Functional Connectivity and Executive Function in Agenesis of the Corpus Callosum

Background: As callosal deficits have been consistently reported in autism, studying individuals with agenesis of the corpus callosum (AgCC) provides unique insight into how disruption of interhemispheric and other long-range connectivity affects behavior and cognition. Moreover, individuals with AgCC may have symptoms in the autism spectrum, suggesting an important mechanistic overlap. Objectives: Here, we examine resting-state functional connectivity of magnetoencephalography (MEG) data in people with anomalies of the corpus callosum (partial absence: pAgCC; complete absence: cAgCC) using novel source-space reconstruction techniques and imaginary coherence. We also compare functional connectivity to performance on tasks of processing speed and executive function, cognitive deficits shared by many AgCC and autistic individuals. Methods: Fourteen participants with AgCC (8 cAgCC, 6 pAgCC) and fourteen IQ and age-matched control subjects enrolled in this study. All individuals with AgCC were evaluated with the Weschler Adult Intelligence Scale III and Delis-Kaplan Executive Function Scale (D-KEFS). Four minutes of resting-state (alert; eyes closed) data was collected using a 275-channel biomagnetometer (VSM MedTech) using a sampling rate of 600Hz. A single, artifact-free epoch (60s) was selected from this period and neural activity during the resting session in the alpha band (8-12Hz) was computed using adaptive spatial filtering techniques (beamforming). Functional connectivity was estimated using imaginary coherence (IC), a metric that overcomes estimation biases in EEG/MEG source-space reconstructions. Connectivity maps between groups were compared (non-parametric unpaired t-test) and voxelwise correlations between IC and D-KEFS scores were computed across AgCC participants. Corrections for multiple comparisons were computed using a false discovery rate (FDR). Results: Our results show significant underconnectivity in the brains of AgCC individuals in two key cortical hubs: left dorsolateral pre-frontal cortex (DLPFC; p<0.05 FDR corrected) and right posterior parietal cortex (PPC; p<0.1, FDR corrected). For both DLPFC and PPC, there is a trend for the pAgCC group to have greater interhemispheric connectivity relative to the cAgCC group. We found that in the right PPC, the AgCC group trends toward greater intrahemispheric connectivity and this finding appears to be equally contributed to by both the cAgCC and pAgCC group. These data suggest that not only is there decreased interhemispheric connectivity in AgCC, but that it is specifically related to critical integrative brain regions. Functional connectivity was related to performance on neuropsychological tests. The degree of connectivity of higher-order auditory regions in the left hemisphere (middle and superior temporal gyrus) positively correlated with verbal processing speed. The degree of connectivity of left DLPFC correlated with problem-solving abilities and second-order language comprehension, as measured by the Tower and Proverbs tests within DKEFS. Conclusions: Congenital disruption of callosal development can result in faulty functional coupling of higher order cortical fields. This can be measured during the brain's "resting state". That performance on cognitive tests correlates with connectivity in critical areas supports the concept that connectivity is a key component of performance generally. These insights may help us to target interventions to AgCC individuals and may help us to understand and treat related disorders of connectivity such as autism and schizophrenia.