The Aetiology of Social Anxiety Disorder
Social anxiety disorder (SAD) involves persistent and excessive fear of negative evaluation or judgement in social situations (Diagnostic and Statistical Manual of Mental Disorders [5th ed.; DSM–5]; American Psychiatric Association [APA], 2013). In interpersonal and/or performance-based social situations, an individual fears that s/he will behave in a way that will be humiliating or embarrassing; hence, such situations are avoided or endured with intense anxiety. SAD is associated with difficulties across a range of life domains, and it has been proposed that the disorder may lie on a number of spectrums, including social fear and avoidance, body-focused concerns, affective dysfunction, and social deficits (Stein, Ono, Tajima, & Muller, Reference Stein, Ono, Tajima and Muller2004). SAD typically has an early onset (median = 13 years of age in Australia) and chronic course (McEvoy, Grove, & Slade, Reference McEvoy, Grove and Slade2011; Ruscio et al., Reference Ruscio, Brown, Chiu, Sareen, Stein and Kessler2008). In Australia, SAD has 12-month and lifetime prevalence rates of 4.2% and 8.4% respectively, and close to 70% of individuals with SAD have experienced a comorbid disorder during their lifetime (Crome et al., Reference Crome, Grove, Baillie, Sunderland, Teesson and Slade2015). SAD is associated with considerable suffering and functional impairment (e.g., relationships, employment), yet the majority of people with SAD do not seek treatment (Crome et al., Reference Crome, Grove, Baillie, Sunderland, Teesson and Slade2015).
No clear pathway has been identified in the development of SAD; however, a range of factors have been proposed to increase risk of developing the disorder. Genetic, neurobiological, and temperamental factors have been implicated in the development of SAD (e.g., Kimbrel, Reference Kimbrel2008; Tillfors, Reference Tillfors2004), with a recent meta-analysis of twin studies reporting heritability estimates of 41% across age groups (Scaini, Belotti, & Ogliari, Reference Scaini, Belotti and Ogliari2014), indicating that shared emotional vulnerability is significant in disorder development. However, environmental factors also appear to account for a substantial proportion of variance (54% for non-shared environmental factors across age groups; Scaini et al., Reference Scaini, Belotti and Ogliari2014). Models of the aetiology of SAD propose that key environmental influences on the development of SAD include parenting factors, traumatic life events, and aversive social experiences (Beidel & Turner, Reference Beidel and Turner2007; Elizabeth et al., Reference Elizabeth, King, Ollendick, Gullone, Tonge, Watson and Macdermott2006; Hudson & Rapee, Reference Hudson and Rapee2000; Lampe, Reference Lampe2009; Neal & Edelmann, Reference Neal and Edelmann2003; Ollendick & Hirshfeld-Becker, Reference Ollendick and Hirshfeld Becker2002; Rapee & Spence, Reference Rapee and Spence2004; Wong & Rapee, Reference Wong, Rapee, Ranta, La Greca and Garcia-Lopz2015). Moreover, given that SAD typically develops in late childhood and early adolescence, the nature of early environmental factors is likely to be critical in understanding of the aetiology of the disorder and facilitating the development of more effective treatments (Wong & Rapee, Reference Wong, Rapee, Ranta, La Greca and Garcia-Lopz2015).
Aims and Methodology
Given the theoretical and empirical significance of environmental factors in the aetiology of SAD, this paper aimed to review empirical evidence for the role of parenting factors, traumatic life events, and aversive social experiences in the development of SAD, including limitations of current research and suggested future directions. A comprehensive literature search was conducted among articles indexed in the PsycInfo, Medline, and PubMed. The keywords employed included: social anxiety, SAD, social phobia, aetiology, etiology, development, model, environmental, parenting, modelling, parental style, attachment, trauma, abuse, maltreatment, neglect, negative experiences, aversive experiences, emotional abuse, physical abuse, sexual abuse, social trauma, peers, victimisation, bullying, imagery, posttraumatic symptoms, and shame. Reference lists of relevant articles were also closely examined for additional papers.
Parenting Factors
There is evidence to suggest that parent-child interaction may be a predisposing factor that influences risk of developing SAD (Higa-McMillan & Ebesutani, Reference Higa-McMillan, Ebesutani, Alfano and Beidel2011). While a recent meta-analysis reported that parenting accounted for only 4% of the variance in child anxiety (McLeod, Wood, & Weisz, Reference McLeod, Wood and Weisz2007), this review was not specific to SAD. Moreover, the early age of disorder onset (McEvoy et al., Reference McEvoy, Grove and Slade2011) suggests that parental factors may play a larger role in the development of SAD relative to other anxiety disorders (Bogels, van Oosten, Muris, & Mulders, Reference Bögels, van Oosten, Muris and Smulders2001). Ollendick and Hirshfeld-Becker (Reference Ollendick and Hirshfeld Becker2002) propose that key parental influences on the development of SAD include behavioural modelling, parenting style, and attachment quality. While it appears that such factors are significant in the development of SAD, it remains unclear whether these parental behaviours occur in response to a child's temperament or anxious behaviours, are causal antecedents of social anxiety, or some interaction of the two (Wong & Rapee, Reference Wong, Rapee, Ranta, La Greca and Garcia-Lopz2015). It is likely that these factors have a complex and reciprocal relationship, such that a child's anxious or inhibited behaviour may interact with parental factors in a way that amplifies socially anxious affective, cognitive, and behavioural patterns.
Behavioural Modelling
Observational learning and information transfer from parent to child regarding social concerns and social avoidance have been proposed to contribute to the development of SAD (Hudson & Rapee, Reference Hudson and Rapee2000). Parents of SAD children may model social concerns, express fears of negative evaluation, and catastrophise potential negative social outcomes. Such modelling may build a perception of social threat and an expectation that others may judge them negatively. Consistent with this theory, SAD individuals report that their parents tended to highlight the importance of other's opinions and emphasise the importance of appearance, (Bruch, Reference Bruch1989; Bruch & Heimberg, Reference Bruch and Heimberg1994). However, there are no known observational studies to date that have documented specific modelling of social concerns among parents of SAD individuals.
Furthermore, SAD individuals are more likely to have anxious parents, who tend to be less socially active, more inhibited, and less engaged with their children (Bandelow et al., Reference Bandelow, Torrente, Wedekind, Broocks, Hajak and Rüther2004; Festa & Ginsburg, Reference Festa and Ginsburg2011; Knappe, Beesdo et al., Reference Knappe, Beesdo, Fehm, Hofler, Leib and Wittchen2009; Woodruff-Borden, Morrow, Bourland, & Cambron, Reference Woodruff-Borden, Morrow, Bourland and Cambron2014). In this way, anxious parents may model social avoidance, restrict or prevent social engagement or opportunities for the development of social skills, and thus perpetuate a cycle of social fear (e.g., Bruch & Heimberg, Reference Bruch and Heimberg1994; Caster, Inderbitzen, & Hope, Reference Caster, Inderbitzen and Hope1999).
