Date of Award

Spring 5-15-2020

Author's School

McKelvey School of Engineering

Author's Department

Biomedical Engineering

Degree Name

Doctor of Philosophy (PhD)

Degree Type



Psychosis refers to a debilitating set of symptoms that impacts individuals, their communities, and society at large. Current psychiatric nosology treats psychosis as a categorical construct. However, recent evidence suggests that a dimensional approach that cuts across extant nosological boundaries may more accurately represent the underlying phenomena contributing to dysfunction in psychosis. One putative domain of transdiagnostic variation is cognitive control, a construct that refers to the set of functions that enable and support goal-directed behavior and regulation of one’s thoughts and actions. Previous analyses in both healthy individuals and individuals with psychosis have led to a number of findings in structural, resting-state, and task magnetic resonance imaging, however it remains unclear how results relate across modalities and along the psychosis spectrum in support of cognitive control. To address this, the present work first used data-driven analysis methods to identify multimodal correlates of cognitive control in a healthy community cohort. Next, these results were replicated using both predictive and independent analysis methods in an independent healthy community cohort from the same study. Analyses were then extended to individuals with psychosis. Results from the first healthy cohort were used to predict cognitive control performance in a transdiagnostic psychosis cohort consisting of healthy controls, persons with bipolar disorder, and persons with schizophrenia. Finally, an independent analysis in the psychosis cohort was performed to identify novel patterns of variation. Results identified a set of replicable findings in the healthy population that suggest positive associations across modalities and included contributions from known cognitive control regions, canonical restingstate network organization, as well as strong contributions from visual regions. Analyses using results from the healthy cohort to predict performance in the psychosis cohort identified significant relationships in two out of five modalities, further supporting transdiagnostic conceptualizations of psychosis. Independent analysis of the psychosis cohort identified neural contributions that were highly similar to those found in the healthy cohort and also significantly correlated with cognitive control performance. Together, findings support transdiagnostic conceptualization of psychosis and provide targets for future study and may aid efforts to move beyond the existing categorical nosology and improve diagnosis and treatment of psychosis.


English (en)


Deanna M. Barch

Committee Members

Dennis Barbour, Vince Calhoun, Sanmay Das, Nico Dosenbach,


Permanent URL: https://doi.org/10.7936/5fgy-5717