Age-dependent spatial transfer of cortical harmonics to scalp EEG in infancy

Abstract

Infancy is a period of rapid change in cortical geometry and head volume conduction, raising a central question for developmental neuroimaging: how should spatial brain coordinates be defined so that they remain interpretable across age? In scalp electrophysiology, this question has two coupled aspects. First, age-dependent forward propagation through the head determines which cortical spatial patterns are expressed at the scalp. Second, cortical Laplace-Beltrami (LB) eigenmodes computed independently on age-specific templates need not remain directly comparable by mode index, even when they represent related low-dimensional structure. Using age-specific infant anatomical templates distributed through MNE-Python, we computed age-specific cortical LB eigenmodes, age-specific EEG forward models, and the corresponding forward-projected cortical harmonic dictionaries. We then quantified (i) age-dependent forward transfer to the scalp, (ii) cross-age mismatch in cortical basis, coefficient, and sensor spaces, and (iii) the improvement obtained from a simple local tracking procedure. Forward-projected scalp gain was consistently concentrated in lower cortical harmonic orders across infancy, with a broadly stable coarse-to-fine profile across age, although the absolute gain varies over age. In cortical mode space, independently computed neighboring-age cortical LB bases were only locally comparable by nominal mode index, exhibiting local reordering, mode-index drift, and coefficient leakage into nearby modes. This non-equivalence had practical consequences: neighboring-age infant dictionaries were not fully interchangeable across age in low-dimensional sensor space, and a fixed adult-derived basis was systematically suboptimal for recovering infant same-index harmonic coordinates, with the largest penalties in early-to mid-infancy. Sequential Procrustes tracking substantially improved same-index neighboring-age consistency. Together, these results motivate age-parameterized or explicitly tracked cortical harmonic coordinates for longitudinal developmental EEG and, more broadly, for lifespan analyses that seek geometrically meaningful spatial coordinates across changing brains.