A range of evidence suggests that SAD parents are more likely to model avoidant behaviour. For example, adults with SAD frequently report that their parents were more socially isolated and discouraged sociability both inside and outside of the family (Anhalt & Morris, Reference Anhalt and Morris2008; Bruch, Reference Bruch1989; Bruch & Heimberg, Reference Bruch and Heimberg1994; Festa & Ginsburg, Reference Festa and Ginsburg2011; Rapee & Melville, Reference Rapee and Melville1997). Moreover, mothers of SAD adults report that they were more controlling of the socialisation experiences of their children (Rapee & Melville, Reference Rapee and Melville1997). Such modelling was demonstrated in a study by Barrett, Rapee, Dadds, and Ryan (Reference Barrett, Rapee, Dadds and Ryan1996), who found that anxious children were more likely to interpret ambiguous scenarios as threatening, and to adopt more avoidant coping strategies after discussion with their parents compared to non-anxious children (Barrett et al., Reference Barrett, Rapee, Dadds and Ryan1996). The authors suggest that families may provide the context in which children learn to interpret and respond to situations as threatening, and that anxious cognitive and behavioural patterns may be modelled, supported, and reinforced by parents of anxious children.
Parenting Style
Parental over-control and over-protection have demonstrated a moderately consistent association with the development of anxiety disorders (Rapee & Spence, Reference Rapee and Spence2004). Adults with SAD tend to perceive their parents as being over-controlling and restricting of their autonomy (e.g., Arrindell, Emmelkamp, Monsma, & Brilman, Reference Arrindell, Emmelkamp, Monsma and Brilman1983; Arrindell et al., Reference Arrindell, Kwee, Methorst, van der Ende, Pol and Moritz1989; Bandelow et al., Reference Bandelow, Torrente, Wedekind, Broocks, Hajak and Rüther2004; Festa & Ginsburg, Reference Festa and Ginsburg2011; Gulley, Oppenheimer, & Hankin, Reference Gulley, Oppenheimer and Hankin2014; Rapee & Melville, Reference Rapee and Melville1997). While many of these studies have utilised retrospective self-report, there is some behavioural data to support these findings. For example, in further analyses of the study by Barrett and colleagues (Reference Barrett, Rapee, Dadds and Ryan1996), the authors found that parents of anxious children listened and agreed less with their child, indicating less support for autonomous thought and action (Dadds, Barrett, Rapee, & Ryan, Reference Dadds, Barrett, Rapee and Ryan1996). Moreover, Hudson and colleagues observed that mothers of anxious children were more intrusive and involved during a collaborative task or discussion of experiences involving negative affective states (Hudson & Rapee, Reference Hudson and Rapee2001, Reference Hudson and Rapee2002; Hudson, Comer & Kendall, Reference Hudson, Comer and Kendall2008). Specific to social anxiety, two observational studies have found that parents of socially anxious children displayed more controlling behaviours during a challenging task with their children (Greco & Morris, Reference Greco and Morris2002; Rork & Morris, Reference Rork and Morris2009). Importantly, this relationship has been supported by a longitudinal study in which higher observed maternal over-control at 7 years of age predicted increased symptoms of social anxiety and higher rates of SAD diagnosis during adolescence (Lewis-Morrarty et al., Reference Lewis-Morrarty, Degnan, Chronis-Tuscano, Rubin, Cheah, Pine, Henderon and Fox2012). Closely related to over-control, SAD individuals frequently report that their parents were over-protective (e.g., Arrindell et al., Reference Arrindell, Emmelkamp, Monsma and Brilman1983; Knappe, Lieb et al., Reference Knappe, Lieb, Beesdo, Fehm, Low, Gloster and Wittchen2009; Knappe, Beesdo-Baum, Fehm, Leib, & Wittchen, Reference Knappe, Beeso-Baum, Fehm, Leib and Wittchen2012; Parker, Reference Parker1979; Spokas & Heimberg, Reference Spokas and Heimberg2009; Taylor & Alden, Reference Taylor and Alden2006). Indeed, an over-protective parental style has also been associated with increased risk for the development of SAD among adolescents who were still living with their parents (Lieb et al., Reference Lieb, Wittchen, Hoefler, Fuetsch, Stein and Merikangas2000).
Over-controlling and over-protective parenting may restrict the development of self-efficacy and social autonomy, in addition to reducing opportunities for the development of social skills. Such parenting also communicates to a child that they lack the capacity to cope with challenging situations and require protection from a dangerous world (Hudson & Rapee, Reference Hudson and Rapee2000). In contrast, challenging paternal behaviour (in which a child is playfully encouraged to push their limits) appears to protect against the development of SAD in 4-year-olds (Majdandzic, Moller, de Vente, Bogels, & van den Boom, Reference Majdandzic, Moller, de Vente, Bogels and van den Boom2014). However, not all studies have found a link between SAD and over-controlling or over-protective parenting, suggesting that this is a possible but not necessary factor in the development of SAD (e.g., Arkinsola & Udoka, Reference Arkinsola and Udoka2013; Bogels et al., Reference Bögels, van Oosten, Muris and Smulders2001; Vreeke, Muris, Mayer, Huijding, & Rapee, Reference Vreeke, Muris, Mayer, Huijding and Rapee2013).
In addition, some SAD individuals recall their parents being less warm, as well as more rejecting, critical, and shaming, in comparison to healthy controls (Arrindel et al., Reference Arrindell, Emmelkamp, Monsma and Brilman1983; Bandelow et al., Reference Bandelow, Torrente, Wedekind, Broocks, Hajak and Rüther2004; Bruch & Heimberg, Reference Bruch and Heimberg1994), and these parental attributes have been associated with higher social anxiety symptoms (Anhalt & Morris, Reference Anhalt and Morris2008; Caster et al., Reference Caster, Inderbitzen and Hope1999; Festa & Ginsburg, Reference Festa and Ginsburg2011; Knappe et al., Reference Knappe, Beeso-Baum, Fehm, Leib and Wittchen2012; Knappe, Lieb et al., Reference Knappe, Lieb, Beesdo, Fehm, Low, Gloster and Wittchen2009; Mothander & Wang, Reference Mothander and Wang2014; Rudoph & Zimmer-Gembeck, Reference Rudoph and Zimmer-Gembeck2014; Spokas & Heimberg, Reference Spokas and Heimberg2009). Consistent with these findings, an observational study found that parents of SAD children provided less positive feedback and more negative feedback compared to parents of non-anxious children during a collaborative task (Hummell & Gross, Reference Hummell and Gross2001), and lower parental emotional warmth has been found to predict higher persistence of SAD (Knappe, Beedso et al., Reference Knappe, Beesdo, Fehm, Hofler, Leib and Wittchen2009). Indeed, it is possible that frequent exposure to negative feedback within the family may lead some children to become hypervigilent and preoccupied by the perceived likelihood of negative social evaluation. While these factors are not consistently related to SAD (e.g., Rork & Morris, Reference Rork and Morris2009), they may be more pronounced where the parent has a history of an anxiety disorder themselves (e.g., Lieb et al., Reference Lieb, Wittchen, Hoefler, Fuetsch, Stein and Merikangas2000), or in combination with other aetiological factors, whether biological or environmental.
