Cookies on this website

We use cookies to ensure that we give you the best experience on our website. If you click 'Accept all cookies' we'll assume that you are happy to receive all cookies and you won't see this message again. If you click 'Reject all non-essential cookies' only necessary cookies providing core functionality such as security, network management, and accessibility will be enabled. Click 'Find out more' for information on how to change your cookie settings.

Using computer models of behaviour, we aim to better understand anxiety and depression, and to guide the development of novel treatments.

These graphs show probability distributions of computational parameters that guided a person's behaviour in a decision making task. They were given feedback on their performance in the form of gaining point ('rew') or losing points ('loss'). The person used this feedback to guide their future decisions in the task. The person's learning rate ('LR') was the rate at which they updated their beliefs about the task. The person's temperature ('Beta') described the extent to which they used those beliefs to guide their actions in the task.

What our lab does

In the Computational Psychiatry Lab we use computer models to investigate why certain people are prone to developing psychiatric illnesses and how we might best help them. We use the computer models to help us understand how patients think and learn about their world and how they arrive at decisions.

As an example, we know that people with depression seem to be more influenced by negative than positive events and that this negative focus seems to maintain the symptoms of depression. In one of our current projects we ask a group of people with depression and a group of people who are not depressed to complete a simple task in which they might win or lose money and we measure how much their decisions in the task are influenced by the wins and losses. For this project we have developed a simple computer program which also plays the task— we can tweak how the program completes the task until it behaves as much as possible like the depressed patients or like the people who are not depressed. By comparing the internal computations of the “depressed” and the “non-depressed” computer programs we can begin to understand differences in how depressed and non-depressed people think when they are learning. We can then try and change these thinking styles, using therapy or medication, to see if this also helps patients feel better.

Where to find out more about Computational Psychiatry

For monthly Computational Psychiatry Webinars visit the Transcontinental Computational Psychiatry Workgroup webpage.


[26/07/2017] Our work "Characterising and Engaging a Computational Treatment Target relevant to Depression" received a Clinical Poster Award at the British Association of Psychopharmacology Summer Meeting.

Our team

Selected publications

Selected publications

Related research themes