Data-Driven Profiles of AttentionDeficit / Hyperactivity Disorder Using Objective and Ecological Measures of Attention, Distractibility and Hyperactivity

Abstract:

In the past two decades, the traditional subtypes of Attention-Deficit/Hyperactivity Disorder (ADHD) have been criticized for having substantial variability in symptom manifestation, clinical course, and treatment response. In the present study, we questioned whether an objective and ecological assessment of attentional control, impulsivity, and hyperactivity, the core symptom domains on which ADHD diagnosis is currently based, could yield similar phenotypic profiles to those defined by DSM-5 criteria. 110 Spanishspeaking children and adolescents (6–16 years) with ADHD (n = 57) and typically developing (n = 53) completed AULA, a continuous performance test embedded in virtual reality. We found that ADHDCombined and ADHD-Inattentive subtypes exhibited the same performance profile. Then, we applied hybrid hierarchical k-means clustering algorithms to AULA’s main outcome measures. A five-cluster structure was the most optimal solution based on several validation indices. We identified two ADHD phenotypes sharing attention impairments and hyperactivity but with an opposing performance profile on processing speed (PS) and response inhibition domains; two normative groups with average and high performance; and one profile with relatively intact performance but poor sustained attention and slow PS. DSM-5 subtypes cut across cluster profiles. Our findings might suggest that PS and response inhibition, but not attentional processes and gross-motor activity, are useful domains to distinguish between ADHD subpopulations. This study highlights the poor feasibility of traditional categorical systems to parse ADHD heterogeneity and the added value of VR-based neuropsychological assessment to obtain an objective and less biased characterization of cognitive functioning in individuals with and without ADHD.