Attachment
An insecure attachment style has been associated various types of psychopathology (DeKlyen & Greenberg, Reference DeKlyen, Greenberg, Cassidy and Shaver2008), including social anxiety during childhood and adulthood. Securely attached individuals are more readily able to engage in positive relationships with peers and join social activities, whereas insecure individuals have weaker social and emotional competence, and thus tend to be less socially engaged, positive and popular at school, as well as more socially anxious in later childhood (Bar-Haim, Dan, Eshel, & Sagi-Schartz, Reference Bar-Haim, Dan, Eshel and Sagi-Schartz2007; Bohlin, Hagekull, & Rydell, Reference Bohlin, Hagekull and Rydell2000; Brumariu & Kerns, Reference Brumariu and Kerns2010; Kerns & Brumariu, Reference Kerns and Brumariu2014). Specifically, an anxious-ambivalent attachment style has been most commonly linked to later development of anxiety disorders, including SAD (Kerns & Brumariu, Reference Kerns and Brumariu2014; Warren, Huston, Egeland, & Sroufe, Reference Warren, Huston, Egeland and Sroufe1997). Anxious-ambivalent attachment is linked to erratic and unpredictable parenting, which is consistent with the variety of reported parenting styles associated with SAD (see Parenting Style). It may be that it is a variable and inconsistent style of parenting, rather than (or in combination with) a particular style of parenting that most increase one's risk of developing SAD.
Warren and colleagues (Reference Warren, Huston, Egeland and Sroufe1997) reported that anxious-ambivalent (but not anxious-avoidant) attachment in infants predicted increased risk for the development of anxiety disorders in adolescents 16 years later, with 38% of these individuals developing SAD. The authors suggest that anxiety arising from the uncertain availability of an attachment figure may be internalised and lead to difficulties in coping with the anxiety of ongoing developmental challenges (Warren et al., Reference Warren, Huston, Egeland and Sroufe1997). Furthermore, prospective studies indicate that an anxious-ambivalent attachment style may increase specific risk of developing SAD (Bar-Haim et al., Reference Bar-Haim, Dan, Eshel and Sagi-Schartz2007; Brumariu & Kerns, Reference Brumariu and Kerns2010). Inconsistency within the caregiver-infant relationship and consequent anxiety regarding the availability of attachment figures experienced by anxious-ambivalent infants may lead to distrust in subsequent social relationships, and an experience of persistent threat of social exclusion or rejection (Vertue, Reference Vertue2003). Correspondingly, SAD adults most commonly demonstrate an anxious-ambivalent (preoccupied) attachment style, which is associated with discomfort in close relationships, difficulty trusting or depending on others, and greater anxiety at the prospect of rejection or abandonment compared to controls (Eikenaes, Pederson, & Wilberg, Reference Eikenaes, Pederson and Wilberg2015; Eng, Heimberg, Hart, Schneier, & Liebowitz, Reference Eng, Heimberg, Hart, Schneier and Liebowitz2001).
Interestingly, two studies have found children with a disorganised attachment style report being more shy and socially anxious compared to controls, and to report higher rates of SAD (Borelli, David, Crowley, & Mayes, Reference Borelli, David, Crowley and Mayes2010; Brumariu & Kerns, Reference Brumariu and Kerns2010). However, both studies were cross-sectional, and no known longitudinal or prospective studies to date have demonstrated this relationship. Nonetheless, these findings are consistent with attachment theory, as children with disorganised attachment are considered to perceive other's behaviour as unpredictable and frightening (Borelli et al., Reference Borelli, David, Crowley and Mayes2010). Indeed, Kerns and Brumariu (Reference Kerns and Brumariu2014) suggest that disorganised children are likely to be at greatest risk of developing anxiety as they lack access to a secure base, having experienced caregivers who are unpredictable, unavailable, or punitive.
Negative and Traumatic Life Experiences
Traumatic or difficult life experiences are another factor that has been linked to the development of SAD. While negative experiences may be precipitating factors that immediately precede SAD onset (Higa-McMillan & Ebesutani, Reference Higa-McMillan, Ebesutani, Alfano and Beidel2011), the weight of theory and empirical evidence suggests that such events are predisposing factors that increase risk of developing many forms of psychopathology, including (but not uniquely) SAD (e.g., Kessler, Davis, & Kendler, Reference Kessler, Davis and Kendler1997; Kimbrel, Reference Kimbrel2008; Rapee & Spence, Reference Rapee and Spence2004; Wong & Rapee, Reference Wong, Rapee, Ranta, La Greca and Garcia-Lopz2015). Indeed, SAD has been associated with substantially less childhood abuse and fewer adverse life events compared to other disorders, such as major depressive disorder or panic disorder (Kessler et al., Reference Kessler, Davis and Kendler1997; Tiet et al., Reference Tiet, Bird, Hoven, Moore, Wu, Wicks, Jensen, Goodman and Cohen2001; Safren, Gershuny, Marzol, Otto, & Poll, Reference Safren, Gershuny, Marzol, Otto and Pollack2002). Nonetheless, Simon et al. (Reference Simon, Herlands, Marks, Mancini, Letamendi, Zhonghe and Stein2009) reported that 70% of a sample of SAD individuals (N = 72) met threshold criteria for at least one type of childhood maltreatment, and Rosellini and colleagues (Reference Rosellini, Rutter, Bourgeois, Emmert-Aronson and Brown2013) reported that 78.8% of a SAD sample (N = 210) could identify the presence of an acute or chronic stressor around the time of SAD onset. Wong and Rapee (Reference Wong, Rapee, Ranta, La Greca and Garcia-Lopz2015) suggest that negative experiences are likely to be of particular import in the development of SAD if they are social or relational in nature (e.g., parental marital discord, emotional, sexual, or physical abuse). Moreover, chronic adversity during sensitive developmental periods may increase risk of developing SAD via changes to the amygdala (Pechtel, Lyons-Ruth, Anderson, & Teicher, Reference Pechtel, Lyons-Ruth, Anderson and Teicher2014) or enhanced cortisol reactivity to stress (Elzinga, Spinhoven, Berretty, de Jong, & Roelofs, Reference Elzinga, Spinhoven, Berretty, de Jong and Roelofs2010).
