Using Immersive Virtual Reality to Increase levels of Self-Compassion in Patients with Depression
By Alina Shrourou, on 25 September 2019
Dr John King and Professor Chris Brewin have recently gained an NIHR i4i Mental Health Challenge Award (an award to develop an innovative technological solution to which will improve the care pathway and outcomes of patients experiencing mental ill-health) for their translational research project, “Treating depression with self-compassion in virtual reality”. In this interview, they discuss the motivation behind their work and provide an overview of the project which they will be starting this autumn.
Please highlight the prevalence and burden of depression.
The community prevalence in the UK is about 3.3%, which translates to approximately 1.7 million adults currently experiencing depression.
It is estimated that the economic burden of depression in the UK is around 12 billion pounds a year through the costs of treatments, losses from patients being unable to work and extended effects on other people who may be involved (carers for example).
However when you consider burden, it is important to think of it not only in terms of cost, but also from a personal perspective. Depression is a debilitating condition that involves a sense of hopelessness and causes huge changes to quality of life. Those changes also extend to anyone close to an individual experiencing depression.
What are the current challenges associated with overcoming depression or other mental health conditions such as anxiety?
Current methods of addressing mental health problems are generally very good, but have a high economic cost. For example, we can be reasonably confident at treating people with depression successfully if we give them the best available talking therapies. CBT delivered to a high standard will help up to 50% of sufferers to recover.
However the problem is that delivering gold standard CBT to every person with a mental health condition is very expensive, so the health service is driven to find alternative ways of delivering efficient treatments. It’s important to find the right treatment for the right people and to have a diverse enough range of treatments for patients so that we can find something that’s a good fit. That in itself creates other challenges because you then have to identify people who are suitable for a briefer form of treatment, for example.
Another difficulty with depression is that it’s a disorder that comes and goes – people may have periods where they are relatively well but if they are under pressure, their depression may come back. It might be that a patient experiences a relapse but are they are unable to reach a therapist because their treatment has finished.
We’re hoping that we will be able to provide something eventually that’s available in patients’ own homes. This could be a huge advantage in just getting them through those ups and downs when they may be feeling very unsupported and alone.
What is the main contributor to depression and how are you targeting this in the treatment that you are investigating?
I don’t think you can say that there’s a single main contributor. Everyone’s life experience is different and everyone has different predispositions – some may be genetic, for example, some socioeconomic.
However, we have identified one contributor that is very common in people experiencing depression, and that’s self-criticism – something which can appear quite early in life and seems often to be a critical factor in terms of the onset and maintenance of depression.
What we have learnt through being part of a much wider network of people working in these fields, is that one way of addressing self-criticism is with self-compassion. It’s not that these things are opposite to each other, but self-compassion can act as a counter-narrative to the negative stories that people tell about themselves when they are prone to self-criticism.
Please provide a brief overview of your project.
Currently, treatments for depression are largely talking therapies or medications. We want to use virtual reality and computer science to provide something new because there’s a real shortage of novel approaches, and we want to meet the unmet medical need of providing an alternative solution that may be more suited to some individuals.
Our approach draws on two sets of scientific findings. One is compassion-focussed work, involving learning to be more kind and self-soothing. There’s a large body of work emerging showing that compassion-focussed therapy can be valuable in helping people with depression and anxiety, and can also create resilience – protective aspects that can help people to withstand the challenges of difficult life events.
The other field is virtual reality and computer science. This derived originally from conversations Chris was having with Mel Slater, a professor of computer science and one of the world’s leading experts in virtual reality, who until very recently was based at UCL. They were discussing some work that Mel had been doing on avatar re-embodiment.
This led to a series of proof-of-concept studies funded by an MRC Translational Medicine award. Publications from this award supported the current development funded by NIHR of a comprehensive clinical intervention using up-to-date technology.
How does the virtual reality (VR) system you are developing generate an illusion of body ownership?
In virtual reality you can give someone a body, an avatar, and you can make it so the movements of the avatar are completely synchronised with the movements of the user. When you look down at yourself in VR, you see a body and notice that when your arms move, its arms move in exactly the same way. If you were to stand in front of a mirror in VR and do various movement exercises, you quickly understand that the avatar is yours.
What Mel and others had observed over the years, is that when you are embodied in a virtual avatar, you start to take on some of the qualities of that avatar. There have been numerous studies looking at what happens when you take on the avatar of a child, to use a simple example. When you do that, your estimations of size tend to become much larger, because things are larger to you from the perspective of a child.
Working with Professor Paul Gilbert, who pioneered compassion in the UK, we came up with a scenario where people who find it very difficult to be kind and compassionate towards themselves, people high in self-criticism, are placed in a virtual environment, given an avatar and are confronted with a very distressed child. We chose a child because we thought that might make it easier for people to be supportive.
We give them a script to go through which spans several stages of support for the child. As they go through the script, the child’s mood improves so they are no longer distressed. We record this whole interaction and then, this where re-embodiment comes in, we re-embody the participant as the child. There is a mirror opposite them in the scenario so that they realise they are now the child.
The left image shows what the participant will see while they are comforting the child, versus the right image where they become the child. These images are from pilot studies and the new technology being created for John and Chris’ work will appear far more life-like.
We then play back their earlier intervention so they get the experience of their own words, voice, movements, being compassionate towards “themselves”. This is a very literal experience of self-compassion which many of the people we worked with have found quite an unusual and novel learning experience. It’s that bit of learning that we are trying to get across.
When they are re-embodied as the child, this change of perspective means that they are seeing through the child’s eyes this other avatar which is clearly them because it’s expressing their words and actions. We hope that this will create a new experience of self-compassion.
How does the virtual reality system you are developing generate an illusion of body ownership?
