MICA: The Translational, Computational and Neurocognitive Basis of Anxiety

One in four of us will suffer from some form of mental health problem in a given year. For the majority of us, this will be some form of anxiety or stress related problem. Some people's anxiety is so crippling that they end up housebound. Anxiety costs the UK economy over £14 billion a year.

We currently have effective clinical treatments, both psychological (i.e. therapy) and pharmacological (i.e. medication), that can treat anxiety. However, we are not currently able to effectively target these treatments to people who will respond. In fact, for most people, the first treatment that they try will not work. Even more worrying is that for up to 25% of sufferers; none of our current treatments will work.

There are at least two reasons that we have these problems targeting treatments:

I) We do not know how our current treatments work at a biological level
II) We do not know how debilitating feelings of anxiety emerge from the underlying biology.

Without a better understanding of both of these issues we will not improve our ability to target treatments. This fellowship therefore seeks to improve our understanding of the biological basis of anxiety symptoms and treatments. It focuses on psychological treatments and aims to better understand 1) the neural circuitry by which psychological treatment works; 2) develop new psychological treatments; and 3) better refine our understanding of the biological bases of anxiety diagnosis. Specifically:

1) Neural Circuitry:

Over the last decade, I have identified a brain circuit which drives feelings of anxiety but we do not know if it changes following treatment. In this project I will scan the brains of anxious people undergoing psychological treatment for anxiety and compare them to anxious people who are not. If we show that this same circuit is important for treatment response, it will enable us to better target treatments to individuals, and reduce the number of people who try treatments that do not work for them.

2) Treatments:

In parallel with this I will attempt to develop a new computerised treatment for anxiety. This is possible because my advances in a new field known as computational psychiatry have given us tools that provide better insight into the computational process of the brain, and how they change in anxiety disorders. I will therefore develop a computerised training procedure that targets these computational processes and test it in a treatment study. If this works, it will provide a quick and cost effective treatment that could be delivered by smartphones and ultimately reduce the number of individuals suffering from anxiety.

3) Diagnosis:

Finally, I will try to improve our understanding of how debilitating feelings of anxiety emerge in the first place. It is increasingly clear that our current diagnoses, which are based on symptom checklists, do not reflect truly separate biological categories. I will therefore search for 'trans-diagnostic' dimensions that drive anxiety but which cut across our current categories. I shall do this by getting thousands of people to complete a simple psychological task online that I have previously linked to debilitating anxiety. In addition, we will collect a wide range of other psychological tasks and questionnaires from these people. Applying state-of-the-art statistical methods to this 'big data' will enable us to identify trans-diagnostic dimensions that drive anxiety symptoms. In the long term these new trans-diagnostic dimensions will improve our ability to determine the biological factors driving symptoms and hence help us predict treatment response and develop new treatments.

In sum, I will combine my unique inter-disciplinary skill-set, with my breakthroughs in delineating the neural and computational basis of anxiety, to develop new and more effective clinical tools for anxiety diagnosis and treatment, with the ultimate goal of improving the quality of life for millions of sufferers.

The goal of this research is to increase our understanding of the neurobiology of pathological anxiety. It has three broad aims. The first aim seeks to determine neurobiological circuitry of response to psychological treatment for anxiety. Individuals will be recruited from NHS psychological treatment services and functional magnetic resonance imaging will be used to examine cortical-subcortical brain connectivity before and after treatment. I predict treatment-related attenuation of a specific cortical-amygdala connectivity pattern relative to an anxious untreated control group. The second aim focuses on developing a novel computerised treatment for anxiety. Computational psychiatry techniques, which I have previously used to model cognitive task performance in anxious individuals, will inform the development of a new computerised treatment regimen (cognitive bias modification). I will develop a user friendly computational psychiatry toolbox, and the efficacy of my new computationally-informed treatment will be assessed in a randomised controlled trial. The final aim focuses on improving diagnosis. A translational task (and associated computational model) linked to pathological anxiety in both humans and rodents will be integrated into a large-scale human cohort with industrial collaborator Cambridge Cognition. Using data from this cohort, structural equation modelling will be used to identify latent trans-diagnostic dimensions that drive anxiety-relevant symptoms and computational-psychiatry-modelled, task performance. In the long-term these new trans-diagnostic dimensions should predict treatment response more effectively. My senior fellowship therefore proposes an ambitious research programme that combines my unique inter-disciplinary clinical, computational and neuroscience skill-set, with my breakthroughs in delineating the neural and computational basis of anxiety, to develop new and more effective clinical tools for anxiety diagnosis and treatment.

By improving our understanding of the neural circuitry driving contemporary psychological treatment response, the outputs of this research will significantly improve the development of novel psychological treatments, and refine our diagnosis of pathological anxiety. This research will therefore impact on a range of beneficiaries, as outlined below:

1) By providing a clear understanding of the neurobiological processes underlying anxiety, this research will significantly and directly benefit patients. Importantly, increasing awareness that anxiety is a treatable "brain circuit" problem has the potential to reduce stigma, which remains a major issue for sufferers of all mental health problems.

2) A second potential impact for patients is the prospect of a clearer understanding of how treatments work (in the later stages of the project). The novel computerised CBM stage of the project may also sow the seeds for new potential treatments for anxiety disorders. In the longer-term, this would have significant value for patients and their families, reducing the overall burden of mental ill health.

3) The novel computerised CBM procedure, which will build on prior work using cognitive tasks to reduce avoidance behaviour, may also be a safe intervention for use in individuals at risk for developing anxiety disorders and could potentially be the first step towards an active resilience promotion strategy (i.e. a 'cognitive vaccine') in healthy individuals with positive implications for mental capitol and well-being.

4) The Computational Psychiatry toolbox should increase adoption of these state-of-the-art tools, and should facilitate computational rigorous research across disciplines. In being Open Source Software, it will also allow others to build on and refine these methods, leading to the development of tools to improve diagnosis and selection of treatments.

5) A more hypothetical and indirect benefit of such a treatment outcome could be a reduction in the overall costs of mental ill health to the economy. Anxiety costs a lot to treat, but it also prevents people from actively participating in the workforce.

6) The general public will also benefit from this fellowship via increased dialogues proposed for communicating the findings to the public via websites and twitter. The incorporated social media aspects will allow direct interaction between the researcher and the public, and help to raise awareness of mental health.

7) The research assistant and postdoc will be trained by Dr Robinson, providing training in quantitative skills (e.g. data analysis, programming languages) required in the UK workforce, whether or not they pursue a career in academia.

8) A future direction of the proposal will be to bring in funding to take the findings to the next stage, including a potential clinical trial of the proposed CBM project. Such a trial has a potential value for businesses in that a putative CBM procedure might be patentable and marketable.

9) The industrial partner, Cambridge Cognition, will benefit from increased links with academic research. More broadly, demonstration of successful industrial-academic collaborations in the field of psychiatry may lead to increased investment in the field.

10) Tackling mental health issues is a major focus of the current UK government, and attracts significant media coverage. I believe this project constitutes good value for money and, given its potential to improve treatment outcomes has the potential to save considerable amounts of money (anxiety costs an estimated 14.2 billion a year in the UK both in terms of direct treatment costs and lost productivity). In other words, there is potential for a large return on investment that can be highlighted by the UK government in the future.