Recent research has demonstrated that high social anxiety is associated with uncertainty about one’s self views and self-concept (Moscovitch et al., 2009; Stopa et al., 2010; Wilson & Rapee, 2006). However, no research has addressed potential mechanisms underlying the link between high social anxiety and low self-certainty nor has research examined whether this relationship is bi-directional. In the current research, I propose a cyclical model in which high social anxiety leads to low self-certainty, which in turn, feeds back into higher levels of social anxiety. I also propose that the relationship between high social anxiety and low self-certainty is mediated by the self-protective self-disclosure patterns employed by socially anxious individuals. In three interconnected studies, I examine the hypothesis that social anxiety, self-disclosure and self-certainty operate in a cyclical model. Study 1 provided a correlational test of the hypothesized feedback model in its entirety and demonstrated that honesty of self-disclosure was the most important and influential mechanism underlying the link between high social anxiety and low self-certainty. Experimentally manipulating the honesty of participants’ self-disclosures in Study 2 demonstrated that dishonest self-disclosures during a social task led to low self-certainty, but only amongst individuals high in trait performance anxiety. Finally, experimentally manipulating self-certainty in Study 3 demonstrated that low self-certainty led to high anticipatory anxiety about an upcoming self-disclosure task. Together, these results elucidate a cyclical maladaptive pattern in which low self-certainty as a result of self-protective self-disclosure leads to high social anxiety and a greater reluctance to self-disclose. Results from the three studies are discussed with respect to their theoretical implications and in relation to clinical applications for individuals with social anxiety disorder.

Recent research has demonstrated that high social anxiety is associated with uncertainty about one’s self views and self-concept (Moscovitch et al., 2009; Stopa et al., 2010; Wilson & Rapee, 2006). However, no research has addressed potential mechanisms underlying the link between high social anxiety and low self-certainty nor has research examined whether this relationship is bi-directional. In the current research, I propose a cyclical model in which high social anxiety leads to low self-certainty, which in turn, feeds back into higher levels of social anxiety. I also propose that the relationship between high social anxiety and low self-certainty is mediated by the self-protective self-disclosure patterns employed by socially anxious individuals. In three interconnected studies, I examine the hypothesis that social anxiety, self-disclosure and self-certainty operate in a cyclical model. Study 1 provided a correlational test of the hypothesized feedback model in its entirety and demonstrated that honesty of self-disclosure was the most important and influential mechanism underlying the link between high social anxiety and low self-certainty. Experimentally manipulating the honesty of participants’ self-disclosures in Study 2 demonstrated that dishonest self-disclosures during a social task led to low self-certainty, but only amongst individuals high in trait performance anxiety. Finally, experimentally manipulating self-certainty in Study 3 demonstrated that low self-certainty led to high anticipatory anxiety about an upcoming self-disclosure task. Together, these results elucidate a cyclical maladaptive pattern in which low self-certainty as a result of self-protective self-disclosure leads to high social anxiety and a greater reluctance to self-disclose. Results from the three studies are discussed with respect to their theoretical implications and in relation to clinical applications for individuals with social anxiety disorder.