptions by others, and thus actually reduce the likelihood of social success, although this may vary between cultures. In this way, safety behaviors overlap with performance deficits (see below).

As noted earlier, several of the safety behaviors used by socially anxious young people are also characteristic features of BI, espe- cially reduced eye contact and limited verbalization. Hence we might speculate that a subset of these behaviors reflect funda- mental temperamental characteristics and are key to the develop- ment of the disorder, while other behaviors develop later as a consequence of heightened social anxiety and are more relevant to its maintenance. Only long-term empirical assessment that as- sesses individual behaviors will be able to shed light on this prediction.

3.5. Social performance deficits

In 2004 we proposed that poor social performance played a role in both the development and maintenance of childhood SAD. The limited evidence available at that time certainly indicated a ten- dency for children with SAD to show poor social performance as judged by others, but it remained unclear whether they actually lacked the ability to perform appropriately or whether they had the skill but their performance was impaired as a consequence of anxiety, leading to diversion of attentional resources from the social task. Hence, it was unclear whether poor social performance played a causal role, or served to perpetuate established high levels of social anxiety. A further unanswered question was whether the negative social expectations and self-perceptions of socially anxious children represent a negative distortion of the situation, or reflect reality.

The majority of studies involving children generally support the view that SAD is associated with relatively poor performance on challenging social tasks. Spence et al. (1999) proposed that a vicious cycle exists in which young people with SAD exhibit poor skill performance in social situations, particularly in interactions with peers, resulting in adverse social outcomes. Such negative out- comes in turn, are proposed to lead to anxiety and then avoidance of such situations, providing less opportunity for the acquisition of social skills, and thus resulting in further poor social performance.

There is good evidence to support this proposition. Compared to non-anxious youth, several studies have shown that socially anxious children do show deficits in social skill performance, as assessed from behavioral observations of roleplay, social tasks in the laboratory and in the real world. For example, compared to a non-anxious control group, youth diagnosed with social phobia have been shown to participate in fewer social interactions and initiate social interactions less frequently in a school setting, to speak fewer words in response to prompts and to have slower speech response latency on read-aloud and roleplay tasks (Alfano et al., 2006; Beidel, Turner, & Morris, 1999; Spence et al., 1999). They are also rated as less socially skilled by their parents (Spence et al., 1999) and independent adult and peer judges (Alfano et al., 2006; Beidel et al., 1999; Blote, Duvekot, Schalk, Tuinenburg, & Westenberg, 2010; Blote, Miers, & Westenberg, 2015; Greco & Morris, 2005; Inderbitzen-Nolan et al., 2007) and to be judged by independent observers as appearing more nervous on social tasks (Miers et al., 2009). Miers, Blote, and Westenberg (2010) examined the social performance of high and low socially anxious adolescents

S.H. Spence, R.M. Rapee / Behaviour Research and Therapy 86 (2016) 50e6758

while giving a speech to a pre-recorded audience. Unfamiliar peer observers rated the high socially anxious youth as performing less well on a range of social skills, and the differences remained sig- nificant after controlling for depression. Although there is an increasing number of studies indicating poor social skills perfor- mance among socially anxiety children and teenagers, we note that some studies have not found this effect (Cartwright-Hatton, Hodges, & Porter, 2003; Tuschen-Caffier, Kuhl, & Bender, 2011). It is not clear why this is the case and whether these differing findings reflect variation in methodology or sample characteristics.

The majority of studies to date examining social skills and social phobia in youth have utilized cross-sectional designs, comparing young people with established social anxiety with non-anxious peers and it cannot be determined whether the poor social per- formance precedes or follows the social anxiety. However, one recent longitudinal study suggests that deficits in social skills may precede increases in social anxiety. Miers et al. (2013) examined predictors of different trajectories of SAD in young people aged 9e17 years who were followed up over a 3e4 year period. They found that an observer rating of poor social skills during a labora- tory speech task predicted a trajectory of initially higher and changing (increasing from age 10e13 years, and then decreasing) rather than moderate and decreasing or sustained low trajectories. As noted above, the study did not identify a consistently increasing trajectory to determine whether poor social skills performance predicted increases in social anxiety in initially non-anxious youth.

As mentioned above BI is associated with several fundamental characteristic avoidance strategies such as poor eye contact and limited verbalisations (Rapee, 2002). These behaviors are likely to interfere with appropriate social interactions and therefore can be said to reflect social performance deficits. Given their very early demonstration (certainly from two years of age (Kagan et al.,1984)), they may be thought to provide a key risk for the development of social anxiety and SAD.

There is some evidence therefore, that child SAD is associated with deficits in social skills performance although it remains un- clear whether these are a consequence of performance interference due to the social anxiety or reflect an actual lack of skills. However, a subset of withdrawal behaviors is likely to be present very early and also constitutes poor social performance. Therefore, high BI children may experience a range of adverse social outcomes as a consequence of their temperamental characteristics and, in turn, such negative experiences should predict increases in social anxi- ety. These points are discussed in the next section. From a thera- peutic point of view it is important to ensure that assessment includes an examination of social skills and subsequent social outcomes. Thus for many young people social skills training forms an important aspect of CBT interventions for SAD (Mesa, Le, & Beidel, 2015; Olivares, Olivares-Olivares, & Macia, 2014; Spence, Donovan, & Brechman-Toussaint, 2000).

4. Developments in the research 2004e2015: extrinsic/ environmental factors

4.1. Peer relationships and the school context

There is substantial evidence that many of the most feared and avoided situations for children and adolescents with SAD relate to the school context, particularly in relation to performing in front of the class (e.g., giving a talk, answering questions, musical or athletic performance), writing on the board, asking for assistance from a teacher, and being with peers during school breaks (see Blote, Miers, Heyne et al., 2015). Positive relationships with peers and the capacity to perform well on a broad range of social tasks at school become increasingly important with age. As they develop

from the preschool years, young people need to be able to make and retain friends, deal with teasing and inappropriate peer behavior, start and hold conversations, join into groups, ask for help, make presentations in class, and refuse unreasonable requests. As they progress into adolescence, they may also need to deal with the challenges of romantic relationships and increased peer influence.

In our 2004 paper we noted, with surprise, that there had been minimal research investigating the experience of adverse social outcomes by children with SAD given that conditioning theories would suggest that social fears could develop as a consequence of aversive learning experiences. We suggested that negative social outcomes such as excessive teasing, criticism, bullying, rejection, ridicule, humiliation, or exclusion by significant others would be associated with the development of social anxiety. The research at the time, however, mainly involved adult retrospective report or cross-sectional analyses. Nonetheless, the limited data led us to propose that a vicious cycle exists in which peers tend to interpret socially anxious behavior in a negative way, and come to dislike and subsequently exclude such children, which in turn serves to confirm the socially anxious child’s worst fears regarding peer so- cial interaction, further enhancing their fear and avoidance.

There is now good evidence that negative social experiences, particularly with peers, play a role in the development and main- tenance of SAD (Blote, Miers, Heyne et al., 2015). Research from Western cultures indicates that socially anxious children tend to have fewer friends, are less well-liked and accepted by peers, have more negative peer interactions, and are more likely to be passively neglected and actively rejected and victimized than non-anxious peers (Blote & Westenberg, 2007; Blote, Kint, & Westenberg, 2007; Erath, Flanagan, & Bierman, 2007; Greco & Morris, 2005; Ranta, Kaltiala-Heino, Pelkonen, & Marttunen, 2009, 2013; Ver- duin & Kendall, 2008). As will be discussed below, there is also convincing evidence that adverse social outcomes are a risk factor for future SAD, not just a consequence of social anxiety, and that socially anxious behavior is perceived by peers in a negative way thus supporting our earlier proposition about the importance of positive social performance. One caveat to the current research is that the vast majority has been conducted with Western pop- ulations. Research indicates that socially withdrawn behavior is not perceived nearly as negatively by peers from East Asian countries and in some cases may even be given positive attributes (Chen & Tse, 2008; Chen et al., 2006). As a result, the negative conse- quences of socially withdrawn behavior described below among Western youth, is generally not seen among East Asian populations. Unfortunately, there is a complete lack of research linking shy be- haviors, peer relationships, and subsequent SAD among non- Western populations.

It is important to determine what it is about socially anxious young people that leads to adverse social outcomes. As described earlier, there is growing evidence that socially anxious young people perform less well on social tasks than non-anxious peers. In turn, poor social performance may then lead peers to respond in a less positive manner towards the anxious child, which leads the anxious young person to expect negative outcomes from future social interactions, perpetuating a vicious cycle (Miers, Blote, & Westenberg, 2011). In keeping with this proposition, Greco and Morris (2005) found that poor social skills (as judged by teach- ers) mediated the association between social anxiety and lack of peer acceptance. Blote, Miers, and Westenberg (2015) also exam- ined the mechanisms through which social anxiety and social performance influence the way that children are judged by their peers. This study demonstrated that socially anxious adolescents showed significantly lower social skills performance on a video- recorded speech task, as judged by unfamiliar adults and peers, compared to non-anxious adolescents and were less liked/more

S.H. Spence, R.M. Rapee / Behaviour Research and Therapy 86 (2016) 50e67 59

rejected by peer judges and rated as less physically attractive than controls. An examination of the inter-relationships between these variables suggested that peer-rated attractiveness and adult-rated social performance partially mediated the association between social anxiety and peer rejection. However, the results showed that there was still a significant direct link between social anxiety and peer rejection over and above the mediating effects, suggesting that other explanatory factors remain to be identified. These authors suggested that there is something about the way in which socially anxious children interact socially that leads them to be judged negatively by others, in terms of judgments of attractiveness, skillfulness and being liked/accepted. Consistent with this propo- sition is a finding by Blote, Bokhorst, Miers, and Westenberg (2012) that socially anxious adolescents were not only more rejected by peer judges, but they were also more likely to be viewed as being “different” from the peer judges, and this assessment of “differ- ence” partially mediated the association between social anxiety and rejection. Perhaps overt behaviors indicative of nervousness may contribute to others’ perceptions of a child being different. This appears to be a particular problem for socially anxious children rather than those with other types of anxiety disorders. Verduin and Kendall (2008) found that unfamiliar peers rated social phobic children as more observably anxious during a videotaped oral presentation task and less liked than non-anxious controls, whereas this effect was not found for children with other anxiety disorders. Peers preferred children who self-reported lower social anxiety and who were rated by unfamiliar peers as being less anxious. Similarly, children described in hypothetical vignettes as showing behaviors demonstrating social reticence and fears of negative evaluation were rated as less likable and more “deserving of victimization” than vignettes of pro-social peers by 12 to 14- year old youth (Luchetti & Rapee, 2014).