Emotional Abuse and Neglect
Compared to other key trauma domains (i.e., sexual abuse and physical abuse/neglect), SAD has been most strongly related to childhood emotional abuse and neglect (Bandelow et al., Reference Bandelow, Torrente, Wedekind, Broocks, Hajak and Rüther2004; Bruce, Heimberg, Blanco, Schneier, & Liebowitz, Reference Bruce, Heimberg, Blanco, Schneier and Liebowitz2012; Gibb, Chelminski, & Zimmerman, Reference Gibb, Chelminski and Zimmerman2007; Iffland, Sansen, Catani, & Neuner, Reference Iffland, Sansen, Cantani and Neuner2012; Kuo, Goldin, Werner, Heimberg, & Gross, Reference Kuo, Goldin, Werner, Heimberg and Gross2011; Pohjavaara, Reference Pohjavaara2004; Simon et al., Reference Simon, Herlands, Marks, Mancini, Letamendi, Zhonghe and Stein2009; Spinhoven et al., Reference Spinhoven, Elzinga, Hovens, Roelofs, Zitma, van Oppen and Penninx2010). This is consistent with reports that parents of SAD individuals are over-controlling, less warm, and more rejecting (see Parenting Style). Among SAD individuals, emotional maltreatment has been associated with greater severity of social anxiety symptomatology and related disability, higher trait anxiety and depression, as well as poorer functioning, resilience, quality of life, and self-esteem (Bruce, Heimberg, Blanco, Schneier, & Liebowitz, Reference Bruce, Heimberg, Blanco, Schneier and Liebowitz2012; Bruce, Heimberg, Goldin, & Gross, Reference Bruce, Heimberg, Goldin and Gross2013; Gibb, Chelminski, & Zimmerman, Reference Gibb, Chelminski and Zimmerman2007; Iffland et al., Reference Iffland, Sansen, Cantani and Neuner2012; Kuo et al., Reference Kuo, Goldin, Werner, Heimberg and Gross2011; Simon et al., Reference Simon, Herlands, Marks, Mancini, Letamendi, Zhonghe and Stein2009). Furthermore, emotional abuse has been found to increase risk of peer rejection, which is frequently a key precipitant for SAD onset (see section entitled Emotional Abuse and Neglect in this article; Lev-Wiesel & Sternberg, Reference Lev-Wiesel and Sternberg2012).
Some studies have found emotional abuse and neglect to be more strongly related to SAD than to physical or sexual abuse (Gibb et al., Reference Gibb, Chelminski and Zimmerman2007; Iffland et al., Reference Iffland, Sansen, Cantani and Neuner2012; Kuo et al., Reference Kuo, Goldin, Werner, Heimberg and Gross2011). In one study, emotional maltreatment was retrospectively reported by 56% (n = 58) of SAD individuals, compared to 30% (n = 29) reporting physical abuse and 19% (n = 18) reporting sexual abuse (Simon et al., Reference Simon, Herlands, Marks, Mancini, Letamendi, Zhonghe and Stein2009). Furthermore, emotional maltreatment by parents has been found to mediate the impact of physical and sexual maltreatment on social anxiety symptoms (Iffland et al., Reference Iffland, Sansen, Cantani and Neuner2012). Whereas other disorders tend to be equally associated with multiple forms of early maltreatment (e.g., posttraumatic stress disorder with emotional, physical, and sexual abuse), it has been suggested at SAD may be specifically related to emotional abuse rather than early trauma per se (Gibb et al., Reference Gibb, Chelminski and Zimmerman2007). Nonetheless, emotional maltreatment has also been frequently linked to other psychopathology (e.g., major depressive disorder; Gibb et al., Reference Gibb, Chelminski and Zimmerman2007), suggesting that it does not solely or specifically increase the risk of developing SAD, and may therefore be a more distal causal factor.
Sexual and Physical Abuse
While SAD appears to be most strongly associated with emotional abuse and/or neglect, sexual abuse has also been found to heighten risk of developing the disorder and increase severity of social anxiety symptomatology, especially among females (Bandelow et al., Reference Bandelow, Torrente, Wedekind, Broocks, Hajak and Rüther2004; Bruce et al., Reference Bruce, Heimberg, Goldin and Gross2013; Chartier, Walker, & Stein, Reference Chartier, Walker and Stein2001; Cougle, Timpano, Sachs-Ericsson, Keough, & Riccardi, Reference Cougle, Timpano, Sachs-Ericsson, Keough and Riccardi2010; Dinwiddie et al., Reference Dinwiddie, Heath, Dunne, Bucholz, Madden, Slutske, Bierut and Martin2000; Feerick & Snow, Reference Feerick and Snow2005; Gibb et al., Reference Gibb, Chelminski and Zimmerman2007; Magee, Reference Magee1999; Nelson et al., Reference Nelson, Heath, Madden, Cooper, Dinwiddie, Bucholz, Glowinski, McLaughlin, Dunne, Statham and Martin2002; Spinhoven et al., Reference Spinhoven, Elzinga, Hovens, Roelofs, Zitma, van Oppen and Penninx2010). For example, the U.S. National Comorbidity Survey (N = 8,098) indicated that those who had experienced molestation or rape were significantly more likely to report SAD (odds ratios ranged from 1.67 to 1.77; Kessler, Davis, & Kendler, Reference Kessler, Davis and Kendler1997). In addition, younger age of sexual abuse onset and greater psychological pressure to participate has been associated with greater social avoidance and distress, as has actual or attempted intercourse (Feerick & Snow, Reference Feerick and Snow2005).
Physical abuse/violence and neglect have also been demonstrated to increase risk of developing SAD and to predict greater social anxiety symptom severity (Bandelow et al., Reference Bandelow, Torrente, Wedekind, Broocks, Hajak and Rüther2004; Binelli et al., Reference Binelli, Ortiz, Muniz, Gelabert, Ferraz, Filho, Crippa, Nardi, Subira and Martin-Santos2012; Bruce et al., Reference Bruce, Heimberg, Blanco, Schneier and Liebowitz2012; Chartier et al., Reference Chartier, Walker and Stein2001; Gibb et al., Reference Gibb, Chelminski and Zimmerman2007; Magee, Reference Magee1999; Pohjavaara, Reference Pohjavaara2004; Spinhoven et al., Reference Spinhoven, Elzinga, Hovens, Roelofs, Zitma, van Oppen and Penninx2010). Indeed, one study found that family violence (but not emotional, physical or sexual abuse) predicted greater social anxiety among Spanish university students (Binelli et al., Reference Binelli, Ortiz, Muniz, Gelabert, Ferraz, Filho, Crippa, Nardi, Subira and Martin-Santos2012), which is consistent with findings that having an aggressive mother or father may increase the risk of developing SAD (odds ratios ranged from 1.34 to 1.77; Kessler et al., Reference Kessler, Davis and Kendler1997). Domestic physical abuse also increases the risk of social peer rejection, which is frequently a precipitating factor in SAD (see section, Negative and Traumatic Life Experiences; Lev-Wiesel & Sternberg, Reference Lev-Wiesel and Sternberg2012), and this relationship may be mediated by social withdrawal or other coping efforts. Nonetheless, both sexual and physical abuse have also been linked to other disorders (e.g., panic disorder and posttraumatic stress disorder; Cougle et al., Reference Cougle, Timpano, Sachs-Ericsson, Keough and Riccardi2010; Stein et al., Reference Stein, Walker, Anderson, Hazen, Ross, Eldridge and Forde1996), indicating that these traumatic experiences do not confer specific risk for SAD.
