Resting-state Alpha and Mu Rhythms Change Shape across Development But Lack Diagnostic Sensitivity for Attention-Deficit/Hyperactivity Disorder and Autism.

Researchers studied brain wave patterns in over 1,600 children aged 5-18. They found that certain brain rhythms change as children develop, but these patterns couldn't distinguish between children with ADHD, autism, and typically developing children. While the brain waves weren't useful for diagnosis, the research reveals new details about how children's brains develop that could help scientists better understand brain function.

This study analyzed electroencephalography (EEG) recordings from 1605 children aged 5-18 years to examine alpha and mu brain rhythms. Using computational methods to isolate these rhythms, researchers found that both exhibit non-sinusoidal waveform shapes that change significantly across development, alongside known frequency changes. The study tested whether these rhythms could serve as diagnostic markers for ADHD and autism spectrum disorder, but found no differences in resting-state features between these conditions and typically developing children. The findings suggest traditional spectral analyses may miss important waveform properties that could inform more sophisticated models of brain rhythm generation.

These findings suggest that resting-state EEG measures of alpha and mu rhythms should not be used for diagnosing ADHD or autism. The developmental changes in waveform shape may inform future research into brain maturation patterns, but current clinical applications are limited.

The study only examined resting-state brain activity and did not assess task-related responses. The computational pipeline, while efficient, may have missed subtle differences. The focus on waveform shape represents a novel approach that requires replication.

In the human brain, the alpha rhythm in occipital cortex and the mu rhythm in sensorimotor cortex are among the most prominent rhythms, with both rhythms functionally implicated in gating modality-specific information. Separation of these rhythms is nontrivial due to the spatial mixing of these oscillations in sensor space. Using a computationally efficient processing pipeline requiring no manual data cleaning, we isolated alpha and/or mu rhythms from electroencephalography recordings performed on 1605 children aged 5-18 years. Using the extracted time series for each rhythm, we characterized the waveform shape on a cycle-by-cycle basis and examined whether and how the waveform shape differs across development.

We demonstrate that alpha and mu rhythms both exhibit a nonsinusoidal waveform shape that changes significantly across development, in addition to the known large changes in oscillatory frequency. This data set also provided an opportunity to assess oscillatory measures for attention-deficit/hyperactivity disorder and autism spectrum disorder. We found no differences in the resting-state features of these alpha-band rhythms for either attention-deficit/hyperactivity disorder or autism spectrum disorder in comparison with typically developing participants in this data set. Although waveform shape is ignored by traditional Fourier spectral analyses, these nonsinusoidal properties may be informative for building more constrained generative models for different types of alpha-band rhythms, yielding more specific insight into their generation.