Decision and Emotion
We build computational models for mood and behaviour that help us understand psychiatric disorders.
What is happiness? We are interested in describing the factors that determine subjective affective states like happiness. We build computational models linking affective states to ongoing experience and quantitatively relating feelings and behaviour. We use neuroimaging, pharmacology, electrophysiology, and smartphone-based data collection to study the relationship between decision making and emotion across the lifespan and in people with psychiatric disorders like depression.
Computational models of affective experience
We find that happiness depends not on how well you are doing, but whether you are doing better than expected. Happiness relates to neural activity in dopamine projection areas and can be manipulated pharmacologically with dopaminergic drugs. We are extending this quantitative understanding of happiness to multiple dimensions relevant to mood disorders including effort, intrinsic reward, future prospects, environmental volatility, confidence, and altruistic behaviour.
Neuromodulatory influences on decision making and mood
We consistently find that dopamine plays a valence-dependent but value-independent role in reward seeking. We are testing how dopaminergic and serotonergic drugs influence mood and behaviour across multiple tasks. Through collaboration at Yale University, we are evaluating how ketamine, a rapid glutamatergic antidepressant, influences mood dynamics during decision tasks. We will test whether mood effects due to serotonergic antidepressants gradually accumulate while ketamine effects are immediate. We will test whether mood dynamics predict later depressive symptoms.
Smartphone-based data collection in mood disorders
We co-developed The Great Brain Experiment, a smartphone app for cognitive science research with over 130,000 users. We remotely tested over 500 individuals with a history of depression, finding that model parameters relate to depression severity. We are building a new app, The Happiness Project, for longitudinal testing. Through collaboration at the National Institute of Mental Health, we are testing 150 adolescents with anxiety and depression to ask how model parameters relate to symptoms over months. We will also test patients with bipolar disorder to test whether a computational parameter for the influence of mood on learning is elevated and can predict future manic and depressive episodes.
- Under the Hood: Using Computational Psychiatry to Make Psychological Therapies More Mechanism-Focused Frontiers in Psychiatry, 11 DOI: 10.3389/fpsyt.2020.00140
- Neurocomputational mechanisms underpinning aberrant social learning in young adults with low self-esteem Translational Psychiatry DOI: 10.1038/s41398-020-0702-4
- View all publications by the Decision and Emotion team