Other Negative Life Events
Other negative events that are social in nature have also been associated with the development of SAD. Compared to healthy controls, SAD has been associated with more frequent reports of death of a parent(s), or other means of separation from parents (Bandelow et al., Reference Bandelow, Torrente, Wedekind, Broocks, Hajak and Rüther2004; Kessler et al., Reference Kessler, Davis and Kendler1997), family conflict, including parental marital discord (Bandelow et al., Reference Bandelow, Torrente, Wedekind, Broocks, Hajak and Rüther2004; Chartier et al., Reference Chartier, Walker and Stein2001), parental divorce or separation (Kessler et al., Reference Kessler, Davis and Kendler1997), lack of close relationship with an adult (Chartier et al., Reference Chartier, Walker and Stein2001), and having to start a new school (Tiet et al., Reference Tiet, Bird, Hoven, Moore, Wu, Wicks, Jensen, Goodman and Cohen2001). Furthermore, SAD individuals were more likely to report having a parent or family member with significant psychopathology, including an anxiety, mood, or substance use disorder (Chartier et al., Reference Chartier, Walker and Stein2001; Kessler et al., Reference Kessler, Davis and Kendler1997; Lieb et al., Reference Lieb, Wittchen, Hoefler, Fuetsch, Stein and Merikangas2000; Tiet et al., Reference Tiet, Bird, Hoven, Moore, Wu, Wicks, Jensen, Goodman and Cohen2001). Such negative relational experiences may be interpreted as indicating that others are dangerous, not trustworthy, or likely to humiliate, reject or ostracise, thereby heightening perceived social threat. However, these experiences do not appear to confer specific risk for SAD and are also linked to other psychopathology (e.g., parental separation and family conflict are associated with both conduct disorder and major depressive disorder; Tiet et al., Reference Tiet, Bird, Hoven, Moore, Wu, Wicks, Jensen, Goodman and Cohen2001).
Aversive Social/Peer Experiences
Social and peer-related negative and traumatic events (e.g., bullying, humiliation, ostracism) are considered to be direct conditioning experiences associated with the onset and development of SAD symptomatology, including increased distress and avoidance related to social situations (Higa-McMillan & Ebesultani, Reference Higa-McMillan, Ebesutani, Alfano and Beidel2011; Ollendick & Hirshfeld-Becker, Reference Ollendick and Hirshfeld Becker2002; Rapee & Spence, Reference Rapee and Spence2004; Wong & Rapee, Reference Wong, Rapee, Ranta, La Greca and Garcia-Lopz2015). Accordingly, peer victimisation is more strongly associated with SAD compared to other anxiety disorders (Cohen & Kendall, Reference Cohen and Kendall2015; McCabe, Antony, Summerfeldt, Liss, & Swinson, Reference McCabe, Antony, Summerfeldt, Liss and Swinson2003) or depression (Ranta et al., Reference Ranta, Kaltiala-Heino, Pelkonen and Marttunen2009), and several studies have found that a substantial proportion of SAD individuals retrospectively report direct traumatic social experiences (e.g., Beidel, Turner & Morris, Reference Beidel, Turner and Morris1999; Hofmann, Ehlers, & Roth, Reference Hofmann, Ehlers and Roth1995; Ost & Hugdahl, Reference Öst and Hugdahl1981; Sansen, Iffland, & Neuner, Reference Sansen, Iffland and Neuner2015; Stemberger, Turner, Beidel, & Calhoun, Reference Stemberger, Turner, Beidel and Calhoun1995). For example, an early study found that 58% of a SAD sample ascribed disorder onset to a traumatic conditioning experience (Öst, Reference Öst1987). Indeed, many SAD adults indicate that such an event was associated with the onset or exacerbation of the disorder (Hackmann, Clark, & McManus, Reference Hackmann, Clark and McManus2000; McCabe et al., Reference McCabe, Antony, Summerfeldt, Liss and Swinson2003; Rosellini, Rutter, Bourgeois, Emmert-Aronson, & Brown, Reference Rosellini, Rutter, Bourgeois, Emmert-Aronson and Brown2013).
Aversive social conditioning experiences may be discrete humiliating events, but are often repetitive, cumulative experiences of chronic social trauma that threaten the basic human need for acceptance and can act as experiences that confirm already developing social threat fears and beliefs (Baumeister & Leary, Reference Baumeister and Leary1995; Tillfors et al., Reference Tillfors, Persson, Willen and Burk2012). Furthermore, traumatic conditioning may occur vicariously via witnessing or hearing about the humiliating social experience of another (Beidel & Turner, Reference Beidel and Turner2007; Ollendick & Hirshfeld-Becker, Reference Ollendick and Hirshfeld Becker2002). Peer victimisation has been found to predict increased social anxiety over time, and may include overt victimisation (e.g., physical or verbal aggression), relational victimisation (e.g., psychological bullying, friendship withdrawal or exclusion, peer neglect, or rejection), or damage to reputation (e.g., spreading rumours; Siegel et al., Reference Siegel, La Greca and Harrison2009). However, there is evidence to suggest that relational victimisation may be the strongest risk factor for SAD (e.g., Landoll, La Greca, Lai, Chan, & Herge, Reference Landoll, La Greca, Lai, Chan and Herge2015; Storch et al., Reference Storch, Masia-Werner, Crisp and Klein2005).
Relational Victimisation
Relational victimisation involves the use of relational assaults to harm a peer (e.g., via exclusion from a social group, or emotional bullying; Seigel et al., Reference Siegel, La Greca and Harrison2009; Storch et al., Reference Storch, Masia-Werner, Crisp and Klein2005). In cross-sectional studies, social anxiety and SAD diagnosis are consistently positively associated with higher levels of relational victimisation and social exclusion, fewer friends and difficulty making new friends, less peer acceptance and more negative interactions in close friendships, as well as more social avoidance and withdrawal (Boulton, Reference Boulton2013; Craig, Reference Craig1998; Demsey & Storch, Reference Demsey and Storch2008; Erath et al., Reference Erath, Flanagan and Bierman2007; Greco & Morris, Reference Greco and Morris2005; Iffland, Sansen, Cantani, & Neuner, Reference Iffland, Sansen, Cantani and Neuner2012; La Greca & Harrison, Reference La Greca and Harrison2005; Landoll et al., Reference Landoll, La Greca, Lai, Chan and Herge2015; Ranta et al., Reference Ranta, Kaltiala-Heino, Pelkonen and Marttunen2009; Rapee & Melville, Reference Rapee and Melville1997; Scharfstein, Alfano, Beidel, & Wong, Reference Scharfstein, Alfano, Beidel and Wong2011; Storch, Brassard, & Masia-Werner, Reference Storch, Brassard and Masia-Werner2003; Storch & Masia-Warner, Reference Storch and Masia-Warner2004; Yen et al., Reference Yen, Yang, Wang, Lin, Liu, Wu and Tang2014). Conversely, lower levels of social anxiety are associated with stronger social support, acceptance, and peer validation, suggesting that these factors may buffer excessive fears of social scrutiny and negative judgement (Festa & Ginsburg, Reference Festa and Ginsburg2011). Prospective studies support these associations, demonstrating that relational victimisation, exclusion, and lower peer acceptance predict greater social anxiety and likelihood of a SAD diagnosis over time, and may have greater impact than overt victimisation (Seigel, La Greca, & Harrison, Reference Siegel, La Greca and Harrison2009; Teachman & Allen, Reference Teachman and Allen2007; Tillfors, Persson, Willen, & Burk, Reference Tillfors, Persson, Willen and Burk2012; Storch, Masia-Werner, Crisp, & Klein, Reference Storch, Masia-Werner, Crisp and Klein2005), especially among girls (Ranta, Kaltiala-Heino, Frojd, & Marttunen, Reference Ranta, Kaltiala-Heino, Frojd and Marttunen2013).
