This review examined critically studies issuing from the cognitive therapy (CT) model claiming to have unveiled cognitive causal factors of social phobia. Additionally, it examined outcome studies of CT-inspired interventions and other treatments having included measurements of cognitive constructs. Overall, we found no evidence consistently supporting the claim that social phobics are characterized by typical cognitive processes. Moreover, we found neither corroborating evidence for a controlling effect of such cognitive processes on social phobic conduct, nor consistent indications that cognitive therapies or techniques effect cognitive changes differently than other approaches. The evidence suggests rather, that cognitive factors change concurrently with other features of psychopathology as part of an overall improvement during or after effective therapy, regardless of therapeutic approach.