Nadine Vietmeier¹; Dr. Nik Dietze²; Prof. Dr. Brunna Tuschen-Caffier³; Prof. Dr. Julia Asbrand⁴
¹ Humboldt-Universität zu Berlin; ² University Clinic of Child and Adolescent Psychiatry and Psychotherapy, University Medical Centre EWL; ³ Albert Ludwig University of Freiburg; ⁴ Friedrich Schiller University Jena

Social anxiety disorder (SAD) is one of the most common mental health conditions in children, characterized by an intense fear of negative evaluation and social avoidance. While cognitive models of adult SAD emphasize the role of attentional biases—both self-focused and externally directed toward social threat—in the maintenance of SAD, the extent and nature of these biases in children remain unclear. Notably, few studies have examined how internal and external attention interact prior to socially stressful situations in children. 

This study employed a multimodal approach to examine attentional biases in children with SAD. We hypothesized that children with SAD would (1) exhibit heightened self-focused attention, indicated by faster reaction times (RTs) to internal stimuli, and (2) demonstrate an initial hypervigilance followed by avoidance of a potentially threatening stimulus (a neutral-looking peer audience), indicated by gaze patterns.

A total of 91 children (43 with SAD, 48 healthy controls [HCs], aged 9–14) participated in a socially evaluative speech task. During the anticipation phase, attentional biases were assessed using an RT experiment and mobile eye-tracking.

Eye-tracking data revealed that children with SAD showed greater early attention allocation to the peer audience—characterized by initial hypervigilance and sustained gaze—compared to HCs during the first 50 seconds of anticipation. Contrary to our expectations, RT data did not show group differences in self-focused attention during the anticipation phase; both groups responded similarly to internal and external stimuli.

These findings offer nuanced insight into attentional processes in childhood SAD. While results support the presence of early-stage external attentional biases, as measured by eye-tracking, they do not provide evidence for increased internal attentional biases under anticipatory stress, as measured by RTs. This discrepancy raises questions about the ecological validity and sensitivity of RT-based paradigms in detecting attentional biases during complex, real-life stressors in children. Future research should move beyond single-modality approaches and incorporate behavioral, self-report, and physiological measures to better capture the dynamic interplay of internal and external attention in children with SAD. A deeper understanding of these mechanisms is critical for refining interventions targeting attentional processes in childhood SAD.