Differences in Resting State EEG Spectral Power in Infants at Elevated Risk for ASD

Background: Clinical diagnosis of autism spectrum disorder (ASD) can be made by two years of age, but experimental evidence suggests earlier emergence of differences in attention and brain activity in infants that later develop ASD. Prior work using resting state EEG has found attenuated spectral power in infants at elevated risk for developing ASD (HR-ASD) compared to normal risk (NR) infants (Tierney, 2012). Additionally, HR-ASD infants demonstrate different developmental trajectories compared to NR infants as indexed by spectral power. Resting state EEG offers an inexpensive, non-invasive measure of cortical activity, with potential to serve as a practicable biomarker of atypical development.

Objectives: Examine differences in resting EEG spectral power between HR-ASD and NR infants throughout the first two years of life.

Methods: EEG data was recorded from 41 HR-ASD and 42 NR infants at 3, 6, 9, 12, 18, and 24 months of age. At each time point, 2 minutes of EEG was recorded in a dimly lit room using a 128-channel HydroCel Geodesic Sensor Net. Infants sat on their parent’s lap while a research assistant blew bubbles to keep them still. EEG data were filtered from 0.1-100 Hz, segmented into 1 second epochs, and baseline corrected. Bad channels were interpolated and epochs containing artifact were removed. EEG recordings that did not have a minimum length of 20s were omitted from further analysis. Spectral power was extracted from frontal regions of the scalp. Delta (2-4 Hz), theta (4-6 Hz), low alpha (6-9 Hz), high alpha (9-13 Hz), beta (13-30 Hz), and gamma (30-50 Hz) frequency bands were defined based on prior research examining EEG power in HR-ASD infants (Tierney, 2012). Both absolute and relative power were compared between groups at each time point.

Results: Preliminary analyses of relative power revealed lower high alpha power in HR-ASD infants at 9 and 18 months of age (p=0.02 and p=0.01, respectively). At 24 months of age, HR-ASD infants demonstrated lower levels of relative delta power (p=0.04) and greater relative theta power (p=0.03) than NR infants. Analyses of absolute power revealed greater low alpha power (p=0.01) and greater theta power (p<0.01) in HR-ASD infants at 24 months.

Conclusions: Preliminary results indicate both absolute and relative differences in spectral power bands between HR-ASD and NR infants. These findings expand upon prior work that has identified differences in delta, theta, and alpha power across development in infants at elevated risk for ASD. These results provide additional insight into typical and atypical neural development in the first two years of life. Ongoing analyses are examining spectral power differences among these developmental trajectories through the incorporation of phenotypic outcome data collected at 36 months of age.