It is perhaps not surprising therefore to find that socially anxious children are more likely to experience victimization by peers, with this finding being evident in multiple cross-sectional studies (Crawford & Manassis, 2011; Gren-Landell, Aho, Andersson, & Svedin, 2011; La Greca & Harrison, 2005; Ranta et al., 2009). Furthermore, there is evidence from prospective studies of a causal relationship with victimization increasing the risk of future social anxiety, although the results have differed in terms of the type of peer victimization that places children at particular risk. For example, Storch, Masia-Warner, Crisp, and Klein (2005) found that relational, but not overt, victimization increased the risk of social phobia, but not social anxiety symptoms, one year later. Similarly, Siegel, La Greca, and Harrison (2009) found that peer victimization, and particularly relational bullying, predicted increases in social anxiety in adolescents over a 2-month period. Although Loukas and Pasch (2013) also found that peer victimization predicted increased social anxiety over a 12-month period, this effect was significant for overt, but not relational victimization. These apparently contra- dictory findings may reflect the possibility that girls and boys respond differently to different types of victimization. For example, Ranta et al. (2013) found that overt/direct victimization at age 15 increased the risk of social phobia at age 17 among boys, whereas for girls relational victimization predicted social phobia. We note that some research has failed to find evidence of increases in social anxiety in response to peer victimization (Tillfors, Persson, Will�en, & Burk, 2012). Differences in results could possibly be explained by methodological differences such as variation in the follow-up period, age group, measures, and whether social phobia or social anxiety symptoms were the outcomes of interest.

Although the balance of evidence to date suggests that peer victimization does increase the risk of social anxiety, it is also clear that its consequences expand to other types of internalizing problems. Reijntjes, Kamphuis, Prinzie, and Telch (2010) conducted

a meta-analysis of 18 longitudinal studies and found that peer victimization predicted future increases in internalizing problems in general (as well as the reverse). More recent studies have also indicated a significant link between peer victimization and future increases in depression (Loukas & Pasch, 2013; Sweeting, Young, West, & Der, 2006; Zwierzynska, Wolke, & Lereya, 2013).

In addition to being a consequence of peer victimization, social anxiety is also likely to increase the chance of victimization, reflecting a reciprocal relationship. Theoretically it makes sense to suggest that socially anxious young people will be at increased risk for future victimization, as a consequence of the peer issues out- lined above, such as fewer friends, lower popularity, lower peer acceptance, and poor social skills performance. As noted, research based on hypothetical vignettes has shown that early teens perceive socially anxious youth as “more deserving” of victimiza- tion (Luchetti & Rapee, 2014). Crawford and Manassis (2011) pro- posed that weaker social skills and high levels of anxiety of socially anxious children means that they tend to be more submissive and less likely to respond assertively or to fight back when victimized. They are also more likely to be alone and less likely to have friends to support and protect them. Bullying behavior from peers is therefore facilitated and perpetuated, and in turn generates further social anxiety as a consequence of fear conditioning, social with- drawal/avoidance and further decrease in the learning of social skills. This vicious cycle reflects that which maintains the ongoing association between quality of peer relationships generally and social anxiety.

Although this proposition makes theoretical sense, to date the empirical evidence is limited. Broadly, longitudinal studies have supported both directions of relationship e internalising predicts later victimization and victimization predicts later internalizing (Reijntjes et al., 2010). However, this research has generally not distinguished specific forms of internalizing. In one of the only studies looking specifically at social anxiety, Ranta et al. (2013) found that direct/overt (but not relational) victimization both predicted and was predicted by social anxiety over a 2-year period but only in boys.

In an attempt to explain the causal mechanisms, we propose here that adverse social outcomes, particularly with peers, increase the risk of developing and perpetuating social anxiety through their impact upon avoidance behaviors and negative beliefs and thoughts relating to peer interactions. Rejected, neglected and victimized children are likely to come to experience social in- teractions as aversive, to be avoided where possible, and to trigger a range of negative beliefs and thoughts about peer relationships. Research into the mediating pathways between adverse social outcomes and social anxiety is in its infancy. However, an inter- esting recent study examined the mechanism of action through which emotional abuse may impact upon both social anxiety and depression, through the development of early maladaptive sche- mas about the self and others (Calvete, 2014). Of particular rele- vance to the development of SAD, Calvete proposed that the experience of maltreatment as a child leads to the development of dysfunctional, and pervasive patterns consisting of memories, emotions, cognitions, and bodily sensations about oneself and re- lationships with others. Some examples include beliefs that one will not receive acceptance and respect from others, failure schema relating to the belief that one has failed, will inevitably fail, and are fundamentally inadequate, and schema relating to an extreme focus on the desires of others and need for acceptance. Calvete (2014) followed up a large sample of adolescents over a 1-year period. Only peer victimization, and not parental emotional abuse, predicted increases in social anxiety. However, consistent with hypotheses, maladaptive schema relating to an extreme focus on the desires of others and need for acceptance were found to

S.H. Spence, R.M. Rapee / Behaviour Research and Therapy 86 (2016) 50e6760

mediate the link between peer abuse and social anxiety. In summary, there is strong evidence that socially anxious youth

tend to experience a range of adverse social outcomes with peers, and that peer victimization in particular increases the risk of future social anxiety. We should also note that there is some evidence that this may also apply to relationships with siblings, given that victimization by siblings has been shown to be associated with SAD in adolescents (Gren-Landell et al., 2011). As noted, these re- lationships are supported in youth from Western countries, but research utilising non-Western samples is sorely needed.

4.2. Parenting factors

Given that the majority of adults with SAD report onset during late childhood or early adolescence, we need to look closely at environmental factors during childhood, particularly within the home and at school that may influence its development. In our 2004 paper we concluded that although studies had demonstrated significant links between parental factors and social phobia, such as over control and protection, these parenting styles were also common to other anxiety disorders. We also noted that most of the research was retrospective in nature, with the few longitudinal studies producing mixed findings. We suggested a complex picture in which child anxiety and parenting style had a two-way direction of influence. There was also minimal research into the role of fa- thers at that time.

Most recent models of child SAD posit that parents play an important role in the early development of anxiety in children and in maintaining or enhancing a behaviorally inhibited temperament. For example, Ollendick and Benoit (2012) and Ollendick, Benoit, & Grills-Taquechel (2014) proposed a complex parent-child interac- tional model to explain the development of SAD. They suggested that a behaviorally inhibited temperament and insecure attach- ment increase the risk of SAD, but this association is influenced by parental behavior that can either buffer or exacerbate the outcome. Furthermore, they proposed that parents’ responses to their child are influenced by their own anxiety and information processing biases, and also by the temperament and behavior of the child in a reciprocal manner. Ollendick, Benoit, Grills-Taquechel, and Weeks (2014) suggested that parents who use supportive, encouraging strategies to guide their children to engage socially and to approach novel situations and who model constructive coping techniques are more likely to facilitate socially confident behavior to assist chil- dren in dealing with novel or challenging situations. In contrast, overly controlling, intrusive, solicitous, or critical/rejecting parenting are posited to perpetuate and exacerbate a behaviorally inhibited temperament. Finally, in explaining the link between parenting practices and child SAD, they suggested that parenting behavior that communicates threat, anticipation of aversive out- comes, is likely to lead to the acquisition of similar cognitive biases in the child, thus explaining inter-generational transmission.

It should be noted that similar models have been proposed to explain the development of child anxiety in general (Negreiros & Miller, 2014) and, as for several other risk factors, parenting is likely to be a general risk rather than a factor relevant specifically to social anxiety.

Although a great deal has been written about the importance of parenting behavior in the development and maintenance of social anxiety, the evidence is relatively weak. In our 2004 paper we noted significant methodological limitations in the research into parenting and social anxiety and lack of consistency in results. These issues remain. More recent reviews have continued to identify primarily cross-sectional studies (longitudinal studies remain the exception and experimental studies are almost non- existent), measures and operationalisation of parenting styles is

inconsistent, and almost no studies have focussed on social anxiety specifically (McLeod, Wood, & Weisz, 2007; Van Der Bruggen, Stams, & B€ogels, 2008; Yap, Pilkington, Ryan, Kelly, & Jorm, 2014).

When reviews of the literature have been conducted despite the methodological limitations, parenting styles predict a relatively small proportion of variance in anxious symptoms (McLeod et al., 2007; Yap et al., 2014). However, evaluating parenting styles very broadly may miss the point because most theoretical models of parenting in child anxiety emphasise the key impact of a control- ling parental style, or more specifically, parental over-protection, that is, parenting that maintains or exacerbates threat expec- tancies (Rapee, Gaston, & Abbott, 2009). In fact, meta-analytic re- views have demonstrated the clearest relationship between over- controlling parenting and child anxiety, with moderate effect sizes (McLeod et al., 2007; Van Der Bruggen, Stams, B, €o, & gels, 2008; Yap et al., 2014). In the only experimental manipulation to date, children (aged 7e13) whose mothers were instructed to control and take over preparation of a speech, displayed signifi- cantly greater anxiety in a subsequent speech than children whose mothers were instructed to promote autonomy (de Wilde & Rapee, 2008).

