Sensory-processing deficits appear crucial to the clinical expression of symptoms of schizophrenia. The visual cortex displays both dysconnectivity and aberrant spontaneous activity in patients with persistent symptoms and cognitive deficits. In this paper, we examine visual cortex in the context of the remerging notion of thalamic dysfunction in schizophrenia. We examined specific regional and longer-range abnormalities in sensory and thalamic circuits in schizophrenia, and whether these patterns are strong enough to discriminate symptomatic patients from controls.

Using publicly available resting fMRI data of 71 controls and 62 schizophrenia patients, we derived conjunction maps of regional homogeneity (ReHo) and fractional amplitude of low-frequency fluctuations (fALFF) to inform further seed-based Granger causality analysis (GCA) to study effective connectivity patterns. ReHo, fALFF and GCA maps were entered into a multiple kernel learning classifier, to determine whether patterns of local and effective connectivity can differentiate controls from patients.

Visual cortex shows both ReHo and fALFF reductions in patients. Visuothalamic effective connectivity in patients was significantly reduced. Local connectivity (ReHo) patterns discriminated patients from controls with the highest level of accuracy of 80.32%.

Both the inflow and outflow of Granger causal information between visual cortex and thalamus is affected in schizophrenia; this occurs in conjunction with highly discriminatory but localized dysconnectivity and reduced neural activity within the visual cortex. This may explain the visual-processing deficits that are present despite symptomatic remission in schizophrenia.

There is an appreciable overlap in the clinical presentation, epidemiology and treatment response of the two major psychotic disorders – schizophrenia and bipolar disorder. Nevertheless, the shared neurobiological correlates of these two disorders are still elusive. Using diffusion tensor imaging (DTI), we sought to identify brain regions which share altered white-matter connectivity across a clinical spectrum of psychotic disorders.

A sample of 41 healthy controls, 62 patients in a clinically stable state of an established psychotic disorder (40 with schizophrenia, 22 with bipolar disorder) were studied using DTI. Tract-based spatial statistics (TBSS) was used in order to study group differences between patients with psychosis and healthy controls using fractional anisotropy (FA). Probabilistic tractography was used in order to visualize the clusters that showed significant differences between these two groups.

The TBSS analysis revealed five clusters (callosal, posterior thalamic/optic, paralimbic, fronto-occipital) with reduced FA in psychosis. This reduction in FA was associated with an increase in radial diffusivity and a decrease in mode of anisotropy. Factor analysis revealed a single white-matter integrity factor that predicted social and occupational functioning scores in patients irrespective of the diagnostic categorization.

Our results show that a shared white-matter dysconnectivity links the two major psychotic disorders. These microstructural abnormalities predict functional outcome better than symptom-based diagnostic boundaries during a clinically stable phase of illness, highlighting the importance of seeking shared neurobiological factors that underlie the clinical spectrum of psychosis.