Abstract
This study is the first to apply the PAC method to explore the rsEEG activity characteristics in HCs and ADHD subtypes, filling the gap in understanding the PAC differences between ADHD subtypes. • Both ADHD subtypes exhibited significantly stronger PAC across multiple frequency bands compared to HCs, with distinct spatial distribution patterns, suggesting the presence of subtype-specific compensatory neural hyperactivation mechanisms. • Brain network analysis not only supports the "delayed maturation theory" of ADHD but also indicates that ADHD-C may involve a shift from a typical small-world network to a more regular network organization. • These findings reveal significant differences in PAC features and brain network structures between HCs and ADHD subtypes, which may reflect the heterogeneity of ADHD subtypes. Abstract Phase-amplitude coupling (PAC) plays a critical role in attention, sensory processing, and working memory—domains often impaired in children with attention deficit/hyperactivity disorder (ADHD). Therefore, PAC is theoretically well-suited for ADHD research. However, the differences in PAC characteristics among children with ADHD subtypes have not yet been thoroughly investigated. This study recorded resting-state electroencephalographic (rsEEG) from 19 healthy controls (HCs), 33 children with predominantly inattentive type (ADHD-I), and 39 with combined type (ADHD-C). We examined intra- and inter-channel PAC differences across different spatial scales and further analyzed PAC-based brain network properties. The results showed that both ADHD subtypes had stronger α-γ PAC than HCs, with ADHD-C exceeding ADHD-I. ADHD-I showed mainly intrahemispheric changes, while ADHD-C involved the left hemisphere and occipital regions. In the α-β band, PAC was significantly higher in ADHD-C than in ADHD-I, mostly in the left brain. ADHD-I also showed increased inter-channel δ-β PAC compared to HCs, with widespread distribution. These findings suggest the presence of compensatory hyperactivation mechanisms in ADHD, particularly in the ADHD-C subtype. Further brain network analysis supported the “delayed maturation theory” of ADHD and indicated that ADHD-C may represent a shift from a typical small-world network architecture to a more regular network organization. Finally, the (Support Vector Machine) SVM classification results further validated the discriminative power of these features in differentiating HCs from ADHD subtypes. Overall, these findings indicate significant differences in PAC strength and brain network topology among ADHD subtypes, suggesting their potential as biomarkers for distinguishing HCs from ADHD subtypes.
Authors
Tang, W., Jiang, J., & Wang, H.
https://doi.org/10.1016/j.ijpsycho.2025.113222