To date, most of the small number of studies that have examined the link between parenting behavior and social anxiety specifically (rather than anxiety in general) have been cross-sectional (see Ollendick et al., 2014 for a review). This is a serious limitation as there is evidence that parents may behave differently towards anxious compared to non-anxious children and therefore the parenting styles described above may be a consequence of trying to deal with an anxious child rather than a cause of the child’s anxiety. In fact, theoretical models predict a reciprocal relationship in which withdrawn child behavior elicits controlling parenting, which in turn increases anxiousness (Dadds & Roth, 2001; Rubin, Coplan, & Bowker, 2009). Consistent with this proposition, early shyness and inhibition have been shown to elicit later parental protectiveness (Edwards, Rapee, & Kennedy, 2010; Rubin, Nelson, Hastings, & Asendorpf, 1999). Combining these ideas, Kiel and Buss (2011) found that a fearful temperament at 2 years elicited protective behaviors from their mothers, which in turn predicted social withdrawal 3 years later.

The influence of parenting is also complicated by evidence that the types of parenting that increase risk of child social anxiety may differ for mothers and fathers. A prospective study (Knappe, Beesdo-Baum, Fehm, Lieb, & Wittchen, 2012) found that maternal overprotection and paternal rejection and lower emotional warmth in adolescence were associated with future risk of SAD. On the other hand, both maternal and paternal overprotection were associated with increased risk of specific phobia and panic disorder. Clearly, fathers and not just mothers play an important role in determining the development of anxiety in children, and it has been suggested that paternal behavior may actually have a stronger influence upon child social anxiety than maternal parenting (B€ogels & Perotti, 2011). In particular, father behavior that encourages in- dependence, curiosity, and competition (termed “challenging parenting”) has been shown to reduce the risk of social anxiety in children (Majdand�zi�c, M€oller, B€ogels, Vente, & Boom, 2014). In contrast, neither maternal nor paternal over-control predicted future social anxiety in that study, and maternal challenging parenting was actually associated with higher future child social anxiety. It was interesting to note that the significant relationships were only evident for the first and not second child in the family.

Emerging evidence also suggests that parenting behavior, in addition to a direct effect, may also interact with other risk factors in the impact on child anxiety. For example, parental over-control was found to moderate the impact of a behaviorally inhibited temperament to increase the risk of social anxiety (Lewis-Morrarty

S.H. Spence, R.M. Rapee / Behaviour Research and Therapy 86 (2016) 50e67 61

et al., 2012). Similarly, an early study showed that a combination of child inhibition and parental intrusiveness predicted the child’s shyness two years later (Rubin et al., 2002).

Clearly the association between parenting behavior and child social anxiety is extremely complex and we await further, longi- tudinal studies to clarify the picture. The limited evidence available to date suggests potentially different patterns of influence for the mother versus father, for different children within the family, and according to the temperament and age of the child. A further complication that has received little attention is the influence of culture, given that parental reactions to withdrawn behaviors are likely to differ between Western and Eastern populations (Chen et al., 1998).

4.3. Trauma and abuse, adverse life events

Consistent with our earlier conclusions (Rapee & Spence, 2004), current literature points to a range of adverse/stressful life events and trauma during childhood that increase the risk of developing SAD. Rather than being specific for SAD, events such as sexual, physical and emotional abuse and neglect, a childhood history of separation from parents, parental marital discord, family violence, childhood illness, and parental psychopathology (including mental health, substance abuse, and criminal behavior) represent trans- diagnostic risk factors for a broad spectrum of disorders in adolescence (B€ogels & Brechman-Toussaint, 2006; McLaughlin et al., 2012). It is not surprising, therefore, to find the experience of such factors during childhood being associated with SAD in adulthood (Acarturk et al., 2009; Bandelow et al., 2004; Bishop, Rosenstein, Bakelaar, & Seedat, 2014; Bruce, Heimberg, Blanco, Schneier, & Liebowitz, 2012; Cougle, Timpano, Sachs-Ericsson, Keough, & Riccardi, 2010; Kuo, Goldin, Werner, Heimberg, & Gross, 2011; Simon et al., 2009). It should be noted, however, that much of the evidence is based on retrospective report from adults who already have a diagnosis of SAD. There are also conflicting findings, with some indication that childhood emotional and not physical abuse is associated with SAD in adults (Kuo et al., 2011; Spinhoven et al., 2010).

In studies involving socially anxious children, cross-sectional research has found higher levels of sexual, physical and emotional abuse and parental neglect (Gren-Landell et al., 2011), but there is a lack of longitudinal research involving young people other than that relating to peer victimization, as summarized above. Prospective studies are needed that examine the impact of exposure to various forms of trauma or adverse life events upon the development of SAD. Eleven years ago we noted that a key chal- lenge for researchers is to determine why some children who experience such adversity go on to develop SAD whereas others do not. This issue remains current and studies are needed that examine the influence of exposure to multiple adversities and the potential moderating effects of intrinsic and extrinsic factors in exacerbating or attenuating the impact of exposure. We also need further investigations into the mediating mechanisms through which exposure to such adverse contexts and events impact upon SAD. Studies such as that outlined above by Calvete (2014) that examined the impact of victimization and maltreatment upon emerging maladaptive schemas, and the subsequent impact upon social anxiety, would be valuable.

4.4. Cultural influences

Social anxiety most likely reflects a fundamental evolutionary process aimed at minimizing intra-species threat through appeasement and sacrifice of status (Gilboa-Schechtman, Shachar, & Helpman, 2014). Consistent with this purpose, variations in

social anxiety and clinically identified SAD have been identified across all societies in which they have been evaluated (Stein & Stein, 2008). Cultural differences might nonetheless affect ways in which social anxiety is expressed and thresholds at which the clinical disorder, SAD, is diagnosed. They may also interact with other risk factors for social anxiety by affecting social reactions to withdrawn behavior. Our model in 2004 was the first to make these explicit predictions.

First, we suggested that the construct of social anxiety itself might differ in its expression between cultures. In particular, we pointed to the syndrome referred to as Taijin Kyofusho (TKS) in Japan, in which individuals display social withdrawal and distress in social contexts due to a core fear that one is causing offense to others (Brockveld et al., 2014; Kim, Rapee, & Gaston, 2008). The behavioral manifestations are very similar to the ways in which social anxiety is displayed among Western populations, but the underlying fear in Western groups is more commonly of being negatively evaluated by others. These differences are consistent with the stronger collectivistic emphasis in East Asian than West- ern cultures. Evidence since this time has largely been consistent with this proposal. It has been shown that TKS can be identified relatively commonly among East Asian populations but is relatively rare within Western populations (Kim et al., 2008).

Second, we predicted that cultural differences could influence the diagnostic threshold for determination of SAD and hence affect prevalence of the disorder. In particular, it was noted that preva- lence estimates for SAD differed markedly between populations, from very low prevalence in Japan and Korea to consistently highest prevalence estimates in the USA (Brockveld et al., 2014). Recall that according to our model, the clinical manifestation of SAD comprises a combination of severity of social anxiousness and the impact that this causes on life. Hence it was predicted that among cultures where social reticence and appeasement are normative and in fact positive characteristics, the impact of social anxiety would be low and concomitantly prevalence of SAD would be low. In contrast, among cultures where social appeasement behaviors hold negative connotations and effects, prevalence of SAD would be high. Recent research has supported this suggestion. In a study describing hy- pothetical vignettes of socially reticent behaviors to 455 students from East Asian countries (Japan and China) and 361 students from Western countries (including Australia, USA, Netherlands, and Germany), students from the Asian countries reported significantly lower predicted life impact from these socially anxious behaviors (Rapee et al., 2011). Further, when describing their own experi- ences, students from Western countries demonstrated a signifi- cantly greater negative correlation between social anxiety and quality of life than the students from Asian countries. Consistently, other research has shown that Chinese youth who are shy often show very positive longer term outcomes (Chen, Chen, Li, & Wang, 2009) which is in contrast to the often poor outcomes shown by shy youth from Western countries (Coplan & Armer, 2005). Students from East Asian countries also report more positive attitudes to- ward socially reticent behavior than students from Western coun- tries (Alden et al., 2010; Heinrichs et al., 2006).

