UC San Diego

Anxiety is distressing, impairing, and the most prevalent mental health problem in children and adolescents. Theoretical models implicate biases in cognitive processes as underlying the development and maintenance of anxiety disorders. Yet little work has been done to test theoretical models of information processing (IP) in clinically anxious youths. The present dissertation sought to examine relationships between youth anxiety and the basic cognitive processes of attention and interpretation using performance-based methodology. The aims of this investigation were to: 1) empirically test the IP model of youth anxiety, specifically whether biased interpretation statistically mediates the association between biased attention and anxiety symptoms, 2) unpack the components of interpretation (i.e., threat valence judgments, speed of responding) on a performance-based task of interpretation bias, and 3) probe the impact of comorbid depressive symptoms on these cognitive processes. Youths (N=26, ages 9 to 17) and their primary caregivers completed diagnostic interviews and a comprehensive self- report battery, and youths completed performance-based assessments of attention and interpretation. In this sample, attention bias towards threat was significantly associated with percentage of negative interpretations endorsed (r=.46, p=.019). However, attention was not significantly related to anxiety symptoms, and the indirect effect of attention on anxiety through interpretation was not statistically significant. Negative interpretations of ambiguous information strongly predicted youth anxiety symptom severity, accounting for 46% of variance in clinician-rated anxiety severity. Deconstructing interpretation into its components, youth response latencies on the interpretation task as measured by threat and benign reaction time indices were not significantly associated with attention. Percentage of negative interpretations endorsed (r=.68, p<.001) and response latencies on the interpretation threat reaction time index (r=.45, p=.022) were significantly related to anxiety severity. Conversely, response latencies on the benign interpretation reaction time index were not related to anxiety. Finally, depressive symptoms were not significantly related to attention or to any of the components of interpretation assessed in this sample of clinically anxious youths. These findings provide preliminary evidence for a relationship between attention and interpretation biases, and stronger evidence for a relationship between interpretation bias and anxiety. Clinical and theoretical implications of this study are discussed