Face emotion recognition differences with respect to frustration in disruptive mood dysregulation disorder (DMDD) and attention-deficit/hyperactivity disorder (ADHD)

Abstract

Objectives: We examined the effect of frustration in emotion recognition across children with ADHD and disruptive mood dysregulation disorder (DMDD) and healthy control (HC) subjects. Methods: Sample group consisted of 43 participants (14 DMDD + ADHD, 14 ADHD, 15 HC subjects, mean = 9.51; SD = 2.10), who completed a novel emotion recognition task with a frustration component that we developed using happy, angry, and neutral faces from the FACES database of Max Planck Institute for Human Development. The task was a two-alternative forced choice paradigm in which participants were asked to identify if a face was happy versus neutral or if a face was angry versus neutral. Frustration, in some blocks, was elicited by providing conflicting feedback and holding back deserved points after some correct answers. Results: In the frustration blocks of the emotion recognition task, there were statistically significant main effects of emotion on accuracy [ F (1,39) = 18.886, P < 0.001, η 2 = 0.326] and the type of feedback in previous trial on the response time [ F (1,39) = 11.019, P < 0.002, η 2 = 0.220], as well as a between-subjects effect of diagnosis on accuracy [ F (2,39) = 6.384, P = 0.004, η 2 = 0.247]. Post hoc analyses revealed that participants were more accurate recognizing happy faces as opposed to angry faces ( P < 0.001), slower after frustrating trials ( P = 0.002), and that participants with DMDD were less accurate than control subjects ( P = 0.017) and participants with ADHD ( P = 0.007). Linear integrated speed-accuracy scores (LISAS), interpreted as reaction time adjusted by error rate, showed a main effect of emotion [ F (1,39) = 12.389, P = 0.001, η 2 = 0.241] and feedback type in previous trial [ F (1,39) = 6.350, P = 0.016, η 2 = 0.140], but there was no between-subjects effect of diagnosis group on LISAS. The post hoc analyses revealed that the reaction time adjusted for error rates when recognizing happy faces were found to be faster as opposed to angry faces ( P = 0.001) and slower in postfrustration trials ( P = 0.016). Conclusions: Children with DMDD are less accurate overall compared with healthy control subjects and children with ADHD in the novel task that involves emotion recognition. Their impairment becomes more pronounced after frustrating trials, which emerge as a distinctive feature between participants with pure ADHD and those with ADHD/DMDD comorbidity.