Developing concurrent transcranial magnetic stimulation and electroencephalography to study prefrontal cortex neurophysiology in people with schizophrenia.

Cognitive impairment such as deficits in working memory - the ability to retain and manipulate information over a brief period - are core features of schizophrenia. Such deficits may result from dysfunctional cortical inhibition in the dorsolateral prefrontal cortex (DLPFC). One method potentially suited to studying inhibitory circuits in the DLPFC is concurrent transcranial magnetic stimulation and electroencephalography (TMS-EEG). Merging these methods is technically challenging, resulting in artifacts which obscure recording of TMS-evoked neural activity. Furthermore, little is known about the mechanisms that underlie TMS-evoked cortical potentials (TEPs) from the DLPFC. The broad aim of this thesis was to develop and validate TMS-EEG methods to study DLPFC neurophysiology in people with and without schizophrenia. Five studies are reported. The first describes EEG artifacts following TMS. Phantom heads (melons) and human participants were used to investigate the effects of different experimental arrangements (stimulators, pulse types, stimulation site, intensity, paired-pulse conditions) on TMS-evoked EEG artifacts. This study demonstrated the pervasive nature of TMS-evoked scalp muscle artifacts over the DLPFC. In the second study, independent component analysis (ICA - a method of blind source separation) was assessed to identify and remove artifacts from EEG recordings following TMS over the DLPFC. Five sub-types of artifact were identified including muscle, blink and auditory artifacts. We provided evidence that each of these artifacts could be removed with reasonable confidence using ICA, revealing otherwise obscured TMS-evoked neural activity. In study three and four we examined the underlying mechanisms of TEPs. We compared suppression of TEPs with motor evoked potentials (MEPs) following long-interval cortical inhibition (LICI - a paired-pulse TMS paradigm) over the motor cortex (study three) and variations in TEP LICI over the DLPFC between individuals (study four). We demonstrated that modulation of LICI strength differed between TEP peaks, suggesting early peaks (P30, N40) reflected excitatory neurotransmission, whereas latter peaks (N100) reflected the inhibitory mechanism responsible for LICI. In the final study, we compared TMS-evoked DLPFC network properties between people with and without schizophrenia. People with schizophrenia displayed a reduced N100 and reduced TMS-evoked high frequency oscillations in fronto-parietal and interhemispheric networks compared with controls. Importantly, TMS-evoked gamma oscillations (30-45 Hz; dependent on cortical inhibition) in the DLPFC were particularly reduced in a sub-group of schizophrenia participants with low working memory capacity. These findings support impaired inhibitory neurotransmission in the DLPFC of people with schizophrenia and suggest the ability of the DLPFC to generate high frequency oscillations may contribute to schizophrenia-related working memory deficits. This thesis describes the application of TMS-EEG to study cortical neurophysiology in both healthy and disease states. The findings demonstrate that TMS-evoked neural activity can be recorded from the DLPFC following artifact removal and provide insight into inhibitory mechanisms within the DLPFC. Moreover, alterations in DLPFC function assessed using TMS-EEG may underlie reduced working memory capacity in people with schizophrenia. These findings have significant implications for the development of TMS-EEG as a neurophysiological technique, our knowledge of inhibitory mechanisms in the human cortex and our understanding of working memory deficits in schizophrenia.