ABSTRACT
There is a substantial and robust body of evidence describing cerebellar abnormalities in autism
spectrum disorder (ASD). In parallel, functional connectivity abnormalities have been extensively documented
in the brains of individuals with ASD. While most such studies focused on cortico-cortical connections, studies
that did look at cortico-cerebellar connections document multiple abnormalities in ASD, using low temporal
resolution fMRI. Yet, to date, there are no studies of cortico-cerebellar functional connectivity abnormalities in
ASD using high temporal resolution methods. Mapping out functional connectivity abnormalities in ASD with
high temporal specificity is critical, as functional connectivity abnormalities can differ substantially, and may
even manifest in opposite directions, in different frequency bands, as previously shown by our group. This gap
is largely due to the challenges associated with source localizing the sensors data from such methods to the
cerebellum. To put our understanding of cortico-cerebellar functional connectivity abnormalities on par with our
understanding of functional connectivity abnormalities in ASD more generally, it is necessary to extend high
temporal resolution studies of abnormal functional connectivity in ASD to the cerebellum.
This proposal is motivated by the fact that has recently been shown by our collaborator and co-
investigator, MSH, and his colleagues, that MEG sensor data can in fact be source localized reliably and
robustly to the Crus I and Crus II areas of the cerebellum, and that MEG is superior in that aspect to EEG.
Given the extensive evidence of abnormalities in Crus I and Crus II in ASD, and the roles of Crus I and Crus II
in cognitive domains that are relevant to the ASD phenotype, these state-of-the-art developments create an
exciting opportunity to gain a deeper insight into cortico-cerebellar circuitry abnormalities in ASD.
We propose to do this using a “proof of concept” design, where these novel approaches will be applied
to investigate resting state cortico-cerebellar functional connectivity abnormalities in ASD using an existing
MEG resting-state dataset that consists of 72 behaviorally characterized ASD participants ages 7-21, and 79
age, non-verbal IQ, and handedness matched typically developing (TD) participants. Specifically, we propose
to compare the TD and ASD group on cortico-cerebellar functional connectivity, between Crus I & II in the
cerebellum, and cortical areas known to have abnormal responses in ASD, at the theta through beta frequency
bands (~3Hz to ~30Hz), using both coherence (Aim 1) and phase amplitude coupling (Aim 2A). We will also
test for correlations between the coherence and phase amplitude measures (Aim 2B). Lastly, we will test
whether the individual behaviorally assessed ASD severity measures are correlated with the functional
connectivity abnormalities documented as part of Aims 1 and 2.