Cortical Morphology and Links to Resting-State Oscillatory Activity in Autism Spectrum Disorders

Background: Our laboratory recently reported that children with autism spectrum disorders (ASD) exhibit increased resting-state alpha power in temporal and parietal regions, with greater alpha activity associated with greater symptom severity (SRS score). Other recent work has shown reduced cortical thickness (CT) in temporal and parietal regions in adults with ASD. Both alpha activity and CT have been postulated to depend on the structural and functional integrity of cortical columns.

Objectives: The present study examined CT in children with ASD and investigated CT as a potential anatomical correlate of increased resting state alpha activity. It was hypothesized that compared to typically developing controls, children with ASD would exhibit reduced CT, primarily in temporal and parietal regions. In addition, analyses in the ASD group probed associations between CT and resting-state alpha power.

Methods: Participants were medication-free 6- to 14-year-old children (20 with ASD, 20 controls). These children were a sub-sample from our previous magnetoencephalography (MEG) study who also completed a 3T structural MRI scan. MEG eyes-closed resting state data were processed by projecting each individual’s raw surface activity into source space with a regional source model and applying a Fast Fourier Transform to artifact-free two-second epochs of the continuous data at each of 15 regional sources. Individual spectra were averaged and alpha (8-12 Hz) power quantified. Structural MRI provided T1-weighted images and, using Freesurfer, CT measures were obtained. Whole-brain general linear model (GLM) group analyses of CT were corrected for multiple comparisons and cluster-wise probabilities computed to indicate the likelihood that a given cluster appeared by chance.

Results: Analyses revealed reduced CT in the ASD group in a left frontal region that included portions of the superior and middle frontal gyri (cluster-wise probability = .01) as well as in left and right temporal and parietal regions (all cluster-wise probabilities = .0001). Temporal and parietal regions included bilaterally the superior temporal sulcus, portions of superior, middle, and inferior temporal gyri, and the inferior parietal lobule. CT was also reduced in the ASD group in the left fusiform and parahippocampal gyri, and left temporal pole. Regression analyses in the ASD group revealed associations between increased alpha power in temporal and parietal regional sources and reduced CT in bilateral portions of the superior temporal gyrus, inferior parietal lobule, and precuneus, as well as left superior temporal sulcus and right fusiform gyrus.

Conclusions: Results demonstrated reduced CT in children with ASD, particularly in bilateral temporal and parietal regions. CT reductions were associated with increased alpha power, suggesting a link between atypical brain structure and function in ASD. The regions exhibiting reduced CT and increased alpha power overlap with regions implicated in social processing and default-mode network activity, in which aberrant activation patterns have been observed in ASD. Thus, present findings may point to a common dysfunction in cortical circuitry underlying multiple neural abnormalities in ASD.