By Carol Wilkinson

One way of studying brain activity and network formation across development is through electroencephalography (EEG). EEG can non-invasively and relatively cheaply measure cortical electrical activity and oscillatory activity generated by the brain. We recently analyzed longitudinal EEG data collected from almost 600 healthy infants across 4 studies. We set out to characterize and model the developmental trajectories of distinct features observed in the power spectra—the EEG readouts showing breakdown of brain waves in different frequency ranges. Our initial goal was simply to know how the power spectrum changes across early development, so that we could better interpret our separate studies of infants and toddlers with neurodevelopmental disorders. However, we were surprised to observe that several features in the power spectra of typically developing infants– namely, peaks in the theta/alpha and beta range – were transient in nature. We think understanding these transient features may give us clues about the timing of key developmental steps in circuit maturation.

For example, most infants 2-4 months of age exhibit an alpha peak at 9.5Hz in addition to a theta peak around 5.5Hz. However, by 6 months, the 9.5Hz peak has disappeared, and the vast majority of infants only have a single peak between 5-7Hz. We are very curious about what produces and leads to the disappearance of this 9.5Hz peak, and whether alterations in the timing of its loss have developmental implications. In a separate EEG study in children (not infants) with Down syndrome, we observe that 40% of children exhibit both a theta and alpha peak that qualitatively is very similar to what we observe in typically developing 2-4 month olds. We wonder whether the transient nature of the 9.5Hz peak could represent the timing of transient connections between the thalamus and subplate neurons, known to be critical for thalamocortical maturation. We would love to collaborate with those doing animal research or infant MRI work to further test this hypothesis.

In the beta range, we observe a couple of intriguing changes in peaks: First, we observed the emergence of a low beta peak (12-20Hz) starting in some infants around 9-12 months of age. This was interesting because the timing of emergence reminded us of developmental anesthesia studies finding that anesthesia-induced frontal alpha coherence (coherence being the synchronization of oscillations across a brain region) is not established until after 10 months. We hypothesized that emergence of a low beta peak was related to the maturation of GABAergic interneuron networks and thalamocortical connectivity that is required for both the generation of cortical beta oscillations and anesthesia-induced frontal alpha coherence.

In a fun collaboration with colleagues in anesthesia, we analyzed EEG data collected from a small group of healthy infants before and during anesthesia. We found that infants with a low beta peak before anesthesia had higher anesthesia-induced alpha coherence than those without a beta peak, suggesting that the emergence of these low beta oscillations may directly reflect thalamocortical loop maturation.

Finally, we observed a transient high amplitude beta peak around 30Hz, which peaks in amplitude at 7-months-old, but has largely resolved by 36 months. The developmental circuitry underlying this transient peak is still unclear. But we do know that similar beta peaks are induced by GABAergic epilepsy medications. Also, we have observed a strikingly similar peak in preschool-aged children with Fragile X syndrome.

Altogether we think that these findings will help provide context to cross-sectional work occurring at these early ages and provide a foundation to compare developmental trajectories of various neurodevelopmental disorders including autism, ADHD, and rare neurogenetic disorders. We hope sharing our observations and hypotheses with the broader HBI community will spur future collaborations and brainstorming to help us better understand the circuit level changes leading to alterations in oscillations through early development.

Carol Wilkinson is an Assistant Professor and Attending Physician in Developmental Behavioral Pediatrics at Boston Children’s Hospital.

Learn more in the original research article:
Developmental trajectories of EEG aperiodic and periodic components in children 2-44 months of age.
Wilkinson CL, Yankowitz LD, Chao JY, Gutiérrez R, Rhoades JL, Shinnar S, Purdon PL, Nelson CA.  Nat Commun. 2024 Jul 10;15(1):5788. doi: 10.1038/s41467-024-50204-4. PMID: 38987558; PMCID: PMC11237135.