Structural cerebellar connectivity in schizophrenia: Support for the cognitive dysmetria theory

626 

Abstract Submission 

Teresa Gomez¹, Sivan Jossinger², John Kruper¹, Adam Richie-Halford³, Michal Ben-Schachar², Jason Yeatman³, Ariel Rokem¹

¹University of Washington, Seattle, WA, ²Bar-Ilan University, Ramat Gan, Israel, ³Stanford University, Stanford, CA

Teresa Gomez  
University of Washington
Seattle, WA

Sivan Jossinger  
Bar-Ilan University
Ramat Gan, Israel

John Kruper  
University of Washington
Seattle, WA

Adam Richie-Halford  
Stanford University
Stanford, CA

Michal Ben-Schachar  
Bar-Ilan University
Ramat Gan, Israel

Jason Yeatman  
Stanford University
Stanford, CA

Ariel Rokem  
University of Washington
Seattle, WA

The cognitive dysmetria theory of schizophrenia (SZ) posits that the core cognitive deficits arise from dysfunctions of cortical-thalamic-cerebellar (CTC) circuits [1]. This theory has received empirical support from fMRI studies, which found increased connectivity in CTC in individuals with SZ [2]. In the present study, we focused on properties of the white matter tissue of the superior cerebellar peduncles (SCPs), a key component of the CTC circuit.

We analyzed diffusion MRI (dMRI) data from the UCLA Consortium for Neuropsychiatric Phenomics LA5c Study [3]. The sample includes 272 subjects (ages: 21-50): 49 with SZ, 49 with bipolar disorder, 41 with ADHD, and 123 healthy controls (HC). DMRI data were collected with 2 mm isotropic resolution and 1,000 s/mm2 b-value in 64 directions. The data were processed using QSIPrep. The SCPs were identified in each individual using pyAFQ (https://yeatmanlab.github.io/pyAFQ/) and anatomical criteria that capture the known decussation of this bundle [4] (Figure 1). Because the sample contains data of varying quality and the cerebellum was not always covered, visual QC of each subject's SCP was conducted by two observers (TG and AR, blind to group). A subject's data were included only if both observers considered SCP delineation to be in the correct anatomical location and with the expected decussation. This included 46 SZ, 39 bipolar, 35 ADHD, and 94 HC. Samples of matched case/control were constructed from by simultaneously matching on age, sex and data quality (quantified as raw neighbor correlations). We focused on tract profiles of the mean diffusivity (MD) and fractional anisotropy (FA) calculated with DTI. Statistical differences were evaluated using tractable (https://yeatmanlab.github.io/tractable/), which models the tract profiles using generalized additive models (GAMs), accounting for the shape of the tract profiles along with variability among participants [5]. Model covariates included age, sex, and raw neighbor correlations.

Statistically significant differences were found in MD tract profiles of the left SCP, where individuals with SZ had lower MD than the matched controls (p<0.05; Figure 2). No other significant differences were found. In particular, individuals with ADHD and bipolar were no different from matched controls in SCP tissue properties.

Previous literature has associated schizophrenia with global abnormalities in diffusion measures, primarily measured as a reduction in FA [6]. Here, we found relatively decreased MD in the SCP, a component of the CTC. Lower MD could indicate increased myelination in SZ and therefore increased connectivity. increased density and directional coherence (but not axonal diameter) may also have similar effects on MD. Thus, these results may be in line with previous fMRI results that found increased functional connectivity in the CTC in individuals with SZ [2], and further supports the cognitive dysmetria theory of SZ.

Psychiatric (eg. Depression, Anxiety, Schizophrenia) ¹

Diffusion MRI Modeling and Analysis ²

Subcortical Structures

White Matter Anatomy, Fiber Pathways and Connectivity

Informatics Other

Cerebellum

Psychiatric

Schizophrenia

Statistical Methods

Tractography

White Matter