Many people with psychosis suffer from persecutory delusions—beliefs that terrible things will happen to them in everyday situations, such as people trying to harm them. The disorder causes social anxiety, which can be exacerbated by other symptoms, such as hearing voices. All of this makes ordinary activities such as shopping or going to the doctor challenging. Often a person just withdraws entirely from social contact. In a vicious cycle, the ensuing isolation and rumination can exacerbate other symptoms, including those causing the withdrawal.

The idea behind a virtual-reality system called gameChange is to help patients learn to feel safer, allaying social anxiety by putting them in simulations of situations they fear in which their worst dread does not materialize.

Last month, clinical psychologist Daniel Freeman of the University of Oxford and his colleagues launched a clinical trial of gameChange, the biggest such trial to date of a VR treatment for schizophrenia. It will enroll 432 people with psychosis from five National Health Service (NHS) centers across the U.K. Researchers will assess participants’ avoidance and distress in real-world situations, using an established measure, before and after treatment and then do so again six months later. The hope is that the treatment will reduce participants’ anxiety, which will, in turn, improve other symptoms, particularly persecutory delusions. Freeman co-founded an Oxford spin-off company, Oxford VR, to develop and commercialize the technology. And if the trial is successful, gameChange could be rolled out by the NHS.

The theory behind gameChange is based on active-learning therapies. People are coached through real-world situations they have difficulties with, so that they learn to think, feel and behave differently. Here, patients learn that their social fears are unfounded, which then reduces their anxieties when they encounter those situations in the future. “You only know when someone’s overcome a mental problem when they go back into these situations and can focus on the task at hand rather than having difficult thoughts or emotions occurring,” Freeman says. “VR is brilliant, because, consciously, the person knows it’s not real, so they can try stuff they wouldn’t in the real world—but most of your brain treats the virtual world as real. Therefore, the learning transfers. That’s the beautiful therapeutic mechanism.”

VR has been used in mental health for more than 20 years—mostly for anxiety disorders, including post-traumatic stress disorder—either in the form of exposure therapy or cognitive coaching similar to the kind employed in gameChange. A few groups have tried VR for psychosis but have approached it as a tool, with a therapist present to actually deliver the therapy. “What’s important in gameChange is that it’s almost self-applied, so you don’t need the workforce you need on one-to-one programs,” says Mar Rus-Calafell, a clinical psychologist at Ruhr University Bochum in Germany, who has worked with Freeman on other projects but is not involved in the gameChange project. This means that if the treatment is as effective as Freeman hopes, it could greatly increase health services’ capacity to help many more people with psychosis.

Freeman and his colleagues first tested this automated approach in a trial of a therapy for fear of heights, which simulates situations such as standing by a high ledge while participants are guided by a virtual coach. The vast majority of those who underwent the trial rated themselves as free of fear of heights at the end of the two-week treatment—and this assessment was maintained in a follow-up two weeks later. Because most people have some reaction to heights, Freeman thought the trial would be a good demonstration to convince people of VR’s potential to treat mental health problems, but the intention was always to tackle schizophrenia.

During the gameChange treatment, participants select from six everyday scenarios, such as going to a café or catching a bus. As they progress, they can, at their own pace, choose to make the experience more challenging. Doing so increases the number and proximity of other people, providing exposure to situations that can potentially make patients more anxious. Several organizations are collaborating on the trial, including parts of the NHS and the McPin Foundation, a charity that promotes the involvement of people with mental health problems in research. Patient groups chose the scenarios and helped to decide what happens in them. “We’ve done lots of user testing with patients as well, getting feedback about would be helpful, what they don’t like,” Freeman says. “Throughout the work, it’s been strongly influenced by patients.”

Virtual therapist guides the patient. Credit: University of Oxford/Oxford VR

I am trying the system out in Freeman’s office at Oxford’s psychiatry department. After donning the headset and grabbing the controllers, I enter the simulation to find myself in a spacious waiting room with lush foliage visible through large windows. Turning around, I see a computer-graphics representation of a woman behind me, slightly cartoonish but detailed and three-dimensional. This is the “therapist.” Casually dressed, with a relaxed, reassuring manner, her voice is measured and soothing as she guides me through interacting with menu bars floating in front of me.

First, I rate how confident I feel in social situations. Then I’m presented with six everyday scenarios to choose from. After I select “street,” we are transported to a street scene with a few parked cars. The therapist reassures me that I won’t have long to wait before my taxi arrives. A youth in a hoodie walks along the pavement toward me but passes with scarcely a glance in my direction. These virtual “others” interact more as the level of challenge increases, Freeman tells me. I’m not 100 percent at ease myself, so I can imagine individuals who suffer from persecutory delusions really struggling with multiple people looking directly at them as they wait patiently on a pavement.

The challenge level increases when I enter the doctor’s waiting room. Several distressed-seeming virtual patients occupy chairs lining the room. Suddenly, slips of paper are blown above my head by a fan. I use my virtual hands to pluck them from the air and return them to their place on the receptionist’s desk. Again, these kinds of unexpected events crop up more frequently at higher challenge levels.

The researchers are collecting economic data to assess cost-effectiveness (including training, equipment and travel expenses, staff time, savings in terms of services not utilized by patients who do well and comparing cost of treatment to improvement in quality of life). They are also talking to NHS staff about best ways to deploy the technology in clinical settings. “The benefit is not only going to be for people with psychosis but also mental health professionals,” Rus-Calafell says. “There’s still a traditional view that psychosis is difficult to treat or even that just medication is enough, which needs to change.”

Ultimately, Freeman believes VR could be applied to a range of mental health problems, and Oxford VR is already working on treatments for other conditions. “GameChange could show how you can automate psychological treatment and get it out to health care systems at scale,” Freeman says. “If we can crack that, it will show the way for many other conditions.”

Yet another outdoor setting for a patient to experience. Credit: University of Oxford/Oxford VR

Social anxiety is just part of psychosis. “Maybe other processes in psychosis should be treated with longer, one-to-one therapy,” Rus-Calafell says. “But we need to see the results.” Further, previous researchers were cautious that people with psychosis might experience adverse effects related to a reduced ability to tell the difference between the virtual and the real. This concern has yet to materialize in VR research on psychosis so far, but trained clinicians have always been present in these studies. “In the Oxford work, there appears to be more autonomous exposure, and that could bring risks,” says psychologist Albert Rizzo of the University of Southern California, who is not involved in the work. “It’s something we should test very carefully. And from the past work by this lab, I’m confident they have safety protocols in place.”

Perhaps the main concern, though, is that automated, software-based systems might rob patients of the human connection that is an important part of healthcare. “Of course, there’s a question about which elements tech can’t provide that are important,” Freeman says. “We need skilled clinicians at moments of acute crisis or suicidal risk and all the really difficult life events people have.” But Freeman insists the aim is not to replace therapists. “We need more therapists, not fewer, but in immersive VR, you can deliver really good treatments that are helpful for many people,” he says. “Not everybody; mental health is far too complicated for there to be one solution.” The results of the gameChange trial will likely play a part in determining whether VR becomes one of the available options. “The onus is on us to show that it produces good clinical benefits and that patients want it,” Freeman says.