Harvard Catalyst Profiles

Converging lines of evidence support the hypothesis that deviations from typical brain structure development take place prior to psychosis onset, while ‘big data’ neuroimaging studies of adults with psychosis find subtle, widespread gray matter disruptions in the brain. In this proposal, we will synergize knowledge about normative structural neurodevelopment and findings of structural brain aberrations in adults with psychosis to develop cost-effective brain-based markers of psychosis risk in youth. To improve identification of those at greatest risk, we leverage results from large-scale structural neuroimaging studies of psychosis to create a ‘Psychosis Neuroimaging Score’, a cumulative summary score that reflects one’s psychosis liability. We are first transporting the Psychosis Neuroimaging Score to youth by incorporating crucial aspects of structural brain development. We are characterizing the normative developmental trajectory of the Psychosis Neuroimaging Score by harmonizing many archival datasets of normative development (N>5,000, 2-30 years old). We are evaluating how greater age-associated deviation from the aggregate Psychosis Neuroimaging Score differentiates youth with psychosis spectrum symptoms from typically developing youth in the Philadelphia Neurodevelopmental Cohort (N=1209, 10-22 years old). We then will examine how greater age-associated deviation from the aggregate Psychosis Neuroimaging Score predicts distinct developmental trajectories associated with psychotic-like experiences in youth from the Adolescent Brain and Cognitive Development Study (N=11,875). We will also assess the extent to which known psychosis risk factors (e.g., family history of psychosis, obstetric complications, trauma) contribute to characterization of these trajectories. Finally, we propose to use measurement-in-error modeling to establish a functional relationship between Psychosis Neuroimaging scores generated from 3T MRI scans and those generated using low-field MRI scans in a community sample of youth. Results from this study will allow us to create more affordable, clinically accessible biological indicators of severe psychopathology, ultimately improving identification of young people at greatest risk and allowing earlier more effective interventions. Student role: The student involved in this project will have the opportunity to assist with data collection (behavioral and MRI), data processing, cleaning, visualization, and analyses of developmental effects of 3T and low-field MRI data. There will be opportunities to contribute to manuscripts and poster presentations.

Adolescence is a dynamic developmental stage in which health trajectories can pivot, in both negative and positive directions. Understanding normative adolescent development thus provides an essential template for detecting early deviations from maturational trajectories, along with actionable windows in which intervention can maximally impact health trajectories. Normative models of sleep physiology over adolescence carries high potential value for early detection of brain-based health risk and subsequent effective intervention. Sleep macro- and micro-architecture undergoes dramatic change over adolescence; plays an active role in sculpting the structural and functional maturation of the brain; predicts long-term brain, behavior, and health trajectories; and is modifiable with non-invasive biobehavioral intervention. To improve sleep-based risk assessment in adolescents, we are creating normative growth charts for sleep physiological features in typically developing young people using two complementary sleep measurement methods: gold-standard polysomnography and a wearable sleep recording device (Dreem3 headband). First, we are examining age trends in sleep physiology in typically developing young people with polysomnography data harmonized across multiple existing cohorts. In tandem, we plan to use the Dreem3 headband to collect home sleep recordings in typically developing young people. A subset of participants will also complete baseline polysomnography and we aim to recapitulate the age-related trends observed in the gold standard measure. We then plan to examine key modifiers of age trends. Completion of our aims will provide an essential template for the study of adolescent sleep physiology, a tool to improve detection of suboptimal sleep in the real-word, and an empirical basis for developmentally sensitive sleep interventions. Student role: The student involved in this project will assist with data processing, cleaning, visualization, and analyses of developmental effects of polysomnography data and data from Dreem3 EEG headbands. There will be opportunities to contribute to manuscripts and future grant submissions.