Identifying Neural Signatures of Auditory-Predictive Processing in Schizophrenia: A Multi-Modal Imaging Approach

Understanding how the brain processes and transmits information is one the fundamental challenges for current basic and medical research. Recent evidence suggests two broad classes of processes can be distinguished that seem to support different functions and are characterized by distinct biological correlates: 1) a "feedforward" mode that transmits information based upon the characteristics of the incoming stimulus and 2) a "feedback" mode that is governed by the internal activity of the brain, such as expectations and predictions about events. This distinction may be fundamental for gaining novel insights into how the brain operates under normal circumstances and how changes in these two different modes may contribute to psychiatric disorders, such as schizophrenia (ScZ).
Until now, distinguishing these different brain modes using non-invasive brain imaging, such as functional magnetic resonance imaging (fMRI) or Magnetoencephalography (MEG), has been challenging. However, novel evidence from basic anatomy and biology has suggested that distinct brain waves at different frequencies as well as particular brain layers may support these different brain modes. As a result, we will attempt for the first time to identify these brain modes through using state-of-the-art brain imaging and thus gain a new understanding of how the brain transmits information and how these processes might contribute to ScZ.
In the first part of the project, we will present healthy volunteers sequences of sounds while they watch a movie. During this task, we measure their brain waves with a MEG-machine. In particular, we are interested in finding out whether changes in rhythms of neural activity, so-called "oscillations", may be influenced by the presence of sounds that deviate in duration. In particular, we aim to show that the flow of these oscillations between brain regions will change depending on whether a sound is different or not.
The MEG-recordings will be accompanied by fMRI-measurements at 7 Tesla. In contrast to the majority of fMRI-research which is carried out with a field strength of 3 Tesla, we expect that fMRI-recordings at 7 Tesla reveal novel details about brain activity that cannot be observed with conventional fMRI-machines. In particular, based on our prior work in this area, we expect that we can observe brain activity in different layers which may be crucial for gaining new insights into how the brain uses different channels to communicate.
Based on these new insights, we will then apply this framework to understand changes in brain activity in ScZ-patients and young people who are at high-risk for developing the disorder. ScZ is a common mental disorder which is associated with a range of complaints, including hallucinations and delusions. These symptoms of psychosis are accompanied by pronounced impairments in perception and cognition. A better understanding of cognitive deficits is particularly important because current treatments are unable to improve perception and memory functions which result in difficulties of patients' to organize their lives and maintain employment.
We expect that our ability to distinguish different brain modes will allow us to identify the cause of patients' difficulties in perceiving the world and their problems in organizing their thoughts. Specifically, we will identify whether the problem for patients with ScZ is to register the information coming into the brain or whether their problems lies more in controlling their thoughts and perceptions through prior assumptions which are generated in higher brain areas. As a result, we expect that in addition to identifying the causes of ScZ, this approach may be relevant for novel therapies and early detection and diagnosis as it could inform whether therapies should focus on improving the ability to perceive auditory and visual information as opposed to focussing on the assumptions and thoughts about the world a patient may have.

In this project, combined MEG and high-field 7T fMRI are used to identify electrophysiological signatures and layer-specific information flow of feedforward (FF) vs. feedback (FB) mediated information processing during auditory predictive processing in normal and abnormal brain functioning. To this end, 30 healthy controls, 30 ScZ patients and 30 CHR-participants will be recruited. Data will be obtained during a novel auditory oddball paradigm that involves a quantitative manipulation of predictability of sound sequences.

Specifically, the contributions of spectral and layer-specific fMRI-signatures of FF- and FB-processing towards auditory-predictive processing during normal brain functioning and in ScZ will be identified. For the MEG and 7T fMRI-data, time-frequency representations will be estimated in three regions of interest (ROIs) (A1, A2, IFG). MEG-data will be analyzed for ERF-signals, time-frequency information and Granger causality. For the 7T fMRI data, we will use a DCM model that reflects the parametric variation of predictability seen in the MMN response in the MEG-data. Moreover, MEG and 7T fMRI-data will be complemented by Magnetic Resonance Spectroscopy (MRS) measurements of GABA and Glutamate/Glutamine (Glx) in A1/2 as well as in IFG as well as computational modeling to relate neuroimaging data to alterations in Excitation/Inhibition balance (E/I-balance), a crucial pathophysiological mechanism in ScZ.

The approach is expected to elucidate the role of FF and FB processing during both normal and abnormal brain functioning with high-spatial, spectral and temporal resolution. In addition, fundamental insights into underlying mechanisms of cognitive deficits in ScZ are expected, informed by the relationship and directional influence found between aberrant bottom-up information processing and top-down cognitive impairments.

Understanding the way the brain processes information is a fundamental question in cognitive neuroscience. The proposed project will address this challenge through a state-of-the-art neuroimaging approach that combines MEG and high-field 7T fMRI to examine the neurobiological signatures of feedforward (FF) and feedback (FB)-processing. To this end, we will characterize for the first time comprehensively the relationship between neural synchrony and layer-specific, directional, flow of information across the auditory during normal brain functioning and in a psychiatric disorder.

This approach will be used to understand the neurobiological origins of schizophrenia (ScZ), a debilitating mental illness with a lifetime prevalence of approximately 1% which leads to enormous economical and social costs. This is due to the fact that the pathophysiology is still unclear and the existing treatments are largely ineffective in targeting the pronounced cognitive and physiological dysfunctions. While a considerable amount of data exists on aberrant brain activity in ScZ, the current study goes significantly beyond the state-of-the-art through the application of a unique neuroimaging approach that allows for a layer- and network-specific separation of FF- and FB-processes. Accordingly, we expect that our proposed research will have wide-ranging implications concerning the basic understanding of cognitive deficits in ScZ. In addition, we expect fundamental insights into the organization of large-scale networks as well as into the possibility to use high-field 7T imaging to identify layer-specific BOLD-activity and its relationship to FF- and FB-processing during auditory predictive processing.

In essence, the benefits will be three-fold. Firstly, the project will provide novel insights into the neural mechanisms underlying FF- and FB-processing using a state-of-the-art neuroimaging approach in both healthy individuals, ScZ-patients and in participants at clinical high-risk (CHR) for ScZ. Secondly, our research will elucidate the relationship between sensory predictions, neural oscillations and layer-specific BOLD-activation patterns at an unprecedented level of temporal- and spatial resolution within the same participants, thereby bridging the gap between data from animal electrophysiological recordings and human brain imaging that will inform the development of biomarkers for the early detection and diagnosis of the disorder. Finally, because delineating FF and FB-processing across functional brain networks can more specifically diagnose underlying causes of impaired cognitive processing in ScZ, the results of this project can open up the development of novel pharmacological and behavioural treatment options for these patients, and will therefore be of interest to the ScZ-community.

The academic community will be reached via its standard ways of dissemination at conferences and in high impact journals aiming not only at researchers in systems and cognitive neuroscience but also researchers in the field of ScZ as well as the wider academic audience interested clinically or generally in neural synchrony and the application of MEG. The wider public will be informed in an appropriate manner via internet, radio, television, and specific publications in outlets aimed at such an audience.

Finally, we will specifically target the pharmaceutical industry via established networks within the Institute of Neuroscience and Psychology (INP) to disseminate our findings to users in clinics and using their feedback to identify potential attendees for a dedicated workshop to disseminate our findings in concentrated form and to identify potential synergies for the future.