Delay Discounting as a Transdiagnostic Neural Phenotype

One of the most exciting applications of clinical neuroscience is to explain and predict individual differences in psychiatric symptoms. However, reproducibility remains a central challenge to this work because studies often report conflicting results and growing evidence suggests that most neuroimaging studies do not have adequate power to detect reliable effects. My dissertation aims to address reproducibility in clinical neuroscience and demonstrate a symptom-specific link to a neurobehavioral dimension across psychiatric disorders. In particular, my work focuses on delay discounting, one of the recommended measures for brain-behavior phenotyping in transdiagnostic research. Delay discounting involves choosing between a smaller sooner reward and a larger reward available after a longer delay and has been thought to capture impulsivity and self-control. In Chapter 1, I use a meta-analytic approach to identify a set of analytic strategies that produce reliable neural effects across delay discounting fMRI experiments. In Chapter 2, I use two large independent fMRI datasets, one that consists of psychiatrically healthy adults and another that consists of individuals with mood and psychotic disorders, to systematically evaluate these neural effects and their links to individual differences in discounting. Both Chapters 1 and 2 conclude that contrary to many claims, brain activity does not differ when choosing smaller sooner versus larger later rewards, or that “impulsive” and “future-oriented” individuals recruit different brain regions when faced with these choices. Instead, computationally derived variables, such as subjective value, drive reliable signals during this task that vary as a function of individual differences in discounting. These results suggest that neural signals during delay discounting may be more relevant to motivation and reward valuation, rather than impulsivity and self-control. In Chapter 3, I implement this approach and demonstrate a transdiagnostic link between value-related brain activity and motivational symptoms across individuals with mood and psychotic disorders. The implications of this work are two-fold. First, it highlights the importance of systematically evaluating existing analytic approaches to improve reproducibility in clinical neuroscience. Second, it shows that delay discounting paradigms reliably elicit value-related activity in the brain, which varies with individual differences in motivational symptoms across disorders.