Cerebellar Tdcs Modulates ASD-Relevant Circuits and Behaviors

Background: Differences in cerebellar structure and function are consistently implicated in autism spectrum disorder (ASD), with converging findings highlighting right lobule VII (RVII, which includes Crus I and Crus II) as a region of interest in ASD. In humans, RVII is part of fronto-parietal and default mode (DMN) networks, and is engaged during social cognition and mentalizing tasks. Our previous work has shown that grey matter volumes in RVII correlate with core ASD symptoms. In rodent models, specific disruption of RVII is sufficient to produce ASD-like behaviors, including social deficits, repetitive behaviors, and cognitive inflexibility. Crucially, stimulation of RVII rescued social deficits in a Tsc1 ASD mouse model. This suggests that RVII has a critical modulatory role on social behaviors.

Objectives: We aimed to clarify the role of cerebellar RVII in ASD through combining functional neuroimaging with non-invasive neuromodulation in typically-developing adults and adults with ASD.

Methods: Study 1 examined the impact of neuromodulation with anodal (“excitatory”) or sham transcranial direct current stimulation (tDCS) targeting RVII on functional connectivity in 33 typically developing adults (11 males, 22 females, age 23.8±2.8 years; 18 in anodal group, 15 in sham group). In Study 2, 11 typically-developing adults (all males; age 20.9±2.5 years) and 4 adults with ASD (all male; 28.2±14.0 years) underwent simultaneous tDCS-fMRI while performing a social ball-playing task. Each participant completed three sessions, receiving anodal, cathodal (“inhibitory”) or sham tDCS in separate sessions. In both studies, participants received 20min of 1.5mA tDCS targeting RVII using the NeuroConn MR tDCS system. Functional MRI data was acquired pre-, during- and post-tDCS.

Results: In Study 1, functional connectivity prior to neuromodulation replicated previous findings showing that RVII is functionally connected to both the fronto-parietal cognitive control network and the DMN. Post-tDCS, functionally connectivity increased between RVII and DMN nodes, including the medial prefrontal cortex and the precuneus. These regions tend to show reduced functional connectivity in children with ASD. During tDCS, functional connectivity was altered between RVII and a region in the inferior parietal lobule that shows atypically increased functional connectivity in ASD populations. In Study 2, cerebellar tDCS modulated performance on the social ball-playing task, with anodal tDCS modulating learning of the patterns of social reciprocity amongst the players. Activation patterns provided further support for the hypothesis that the cerebellum is important in social learning.

Conclusions: Our findings indicate that the cerebellum is part of the distributed neural circuits implicated in ASD, and further suggest that the cerebellum modulates ASD-relevant networks and social learning. These results clarify the role of cerebellar RVII in ASD, and provide further evidence that cerebellar dysfunction could lead to social deficits in autism. The translational potential of RVII as a target for non-invasive neuromodulation in ASD will be discussed.