James F. Cavanagh, Ph.D. - US grants

[unreadable] DESCRIPTION (provided by applicant): The proposal aims to investigate how stress is internalized to affect cognitive functioning and increase the risk for Major Depressive Disorder (MOD). Dysfunctional stress reactivity may be a risk factor for MOD, yet mechanisms underlying this process remain unexplained. Both stress and MOD have been identified as compromising factors on "higher level" cognitive control systems and "lower level" sub-cortical systems. Similarly, both stress and MDD may alter an individual's sensitivity to reward and punishment (reinforcement). This proposal aims to identify the roles of frontal cortical and striatal systems in stress and MDD during reinforcement learning. Two participant groups will be tested: a non-depressed (n=75) and a depressed (n=25) group. All participants will first complete a probabilistic reward learning task. Following that task, the depressed group and a randomly assigned two thirds of the non-depressed group will complete a similar task under social stress. The other non-depressed participants will complete the same task under standard conditions. Electroencephalographic (EEC) recordings will be obtained during the tasks, which will allow objective measurement of neural activities reflective of reinforcement learning. Facets of acute stress reactivity (emotional and cortisol reactivity) will be investigated as moderators of the stress-learning link. This design will allow consideration of: 1) how specific neural systems become functionally compromised during high stress reactivity and 2) whether these same systems are compromised in depression. PUBLIC HEALTH RELEVANCE: An understanding of the specific way that stress alters reinforcement sensitivity may reveal distinct neural mechanisms mediating the translation of prolonged stress into ongoing affective distress. The combined use of physiological and computational models to understand how stress, emotion, and cognitive processes contribute to mental disorders such as major depression is innovative, and in line with the high priority statements of the NIMH regarding combined approaches. Future directions may reveal objective risk factors for dysfunctional stress reactivity, providing tailored approaches of diagnosis and treatment for MDD. [unreadable] [unreadable] [unreadable]

Although considered a trans-diagnostic phenotype, anhedonia can emerge from deficits in motivation, valuation, or hedonic appreciation, each of which reflect different neural processes and are differently expressed across individuals. There is a critical need to refine the construct of anhedonia in order to improve treatment. Our long-term goal is to combine computational, imaging, and causal manipulations to define a translational biomarker of diminished valuation in anhedonia. In this proposal we identify how the EEG response known as the Reward Positivity (RewP) is a candidate biomarker specific to value-based deficiencies in anhedonia. The RewP is only elicited by the presentation of a rewarding outcome, it is decreased in depression, and it scales with the central feature of reinforcement learning models, the positive reward prediction error (+RPE). Importantly, this same neural response can be elicited in rodents using the same learning task as in humans. The objective of this proposal is to test whether induced emotion, depressed mood, and learned helplessness (in mice) directly diminish +RPE coding in the RewP. The rationale for this approach is that electrophysiology is a highly promising tool for identifying mechanisms of complex behaviors and translating these mechanisms between species. In Aim 1, we will determine if induced emotion and +RPE have independent or interactive effects on the RewP. In Aim 2 we will recruit depressed participants and determine if anhedonia and +RPE have independent or interactive influences on the source-level generators underlying the RewP (using MEG). In Aim 3 we will use the same task in a mouse model with infralimbic recordings; we will then test the causal diminishment (learned helplessness) and recovery (fluoxetine) of this mechanism. This proposed research is innovative because we have identified a computational function tightly tied to a neural response that directly addresses the disease-specific phenotype in human patients and is capable of being assessed, manipulated, and recovered within a rodent model. This contribution is expected to be significant because it will advance a translational mechanism for deficient valuation in anhedonia. Upon completion of these aims, the expected outcome will validate the RewP as a sensitive and specific mechanism of aberrant valuation in anhedonia. In line with the RDoC framework, our use of computational modeling will allow us to algorithmically contrast multiple sub-constructs of approach motivation in the positive valence systems domain. The translational computational psychiatry approach advanced here links circuit-level dysfunction, aberrant computations, and trans-diagnostic behavioral phenotype. The successful completion of the aims advanced here will create what we think is the most promising path for combining these strengths into a computationally-inspired, mechanistically tested, translatable model of aberrant valuation in anhedonia. This novel candidate biomarker will be translatable between species and testable in an outpatient clinic.