5. Implications for a model of SAD in children and adolescents

It is clear from the discussion above that many of the features of the original model proposed by Rapee and Spence (2004) continue to be supported, although research in the intervening period means that the model requires some updating. The evidence reviewed indicates that the following points continue to be important in understanding the etiology and maintenance of SAD:-

S.H. Spence, R.M. Rapee / Behaviour Research and Therapy 86 (2016) 50e6762

a. Risk factors are likely to have different impact at different ages. For example, twin data indicate a greater heritable basis to SAD among youth than among adults (Scaini et al. (2014).

b. Risk factors can interact e there is increasing evidence that some risk factors may combine to increase the impact upon the development of SAD, and others may have a protective effect that buffers the impact of a particular risk factor. For example, evidence suggests that the impact of behavioral inhibition or genetic risk upon future SAD can be influenced by parenting style.

c. Different pathways and combinations of factors can result in SAD (the principle of equifinality). The evidence reviewed above reveals a myriad of factors that can influence the development and maintenance of SAD and there are clearly multiple path- ways that can lead to SAD.

d. Any one risk factor can result in various outcomes, not just SAD (the principle of multifinality), thus we need to examine transdiagnostic vs disorder specific factors. There are several examples of this in the studies reviewed. For example, an over- controlling parenting style, emotional and sexual abuse, peer victimization, and a temperament of behavioral inhibition have all been shown to be associated with the development of other types of anxiety and depression, not just with SAD. In fact, identification of broad factors that are specifically relevant to the development of SAD is rare and it is more likely that subtle nuances in the ways in which such factors occur or are experi- enced will influence variations in the expression of anxiousness. For example, parent modelling of anxiety is likely to be a broad risk that is relevant across various forms of anxiety, but specific modelling of shyness and hesitation during interactions with strangers might more specifically predict social anxiety (de Rosnay, Cooper, Tsigaras, & Murray, 2006).

e. Risk factors can be transactional and reciprocal, with young people influencing their own social outcomes which, in turn, impact upon the young person. Again, there are important il- lustrations of this point in the recent literature. The findings that social anxiety in children can influence the use of particular parenting strategies and increase the risk of future peer victimization are both examples of the ways in which social anxiety can contribute to its own magnification.

f. It is extremely difficult to distinguish between causal and maintaining factors from the research to date. Although, theo- retically, these may not be the same, it appears from the liter- ature reviewed that many of the identified risk factors apply to both the development and maintenance of SAD. Although there is a strong theoretical rationale to suggest that some variables increase the risk of developing SAD, there is insufficient evi- dence from prospective research to enable us to draw conclu- sions about their causal role. For example, there is a lack of evidence to determine whether safety behaviors represent maintaining factors that act to perpetuate existing social anxiety or lead to its onset.

Fig. 1 presents an update to our previous theoretical model that illustrates proximal and distal risk factors, moderating and medi- ating processes that are proposed to impact on the development of SAD. Although some of the linkages are still speculative, there is increasing evidence to support an interactional model, in which environmental factors interact with instrinsic elements, particu- larly with the child’s temperament, to increase the risk of devel- oping SAD. Thus children with a genetic or biological predisposition to SAD and/or behaviorally inhibited temperament are proposed to be at particular risk for the development of SAD if they also expe- rience environmental risk factors for SAD such as an abusive family situation, or repeated victimization by peers. These environmental

and instrinsic factors are proposed to interact to influence the young person’s acquisition of social skills, interpersonal problem solving and emotional regulation skills.

Adverse life and learning experiences, for those who are intrinsically vulnerable, are also proposed to result in maladaptive schema and beliefs relating to the self and others. Through their social experiences, young people come to believe that they are deficient, stupid, and unattractive, with little ability to control the outcomes of social situations. Other people come to be regarded as highly critical, with extremely high standards, and are perceived as being like an audience observing their every action. The model posits that maladaptive beliefs about the self and others then contribute to a range of cognitive biases and distortions before, during and after challenging social interactions, including biases in attention, in expectations, interpretations, and evaluations. In turn, these cognitive events are proposed to lead to the behavioral, so- matic and emotional responses of anxiety. We suggest that increased vigilance to social situations and expectations that one will perform in a humiliating or embarrassing way, that others will appraise and respond negatively, and that the outcome will be terrible will all contribute to the experience of anxiety. During so- cial tasks, high levels of self-focussed attention, and distraction away from the social task are likely to impair social performance. When a social interaction finishes, a tendency to interpret the response of others and the quality of one’s own performance as being worse than they actually were is also likely to generate anxiety. Even if a young person’s social performance is less competent than that of peers, and there may be some validity to the poor evaluations, a tendency to self-evaluate even more negatively is likely to contribute to anxiety. Similarly, a tendency to dwell on the situation and to recall adverse social outcomes in imagery may contribute to symptoms of anxiety. Avoidance of social interactions in the future is likely to be a behavioral manifestation of anxiety, which in turn is proposed to reduce opportunities to acquire social skills and to limit the development of positive, protective peer re- lationships. Thus, a vicious cycle is established leading to increasing levels of social anxiety. Ultimately, the diagnostic criteria for SAD will be met when social anxiety is sufficiently severe and combined with increasing impairment. In turn, degree of impairment will be affected by variables such as the individual’s life goals, age, gender and culture.

6. Implications for treatment and prevention

There is increasing evidence that traditional cognitive behavior therapies that are used in the treatment of most forms of anxiety disorder are less successful for SAD than for other types of anxiety, in both children and adults (Hudson, Keers et al., 2015; Hudson, Rapee, et al., 2015; Norton & Price, 2007). These interventions typically include education about anxiety and emotions, emotional regulation strategies such as relaxation, identification and modifi- cation of maladaptive thoughts, problem solving skills training, and graded exposure to feared situations. Many also include a parent training component designed to teach parents how to encourage their child to approach rather than avoid feared situations, and how to model appropriate behavior in social situations. As such, tradi- tional CBT for child anxiety aims to address a range of factors that have been implicated in the development and maintenance of anxiety disorders in general.

One possible explanation for the relatively poor results of the generic approach in the treatment of SAD is that is has not focussed sufficiently upon the cognitive and behavioral components that involved in its development and maintenance. If, after treatment, a child is still entering feared situations with poor social perfor- mance, biases in attention and self-focus, significant maladaptive

Fig. 1. A model of the development of social anxiety and social anxiety disorder.

S.H. Spence, R.M. Rapee / Behaviour Research and Therapy 86 (2016) 50e67 63

schema about themselves and their relationships, and is engaging in negative cognitive expectations, misinterpretations, and post- event processing, then it is perhaps not surprising that they will continue to show social anxiety.

It is possible that interventions that aim to change factors that contribute significantly to the development and maintenance of SAD may be more effective than those that focus on processes that are common across anxiety disorders. It would seem important to ensure that treatment of childhood SAD includes elements to change the modifiable distal and proximal risk and protective fac- tors, and their mediating processes. Where appropriate, in- terventions for socially phobic children may need to include elements to teach social skills and to reduce victimization. In terms of cognitive content, elements may be required to assist children to identify and tackle maladaptive beliefs about themselves and others, and to reduce biased attentional processes, interpretations, and cognitions relating to expectations and evaluations (as outline in the model above).

Over the past decade researchers have started to examine SAD- specific interventions for youth. For example, Klein et al. (2015) demonstrated the feasibility of a brief intervention to teach chil- dren with SAD to use positive rather than negative interpretations of social threat scenarios. The results showed that children were able to learn to reduce the negative bias, with an associated reduction in social anxiety as assessed by parents, but not children. Similarly, Vassilopoulos, Banerjee, and Prantzalou (2009) showed that it was possible to train socially anxious preadolescents from a non-clinical sample to make more benign interpretations of ambiguous social cues, with associated reductions social anxiety. Melfsen et al. (2011) developed and evaluated a cognitive inter- vention for 8e14 year olds with SAD, based on the Clark and Wells (1995) model, and reported positive outcomes. Similarly, relatively strong outcomes have been reported for interventions that include a strong component of social skills training in the treatment of SAD

in youth (Beidel, Turner, & Morris, 2000; Spence et al., 2000). To date, there has been an absence of studies with children and ado- lescents to examine whether SAD-specific intervention is actually more effective than the traditional, generic CBT approach. With adults, however, Rapee et al. (2009) showed significantly larger effects of an intervention that incorporated components to specif- ically target theoretically maintaining factors of SAD compared with a more traditional cognitive behavioral intervention.

7. Summary and conclusion

Since our 2004 paper, research relating to the etiology of SAD has advanced substantially. There remains strong empirical support for our original model, but new evidence enables us to clarify some important points and to include more detail about moderating and mediating factors. It is clear that there are many pathways to the development of SAD, with a wide range of risk and protective fac- tors having an influence. The model that we present here highlights the complex pathways towards the development of SAD, and the interplay between environmental and intrinsic factors. Gradually, evidence is emerging to explain how risk and protective factors can interact to increase or decrease the risk of SAD and how different factors may influence risk at different stages of development. Although research has expanded significantly since 2004, there remain many unanswered questions and several elements of our model are still speculative. Longitudinal studies over the next decade will undoubtedly clarify strengths and weaknesses in the model. Such information will be of particular value in informing the development of more effect treatments for SAD and in the design of preventive intervention.

References

Abidin, R. R. (1992). The determinants of parenting behavior. Journal of Clinical Child Psychology, 21(4), 407e412.

S.H. Spence, R.M. Rapee / Behaviour Research and Therapy 86 (2016) 50e6764

Acarturk, C., Smit, F., de Graaf, R., van Straten, A., ten Have, M., & Cuijpers, P. (2009). Incidence of social phobia and identification of its risk indicators: A model for prevention. Acta Psychiatrica Scandinavica, 119(1), 62e70.

Alden, L., Oh, K. Y., Heinrichs, N., Rapee, R. M., Bogels, S. M., Schreier, S. S., et al. (2010). Social anxiety and social norms in individualistic and collectivistic countries. Depression and Anxiety, 27(12), 1128e1134.

Alfano, C. A., Beidel, D. C., & Turner, S. M. (2006). Cognitive correlates of social phobia among children and adolescents. Journal of Abnormal Child Psychology, 34(2), 189e201.

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: American Psychiatric Association.

Bandelow, B., Torrente, A. C., Wedekind, D., Broocks, A., Hajak, G., & Ruther, E. (2004). Early traumatic life events, parental rearing styles, family history of mental disorders, and birth risk factors in patients with social anxiety disorder. European Archives of Psychiatry and Clinical Neuroscience, 254(6), 397e405.

Barker, T. V., Reeb-Sutherland, B., Degnan, K. A., Walker, O. L., Chronis-Tuscano, A., Henderson, H. A., et al. (2015). Contextual startle responses moderate the relation between behavioral inhibition and anxiety in middle childhood. Psy- chophysiology, 52(11), 1544e1549.