While a number of studies demonstrate that social anxiety is associated with active relational victimisation, SAD may also be associated with social neglect and isolation, such that some socially anxious children may simply be overlooked or ignored rather than actively disliked, manipulated, or rejected (Beidel & Turner, Reference Beidel and Turner2007; Hudson & Rapee, Reference Hudson and Rapee2000). Furthermore, the experience of social failure associated with social neglect may exacerbate existing tendencies for self-isolation, generating a vicious cycle (Adalbjarnardottir, Reference Adalbjarnardottir1995; Beidel & Turner, Reference Beidel and Turner2007; Hudson & Rapee, Reference Hudson and Rapee2000; Rubin & Mills, Reference Rubin and Mills1988).
Overt Victimisation
Perhaps unsurprisingly, overt victimisation (e.g., physical or verbal assault) also appears to be associated positively with fears of negative evaluation, social avoidance, and the development of SAD — albeit less strongly than relational victimisation (Storch et al., Reference Storch, Brassard and Masia-Werner2003; Storch & Masia-Werner, Reference Storch and Masia-Warner2004). Thus, cross-sectional studies demonstrate that SAD diagnosis and higher social anxiety levels are associated with more negative peer interactions, verbal aggression, and direct victimisation (Craig, Reference Craig1998; Erath, Flanagan, & Bierman, Reference Erath, Flanagan and Bierman2007; Landoll et al., Reference Landoll, La Greca, Lai, Chan and Herge2015; Ranta, Kaltiala-Heino, Pelkonen, & Marttunen, Reference Ranta, Kaltiala-Heino, Pelkonen and Marttunen2009; Yen et al., Reference Yen, Yang, Wang, Lin, Liu, Wu and Tang2014). Adolescents with SAD report more frequent overt victimisation than those without SAD (Gren-Landell, Aho, Andersson, & Svedin, Reference Gren-Landell, Aho, Andersson and Svedin2011), and adults who report overt bullying in childhood demonstrate higher prevalence of SAD (48.1%) compared to those who did not report childhood bullying (26.4%) (Gladstone, Parker, & Malhi, Reference Gladstone, Parker and Malhi2006). Importantly, prospective studies also demonstrate that overt victimisation (physical or verbal aggression) increases risk of developing SAD and predicts greater social anxiety symptomatology over time (Loukas & Pasch, Reference Loukas and Pasch2013; Storch et al., Reference Storch, Masia-Werner, Crisp and Klein2005; Vernberg, Abwender, Ewell, & Beery, Reference Vernberg, Abwender, Ewell and Beery1992), especially among boys (Marteinsdottir, Svensson, Svedberg, Anderberg, & von Knorring, Reference Marteinsdottir, Svensson, Svedberg, Anderberg and von Knorring2007; Ranta, Kaltiala-Heino, Frojd, & Marttunen, Reference Ranta, Kaltiala-Heino, Frojd and Marttunen2013; Storch et al., Reference Storch, Brassard and Masia-Werner2003).
In addition, social anxiety appears to be particularly associated with verbal bullying (e.g., teasing or name calling). Thus, SAD adults are more likely to report a history of frequent teasing compared to individuals with panic disorder or obsessive-compulsive disorder (McCabe, Miller, Laugesen, Antony, & Young, Reference McCabe, Miller, Laugesen, Antony and Young2010), and a history of childhood teasing has been associated with greater social anxiety symptomatology (McCabe et al., Reference McCabe, Miller, Laugesen, Antony and Young2010; Roth, Coles, & Heimberg, Reference Roth, Coles and Heimberg2002). Indeed, one study found that 92% of a sample of adults with SAD reported experiencing severe teasing in childhood (McCabe, Antony, Summerfeldt, Liss, & Swinson, Reference McCabe, Antony, Summerfeldt, Liss and Swinson2003). Furthermore, there is some evidence to suggest that cyber victimisation (i.e., overt victimisation via the internet or other forms of electronic media) is also associated with social anxiety (Landoll et al., Reference Landoll, La Greca, Lai, Chan and Herge2015). However, findings are preliminary and suggest that relational victimisation remains a stronger predictor (Landoll et al., Reference Landoll, La Greca, Lai, Chan and Herge2015).
The Relationship of Aversive Social Experiences and SAD
A number of studies suggest that social anxiety has a bidirectional relationship with social victimisation and rejection (e.g., Blöte, Bokhorst, Miers, & Westenberg, Reference Blöte, Bokhorst, Miers and Westenberg2012; Gazelle et al. Reference Gazelle, Putallaz, Li, Grimes, Kupersmidt and Coie2005; Ranta et al., Reference Ranta, Kaltiala-Heino, Frojd and Marttunen2013; Siegel et al., Reference Siegel, La Greca and Harrison2009; Tillfors et al., Reference Tillfors, Persson, Willen and Burk2012). This appears to be a vicious feedback cycle in which socially anxious individuals are more likely to be rejected or victimised (potentially due to poorer social performance; Blöte, Miers, & Westenberg, Reference Blöte, Miers and Westenberg2015; Greco & Morris, Reference Greco and Morris2005), which in turn increases social threat and avoidance, leading to further rejection and victimisation (e.g., Pabian & Vandebosch, Reference Pabian and Vandebosch2015). Furthermore, socially anxious children are more likely to experience commonplace social events as distressing or traumatic, thereby increasing their anxiety. For example, Beidel et al. (Reference Beidel, Turner and Morris1999) described the most distressing events reported by SAD children as having to perform in front of others, being asked to answer a question by their teacher, being spoken to by a popular kid, and having to talk to someone on the telephone.
Interestingly, socially aversive experiences are also reported by individuals without SAD (Stemberger et al., Reference Stemberger, Turner, Beidel and Calhoun1995), and not all SAD individuals recall discrete socially traumatic events (Ost, Reference Öst1987), suggesting that this is not a sufficient or necessary cause for the disorder. It is likely that SAD results from a complex interaction of such conditioning experiences, with biological factors, cognitive factors, parenting factors, and other negative experiences (Beidel & Turner, Reference Beidel and Turner2007). For example, coping style has been found to moderate the relationship between peer victimisation and social anxiety (Boulton, Reference Boulton2013), and self-esteem has been found to partially mediate the association of peer relationships with social anxiety among adolescents (Bosacki, Dane, Marini, & YLC-CURA, Reference Bosacki, Dane, Marini and YLC2007), consistent with the negative and unstable self-concept shown in SAD (Wilson & Rapee, Reference Wilson and Rapee2006).
Posttraumatic Symptomatology and Imagery in the Aetiology of SAD
Aversive social/peer experiences may be experienced as traumatic, and thus have been found to trigger posttraumatic symptoms among SAD individuals, including intrusive re-experiencing, avoidance, negative alterations in cognitions and mood, and hyperarousal symptoms (Carleton, Pelso, Collimore & Asmundson, Reference Carleton, Pelso, Collimore and Asmundson2011; Erwin, Heimberg, Marx, & Franklin, Reference Erwin, Heimberg, Marx and Franklin2006). Erwin et al. (Reference Erwin, Heimberg, Marx and Franklin2006) found that a third of SAD individuals in their sample would have met criteria for Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; APA, 2000) posttraumatic stress disorder (PTSD) if the socially traumatic event satisfied criterion A (witnessing or experiencing death, serious injury or threat to one's own or another's physical integrity). In contrast to controls, SAD patients in this study reported reacting to memories of past socially aversive events with hyperarousal and avoidance (Erwin et al., Reference Erwin, Heimberg, Marx and Franklin2006). SAD patients also endorsed significant re-experiencing symptoms in response to memories of socially traumatic events (though no more than controls), and the authors suggest that these symptoms may be ‘distorted recollection of the past event, potentially contributing to distorted beliefs and interfering with the processing of socially stressful events’ (Erwin et al., Reference Erwin, Heimberg, Marx and Franklin2006, p. 909).
