Background: Long-range white matter connections are thought to be reduced in the brains of people with autism, contributing to the behavioral and cognitive phenotype of this developmental disorder. The clearest evidence for this underconnectivity hypothesis comes from structural, diffusion, and functional magnetic resonance studies of the corpus callosum in autism. A recent study found evidence that corresponding voxels of cortex in the left and right hemispheres showed reduced functional coupling during resting-state neuroimaging, consistent with reduced connectivity mediated via the corpus callosum (Anderson et al., 2010).
Objectives: Here we corroborate this finding and extend it to an examination of interhemispheric coherence between all regions of cortex, at the whole-brain level. We compare all homotopic and heterotopic coherence values to a large normative sample of neurotypical subjects (Biswal et al., 2010).
Methods: Two independent groups of high-functioning adults with autism (N=11 and 19) and their matched neurotypical controls (N=12 and 20; each paired group imaged on a different scanner with different parameters) were contrasted and compared to reference data from 700 individuals. Diagnosis of an ASD was made following administration of the Autism Diagnostic Observation Schedule (ADOS) and the Autism Diagnostic Interview - Revised (ADI-R), together with clinical interview. Resting-state blood oxygenation level-dependent (BOLD) coherence was computed between all pairwise cortical regions, which were defined in each individual brain using the automated surface-based segmentation provided by FreeSurfer (Fischl et al., 2004). The coherence values were computed by multitaper analysis using the Chronux toolbox (Mitra and Bokil, 2008), providing a frequency- and phase-dependent measure of the strength of coupling between pairs of cortical regions.
Results: We found significantly reduced coherence in both autism groups compared to their respective controls, and compared to the normal reference set. This abnormality was evident across most homotopic regions, and robust to frequency band. By contrast, heterotopic coherence did not differ notably between groups. Homotopic coherence was also reduced when the analysis was performed between corresponding voxels in both hemispheres, instead of an analysis based on segmented cortical regions.
Conclusions: These findings provide additional evidence for long-distance underconnectivity in autism, and point to a specific functional deficit resulting from abnormal callosal connectivity.
Anderson et al. (2010). Decreased Interhemispheric Functional Connectivity in Autism. Cerebral Cortex.
Biswal et al. (2010). Toward discovery science of human brain function. Proceedings of the National Academy of Sciences. 107 (10).
Fischl et al. (2004). Automatically parcellating the human cerebral cortex. Cereb Cortex, 14 (1).
Mitra and Bokil (2008). Observed brain dynamics, Oxford University Press.
See more of: 4th Oral Brain Imaging in ASD temporary
See more of: Brain Structure & Function