When you embody someone in an avatar, they respond to things happening to that avatar as though it’s happening to themselves. The brain is operating on two levels here: people are aware that this is something outside themselves, yet they respond quite automatically. If you threaten the avatar in VR for example, the person feels threatened themselves quite instinctively – it’s an automatic response. We are trying to use that involuntary reaction created by embodiment, so that the person will come to have these natural experiences of being reassured.
Would it have the same effect if you used the technique with a video recording rather than VR?
The difference between a video recording and VR experience is that there is an irresistible salience to VR that you can’t avoid being influenced by.
To contextualise that – I recently tried out one of the headsets that we’re going to be using in our work. I tried the technology on a rollercoaster experience, which is a common VR demo due to the fast movements and visual flow. In this situation, the imagery was so realistic that I had to remove the headset because I thought I might fall off my chair! It’s that level of instant, forced imagery you can’t ignore, which makes it different to watching a recording.
In our pilot study we looked at the effect of the person just watching themselves being compassionate to the child from the outside, which is much more like conventional interventions such as video or role play techniques – and it did not have the same effect as seeing it though the child’s eyes.
We think it’s that perspective take on actually being in the child and seeing it through your own eyes that makes a much bigger impact – that’s what we’re trying to leverage through VR, as it is something you can’t do in any other way.
Please provide an overview of the sub-contractor for this project. How are they helping to accelerate translation?
The strength of outsourcing the development to an industrial sub-contractor is that they are bang up to date with the fastest and most advanced technology. Previously when we were lab based and trying to develop our technology, the progress was reasonably slow and the headset was clunky, weighing several kilograms – not quite the effortless product we want to come out with.
Now that we are outsourcing this development, we’re expecting that process to be very different this time because they are familiar with the latest technology and software libraries and have a lot more background knowledge on how to solve various problems. The agile sense of being able to work quickly and thus being able to take advantage of new technologies is really attractive to us.
If you have experienced VR before, you may have noticed that there is often a lag between your movements and the response of your avatar, or other features not working properly. When I tested the headsets we will use in our study, not only were they incredibly lightweight, but also absolutely rock solid in terms of wherever I looked, there was no visual delay. It is also a totally standalone product, where all you need to do is have a look around the room to calibrate it, and then it’s ready to go.
We’re going to be using commercially available hardware so our industry partners are involved in the software side only. We want our end users to be able to use standard VR headsets available to consumers everywhere as that makes it much easier to disseminate what we are doing later on. Ideally, clinicians or consumers will be able to download our software via an app, and then use the software with whatever VR hardware they have available.
What challenges did you come across when developing a translational pathway for your work?
It was a big learning process for us. We come from a tradition which is not focussed on commercialisation, rather the tradition in psychology and mental health is that you work hard to develop new ways of treating people, most likely based around talking therapies and changes to currently existing ones, and you publish that. You might give workshops to teach people how to do it or produce materials to help disseminate that work, but there’s not much in the tradition in the talking therapies of commercialising.
Our background didn’t lend itself to understanding some of the things that were being asked of us when we were starting to apply for these streams of funding. Things like: how are you going to manage the intellectual property (IP)? What is your commercialisation plan? These have never been things that we needed to worry about before, so that was a big challenge for us.
How did the UCL TRO help you with your project?
The TRO were essential to us and we certainly would not be in the position we are now with an NIHR i4i award to start our work, if it wasn’t for the TRO.
There’s a culture to understanding translation, and the Translational Research Group within the TRO were people who come from that background and understood what our priorities should be.
Not only did we need to explain the science to potential funders and our hopes for the future, but we also had to explain all the other things relating to translation to them – including commercialisation. The TRO were invaluable because they have a huge amount of experience in doing exactly that. They knew which regulatory bodies we would need to engage with, what the priorities and risks that the funders would be looking for us to manage, and they also had contacts, linking us up with the right people who would be able to advise us in various areas. For example, they introduced us to UCLB, who ended up being central to our conversations with the commercial sub-contractor and making sure that the funder knows what the basis of the IP is going forward.
What are the next steps in your research?
In this particular grant, the award is funding us to develop the technology which we expect to take just under a year. During that process we will introduce the personalised avatars, so that we can give the avatars the faces of the patients using them. That should make the intervention much more potent in terms of people’s identification with the avatars. There are some risks associated with that, so in the first year we will be including a lot of PPI (Public Patient Involvement) input to make sure that what we develop is safe and tolerable to fit expectations of the end users.
After the year, we’ve got two trials to run. The first is a feasibility trial where we will be looking at practicalities, tolerability and safety in a fairly small scale study in a single London IAPT (Improving Access to Psychological Treatments) – a fast track NHS treatment service for people with mild to moderate depression and anxiety.
Following this, we will be doing a larger trial where we recruit people to use the final version of our technology. This time, we will be getting a sense of its effectiveness rather than feasibility. It’s not a pure efficacy trial as we’ll still be recruiting fairly broadly and we will not be targeting specific patients at that point, but would focus on assessing how effective our therapy is compared to other treatments for depression provided by the NHS.
Although that would be the end point of this particular project, during this time we would aim to go back to the funder and look at doing a more closely-defined efficacy study, to help move us towards commercialisation.
About Dr King and Professor Brewin
John King (left) is a clinical psychologist at UCL. Although he started his career in neuroscience, his interest has shifted over the last 10 years where more recently, he has been investigating ways we can help difficult and large scale problems including depression and anxiety.
Chris Brewin (right) has been a practising clinical psychologist for the last 30 years and after 19 years at UCL retired, becoming an emeritus professor. His current research interests involve looking at how VR can be used to create experiences in patients that you couldn’t create any other way – in particular, applying some of the insights into treatments for compassion.