Beesdo-Baum, K., Knappe, S., Fehm, L., Hofler, M., Lieb, R., Hofmann, S., et al. (2012). The natural course of social anxiety disorder among adolescents and young adults. Acta Psychiatrica Scandinavica, 126(6), 411e425.

Beesdo, K., Bittner, A., Pine, D. S., Stein, M. B., Hofler, M., Lieb, R., et al. (2007). Incidence of social anxiety disorder and the consistent risk for secondary depression in the first three decades of life. Archives of General Psychiatry, 64(8), 903e912.

Beidel, D. C., & Turner, S. M. (2007). Clinical presentation of social anxiety disorder in children and adolescents. Beidel, Deborah C; Turner, Samuel M (2007) Shy children, phobic adults: Nature and treatment of social anxiety disorders (2nd ed) (pp. 47e80) xiii, 398 pp (pp. 47e80). Washington, DC, US: American Psy- chological Association; US.

Beidel, D. C., Turner, S. M., & Morris, T. L. (1999). Psychopathology of childhood social phobia. Journal of the American Academy of Child and Adolescent Psychi- atry, 38(6), 643e650.

Beidel, D. C., Turner, S. M., & Morris, T. L. (2000). Behavioral treatment of childhood social phobia. Journal of Consulting and Clinical Psychology, 68(6), 1072e1080.

Bishop, M., Rosenstein, D., Bakelaar, S., & Seedat, S. (2014). An analysis of early developmental trauma in social anxiety disorder and posttraumatic stress disorder. Annals of General Psychiatry, 13. May 2014, ArtID 16, 13.

Black, J. J., Clark, D. B., Martin, C. S., Kim, K. H., Blaze, T. J., Creswell, K. G., et al. (2015). Course of alcohol symptoms and social anxiety disorder from adolescence to young adulthood. Alcoholism: Clinical and Experimental Research, 39(6), 1008e1015.

Blote, A. W., Bokhorst, C. L., Miers, A. C., & Westenberg, P. (2012). Why are socially anxious adolescents rejected by peers? The role of subject-group similarity characteristics. Journal of Research on Adolescence, 22(1), 123e134.

Blote, A. W., Duvekot, J., Schalk, R. D., Tuinenburg, E. M., & Westenberg, P. (2010). Nervousness and performance characteristics as predictors of peer behavior towards socially anxious adolescents. Journal of Youth and Adolescence, 39(12), 1498e1507.

Blote, A. W., Kint, M. J., & Westenberg, P. (2007). Peer behavior toward socially anxious adolescents: Classroom observations. Behaviour Research and Therapy, 45(11), 2773e2779.

Blote, A. W., Miers, A. C., Heyne, D. A., Clark, D. M., & Westenberg, P. (2014). The relation between social anxiety and audience perception: Examining Clark and Wells’ (1995) model among adolescents. Behavioural and Cognitive Psychother- apy, 42(5), 555e567.

Blote, A. W., Miers, A. C., Heyne, D. A., & Westenberg, P. (2015). Social anxiety and the school environment of adolescents. In K. Ranta, A. M. La Greca, L.-J. Garcia- Lopez, & M. Marttunen (Eds.), Social anxiety and phobia in adolescents: Devel- opment, manifestation and intervention strategies (pp 151e181) xi, 325 pp Cham, Switzerland (pp. 151e181). Switzerland: Springer International Publishing.

Blote, A. W., Miers, A. C., & Westenberg, P. (2015). The role of social performance and physical attractiveness in peer rejection of socially anxious adolescents. Journal of Research on Adolescence, 25(1), 189e200.

Blote, A. W., & Westenberg, P. (2007). Socially anxious adolescents’ perception of treatment by classmates. Behaviour Research and Therapy, 45(2), 189e198.

B€ogels, S. M., & Brechman-Toussaint, M. L. (2006). Family issues in child anxiety: Attachment, family functioning, parental rearing and beliefs. Clinical Psychology Review, 26(7), 834e856.

B€ogels, S. M., & Perotti, E. C. (2011). Does father know best? A formal model of the paternal influence on childhood social anxiety. Journal of Child and Family Studies, 20(2), 171e181.

Brockveld, K. C., Perini, S. J., & Rapee, R. M. (2014). Chapter 6-social anxiety and social anxiety disorder across cultures social anxiety (3rd ed., pp. 141e158). San Diego: Academic Press.

Bruce, L. C., Heimberg, R. G., Blanco, C., Schneier, F. R., & Liebowitz, M. R. (2012). Childhood maltreatment and social anxiety disorder: Implications for symptom severity and response to pharmacotherapy. Depression and Anxiety, 29(2), 131e138.

Bruce, S. E., Yonkers, K. A., Otto, M. W., Eisen, J. L., Weisberg, R. B., Pagano, M., et al. (2005). Influence of psychiatric comorbidity on recovery and recurrence in generalized anxiety disorder, social phobia, and panic disorder: A 12-year prospective study. The American Journal of Psychiatry, 162(6), 1179e1187.

Buckner, J. D., Schmidt, N. B., Lang, A. R., Small, J. W., Schlauch, R. C., &

Lewinsohn, P. M. (2008). Specificity of social anxiety disorder as a risk factor for alcohol and cannabis dependence. Journal of Psychiatric Research, 42(3), 230e239.

Buckner, J. D., & Turner, R. J. (2009). Social anxiety disorder as a risk factor for alcohol use disorders: A prospective examination of parental and peer in- fluences. Drug and Alcohol Dependence, 100(1), 128e137.

Burkhouse, K. L., Gibb, B. E., Coles, M. E., Knopik, V. S., & McGeary, J. E. (2011). Se- rotonin transporter genotype moderates the link between children’s reports of overprotective parenting and their behavioral inhibition. Journal of Abnormal Child Psychology, 39(6), 783e790.

Burstein, M., He, J.-P., Kattan, G., Albano, A. M., Avenevoli, S., & Merikangas, K. R. (2011). Social phobia and subtypes in the National Comorbidity Survey- adolescent supplement: Prevalence, correlates, and comorbidity. Journal of the American Academy of Child & Adolescent Psychiatry, 50(9), 870e880.

Calvete, E. (2014). Emotional abuse as a predictor of early maladaptive schemas in adolescents: Contributions to the development of depressive and social anxiety symptoms. Child Abuse & Neglect, 38(4), 735e746.

Canino, G., Shrout, P. E., Rubio-Stipec, M., Bird, H. R., Bravo, M., Ramirez, R., et al. (2004). The DSM-IV rates of child and adolescent disorders in Puerto Rico: Prevalence, correlates, service use, and the effects of impairment. Archives of General Psychiatry, 61(1), 85e93.

Caouette, J. D., & Guyer, A. E. (2014). Gaining insight into adolescent vulnerability for social anxiety from developmental cognitive neuroscience. Developmental Cognitive Neuroscience, 8, 65e76.

Cartwright-Hatton, S., Hodges, L., & Porter, J. (2003). Social anxiety in childhood: The relationship with self and observer rated social skills. Journal of Child Psy- chology and Psychiatry, 44(5), 737e742.

Chen, X., Chen, H., Li, D., & Wang, L. (2009). Early childhood behavioral inhibition and social and school adjustment in Chinese children: A 5-Year longitudinal study. Child Development, 80(6), 1692e1704.

Chen, X., DeSouza, A. T., Chen, H., & Wang, L. (2006). Reticent behavior and expe- riences in peer interactions in Chinese and Canadian children. Developmental Psychology, 42(4), 656e665.

Chen, X., Hastings, P. D., Rubin, K. H., Chen, H., Cen, G., & Stewart, S. L. (1998). Child- rearing attitudes and behavioral inhibition in Chinese and Canadian toddlers: A cross-cultural study. Developmental Psychology, 34(4), 677e686.

Chen, X., & Tse, H. C.-H. (2008). Social functioning and adjustment in Canadian-born children with Chinese and European backgrounds. Developmental Psychology, 44(4), 1184e1189.

Chronis-Tuscano, A., Degnan, K. A., Pine, D. S., Perez-Edgar, K., Henderson, H. A., Diaz, Y., et al. (2009). Stable early maternal report of behavioral inhibition predicts lifetime social anxiety disorder in adolescence. Journal of the American Academy of Child & Adolescent Psychiatry, 48(9), 928e935.

Clark, D. M., & Wells, A. (1995). A cognitive model of social phobia. In R. G. Heimberg, & M. R. Liebowitz (Eds.), Social phobia: Diagnosis, assessment, and treatment (pp. 69e93). New York, NY, US: Guilford Press.

Clauss, J. A., & Blackford, J. U. (2012). Behavioral inhibition and risk for developing social anxiety disorder: A meta-analytic study. Journal of the American Academy of Child and Adolescent Psychiatry, 51(10), 1066e1075.e1061.

Coplan, R. J., & Armer, M. (2005). Talking yourself out of being shy: Shyness, expressive vocabulary, and socioemotional adjustment in preschool. Merrill- Palmer Quarterly (1982), 51(1), 20e41.

Cougle, J. R., Timpano, K. R., Sachs-Ericsson, N., Keough, M. E., & Riccardi, C. J. (2010). Examining the unique relationships between anxiety disorders and childhood physical and sexual abuse in the National Comorbidity Survey-replication. Psychiatry Research, 177(1e2), 150e155.

Crawford, A. M., & Manassis, K. (2011). Anxiety, social skills, friendship quality, and peer victimization: An integrated model. Journal of Anxiety Disorders, 25(7), 924e931.

