Convergent Functional Connectivity Alterations in ADHD Revealed by EEG: A Systematic Review Across Methods, States and Spatial Scales.

Functional brain connectivity (FC) research in attention-deficit/hyperactivity disorder (ADHD) has been dominated by functional magnetic resonance imaging (fMRI). However, electroencephalography (EEG) offers a more clinically accessible alternative with lower cost, easier in-practice implementation, and better temporal resolution. Despite these advantages, clinical translation of FC findings has been hindered by substantial methodological heterogeneity. It is unclear whether coherent, disorder-relevant patterns can be extracted across this diverse literature. We conducted a systematic review and narrative synthesis of EEG FC studies in ADHD, examining whether consistent alterations emerge across analytic families (linear/non-linear; lagged/non-lagged), spatial domains (sensor-/source-spaced) and recording conditions (rest/task). Thirty-seven studies meeting inclusion criteria were identified, encompassing paediatric, adolescent, and adult samples (n=4038, 65.8% male, 30.0% female, 4.2% not specified). Despite pronounced methodological diversity, several convergent themes emerged. Across approaches, ADHD was commonly characterised by reduced posterior integration with relative frontal compensation, altered interactions between large-scale networks, fragmented or compartmentalised network organisation, and a shift toward increased local connectivity alongside weakened global integration. These patterns were most robust in slower frequency bands and often accentuated under cognitive or affective demands. Importantly, these EEG-derived network features closely parallel established fMRI findings of large-scale functional dysconnectivity in ADHD. Together, these results demonstrate that meaningful, region-level network signatures of ADHD can be identified across heterogeneous EEG FC methodologies. These findings support the viability of EEG as a translational tool for characterising brain network dysfunction in ADHD, while also highlighting the importance of standardised, lag-aware, and source-informed approaches for future clinical and biomarker-oriented research.