Children's Cognitive Responses to the Somatic Symptoms of Panic

The purpose of this study was to test Nelles and Barlow's (1988) hypothesis that spontaneous panic attacks are rare or nonexistent prior to adolescence as children lack the ability to make the internal, catastrophic attributions (i.e., thoughts of losing control, going crazy, or dying) characteristic of panic according to the cognitive model (Clark, 1986). Conceptions of panic attacks, including the understanding of symptoms and causes, and cognitive interpretations of the somatic symptoms of panic were examined in children from Grades 3, 6, and 9. A significant main effect for grade was found for conceptions of panic attacks, with third graders receiving significantly lower scores than sixth and ninth graders. However, the majority of all children, regardless of age, tended to employ internal (e.g., I'd think I was scared or nervous) rather than external (e.g., I'd think I was feeling that way because of the temperature or the weather) explanations of panic attacks. No significant grade differences were found for the tendency to make internal versus external and catastrophic versus noncatastrophic attributions in response to the somatic symptoms of panic. When presented with panic imagery in a panic induction phase, children, regardless of age, made more internal and noncatastrophic attributions. Finally, internal attributional style in response to negative outcomes and anxiety sensitivity were found to be significant predictors of internal, catastrophic attributions. The challenge that these findings pose to Nelles and Barlow's hypothesis, and their relevance for understanding children's cognitive interpretations of panic symptomatology are discussed.