SLEEP EEG DYNAMICS REFLECT DIVERGENT NEURODEVELOPMENTAL TRAJECTORIES IN INFANTS

Emergence of feature-rich neuronal network dynamics follows individualized maturational trajectories of brain development, deviations from which may precede neurodevelopmental disorders (NDD). Following the critical brain hypothesis, healthy maturation promotes a progressive reorganization toward stable near-critical neuronal dynamics, yet their evolution in early life remains poorly characterized. Sleep provides a stable window to assess these processes, as slow-wave activity during NREM sleep reliably indexes early cortical maturation.
We analyzed longitudinal sleep EEG from 35 infants (14 NDD), each with two recordings in the first year of life from an open Physionet dataset. Artifact-free, bipolar-referenced EEG segments of 2.5 minutes were extracted during N3 sleep (7 channels). Neuronal dynamics in delta (0.5–4 Hz) and theta (4–8 Hz) bands were quantified using the bistability index (BiS), detrended fluctuation analysis (DFA), and spectral power, and analyzed using linear mixed-effects models including group (typically developing, TD, vs NDD) and subject as a random factor.
In delta, the mixed-effects model in TD showed a negative effect of BiS (coef=−0.108, p=0.026) and positive effects of DFA (coef=0.357, p=0.003) and power (coef=0.160, p<1e−7), with no significant interactions in NDD. In theta, BiS decreased with age in TD (coef=−0.158, p=0.008), whereas power showed a strong age-related increase (coef=1.697, p<1e−7) and a significant group interaction, suggesting altered progression in NDD.
EEG markers of slow-wave activity capture age-dependent brain maturation in infants, suggesting a normalization of brain dynamics closer to a critical-like regime, while deviations from typical trajectories may reflect early neurodevelopmental divergence.