Clark and Wells’ prominent model of social anxiety disorder (SAD) assumes that cognitive variables such as negative expectations or dysfunctional cognitions play a central role in the symptomatology of SAD. In contrast to adults, it is less clear how well the cognitive model can be applied to children and adolescents. A network analysis with seven nodes was conducted to explore the importance of cognitive variables and their interaction with symptoms of SAD based on N = 205 children and adolescents (8–18 years, M = 11.54 years). Cognitive variables had a high but differential impact within the positively connected network of SAD. Dysfunctional cognitions were most strongly connected within the network. Dysfunctional cognitions, as predicted by Clark and Wells’ model, seem to act as a hub affecting several symptoms. The association between negative expectations and avoidance indicates that negative expectations may particularly contribute to the maintenance of SAD.

Background Silence in certain situations represents the core symptom of selective mutism (SM). However, it is unclear what additional symptoms are part of this disorder. Although knowledge of symptoms is essential for diagnostics and intervention, to date, only scarce research exists on circumscribed symptoms of SM. Given the large overlap between SM and social anxiety disorder (SAD), it remains also unclear which symptoms can differentiate both disorders. Methods A network analysis of potential symptoms of SM was performed based on a mixed sample of N = 899 children and adolescents with and without indication of SM (n = 629 with silence in certain situations). In a preliminary analysis, we demonstrated that children with and without silence in certain situations do not differ with respect to their network structure, justifying an analysis on the entire mixed sample. Possible communities (symptom clusters) within the network and thus potential latent variables were examined, and symptoms were analyzed in terms of their centrality (the extent to which they are associated with other symptoms in the network). To investigate the differentiability of symptoms of the SM network from symptoms of SAD, we computed a network that additionally contains symptoms of SAD. Results In the resulting network on symptoms of SM, silence was, as expected, the symptom with the highest centrality. We identified two communities (symptom cluster): (1) symptoms associated with the fear response of freezing, (2) symptoms associated with speech production and avoidance. SM network symptoms and SAD symptoms largely formed two separate symptom clusters, with only selectivity of speaking behavior (more talkative at home and taciturn or mute outside the home) falling into a common cluster with SAD symptoms. Conclusions Silence appears to have been confirmed by analysis as a core symptom of SM. Additional anxiety-related symptoms, such as avoidance behavior or motor inhibition associated with freezing, seem to co-occur with silence. The two communities of SM potentially indicate different mechanisms of silence. The symptoms of SM appear to be distinguishable from those of SAD, although there seems to be overlap in terms of difficulty speaking in situations outside the home.