Dadds, M. R., & Roth, J. H. (2001). Family processes in the development of anxiety problems. In M. W. Vasey, & M. R. Dadds (Eds.), The developmental psychopa- thology of anxiety (pp. 278e303). Oxford University Press.

Del Piero, L. B., Saxbe, D. E., & Margolin, G. (2016). Basic emotion processing and the adolescent brain: Task demands, analytic approaches, and trajectories of changes. Developmental Cognitive Neuroscience, 19, 174e189.

Detweiler, M. F., Comer, J. S., Crum, K. I., & Albano, A. M. (2014). Social anxiety in children and adolescents: Biological, developmental, and social considerations. In S. G. Hofmann, & P. M. DiBartolo (Eds.), Social anxiety: Clinical, developmental, and social perspectives (3rd ed) (pp 253e309) xxvi, 818 pp San Diego, CA, US (pp. 253e309). US: Elsevier Academic Press.

Edwards, S. L., Rapee, R. M., & Kennedy, S. (2010). Prediction of anxiety symptoms in preschool-aged children: Examination of maternal and paternal perspectives. Journal of Child Psychology & Psychiatry, 51(3), 313e321.

Eley, T. C., Rijsdijk, F. V., Perrin, S., O’Connor, T. G., & Bolton, D. (2008). A multivariate genetic analysis of specific phobia, separation anxiety and social phobia in early childhood. Journal of Abnormal Child Psychology, 36(6), 839e848.

Elizabeth, J., King, N., Ollendick, T. H., Gullone, E., Tonge, B., Watson, S., et al. (2006). Social anxiety disorder in children and youth: A research update on aetiological factors. Counselling Psychology Quarterly, 19(2), 151e163.

Erath, S. A., Flanagan, K. S., & Bierman, K. L. (2007). Social anxiety and peer relations in early adolescence: Behavioral and cognitive factors. Journal of Abnormal Child Psychology, 35(3), 405e416.

Ford, T., Goodman, R., & Meltzer, H. (2003). The British child and adolescent mental health survey 1999: The prevalence of DSM-IV disorders. Journal of the Amer- ican Academy of Child and Adolescent Psychiatry, 42(10), 1203.

S.H. Spence, R.M. Rapee / Behaviour Research and Therapy 86 (2016) 50e67 65

Fox, N. A., Henderson, H. A., Marshall, P. J., Nichols, K. E., & Ghera, M. M. (2005). Behavioral inhibition: Linking biology and behavior within a developmental framework. Annual Review of Psychology, 56(1), 235e262.

Fox, A. S., & Kalin, N. H. (2014). A translational neuroscience approach to under- standing the development of social anxiety disorder and its pathophysiology. The American Journal of Psychiatry, 171(11), 1162e1173.

Fox, N. A., Nichols, K. E., Henderson, H. A., Rubin, K., Schmidt, L., Hamer, D., et al. (2005). Evidence for a gene-environment interaction in predicting behavioral inhibition in middle childhood. Psychological Science, 16(12), 921e926.

Frenkel, T. I., Fox, N. A., Pine, D. S., Walker, O. L., Degnan, K. A., & Chronis-Tuscano, A. (2015). Early childhood behavioral inhibition, adult psychopathology and the buffering effects of adolescent social networks: A twenty-year prospective study. Journal of Child Psychology and Psychiatry, 56(10), 1065e1073.

Gilboa-Schechtman, E., Shachar, I., & Helpman, L. (2014). Evolutionary perspective on social anxiety. In S. G. Hofmann, & P. M. DiBartolo (Eds.), Social anxiety: Clinical, developmental, and social perspectives (3rd ed., pp. 599e614). New York: Academic Press.

Greco, L. A., & Morris, T. L. (2005). Factors influencing the link between social anxiety and peer acceptance: Contributions of social skills and close friendships during middle childhood. Behavior Therapy, 36(2), 197e205.

Gren-Landell, M., Aho, N., Andersson, G., & Svedin, C. G. (2011). Social anxiety disorder and victimization in a community sample of adolescents. Journal of Adolescence, 34(3), 569e577.

Guyer, A. E., Choate, V. R., Detloff, A., Benson, B., Nelson, E. E., Perez-Edgar, K., et al. (2012). Striatal functional alteration during incentive anticipation in pediatric anxiety disorders. The American Journal of Psychiatry, 169(2), 205e212.

Guyer, A. E., Jarcho, J. M., P�erez-Edgar, K., Degnan, K. A., Pine, D. S., Fox, N. A., et al. (2015). Temperament and parenting styles in early childhood differentially influence neural response to peer evaluation in adolescence. Journal of Abnormal Child Psychology, 43(5), 863e874.

Hamm, L. L., Jacobs, R. H., Johnson, M. W., Fitzgerald, D. A., Fitzgerald, K. D., Langenecker, S. A., et al. (2014). Aberrant amygdala functional connectivity at rest in pediatric anxiety disorders. Biology of Mood & Anxiety Disorders, 4(1), 15e15.

Hattingh, C. J., Ipser, J., Tromp, S. A., Syal, S., Lochner, C., Brooks, S. J., et al. (2012). Functional magnetic resonance imaging during emotion recognition in social anxiety disorder: An activation likelihood meta-analysis. Frontiers in Human Neuroscience, 6, 347.

Heimberg, R. G., Brozovich, F. A., & Rapee, R. M. (2010). A cognitive behavioral model of social anxiety disorder: Update and extension. In S. G. Hofmann, & P. M. DiBartolo (Eds.), Social anxiety (2nd ed., pp. 395e422). San Diego: Aca- demic Press.

Heinrichs, N., Rapee, R. M., Alden, L. A., B€ogels, S., Hofmann, S. G., Ja Oh, K., et al. (2006). Cultural differences in perceived social norms and social anxiety. Behaviour Research and Therapy, 44(8), 1187e1197.

Henderson, H. A., Pine, D. S., & Fox, N. A. (2015). Behavioral inhibition and devel- opmental risk: A dual-processing perspective. Neuropsychopharmacology, 40(1), 207e224.

Hirshfeld-Becker, D. R., Biederman, J., Henin, A., Faraone, S. V., Davis, S., Harrington, K., et al. (2007). Behavioral inhibition in preschool children at risk is a specific predictor of middle childhood social anxiety: A five-year follow-up. Journal of Developmental and Behavioral Pediatrics, 28(3), 225e233.

Hirshfeld-Becker, D. R., Micco, J. A., Wang, C. H., & Henin, A. (2014). Behavioral in- hibition: A discrete precursor to social anxiety disorder? In J. W. Weeks (Ed.), The Wiley Blackwell handbook of social anxiety disorder (pp 133e158) xx, 621 pp (pp. 133e158). Wiley-Blackwell.

Hodson, K. J., McManus, F. V., Clark, D. M., & Doll, H. (2008). Can Clark and Wells’ (1995) cognitive model of social phobia be applied to young people. Behavioural and Cognitive Psychotherapy, 36(4), 449e461.

Hudson, J. L., Keers, R., Roberts, S., Coleman, J. R. I., Breen, G., Arendt, K., et al. (2015). Clinical predictors of response to cognitive-behavioral therapy in pediatric anxiety disorders: The genes for treatment (GxE) study. Journal of the American Academy of Child & Adolescent Psychiatry, 54(6), 454e463.

Hudson, J. L., Rapee, R. M., Lyneham, H. J., McLellan, L. F., Wuthrich, V. M., & Schniering, C. A. (2015). Comparing outcomes for children with different anx- iety disorders following cognitive behavioural therapy. Behaviour Research and Therapy, 72, 30e37.

Inderbitzen-Nolan, H. M., Anderson, E. R., & Johnson, H. S. (2007). Subjective versus objective behavioral ratings following two analogue tasks: A comparison of socially phobic and non-anxious adolescents. Journal of Anxiety Disorders, 21(1), 76e90.

Jin, R., Kessler, R. C., Walters, E. E., Demler, O., Merikangas, K. R., & Berglund, P. (2005). Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey replication. Archives of General Psychiatry, 62(6), 593e602.

Kagan, J., Reznick, J. S., Clarke, C., Snidman, N., & Garcia-Coll, C. (1984). Behavioral inhibition to the unfamiliar. Child Development, 55(6), 2212e2225.

Kennedy, S. J., Rapee, R. M., & Edwards, S. L. (2009). A selective intervention pro- gram for inhibited preschool-aged children of parents with an anxiety disorder: Effects on current anxiety disorders and temperament. Journal of the American Academy of Child and Adolescent Psychiatry, 48(6), 602.

Kessler, R. C., Avenevoli, S., Costello, E., Georgiades, K., Green, J. G., Gruber, M. J., et al. (2012). Prevalence, persistence, and sociodemographic correlates of DSM-IV disorders in the National Comorbidity Survey replication adolescent supple- ment. Archives of General Psychiatry, 69(4), 372e380.

Kiel, E. J., & Buss, K. A. (2011). Prospective relations among fearful temperament, protective parenting, and social withdrawal: The role of maternal accuracy in a moderated mediation framework. Journal of Abnormal Child Psychology, 39(7), 953e966.

Kimbrel, N. A. (2008). A model of the development and maintenance of generalized social phobia. Clinical Psychology Review, 28(4), 592e612.

Kim, J., Rapee, R. M., & Gaston, J. E. (2008). Symptoms of offensive type Taijin- kyofusho among Australian social phobics. Depression and Anxiety, 25, 601e608.

Klein, A. M., Rapee, R. M., Hudson, J. L., Schniering, C. A., Wuthrich, V. M., Kangas, M., et al. (2015). Interpretation modification training reduces social anxiety in clinically anxious children. Behaviour Research and Therapy, 75, 78e84.