Moreover, there is an increasing body of literature suggesting that negative self-imagery plays a key role in the development and maintenance of SAD and that such imagery may be rooted in experiences of social trauma and threatened social belongingness (Clark & Wells, Reference Clark, Wells, Heimberg, Liebowitz, Hope and Schneier1995; Hirsch, Clark, Mathews, Williams, & Morrison, Reference Hirsch, Clark, Mathews, Williams and Morrison2006; Rapee & Heimberg, Reference Rapee and Heimberg1997; Sansen et al., Reference Sansen, Iffland and Neuner2015; Schreiber & Steil, Reference Schreiber and Steil2013). Negative self-imagery appears to maintain social anxiety by facilitating and reinforcing negative interpretations of ambiguous social information (Hirsch, Clark, & Mathews, Reference Hirsch, Clark and Matthews2006), increasing access to a negative working model of the self (Hulme, Hirsch, & Stopa, Reference Hulme and Hirsch2012), and inhibiting retrieval of positive/realistic imagery, memories, and working models of the self (e.g., Hirsch et al., Reference Hirsch, Clark, Williams, Morrison and Mathews2005; Hulme et al., Reference Hulme and Hirsch2012; Stopa & Jenkins, Reference Stopa and Jenkins2007). Furthermore, utilising attentional resources to focus on negative self-imagery prevents SAD individuals from noticing neutral or positive feedback, or that their feared outcome may not be occurring. Hence, these processing biases prevent disconfirmation of feared social outcomes and impede updating of negative self-images with more realistic ‘pictures’ based on benign or positive experiences, or lessening the affective response associated with negative images, thereby maintaining social anxiety (Wild, Reference Wild and Stopa2009). Finally, focus on negative self-referent imagery increases anxiety and engagement in safety behaviours, thereby undermining social performance and increasing the actual and perceived likelihood of negative feedback (e.g., Hirsch, Meynen, & Clark, Reference Hirsch, Meynen and Clark2004; Hirsch, Clark, Mathews, & Williams, Reference Hirsch, Clark, Mathews and Williams2003).
Importantly, Hackmann et al. (Reference Hackmann, Clark and McManus2000) found that negative, distorted self-imagery experienced by SAD individuals tends to be linked in theme and content to earlier socially traumatic or aversive events (e.g., teasing, rejection) and that these events cluster around the onset or exacerbation of social anxiety symptoms. Such aversive social events in formative years appear to be laid down in memory and incorporate any negative meanings or catastrophic interpretations made at the time (Hirsch, Clark, & Matthews, Reference Hirsch, Clark and Matthews2006). The ensuing negative distorted self-images that are triggered by and recur in social situations (Hackmann, Surawy, & Clark, Reference Hackmann, Surawy and Clark1998) may be considered comparable to intrusive re-experiencing symptomatology characteristic of PTSD (Erwin et al., Reference Erwin, Heimberg, Marx and Franklin2006). Furthermore, recent research has yielded very strong treatment outcomes for SAD using imagery rescripting interventions that aim to update traumatic memories for SAD (e.g., Lee & Kwon, Reference Lee and Kwon2013; Nilsson, Lundh, & Viborg, Reference Nilsson, Lundh and Viborg2012; Norton & Abbott, Reference Norton and Abbott2016; Reimer & Moscovitch, Reference Reimer and Moscovitch2015; Wild, Hackmann, & Clark, Reference Wild, Hackmann and Clark2008), suggesting that such memories and associated intrusive imagery may be key precipitating and maintaining factors for the disorder.
SAD as a Shame-Based Disorder
Shame is a self-conscious emotion in which the individual experiences their global self as defective, thus wishes to hide, disappear, or die (Lewis, Reference Lewis, Lewis, Haviland-Jones and Barrett2008), and has been defined as ‘an intensely painful feeling or experience of believing we are flawed and therefore unworthy of acceptance and belonging’ (Brown, Reference Brown2006, p. 45). Despite being conceptually distinct, there is clear overlap between social anxiety and shame (Gilbert, Reference Gilbert2000, Reference Gilbert and Weeks2014), and a number of studies have demonstrated a correlation between these constructs (e.g., Fergus, Valentiner, McGrath, & Jencius, Reference Fergus, Valentiner, McGrath and Jencius2010; Gilbert, Reference Gilbert2000; Hedman, Strom, Stunkel, & Mortberg, Reference Hedman, Strom, Stunkel and Mortberg2013; Matos, Pinto-Gouveia, & Gilbert, Reference Matos, Pinto-Gouveia and Gilbert2013). Indeed, while SAD is classified as an anxiety disorder, it is arguable that shame is as central as anxiety to the experience of individuals with severe social anxiety, and thus their condition may also be conceptualised as a shame disorder (Herman, Reference Herman, Dearing and Tangney2011). Social anxiety appears to be strongly associated with internal shame, involving internalised negative judgments of oneself as undesirable, deficient, inferior, inadequate, weak, disgusting, or bad (Matos et al., Reference Matos, Pinto-Gouveia and Gilbert2013), but has also been linked to external shame (concern that one will be judged negatively for appearing this way in the eyes of others; Gilbert, Reference Gilbert and Weeks2014). Furthermore, it is plausible that the aetiological factors reviewed above (negative parenting, difficult life events, aversive social experiences) may influence the development of SAD via the internalisation of shame-based cognitive and affective states.
