Social Phobia 355
In social phobia the maladaptive schema activated by anticipated or actual expo-
sure to a social evaluative situation involve negative beliefs about the inadequacy of one’s
social ability, the threatening or critical nature of social interactions, and a negative self-
image in which socially anxious individuals assume they make a negative impression
on others (D. M. Clark, 2001). In essence, the schematic organization in social phobia
revolves around issues related to the social self. Evidence of an explicit interpretation
bias for social threat on retrospective measures (see Hypothesis 1) supports the predic-
tions of activated negative social self- schemas in Hypothesis 2. However, there are three
other lines of research that have directly addressed this issue.
First, a number of early questionnaire studies have reported a significant increase in
negative social evaluative cognitions that is specific to social phobia compared to other
anxiety disorders or nonclinical controls (e.g., Becker et al., 2001; Beidel et al., 1985;
Turner & Beidel, 1985; Turner et al., 1986). Based on think-aloud and questionnaire
measures of thoughts generated by socially anxious individuals after they participated
in a 7–8 minute conversation, Stopa and Clark (1993) found that the socially anxious
group had significantly more negative self- evaluative cognitions and believed their nega-
tive thoughts more than individuals with other anxiety disorders (see also Magee &
Zinbarg, 2007). A more recent questionnaire study suggests that any experience that
involves conspicuousness or heightened self- consciousness, whether positive or negative,
might be related to social anxiety (Weeks et al., 2008).
A second body of research has shown that social phobia may be characterized by
implicit (i.e., automatic or unintended encoding and retention processes) memory and
associative bias that reflects activation of negative social self- schemas. Using a variety of
experimental paradigms, socially anxious individuals have shown an implicit memory
bias for social threat sentences or videos (Amir et al., 2003; Amir, Foa, & Coles, 2000)
but not previously presented social threat words (Lundh & Öst, 1997; Rapee et al.,
1994; Rinck & Becker, 2005). Too few studies have employed the Implicit Association
Test (IAT) to provide clear results but there is some suggestion that high social anxiety is
associated with less positive implicit self- esteem (Tanner et al., 2006), although de Jong
(2002) did not find this difference. At this point, we can only conclude that support for
Hypothesis 2 from research on implicit processes is weak and inconsistent at best.
Much of the research relevant to Hypothesis 2 has investigated the presence of a
negative self-image in social phobia which involves “processing of the self as a social
object” (D. M. Clark & Wells, 1995, p. 72). In the current model this negative social
self-image reflects activation of maladaptive social self- schemas. A number of findings
are consistent with this formulation. Based on a semistructured interview, individuals
with social phobia reported significantly more spontaneous negative images of how they
might appear to others when recalling a recent episode of social anxiety than low socially
anxious individuals (Hackmann et al., 1998). Mansell and Clark (1999) found that only
high socially anxious individuals had a significant correlation between perceived body
sensations while giving a speech and self- ratings of how anxious they thought they
appeared to others. When socially anxious volunteers were randomly assigned to hold
a negative or control image in their mind while interacting briefly with a confederate,
the negative self-image condition elicited significantly more anxiety, greater use of safety
behaviors, poorer social performance, and beliefs that they appeared more anxious and
performed less well with the confederate (Hirsch, Meynen, & Clark, 2004; see also
Hirsch, Clark, Williams, & Morrison, 2005).