Fronto-Thalamic Structural Connectivity Associated With Schizotypy, a Psychosis Risk Phenotype, in Nonclinical Subjects

Abstract

Background and Hypothesis Schizotypy is a risk phenotype for the psychosis spectrum and pilot studies suggest a biological continuum underlying this phenotype across health and disease. It is unclear whether this biological continuum might include brain structural associations in networks altered in schizophrenia spectrum disorders, such as the fronto-thalamo-striatal system or nodes of the default mode network, such as the precuneus. Study Design In this study, we analyze a large multi-center cohort of 673 nonclinical subjects phenotyped for schizotypal traits (using the Schizotypal Personality Questionnaire-Brief version) using tract-based spatial statistics of diffusion tensor imaging data, as well as voxel-based morphometry (VBM) analysis of regional brain volumes and gyrification analysis of early neurodevelopmental markers of cortical folding on T1-weighted MRI. Study Results We identify significant (P < .05 family-wise error corrected) associations of schizotypy with major fiber tract fractional anisotropy: positive (cognitive-perceptual) schizotypy correlated negatively with the left anterior thalamic radiation (a principal thalamo-frontal projection), left uncinate fasciculus and cingulum, while negative (interpersonal) schizotypy correlated positively with left anterior thalamic radiation, cingulum, and the anterior corpus callosum, and disorganized schizotypy correlated negatively with right cingulum, and superior and inferior longitudinal fasciculi. VBM analyses showed a negative correlation of gray matter with negative schizotypy in the left cerebellum, while gyrification in the inferior parietal cortex correlated positively with negative (interpersonal) schizotypy. Conclusions These findings pave the way for a neural network conceptualization of schizotypy as a psychosis proneness trait across the general population, showing associations with fronto-subcortical and frontotemporal systems as structural substrates of this risk phenotype.