LAUSR

Abstract Background Auditory hallucinations are a prevalent, debilitating symptom of schizophrenia (Sz). Lack of detailed phenomenological assessments of perceptual disturbances in large psychiatric imaging datasets limits our ability to disentangle the underlying neural mechanisms of hallucinations. Our study investigates how changes in local functional communication dynamics may be associated with wide-ranging auditory disturbances in Sz. Methods Local functional connectivity was estimated using regional homogeneity (ReHo) analysis of resting fMRI data, which quantifies synchronization of fMRI activity of a voxel to its neighboring voxels. Resting fMRI data of 99 Sz patients was analyzed (mean age=36.2±13.3 y, sex=71/28 m/f); Auditory perceptual disturbance in the past week was estimated using the auditory perception state (APS) subscale score of the recently validated Auditory Perceptual Trait and State Scale (http://www.mdbrain.org/APTS.pdf). Voxelwise regression analysis of ReHo was performed including APS score as a regressor of interest. Significant results were thresholded using AFNI’s 3dClustSim with autocorrelation function option to yield corrected p<0.05, corresponding to cluster-size threshold of 49 voxels at voxelwise p=0.001. Results Higher APS scores were associated with reduced ReHo in clusters in left putamen, right putamen, left temporoparietal junction, and right hippocampus. In a follow-up analysis using these clusters as seeds in whole-brain resting-state functional connectivity analysis (rsFC) analysis, higher APS scores were significantly associated with reduced rsFC between the right putamen seed and clusters in the contralateral putamen and auditory cortex. Discussion Our findings are consistent with those of a prior study that reported abnormal ReHo in left and right putamen of a unmedicated first-episode Sz patients (Cui et al. 2016). However, in that small sample (n=32), striatal ReHo was elevated relative to controls, and AH severity was not significantly correlated with striatal ReHo measures. Our study investigated ReHo in a large sample of chronic, medicated patients (all except 4 were taking antipsychotic medication at time of study). While it is widely accepted that striatal signaling is disrupted in Sz, future work is needed to better understand how striatal signaling deficits may change over the course of illness and how this relates to particular symptoms such as hallucinations. Implications for development of novel therapies that account for these nuanced findings will be discussed.