Cortical thickness and oscillatory phase resetting: a proposed mechanism of salience network dysfunction in...

Schizophrenia is characterised by both electrophysiological abnormalities and consistent changes in the structure of cortical grey matter. But the relationship between these two observations is largely unknown. Structural changes reported in schizophrenia include reduced grey matter volume, thickness and surface area in several cortical regions, but most frequently in the insula and anterior cingulate cortex. These two regions together constitute an intrinsic brain circuit known as the "Salience Network", which has a key role in stimulus processing. During stimulus processing tasks, evoked activity is noted using electroencephalography (EEG). Phase resetting of ongoing oscillations contributes to this evoked activity. Neuronal oscillations play a crucial role in cerebral recruitment during cognitive tasks, and influencing the oscillatory phase can modulate cortical excitability and the transition between various cognitive states. At a network level, such a transition or switch is thought to be enabled by the Salience Network. In this study, we investigated the relationship between the cortical thickness in the Salience Network (measured using MRI) and the degree of phase resetting observed during an oddball task (measured using EEG) in 18 medicated male patients in a clinically stable phase of schizophrenia and 20 age and gender matched healthy controls. We obtained a measure of partial phase resetting after a stimulus is presented, and a second measure representing mean evoked activity, using the methods proposed by Martinez-Montes. Using MRI analysis, we have firstly shown that there is a significant loss of cortical thickness of regions that constitute the Salience Network in patients with schizophrenia. EEG analysis revealed that in healthy controls, the expected relationship between phase resetting and evoked electrical activity is observed, but in patients with schizophrenia the theta phase resetting is a weak predictor of the activity evoked by attending to a target stimulus, though the difference between the groups did not reach statistical significance in the present sample. Furthermore, in patients with schizophrenia the reduced thickness in the Salience Network is associated with the inefficient phase resetting of theta oscillations. Our findings suggest that the grey matter reduction seen in the Salience Network in patients with schizophrenia has substantial functional consequences. In particular, the structural defect of the insula that is seen in schizophrenia is likely to be associated with less efficient recruitment of brain circuits for processing information. This implies a possible mechanism by which disruptions in the intrinsic Salience Network can result in a general disturbance in salience detection seen in schizophrenia.