Anxiety is one of the most common mental health issues, and debilitating anxiety symptoms can be present across a spectrum of diseases and disorders. Many symptoms of anxiety manifest in the body (such as a racing heart rate or shortness of breath), and an impaired ability to monitor bodily symptoms is hypothesised to exist across a range of psychiatric illnesses, and in particular within anxiety. Whilst the severity of these symptoms can vary greatly across individuals, this miscommunication between the brain and body is thought to be a key component, where bodily perceptions (termed ‘interoception’) may be under-, over- or mis-interpreted to perpetuate continued anxiety. Therefore, the main aim of this project was to measure specific body perceptions in people with differing levels of anxiety from the general population, before future work moves towards applications in disease or disorder-specific populations. This project took place within the wider framework of the Translational Neuromodeling Unit (TNU), a joint institution of the University of Zurich and ETH Zurich, where the goal is to develop and validate mathematical models that allow us to understand and treat a range of psychiatric conditions.
The body perceptions that were the focus of this project centred around breathing. Whilst altering and evaluating the internal state of our body is challenging, our breathing offers a viable pathway for accessing interoception. While anxiety is known to be related to altered perceptions of breathing, it is not yet known where in this perceptual model anxiety exerts its effect. Therefore, my research developed and utilised novel tasks that specifically measured a number of important aspects of breathing perception, including accuracy detecting small changes in breathing resistance, self-awareness and insight into interoceptive abilities, and how quickly an individual can learn about their breathing within a changing environment. Functional magnetic resonance imaging (fMRI) of the brain was also used in conjunction with a novel breathing learning task, to observe both how the brain dynamically adapts its perception of breathing and how this process may be impaired with anxiety. These rich datasets allowed us to apply state-of-the-art computational models to investigate what, where and how anxiety can interrupt the brain’s processing of internal bodily states.
In this project we found that significant disparities in breathing perceptions exist in individuals with different levels of anxiety. These disparities were apparent in subjective experiences (assessed with questionnaires), meta-cognitive evaluation (characterised with computational models of behavioural responses), and brain activity (measured with fMRI). These results now provide us with a rich characterisation and pointers towards potential mechanisms of how anxiety may change our body perceptions within a healthy population. Future projects that will directly follow on from this research will use this knowledge to investigate how these relationships may further change in individuals with clinically-significant levels of anxiety, and also how current treatments of anxiety (pharmacotherapy and exercise treatment) act on these body perceptions.
