Exploring Atypical Connectivity in Autism Using Graph Theory and Electroencephalography

Thursday, May 12, 2016: 11:30 AM-1:30 PM
Hall A (Baltimore Convention Center)
R. A. Bethlehem1, M. G. Kitzbichler2, J. Freyberg1, E. Ruzich1, S. K. Crockford1 and S. Baron-Cohen1, (1)Autism Research Centre, University of Cambridge, Cambridge, United Kingdom, (2)Department of Psychiatry, Brain Mapping Unit, University of Cambridge, Cambridge, United Kingdom

Atypical neural connectivity has been proposed to be a potential hallmark of autism, with hypotheses including decreased fronto-posterior and enhanced parietal-occipital connectivity, reduced long-range and increased short-range connectivity, and temporal binding deficits. However, empirical findings vary substantially depending on the aspects of connectivity examined, the developmental stage of the individual, the spatial and temporal scales, task versus no-task conditions, how motion artefacts are handled, and the specific neural systems under consideration.


To explore the atypical connectivity hypothesis of autism with functional connectivity (FC) analysis of resting-state electroencephalography (EEG) data, using graph theoretical analysis, a non-biased, model-free, data-driven approach.


Resting-state EEG data (4-minute with 1-minute alternating eyes-closed, eyes-open) were recorded from 14 adults with autism and 34 neurotypical controls matched for age and IQ. We used data-driven graph theory methods to study FC during the eyes-closed condition of these data. Functional connectivity was estimated using envelope correlations that are more robust against the problem of volume conduction in EEG. In contrast to fMRI, EEG data arguably suffer less from head-motion artefacts and provide a much higher temporal resolution to study network dynamics. 


There were weak whole-brain effects combined with heterogeneous local effects. Consistent with prior literature, differences were found in the alpha and beta frequency range. Specifically, in the alpha frequency range there was a trend towards increased clustering of positive correlations in occipital regions in the neurotypical group, relative to the autism group. Conversely, in the beta frequency range there was increased clustering of negative correlations in central and frontal regions in the neurotypical group relative to the autism group. Most prominently we found higher absolute coupling between frontal and occipital regions in the gamma frequency range in the control group. As the sign of this coupling in the gamma frequency is negative, it means there was less decreased coupling and this is thus interpreted as relative increased functional connectivity between occipital and frontal regions in ASC, that was most prominent in frontal regions.


The results reveal no clear whole-brain connectivity pattern associated with autism. There are numerous local differences between the neurotypical and autism groups. Most prominently there was a slight but robust increase in functional connectivity between occipital and frontal regions in ASC in the gamma frequency range. Gamma osciliations have previously been associated with a possible excitation-inhibition balance in visual cortex and it is possible the present hyperconnectivity during resting state conditions is the result of a similar excitation-inhibition imbalance.