The Influence of Autism Spectrum Disorder and Attention Deficit Hyperactivity Disorder Symptom Severity on the Facial Emotion Recognition Skills of Adolescents

Background: Autism spectrum disorder (ASD) and attention deficit hyperactivity disorder (ADHD) have a high degree of comorbidity but differential etiology (Leitner, 2014). Previous studies reported deficits in facial emotion recognition (FER) associated with poorer social functioning (Høyland et al., 2017; Kats-Gold, 2007) in both ASD (Lozier et al., 2014) and ADHD (Da Fonseca, 2009) populations. When examined categorically, diagnoses of ASD and ADHD have been associated with differential deficits in FER (Buitelaar et al., 1999; Berrgren et al., 2016); however, despite the disorders’ frequent co-occurrence, no work has examined the relative influence of ADHD and ASD symptom severity on FER within the context of ASD diagnosis (Webb, Neuhaus, & Faja, 2017; Miller et al., 2011).

Objectives: To determine the unique influence of ASD and ADHD symptom severity on FER abilities in adolescents with and without ASD.

Methods: Using a standardized assessment of FER (DANVA-2; Nowicki, 2004), data were collected from 86 adolescents with and without ASD (Table 1). ADOS-2 (Lord et al., 2012) Comparison Score was used to measure ASD symptom severity. Parents of participants reported adolescent’s ADHD (Combined Type) symptom severity via the CASI-5 (Gadow and Sprafkin, 2013). A One-Way ANOVA determined if FER and symptom severity differed between diagnostic groups. Pearson correlations assessed relationships between ASD and ADHD symptom severity and FER in each group.

Results: FER performance did not differ as a function of ASD diagnosis (F (1,84) = 1.48, p = .23). For the ASD group, ASD symptom severity was associated with greater FER errors (r (38) = .45, p < .01; Figure 1), with the relationship driven primarily by errors for happy (r (38) = .40, p = .01) and fearful facial expressions (r (38) = .52, p < .01). For non-ASD participants, ADHD symptom severity corresponded with greater FER errors (r (44) = .38, p < .01; Figure 1), with the relationship driven primarily by errors for high-intensity (r (44) = .50, p < .001) and sad expressions (r (44) = .34, p = .02).

Conclusions: These results support previous findings of disorder-specific FER deficits associated with ASD and ADHD. For individuals with ASD, FER errors were associated with ASD, but not ADHD symptom severity. FER errors for happy (Berrgren et al., 2016) and fearful (Ashwin et al., 2006) expressions were particularly associated with more severe ASD symptom presentation. Conversely, for non-ASD participants, FER errors were associated with ADHD symptom severity but not ASD symptom severity. This finding is particularly interesting given that the ASD group demonstrated greater overall levels of ADHD symptom severity relative to the non-ASD group. Additionally, in the non-ASD group, participants with greater ADHD symptom severity made more FER errors for high-intensity emotions, indicating greater relative impairments on less difficult FER trials. Overall, these findings suggest that once ASD symptoms reach a threshold of clinical significance for diagnosis, they may be more influential in contributing to deficits in FER than ADHD symptoms for adolescents.