Thought disorder and social cognition in clinical risk states for schizophrenia

点击翻译按钮获取中文摘要

Project Summary In an effort to intervene before psychosis onset and prevent morbidity, a major recent focus in schizophrenia research has been the identification of young people during a putative prodromal period, so as to develop safe and effective interventions to modify disease course. Over the past decade, studies at Columbia and elsewhere have evaluated clinical high-risk (CHR) individuals across a wide range of cognitive processes to try to identify core deficits of schizophrenia evident before psychosis onset. Subthreshold thought disorder and impaired emotion recognition have emerged as profound deficits that predate, rather than follow, psychosis onset and thus may be indicators of schizophrenia liability, consistent with studies in other risk cohorts, including genetic high risk. Further, subthreshold thought disorder and emotion recognition deficit are significantly correlated, suggesting shared neural substrates in temporoparietal regions. This study aims to identify the neural mechanisms that underlie subthreshold thought disorder and emotion recognition deficit in 125 CHR individuals followed prospectively for psychosis outcome. CHR cohorts are enriched with early cases of schizophrenia, as 20-25% develop schizophrenia and related psychotic disorders within 1-2 years. CHR cohorts may be optimal for studying core characteristics of illness as they otherwise have low-level symptoms, less illness chronicity and minimum exposure to antipsychotics. 25 individuals with schizophrenia and 50 healthy volunteers are included for comparison. Subthreshold thought disorder and emotion recognition deficits will be studied across behavioral, physiological and circuit levels. For thought disorder, we will use automated speech analysis approaches developed in collaboration with IBM to identify constituent impairments in semantics and syntax, and a listening task that elicits reliable activation in language circuits. Our automated machine-learning approach to speech analysis, informed by artificial intelligence, derives the semantic meaning of words and phrases by drawing on a large corpus of text, similar to how humans assign meaning to what they read or hear. Emotion recognition will be measured using standard tasks, naturalistic tasks with dynamic face stimuli and parametric face morph tasks that discriminate between perception and appraisal; task-related BOLD activity will be used to identify relevant circuits. Associations with basic sensory impairment will be tested, including novel auditory mismatch negativity paradigms. Resting state functional connectivity (RSFC) methods will be used for circuit-level analysis of language production and emotion recognition across stages of illness, to determine unique and shared substrates of these constructs in early schizophrenia. If successful, this proposal will identify neural targets for remediation of cognitive impairments.