Lifespan EEG reference charts disentangle alpha-band alterations in neurological and neuropsychiatric disorders

Accurate quantification of alpha-band (8-13 Hz) slowing, a putative biomarker in neurological and neuropsychiatric disorders, is hindered by two problems: the conflation of oscillatory activity with aperiodic 1/f components in EEG spectra, and the absence of age-specific normative references in routine clinical EEG. Both problems are compounded by pronounced changes in alpha rhythms across the lifespan. Here we generate lifespan brain reference charts for alpha peak frequency and power, deriving lobe-specific models from raw and 1/f-adjusted spectra in a large outpatient cohort (N = 22,094; age: 1-100 years). We then applied these normative models to inpatients spanning a range of mental disorders. Across all diagnostic groups, alpha power was consistently reduced relative to age-matched norms, regardless of whether spectra were 1/f-adjusted. In contrast, alpha peak frequency showed a striking method-dependent dissociation: conventional analysis of raw spectra suggested alpha slowing, whereas 1/f-adjusted estimates revealed robust alpha acceleration. These findings demonstrate that aperiodic components can mask underlying oscillatory changes, and that 1/f adjustment is essential for interpreting alpha dynamics in clinical populations. More broadly, these findings indicate that the widely reported “alpha slowing” in mental disorders may partly reflect unmodeled shifts in aperiodic neural activity rather than a true deceleration of alpha oscillations.