A few papers representative of current emphases:
- computational learning theories of anxiety and avoidance
- using neurocomputational approaches to understand treatment mechanisms
- improving clinical translation of computational psychiatry methods.
Integrating computational learning theory into theories of altered learning in anxiety
Clinical anxiety has long been viewed as the product of disrupted learning. We are using computational models of learning and choice to describe learning differences in anxiety, with a particular focus on differences in learning under uncertainty.
Neural systems underlying altered uncertainty learning in anxiety
Noradrenergic and serotonergic disruption are hallmarks of anxiety disorders, while computational theories of uncertainty learning describe neuromodulatory effects on learning and neural functioning. Using functional and structural MRI, we are studying how networks involved in uncertainty processing are disrupted during learning in anxiety.
Generative neurocomputational modeling to understand mechanisms of treatment for internalizing disorders
First-line psychotherapies like CBT are hypothesized to act by changing how people learn from and process events. We use neurocomputational approaches to more precisely describe, and improve on, how CBT approaches change learning and effect symptom change.
Improving psychometrics and clinical applicability of computational approaches
Computational modeling promises precise and specific measurement of altered learning processes in psychopathology. Yet, these approaches are technically complex and their reliability and validity are rarely tested. We seek to understand when and how computational approaches improve measurement and to develop best practices for the field.
This page is currently in progress.