Electroencephalographic Examination of Resting State Neural Oscillatory Activity in Young Children with Autism Spectrum Disorder

Background: Previous research has observed atypical neural oscillatory activity in 6-15 year-old children and adolescents with Autism Spectrum Disorder (ASD), using Magnetoencephalography (MEG; Cornew et al., 2012). In particular, participants with ASD exhibited regionally specific elevations in delta, theta, alpha, beta, and gamma band activity at rest, relative to controls. Furthermore, increased temporal and parietal alpha power was associated with greater symptom severity in ASD. These findings have been interpreted as evidence for an imbalance in neural excitation/inhibition in ASD.

Objectives: To investigate neural oscillatory activity in younger children with autism using a task-free “Resting State” procedure during continuous Electroencephalography (EEG) recording.

Methods: Participants are children with ASD aged 3- to 5-years (n=28) and control participants who were developing typically or experiencing non-ASD developmental delays (CON; n=25). ASD diagnostic status was assessed by research-reliable doctoral level clinicians using best clinical judgment based on several measures (ADOS-2, Mullen Scales of Early Learning, Vineland Adaptive Behavior Scales). Participants watched a movie (Inscapes) featuring moving abstract shapes without a narrative, designed to keep young participants still and awake (Vanderwal et al., 2015). Continuous 128-channel EEG (Electrical Geodesics, Inc.) was recorded for approximately 7 minutes per participant. Data were then processed offline, and only participants who produced 70 seconds or more of artifact-free data (ASD: n=17; CON: n=19) were included in further analysis. The final groups of ASD and CON participants were matched on chronological age and nonverbal mental age. Power Spectral Density data were calculated for theta (4 – 7 Hz) and alpha (7 – 11 Hz) activity, for electrodes located over the Orbitofrontal, Dorsolateral Frontal, Central, and Parietal Cortices. ANOVAs with Frequency Band (Theta, Alpha), Region (Orbitofrontal, Dorsolateral Frontal, Central, Parietal), and Hemisphere (Left, Right) as Within-Subjects Factors and Group (ASD, CON) as a Between-Subjects Factor were then conducted.

Results: Significant main effects were observed for Frequency Band (F(1, 34)=191.957, p < .001, η_p²=.850) and Region (F(2.215, 75.310)=153.733, p < .001, η_p²=.819), but not Group (p=.350). Significant main effects were moderated by a significant interaction between Frequency Band and Region, whereby power in the Theta and Alpha bands were less differentiated from one another in the Central and Parietal regions than in other cortical regions (F(1.808, 61.467)=46.098, p < .001, η_p²=.576). No other significant main effects or interactions were observed. No significant correlations were observed between parietal alpha power and social functioning in either participant group.

Conclusions: The preliminary findings of this on-going EEG study of young children do not replicate previously observed MEG-based findings indicating atypical neural oscillatory activity in older children and adolescents with ASD. This may be related to differences in the studies, which include the age of the participants (younger versus older), the need to use video watching (versus eyes closed) with the young children in the current study, differences in EEG versus MEG measurement properties, and/or the currently smaller sample size (n=36 versus n=50) and associated differences in statistical power to detect effects to date in this on-going study.