How might we detect and soothe social anxiety in a timely manner?
Background
Social anxiety greatly affects young people’s lives, but the current solutions in place are inadequate for the rising prevalence of social anxiety. Detection of social anxiety using wearables may provide a novel way of recognising it, which reveals new opportunities for monitoring and treatment; this could significantly improve the lives of those who suffer, as well as benefiting society as a whole and healthcare services.
My interest in health tracking and wearables led me to investigate whether social anxiety in young people can be detected using physiological data collected from a wearable using supervised machine learning. In order to investigate this, I taught myself machine learning. Focus groups and interviews were also conducted with socially anxious young people to identify speculative use cases, these were iterated based on their feedback. Although there are designs, the main focus of this project was the data analysis.
My role
I led the project with supervision from Dr Nejra Van Zalk and Dr David Boyle.
Design process
The design process involved extensive desk research, prototyping hypothesis-driven designs and frequent focus groups with young people to understand what type of intervention would be most beneficial. Prototypes were tested and iteratively improved based on the working group’s feedback. Young people were recruited using posters and the approach for recruitment was inspired by A/B testing. Below is an image of a remote call with the young people working group.
.png)
Data analysis process
In order to investigate the detection of anxiety using wearables, physiological data was recorded using a wearable, while thirteen young people with social anxiety participated in impromptu speech tasks. Following a supervised machine learning approach, various classification algorithms were then used to develop models for three different contexts, investigating the detection of social anxiety and its nature. On the right are images of three wearable data samples from a participant during their impromptu speech.
Learnings
The learnings indicated that socially anxious young people had varied needs and had also formed their own behaviours to soothe their anxiety in some ways, when they weren’t avoiding social situations. These techniques are something that could be facilitated into an app in the future. For example conversation starters, mediations and general distractions prompted by the detection of anxiety.
From speaking to a clinical psychologist it was also clear activities such as social exposure therapy were beneficial for sufferers. Based on conversations with the focus group there seems to be an opportunity to gamify this type of therapy, with points and number of social interactions. This could currently be logged manually, but in the future with we combine location data with social anxiety detection it can probably be inferred.
“By revealing new opportunities for monitoring and treatment of social anxiety, the project is expected to be greatly beneficial for sufferers, the society and healthcare services.”
- Celine Mougenot, Senior Lecturer at RCA & Imperial College
Outcome
The outcome of my investigation was very interesting, the results indicated the real possibility of detecting social anxiety via wearables. Additionally, there are numerous desirable use cases for digital treatment for social anxiety fuelled by wearable detection, such as comforting contextual messages during anticipation of anxiety and conversation starter ideas before an interaction. This research could transform the current approaches to diagnosing, treating and monitoring social anxiety which could have a significant positive social and economic impact. Please don’t hesitate to contact me if you would like more details, and the published paper can be found the published paper can be found here.