On the Computational Structure of Mood and Anxiety Disorders

A symposium at the annual Society of Biological Psychiatry Meeting.
Chair: Quentin Huys

Mood and anxiety disorders are amongst the most important contributors to the burden due to mental illness, but the underlying processes are still only partially understood. In part, this is because of the complexity of the phenomena, which represent an interaction between experience, learning, environments, and neurobiological variation. The tools of computational psychiatry are well-suited to parsing such complex problems, and there have recently been substantial advances in the computational characterisation of important aspects of these disorders. This symposium will draw together this cutting-edge research. It will cover results on how the interaction between mood and reward learning can be the cause of mood fluctuations (Yael Niv, Princeton); on the use of computational models to subtype the processes leading to effort-based decision abnormalities in depression and schizophrenia (Jessica Cooper, Emory); on an authoritative study of cognitive and decision-making paradigms showing a broad reward sensitivity impairment in depression without a learning deficit (Jonathan Roiser, UCL); and d) on the use of computational modelling to deconstruct anxiety behaviour into specific components with clearly defined but specific adaptive purposes (Dominik Bach, Zurich). Overall, the symposium will highlight recent conceptual and empirical advances in our understanding of the computational structure of mood and anxiety disorders.

  • Algorithms for survival: characterising anxiety-like behavioural inhibition
  • Dominik Bach
  • pdf Computational models of effort-based choice in patients with major depression and schizophrenia
  • Jessica Cooper
  • pdf The interaction between mood and reward learning
  • Yael Niv
  • pdf A Computational Approach to Understanding Motivational Symptoms in Depression
  • Jonathan Roiser