Theory and empirical evidence indicates that severe and/or chronic experiences of being abused, rejected, criticised, and shamed during childhood (e.g., via social exclusion, verbal bullying, emotional abuse, punitive or unpredictable parenting) may lead to an internalised negative representation of the self, and thus an increased tendency toward experiencing shame (Claesson & Sohlberg, Reference Claesson and Sohlberg2002; Pinto-Gouveia & Matos, Reference Pinto-Gouveia and Matos2011; Gilbert, Reference Gilbert2003; Gilbert, Allan, & Goss, Reference Gilbert, Allan and Goss1996; Lewis, Reference Lewis, Lewis, Haviland-Jones and Barrett2008; Matos et al., Reference Matos, Pinto-Gouveia and Gilbert2013; Stuewig & McCloskey, Reference Stuewig and McCloskey2005). Internal, stable, global attributions of these negative early experiences are argued to be stored as autobiographical experiences that become the foundation for subsequent self-related beliefs and behaviours (Gilbert, Reference Gilbert2003). For example, a child who experiences frequent harsh criticism from their parents is likely to encode emotionally textured experiences of having elicited criticism from others and being treated as bad, and therefore comes to believe ‘I am bad’(Gilbert, Reference Gilbert2003). Similarly, a child who is excluded by peers is likely to encode emotionally textured experiences of having elicited exclusion from others and being treated as undesirable, and thus comes to internalise the belief that ‘I am undesirable’ (Gilbert, Reference Gilbert2003). In this way, such experiences may be internalised as global evaluations of the self as undesirable, inferior, inadequate, weak, disgusting, or bad. These shame-based schemas (i.e., self-perpetuating dysfunctional cognitive-affective structures; Young, Reference Young1999) are frequently found among SAD individuals (e.g., Calvete, Orue, & Hankin, Reference Calvete, Orue and Hankin2013; Pinto-Gouveia, Castilho, Galhardo, & Cunha, Reference Pinto-Gouveia, Gastilho, Galhardo and Cunha2006), and are readily reactivated by social experiences (Lee, Scragg, & Turner, Reference Lee, Scragg and Turner2001). Furthermore, the negative self-imagery described by SAD individuals frequently demonstrates shame-based as well as fear-based themes, and is rooted in shaming social experiences (Hackmann et al., Reference Hackmann, Clark and McManus2000). This is consistent with findings that memories of shaming experiences may be encoded with the features of trauma memories (e.g., intrusion, affective avoidance, distorted/fragmented cognitive processes, hyperarousal; Matos & Pinto-Gouseia, Reference Matos and Pinto-Gouveia2010) and that such memories can become a central reference point for personal narrative and self-identity (Pinto-Gouveia & Matos, Reference Pinto-Gouveia and Matos2011).
While research has linked greater shame-proneness with negative parenting practices (e.g., Claesson & Sohlberg, Reference Claesson and Sohlberg2002; Webb, Heisler, Call, Chickering, & Colburn, Reference Webb, Heisler, Call, Chickering and Colburn2007), insecure attachment (e.g., Lopez et al., Reference Lopez, Gover, Leskela, Sauer, Shirmer and Wyssmann1997), negative life experiences (e.g., Stuewig & McCloskey, Reference Stuewig and McCloskey2005), and social anxiety (e.g., Fergus et al., Reference Fergus, Valentiner, McGrath and Jencius2010), only one known study to date has specifically investigated the relationship between negative early life experiences, shame, and the development of SAD. Thus, Shahar, Doron, and Szepsenwol (Reference Shahar, Doron and Szepsenwol2014) found that shame and self-criticism mediated the relationship between emotional abuse (but not neglect) and social anxiety symptomatology. Shahar et al. (Reference Shahar, Doron and Szepsenwol2014) suggest that early emotional maltreatment may be internalised as a ‘shame-based cognitive-affective schema’ (p. 1), which is characterised by a global experience of the self as inadequate. The authors argue that self-criticism may also be adopted as a defensive strategy to protect against exposure of perceived flaws and deficiencies in the self, as well facilitating avoidance of the distress associated with reactivated shame states. These findings provide preliminary evidence for the crucial role of shame in the aetiology and maintenance of SAD.
Limitations of Previous Studies
There are a number of limitations in the literature on environmental factors in the aetiology of SAD. Critically, studies to date have been largely retrospective, self-report, and cross-sectional, and there may be bias in the subjective reports of anxious individuals often made many years after the events. Furthermore, most studies have explored the relationship between SAD and each key environmental factor individually; thus, the complex interplay, relative impact, and chronology of such factors remains relatively unexplored. Additionally, many studies report on childhood anxiety, rather than SAD specifically.
Moreover, while a range of negative experiences in childhood (e.g., emotional, physical, sexual maltreatment) have been considered possible precipitating factors for SAD, many theorists suggest that such negative life events increase an individual's predisposition toward developing psychopathology more generally (e.g., via development of negative beliefs about self or others) rather than rather than being specific to SAD (e.g., Rapee & Spence, Reference Rapee and Spence2004). Thus, the specific or general impact of non-peer childhood trauma in the development of SAD remains unclear. Furthermore, no study has yet explored the impact of caregiver response to childhood abuse (e.g., management of abusive situation, support available to the child). Response to trauma is a strong predictor of posttraumatic symptoms (Lieberman, Chu, van Horn & Harris, Reference Lieberman, Chu, van Horn and Harris2011), and hence may mediate the impact of negative life events on subsequent SAD development.
In addition, recent studies have suggested that socially traumatic experiences and their sequelae (e.g., negative distorted self-imagery) may be critical precipitating and maintaining factors in SAD (e.g., Erwin et al., Reference Erwin, Heimberg, Marx and Franklin2006). However, only two studies (Carlton et al., Reference Carleton, Pelso, Collimore and Asmundson2011; Erwin et al., Reference Erwin, Heimberg, Marx and Franklin2006) have demonstrated significant levels of posttraumatic symptomatology in response to socially aversive experiences, and these findings were based on DSM-IV rather than DSM-5 criteria for PTSD. Furthermore, the relation of aversive social experiences to the development of negative distorted self-imagery (a key maintaining factor in SAD; Clark & Wells, Reference Clark, Wells, Heimberg, Liebowitz, Hope and Schneier1995) requires clarification. Finally, only one known study to date has explored the role of internalised shame from early negative experiences in the development of SAD, and the possibility of a shame-based model of SAD has not been considered.
Summary and Future Research Implications
The above reviewed literature implicates a number of key environmental factors in the aetiology of SAD. Specifically, theory and research suggests that the risk of developing SAD is increased by over-controlling, critical and cold parenting, insecure (anxious-ambivalent) attachment, emotional maltreatment, and to a lesser extent, physical and sexual maltreatment, and other relational adversities in childhood, as well as aversive social/peer experiences (especially relational victimisation). Moreover, these factors may lead to posttraumatic reactions, distorted negative self-imagery, and internalised shame-based schemas. However, understanding of the nature, interactions and relative contributions of these factors remains unclear. Moreover, it is likely that SAD develops via a complex interplay of biological and environmental factors, and that there are multiple and diverse pathways to the development of the disorder. Indeed, while environmental factors may serve as risk factors that increase the likelihood of developing SAD, they do not directly cause the disorder, and no single risk factor is sufficient or necessary to the development of SAD.
Future research would benefit significantly from the use of longitudinal prospective studies, greater utilisation of objective observational measures rather than subjective self-report, and assessment of the interplay of the range of environmental risk factors for SAD. Greater study of parenting factors and negative life events specific to SAD is also required. In particular, further investigation is required to understand whether particular traumatic life events increase risk for general psychopathology, or SAD specifically — or the conditions under which either may be the case. Furthermore, the impact of aversive peer experiences on posttraumatic symptomatology (including negative self-imagery) and their role in the maintenance of SAD requires further exploration, including the value of trauma-based interventions for the disorder (e.g., imagery rescripting; Wild et al., Reference Wild, Hackmann and Clark2008). Additionally, as not all SAD individuals report a socially traumatic event associated with onset of the disorder, the role of other precipitating environmental factors warrants investigation. Finally, the role of shame in the development and maintenance of SAD is significantly understudied, and the conceptualisation of SAD as a shame-based disorder is deserving of consideration. Nonetheless, it remains likely that multiple aetiological pathways underlie the development of SAD.
