Computational mechanisms in mental health
Context
Our research aims to bridge the gap between clinical psychology constructs like ‘catastrophising’ or ‘need for control’ and the underlying learning and decision-making mechanisms that cause these reactions. We think there are some individual differences in how people learn about things like control and uncertainty, and how they flexibly respond to changes in the world, which may act as risk factors for conditions such as eating disorders, anxiety disorders, depression, and PTSD.
The research
Our research typically asks participants to do computer-based tasks, then uses computational modelling of learning and decision-making processes (including reinforcement learning and drift diffusion models) to understand the processes which differ between individuals and how they relate to our mental health. One recently funded project has involved creating a task that varies the control individuals have over the outcomes they receive, and mapping how they learn about this control over time to reinforcement learning parameters. In turn, we have related these parameters to continuous traits of eating disorders, and also tested whether different clinical groups have differences in these parameters.
Featured publications
- A series of unfortunate events: Do those who catastrophize learn more after negative outcomes?
- Deep Brain Stimulation in Anorexia Nervosa: Hope for the Hopeless or Exploitation of the Vulnerable? The Oxford Neuroethics Gold Standard Framework
- The recoverability, reliability, and generalizability of reward processing parameters and relation to mental health symptoms
- Reinforcement Learning in Patients With Mood and Anxiety Disorders vs Control Individuals: A Systematic Review and Meta-analysis
- Adaptive learning from outcome contingencies in eating-disorder risk groups.
- Computational Perspectives on Cognition in Anorexia Nervosa: A Systematic Review
- Eating disorder symptoms and control-seeking behavior