Kley, H., Tuschen-Caffier, B., & Heinrichs, N. (2011). Manipulating self-focused attention in children with social anxiety disorder and in socially anxious and non-anxious children. Journal of Experimental Psychopathology, 2(4), 551e570.

Kley, H., Tuschen-Caffier, B., & Heinrichs, N. (2012). Safety behaviors, self-focused attention and negative thinking in children with social anxiety disorder, so- cially anxious and non-anxious children. Journal of Behavior Therapy and Experimental Psychiatry, 43(1), 548e555.

Knappe, S., Beesdo-Baum, K., Fehm, L., Lieb, R., & Wittchen, H.-U. (2012). Charac- terizing the association between parenting and adolescent social phobia. Jour- nal of Anxiety Disorders, 26(5), 608e616.

Knappe, S., Beesdo-Baum, K., Fehm, L., Stein, M. B., Lieb, R., & Wittchen, H.-U. (2011). Social fear and social phobia types among community youth: Differential clinical features and vulnerability factors. Journal of Psychiatric Research, 45(1), 111e120.

Knappe, S., Sasagawa, S., & Creswell, C. (2015). Developmental epidemiology of social anxiety and social phobia in adolescents. In K. Ranta, A. M. La Greca, L.- J. Garcia-Lopez, & M. Marttunen (Eds.), Social anxiety and phobia in adolescents: Development, manifestation and intervention strategies (pp 39e70) xi, 325 pp Cham, Switzerland (pp. 39e70). Switzerland: Springer International Publishing.

Kramer, M., Seefeldt, W. L., Heinrichs, N., Tuschen-Caffier, B., Schmitz, J., Wolf, O. T., et al. (2012). Subjective, autonomic, and endocrine reactivity during social stress in children with social phobia. Journal of Abnormal Child Psychology, 40(1), 95e104.

Kuo, J. R., Goldin, P. R., Werner, K., Heimberg, R. G., & Gross, J. J. (2011). Childhood trauma and current psychological functioning in adults with social anxiety disorder. Journal of Anxiety Disorders, 25(4), 467e473.

La Greca, A. M., & Harrison, H. M. (2005). Adolescent peer relations, friendships, and romantic relationships: Do they predict social anxiety and depression? Journal of Clinical Child & Adolescent Psychology, 34(1), 49e61.

Lawrence, D., Johnson, S., Hafekost, J., Boterhoven De Haan, K., Sawyer, M., Ainley, J., et al. (2015). The mental health of children and adolescents. Report on the second Australian Child and Adolescent Survey of Mental Health and Wellbeing. Can- berra: Department of Health.

Lewis-Morrarty, E., Degnan, K. A., Chronis-Tuscano, A., Pine, D. S., Henderson, H. A., & Fox, N. A. (2015). Infant attachment security and early childhood behavioral inhibition interact to predict adolescent social anxiety symptoms. Child Devel- opment, 86(2), 598e613.

Lewis-Morrarty, E., Degnan, K. A., Chronis-Tuscano, A., Rubin, K. H., Cheah, C. S. L., Pine, D. S., et al. (2012). Maternal over-control moderates the association be- tween early childhood behavioral inhibition and adolescent social anxiety symptoms. Journal of Abnormal Child Psychology, 40(8), 1363e1373.

Loukas, A., & Pasch, K. E. (2013). Does school connectedness buffer the impact of peer victimization on early adolescents’ subsequent adjustment problems? The Journal of Early Adolescence, 33(2), 245e266.

Luchetti, S., & Rapee, R. M. (2014). Liking and perceived probability of victimization of peers displaying behaviors characteristic of anxiety and depression. Journal of Experimental Psychopathology, 5(2), 212e223.

Majdand�zi�c, M., M€oller, E. L., B€ogels, S. M., Vente, D. W., & Boom, v.d. D. C. (2014). Fathers’ challenging parenting behavior prevents social anxiety development in their 4-year-old children: A longitudinal observational study. Journal of Abnormal Child Psychology, 42(2), 301e310.

McLaughlin, K. A., Green, J. G., Gruber, M. J., Sampson, N. A., Zaslavsky, A. M., & Kessler, R. C. (2012). Childhood adversities and first onset of psychiatric disor- ders in a national sample of US adolescents. JAMA Psychiatry, 69(11), 1151e1160.

McLeod, B. D., Wood, J. J., & Weisz, J. R. (2007). Examining the association between parenting and childhood anxiety: A meta-analysis. Clinical Psychology Review, 27(2), 155e172.

Melfsen, S., Kühnemund, M., Schwieger, J., Warnke, A., Stadler, C., Poustka, F., et al. (2011). Cognitive behavioral therapy of socially phobic children focusing on cognition: A randomised wait-list control study. Child and Adolescent Psychiatry and Mental Health, 5(1), 5e5.

Mesa, F., Le, T.-A., & Beidel, D. C. (2015). Social skill-based treatment for social anxiety disorder in adolescents. In K. Ranta, A. M. La Greca, L.-J. Garcia-Lopez, & M. Marttunen (Eds.), Social anxiety and phobia in adolescents: Development, manifestation and intervention strategies (pp. 289e299). Cham, Switzerland: Springer International Publishing.

Miers, A. C., Blote, A. W., Bogels, S. M., & Westenberg, P. (2008). Interpretation bias and social anxiety in adolescents. Journal of Anxiety Disorders, 22(8), 1462e1471.

Miers, A. C., Blote, A. W., Bokhorst, C. L., & Westenberg, P. (2009). Negative self- evaluations and the relation to performance level in socially anxious children and adolescents. Behaviour Research and Therapy, 47(12), 1043e1049.

Miers, A. C., Blote, A. W., de Rooij, M., Bokhorst, C., & Westenberg, P. (2013). Tra- jectories of social anxiety during adolescence and relations with cognition, social competence, and temperament. Journal of Abnormal Child Psychology,

S.H. Spence, R.M. Rapee / Behaviour Research and Therapy 86 (2016) 50e6766

41(1), 97e110. Miers, A. C., Blote, A. W., Heyne, D. A., & Westenberg, P. (2014). Developmental

pathways of social avoidance across adolescence: The role of social anxiety and negative cognition. Journal of Anxiety Disorders, 28(8), 787e794.

Miers, A. C., Blote, A. W., & Westenberg, P. (2010). Peer perceptions of social skills in socially anxious and nonanxious adolescents. Journal of Abnormal Child Psy- chology, 38(1), 33e41.

Miers, A. C., Blote, A. W., & Westenberg, P. (2011). Negative social cognitions in socially anxious youth: Distorted reality or a kernel of truth? Journal of Child and Family Studies, 20(2), 214e223.

Monk, C. S., Telzer, E. H., Mogg, K., Bradley, B. P., Mai, X., Louro, H. M. C., et al. (2008). Amygdala and ventrolateral prefrontal cortex activation to masked angry faces in children and adolescents with generalized anxiety disorder. Archives of General Psychiatry, 65(5), 568e576.

Muris, P., van Brakel, A. M., Arntz, A., & Schouten, E. (2011). Behavioral inhibition as a risk factor for the development of childhood anxiety disorders: A longitudinal study. Journal of Child and Family Studies, 20(2), 157e170.

Natsuaki, M. N., Leve, L. D., Neiderhiser, J. M., Shaw, D. S., Scaramella, L. V., Ge, X., et al. (2013). Intergenerational transmission of risk for social inhibition: The interplay between parental responsiveness and genetic influences. Development and Psychopathology, 25(1), 261e274.

Negreiros, J., & Miller, L. D. (2014). The role of parenting in childhood anxiety: Etiological factors and treatment implications. Clinical Psychology: Science and Practice, 21(1), 3e17.

Norton, P. J., & Price, E. C. (2007). A meta-analytic review of adult cognitive- behavioral treatment outcome across the anxiety disorders. The Journal of Nervous and Mental Disease, 195(6), 521.

Oh, K. J., Shin, Y. J., Moon, K. J., Hudson, J. L., & Rapee, R. M. (2002). Childrearing practices and psychological disorders in children: Cross cultural comparisons of Korea and Australia.

Olivares, J., Olivares-Olivares, P. J., & Macia, D. (2014). Social skills training and treatment of adolescents with generalized social phobia. Behavioral Psychology/ Psicologia Conductual: Revista Internacional Clinica y de la Salud, 22(3), 441e459.

Ollendick, T. H., & Benoit, K. E. (2012). A parent-child interactional model of social anxiety disorder in youth. Clinical Child and Family Psychology Review, 15(1), 81e91.

Ollendick, T. H., Benoit, K. E., & Grills-Taquechel, A. E. (2014). Social anxiety disorder in children and adolescents. In J. W. Weeks (Ed.), The Wiley Blackwell handbook of social anxiety disorder (pp 181e200) xx, 621 pp (pp. 181e200). Wiley- Blackwell.

Pine, D. S., Cohen, P., Gurley, D., Brook, J., & Ma, Y. (1998). The risk for early- adulthood anxiety and depressive disorders in adolescents with anxiety and depressive disorders. Archives of General Psychiatry, 55(1), 56e64.

Ranta, K., Kaltiala-Heino, R., Frojd, S., & Marttunen, M. (2013). Peer victimization and social phobia: A follow-up study among adolescents. Social Psychiatry and Psychiatric Epidemiology, 48(4), 533e544.

Ranta, K., Kaltiala-Heino, R., Pelkonen, M., & Marttunen, M. (2009). Associations between peer victimization, self-reported depression and social phobia among adolescents: The role of comorbidity. Journal of Adolescence, 32(1), 77e93.

Ranta, K., Tuomisto, M. T., Kaltiala-Heino, R., Rantanen, P., & Marttunen, M. (2014). Cognition, imagery and coping among adolescents with social anxiety and phobia: Testing the Clark and Wells model in the population. Clinical Psychology & Psychotherapy, 21(3), 252e263.

