Sensory-learning deficits and conversion to psychosis among individuals at clinical high-risk: a longitudinal model-based fMRI study

点击翻译按钮获取中文摘要

Schizophrenia (SCZ) is among the most severe and burdensome of medical conditions worldwide. SCZ research to date has succeeded in describing brain abnormalities in patients but has failed to clarify the mechanisms underlying hallucinations and other core psychotic symptoms of this disorder. A fundamental function of the brain is to learn regularities in the environment and predict upcoming events and their contingencies (e.g., anticipating hearing a voice when answering the phone). Disruptions in predictive mechanisms of the brain, specifically in prediction-error signaling, correlate with delusion propensity and have recently been hypothesized to underlie hallucinations. However, a causal link between predictive mechanisms and psychotic symptoms has yet to be established. Thus, the aim of this proposal is to define the neural correlates of auditory-verbal predictions (i.e., the expectation of hearing voices depending on varying probabilities of the occurrence of speech stimuli during a novel speech discrimination task) in healthy controls (N=20; each will receive one baseline fMRI scan) and individuals at clinical high risk for psychosis (CHR; N=50; each subject will receive an fMRI scan at baseline and after one year or at conversion to psychosis, whichever comes first), and to establish a direct link between these auditory-verbal predictions and auditory- verbal perceptual abnormalities as well as conversion to psychosis. The CHR phase is of particular interest in the study of pathophysiological processes because such processes can be separated from effects of illness, chronicity, treatment and hospitalizations. Abnormalities identified in this early phase are more likely to be specific to the illness, because they may represent more psychosis-specific upstream events rather than downstream markers of general psychopathology that could be shared, as a common final pathway, by different disorders. We hypothesize that there will be deficits in prediction-error signaling in the auditory cortex at baseline in CHR individuals compared to control subjects, and that greater prediction-error deficits at baseline will predict conversion to psychosis, as well as greater symptoms and worse neurocognitive functioning, at follow up. We also hypothesize that individuals who convert to psychosis will have greater worsening of prediction error deficits over time than non-converters. The contribution of this R21 is significant because it will fill a gap in the SCZ literature regarding the mechanisms of symptom generation and has the potential to help define a novel target (sensory predictions) that is amenable to therapeutic modification (e.g., with learning-based, noninvasive approaches). By including patients with a range of perceptual abnormalities in different sensory domains, this project will also establish the specificity of sensory predictive deficits to AVH and psychosis in general. The predictive deficits in the auditory system will likely be state-dependent and may thus provide an objective measure for predicting the development of psychosis and/or monitoring treatment response.