ADHD is a neurodevelopmental disorder associated with a high rate of comorbidity with anxiety disorders (25-34%). Children with ADHD experience serious adverse outcomes secondary to impairment in executive function, particularly within the domain of working memory (WM), behavioral inhibition (BI), and sustained attention (SA). While executive function deficits in ADHD are well documented, whether and how comorbid anxiety affects cognitive performance are equivocal. One potential explanation is that most studies examine linear relations, yet evidence suggests that anxiety affects performance in a non-linear (quadratic) manner, consistent with the the Yerkes-Dodson Law. The aim of this study was to investigate whether 1) children with ADHD show deficits in WM, BI, and SA relative to typically developing children, 2) comorbid anxiety displays a linear or nonlinear relationship with WM, BI, and SA performance among children with ADHD and 3) between group differences in cognitive performance vary based on levels of anxiety. Linear and non-linear relations between anxiety and cognitive performance were assessed in a sample of 54 boys diagnosed with ADHD and 50 typically developing boys. Anxiety was assessed across dimensions and raters. Results indicate rater and domain-specific effects of comorbid anxiety on cognitive performance. Non-linear relations between children’s self-rated physiological anxiety and Phonological working memory (PHWM), Visuospatial working memory (VSWM), and the Central Executive (CE) were found. Non-linear relations between parent-rated anxiety and PHWM and the CE were also found. However, no significant linear or non-linear effects of anxiety on BI and SA were found. The results indicate that children with moderate self-rated and parent-rated anxiety performed better on WM measures relative to those with low and high levels of self-rated and parent-rated anxiety. The present study was the first to examine and document non-linear effects of anxiety on cognitive performance among children diagnosed with ADHD. Given the results, clinicians should continue to assess anxiety during diagnostic screening in ADHD samples. Treatments should focus on compensating for CE abilities and mitigating high levels of anxiety as it may further impair WM.