Social Anxiety Predicts Aggression in Children with ASD: Clinical Comparisons with Socially Anxious and Oppositional Youth

Our conceptual focus in the present study was on humiliation and rejection (H/R) fears. However, we also explored and note herein effects for MASC public performance (P/P) fears for each group. There was a significant group difference on P/P fears (F (2,56) = 4.39, p = .02, η _p ²  = .14). There was no difference between the HFASD group (M = 47.84, SD = 9.06) and the ODD/CD group (M = 44.75, SD = 11.29), t(56) = .90, p = .37. When collapsed (M = 54.55, SD = 12.54) and compared to the SAD group, the SAD group was higher on P/P fears, t(56) = 19.08, p < .001. In the HFASD group, P/P fears had a significant negative linear effect on aggression (b = −.84, p = .01), but no curvilinear effect, and the overall model was not significant (F (3,15) = 2.99, p = .07). However, it was not possible to differentiate whether the observed linear relationship for P/P fears was qualitatively different from the curvilinear effect for H/R fears or merely an artifact of range restriction in observed P/P fears. In the SAD group, as observed for H/R fears, P/P fears were negatively linearly associated with aggression (b = −.33, p = .01), and the quadratic P/P term did not contribute incrementally. In the ODD/CD group, as observed for H/R fears, neither linear nor quadratic P/P effects predicted aggression scores (F (2,17) = 1.34, p = .29 and F (3,16) = 1.51, p = .25, respectively). In the HFASD group, the inclusion of depression strengthened the relationship between P/P fears and aggression, as was observed for H/R fears. Inclusion of depression did not affect the association of P/P fears and aggression in the SAD or ODD/CD groups.