Rao, P. A., Beidel, D. C., Turner, S. M., Ammerman, R. T., Crosby, L. E., & Sallee, F. R. (2007). Social anxiety disorder in childhood and adolescence: Descriptive psychopathology. Behaviour Research and Therapy, 45(6), 1181e1191.

Rapee, R. M. (2002). The development and modification of temperamental risk for anxiety disorders: Prevention of a lifetime of anxiety? (Vol. 52, pp. 947e957). United States: Elsevier Inc.

Rapee, R. M. (2014). Preschool environment and temperament as predictors of social and nonsocial anxiety disorders in middle adolescence. Journal of the American Academy of Child and Adolescent Psychiatry, 53(3), 320.

Rapee, R. M., & Coplan, R. J. (2010). Conceptual relations between anxiety disorder and fearful temperament. New Directions for Child and Adolescent Development, 2010(127), 17e31.

Rapee, R. M., Gaston, J. E., & Abbott, M. J. (2009). Testing the efficacy of theoretically derived improvements in the treatment of social phobia. Journal of Consulting and Clinical Psychology, 77(2), 317e327.

Rapee, R. M., & Heimberg, R. G. (1997). A cognitive-behavioral model of anxiety in social phobia. Behaviour Research and Therapy, 35(8), 741e756.

Rapee, R. M., Kim, J., Wang, J., Liu, X., Hofmann, S. G., Chen, J., et al. (2011). Perceived impact of socially anxious behaviors on individuals’ lives in western and east asian countries. Behavior Therapy, 42(3), 485e492.

Rapee, R. M., & Spence, S. H. (2004). The etiology of social phobia: Empirical evi- dence and an initial model. Clinical Psychology Review, 24(7), 737e767.

Reijntjes, A., Kamphuis, J. H., Prinzie, P., & Telch, M. J. (2010). Peer victimization and internalizing problems in children: A meta-analysis of longitudinal studies. Child Abuse & Neglect, 34(4), 244e252.

de Rosnay, M., Cooper, P. J., Tsigaras, N., & Murray, L. (2006). Transmission of social anxiety from mother to infant: An experimental study using a social referencing paradigm. Behaviour Research and Therapy, 44(8), 1165e1175.

Rothbart, M. K., Sheese, B. E., Rueda, M. R., & Posner, M. I. (2011). Developing mechanisms of self-regulation in early life. Emotion Review, 3(2), 207e213.

Roy, A. K., Benson, B. E., Degnan, K. A., Perez-Edgar, K., Pine, D. S., Fox, N. A., et al. (2014). Alterations in amygdala functional connectivity reflect early

temperament. Biological Psychology, 103, 248e254. Rubin, K. H., Burgess, K. B., & Hastings, P. D. (2002). Stability and social-behavioral

consequences of toddlers’ inhibited temperament and parenting behaviors. Child Development, 73(2), 483e495.

Rubin, K. H., Coplan, R. J., & Bowker, J. C. (2009). Social withdrawal in childhood. Annual Review of Psychology, 60, 141e171.

Rubin, K. H., Nelson, L. J., Hastings, P., & Asendorpf, J. (1999). The transaction be- tween parents’ perceptions of their children’s shyness and their parenting styles. International Journal of Behavioral Development, 23(4), 937e957.

Scaini, S., Belotti, R., & Ogliari, A. (2014). Genetic and environmental contributions to social anxiety across different ages: A meta-analytic approach to twin data. Journal of Anxiety Disorders, 28(7), 650e656.

Schmitz, J., Kramer, M., Blechert, J., & Tuschen-Caffier, B. (2010). Post-event pro- cessing in children with social phobia. Journal of Abnormal Child Psychology, 38(7), 911e919.

Schmitz, J., Kramer, M., Tuschen-Caffier, B., Heinrichs, N., & Blechert, J. (2011). Restricted autonomic flexibility in children with social phobia. Journal of Child Psychology and Psychiatry, 52(11), 1203e1211.

Schreiber, F., Hofling, V., Stangier, U., Bohn, C., & Steil, R. (2012). A cognitive model of social phobia: Applicability in a large adolescent sample. International Journal of Cognitive Therapy, 5(3), 341e358.

Schreiber, F., & Steil, R. (2013). Haunting self-images? The role of negative self- images in adolescent social anxiety disorder. Journal of Behavior Therapy and Experimental Psychiatry, 44(2), 158e164.

Siegel, R. S., La Greca, A. M., & Harrison, H. M. (2009). Peer victimization and social anxiety in adolescents: Prospective and reciprocal relationships. Journal of Youth and Adolescence, 38(8), 1096e1109.

Simon, N. M., Herlands, N. N., Marks, E. H., Mancini, C., Letamendi, A., Li, Z., et al. (2009). Childhood maltreatment linked to greater severity and poorer quality of life and function in social anxiety disorder. Depression and Anxiety, 26(11), 1027e1032.

Spence, S. H., Donovan, C., & Brechman Toussaint, M. (1999). Social skills, social outcomes, and cognitive features of childhood social phobia. Journal of Abnormal Psychology, 108(2), 211e221.

Spence, S. H., Donovan, C., & Brechman-Toussaint, M. (2000). The treatment of childhood social phobia: The effectiveness of a social skills training-based, cognitive-behavioural intervention, with and without parental involvement. Journal of Child Psychology and Psychiatry, 41(6), 713e726.

Spinhoven, P., Elzinga, B. M., Hovens, J. G., Roelofs, K., Zitman, F. G., van Oppen, P., et al. (2010). The specificity of childhood adversities and negative life events across the life span to anxiety and depressive disorders. Journal of Affective Disorders, 126(1e2), 103e112.

Stein, M. B., & Stein, D. J. (2008). Social anxiety disorder. The Lancet, 371(9618), 1115e1125.

Storch, E. A., Masia-Warner, C., Crisp, H., & Klein, R. G. (2005). Peer victimization and social anxiety in adolescence: A prospective study. Aggressive Behavior, 31(5), 437e452.

Sweeting, H., Young, R., West, P., & Der, G. (2006). Peer victimization and depression in early-mid adolescence: A longitudinal study. The British Journal of Educational Psychology, 76(Pt 3), 577.

Tillfors, M., Persson, S., Will�en, M., & Burk, W. J. (2012). Prospective links between social anxiety and adolescent peer relations. Journal of Adolescence, 35(5), 1255e1263.

Tuschen-Caffier, B., Kuhl, S., & Bender, C. (2011). Cognitive-evaluative features of childhood social anxiety in a performance task. Journal of Behavior Therapy and Experimental Psychiatry, 42(2), 233e239.

Van Der Bruggen, C. O., Stams, G. J. J. M., & B€ogels, S. M. (2008). Research review: The relation between child and parent anxiety and parental control: A meta- analytic review. Journal of Child Psychology and Psychiatry, 49(12), 1257e1269.

Vassilopoulos, S. P., Banerjee, R., & Prantzalou, C. (2009). Experimental modification of interpretation bias in socially anxious children: Changes in interpretation, anticipated interpersonal anxiety, and social anxiety symptoms. Behaviour Research and Therapy, 47(12), 1085e1089.

Verduin, T. L., & Kendall, P. C. (2008). Peer perceptions and liking of children with anxiety disorders. Journal of Abnormal Child Psychology, 36(4), 459e469.

White, L. K., McDermott, J. M., Degnan, K. A., Henderson, H. A., & Fox, N. A. (2011). Behavioral inhibition and anxiety: The moderating roles of inhibitory control and attention shifting. Journal of Abnormal Child Psychology, 39(5), 735e747.

de Wilde, A., & Rapee, R. M. (2008). Do controlling maternal behaviours increase state anxiety in children’s responses to a social threat? A pilot study. Journal of Behavior Therapy and Experimental Psychiatry, 39(4), 526e537.

Williams, L. R., Degnan, K. A., Perez-Edgar, K. E., Henderson, H. A., Rubin, K. H., Pine, D. S., et al. (2009). Impact of behavioral inhibition and parenting style on internalizing and externalizing problems from early childhood through adolescence. Journal of Abnormal Child Psychology, 37(8), 1063e1075.

Wittchen, H.-U., & Fehm, L. (2003). Epidemiology and natural course of social fears and social phobia. Acta Psychiatrica Scandinavica, 108(Suppl. 417), 4e18.

Wittchen, H.-U., Stein, M., & Kessler, R. (1999). Social fears and social phobia in a community sample of adolescents and young adults: Prevalence, risk factors and co-morbidity. Psychological Medicine, 29(2), 309e323.

Wong, Q. J. J., & Rapee, R. M. (2016). The aetiology and maintenance of social anxiety disorder: A synthesis of complimentary theoretical models and formulation of a new integrated model. Journal of Affective Disorders, 203, 84e100.

Yap, M. B. H., Pilkington, P. D., Ryan, S. M., Kelly, C. M., & Jorm, A. F. (2014). Parenting strategies for reducing the risk of adolescent depression and anxiety disorders:

S.H. Spence, R.M. Rapee / Behaviour Research and Therapy 86 (2016) 50e67 67

A Delphi consensus study. Journal of Affective Disorders, 156, 67. Yokoyama, C., Kaiya, H., Kumano, H., Kinou, M., Umekage, T., Yasuda, S., et al. (2015).

Dysfunction of ventrolateral prefrontal cortex underlying social anxiety disor- der: A multi-channel NIRS study. NeuroImage: Clinical, 8, 455e461.

Zwierzynska, K., Wolke, D., & Lereya, T. S. (2013). Peer victimization in childhood and internalizing problems in adolescence: A prospective longitudinal study. Journal of Abnormal Child Psychology, 41(2), 309e323.