Combining Computational Methods, RDoC, and Big Neuroimaging Data to Understand Mechanisms of Neuropsychiatric Symptoms in Alzheimer's Disease - More than 6 million Americans suffer from Alzheimer’s disease (AD), the most common age-associated, neurodegenerative dementia. 80% of AD patients also exhibit neuropsychiatric symptoms (NPS), including depression, anxiety, agitation, aggression, apathy and others. NPS in AD respond poorly to conventional treatments and can lead to severe functional impairment and consequent increased caregiver burden. While NPS occur during “normal” aging, there is profound disease-related degeneration in neurocircuitry in AD that may be a mechanism for the differences in clinical course of NPS and lack of response to conventional NPS treatments in AD patients. Our proposed study aims to identify relationships between the neurocircuitry underlying NPS and AD neurocircuit degeneration that ultimately may drive worse outcomes in AD with NPS. Neurodegeneration in AD first targets hippocampus and temporal regions, then spreads along other nodes of the Default Mode Network (DMN), a key brain circuit implicated in cognition and emotions. Functional magnetic resonance imaging (fMRI) studies have shown that impairments in access and engagement of the DMN with the central executive network (CEN; cognitive processing) and salience network (SN; salience mapping) underly psychopathology. We apply this “Triple Network” model which links neurocircuitry of NPS to AD neurodegeneration to investigate the mechanisms of NPS in AD. In addition, we apply NIMH’s Research Domain Criteria (RDoC) framework, which casts brain disorders as extremes from the normal range of behavior. We propose secondary analyses of RDoC-related measures from Human Connectome Project (HCP) Young Adult and Aging Lifespan datasets, and the HCP Disordered Emotional States, Anxiety and Depression, Alzheimer’s Disease, and Brain Aging and Dementia datasets, using computational methods for big data analysis that inherently embody the principles of RDoC, treating NPS and AD as having extremes from normal values of brain – behavior mappings. First, we identify brain circuits for RDoC negative and positive valence and cognitive system constructs, then map these circuits to behavior and self-report measures. We then construct normative models of brain – behavior mappings in healthy individuals, then apply those models to disentangle the complex interactions between NPS and AD. Our overall hypothesis is that deviations from normative values of RDoC-related Triple Network brain – behavior mappings will elucidate mechanisms of NPS in AD. We maximize scientific rigor via a very large sample size for our study, and by adopting ReproNim practices designed for replicable and generalizable neuroimaging research.
