Voxel-based, brain-wide association study of aberrant functional connectivity in schizophrenia implicates thalamocortical circuitry

Abstract

BACKGROUND: Wernicke's concept of 'sejunction' or aberrant associations among specialized brain regions is one of the earliest hypotheses attempting to explain the myriad of symptoms in psychotic disorders. Unbiased data mining of all possible brain-wide connections in large data sets is an essential first step in localizing these aberrant circuits. METHODS: We analyzed functional connectivity using the largest resting-state neuroimaging data set reported to date in the schizophrenia literature (415 patients vs. 405 controls from UK, USA, Taiwan, and China). An exhaustive brain-wide association study at both regional and voxel-based levels enabled a continuous data-driven discovery of the key aberrant circuits in schizophrenia. RESULTS: Results identify the thalamus as the key hub for altered functional networks in patients. Increased thalamus-primary somatosensory cortex connectivity was the most significant aberration in schizophrenia (P=10(-18)). Overall, a number of thalamic links with motor and sensory cortical regions showed increased connectivity in schizophrenia, whereas thalamo-frontal connectivity was weakened. Network changes were correlated with symptom severity and illness duration, and support vector machine analysis revealed discrimination accuracies of 73.53-80.92%. CONCLUSIONS: Widespread alterations in resting-state thalamocortical functional connectivity is likely to be a core feature of schizophrenia that contributes to the extensive sensory, motor, cognitive, and emotional impairments in this disorder. Changes in this schizophrenia-associated network could be a reliable mechanistic index to discriminate patients from healthy controls.