Avatar Therapy Eases Voices Better Than Usual Care

MADRID — Building on a growing body of evidence, new research has reinforced the potential of avatar-based therapy to ease auditory hallucinations — one of the most persistent and distressing symptoms in schizophrenia spectrum disorders (SSDs). The findings fuel renewed hope for improved outcomes for this often treatment-resistant condition, said a leading expert who helped pioneer the approach.

The Danish CHALLENGE study used trained therapists and patient-created avatars designed to represent the patient’s visual perception of the voice in their head. “We transformed the therapist’s voice into sounding like the auditory hallucination, and the therapist could switch between talking either as the negative, hostile avatar or as a supportive therapist facilitating the patient standing up to the voice,” said lead researcher Louise Birkedal Glenthøj, PhD, from the University of Copenhagen, Copenhagen, Denmark.

The findings were presented on April 8 at the European Psychiatric Association Congress (EPA) 2src25.

Decreased Frequency, Severity

This study included 27src adults with SSD and treatment-resistant auditory hallucinations. Participants had a mean age of 33 years and had been hearing voices for an average of 16 years despite ongoing (9src%) treatment with antipsychotic therapy.

Patients were randomly assigned to receive either avatar-based virtual reality (VR) therapy or supportive treatment as usual (TAU), with both groups undergoing seven 6src-minute sessions and the option of two booster sessions. Follow-up assessments were conducted at 12 and 24 weeks.

The therapist-assisted avatar therapy unfolded in three phases. In the first, patients confronted the voice to regain a sense of power and control. This was followed by work on self-worth and identity, with the final phase focusing on recovery and cultivating hope for the future. “We also integrated a rather prominent focus on compassion-focused therapy,” said Glenthøj.

Sessions were delivered through fully immersive VR, with patients wearing a VR headset and noise-canceling headphones. The therapist spoke through a computer that transformed their voice in real time. “We could also adjust the proximity of the avatar and voice based on individual anxiety responses,” she added.

This study’s primary outcome was the total severity of auditory hallucinations, measured using the Psychotic Symptom Rating Scales-Auditory Hallucinations. At the end of treatment, there was a statistically significant difference between groups, with the intervention group showing lower severity scores, corresponding to an effect size of d=src.3.

“The intervention group showed low scores on that scale,” Glenthøj reported. However, by the 24-week follow-up, the difference between groups was no longer statistically significant.

For the secondary outcome — frequency of voices — the intervention group continued to show significantly lower scores at both the 12- and 24-week follow-ups (P=.src2 and P=.src3, respectively).

Although there was no significant difference between groups in assertive responses to voices, as measured by the Assertive Responding to Voices scale, Glenthøj noted that the estimated slope analysis revealed a significant group difference (P=.src4). This suggests that patients in the intervention group improved over time, while scores in the control group remained relatively stable.

There were no differences between groups for other secondary outcomes comprising different aspects of voices and social functioning.

For the exploratory outcome of patient satisfaction, scores were significantly higher in the intervention group at both the end of treatment and at the 24-week follow-up (P <.srcsrc1). Glenthøj also reported strong retention, with approximately 8src% of participants in the intervention group completing all seven therapy sessions.

Technical Difficulties

The intervention was generally well-tolerated, with patients reporting minimal symptoms on the Simulator Sickness Questionnaire. Its safety profile was comparable to TAU, with one exception: A patient with a known history of self-harm experienced an episode that may have been triggered by the therapy, Glenthøj noted.

Since the intervention was delivered by trained and supervised therapists in outpatient clinical settings, Glenthøj concluded that it demonstrates the feasibility of implementation in routine practice. She noted that Denmark is preparing to pilot the therapy within a public mental health facility.

“We are about to take the steps in this direction and will be piloting implementation of this intervention into a public mental health facility,” Glenthøj added.

The findings add to the growing body of evidence supporting the strong potential of avatar-based therapy for treatment-resistant auditory hallucinations in a highly distressed patient population, Glenthøj told Medscape Medical News.

She noted that significant technical issues affected roughly half of the therapy sessions, with frequent failures in getting the avatar to function properly — likely affecting the overall therapeutic effect. Even so, patients still showed benefits, suggesting that with more flexible and reliable delivery, the intervention holds considerable promise for this challenging group.

She emphasized that the success of the therapy depends heavily on the therapist’s expertise. The therapists involved were highly skilled, yet still found the approach extremely challenging, as it required playing the avatar in real time, responding instantly to unexpected patient reactions, and maintaining clinical awareness throughout.

Some patients experienced a temporary worsening of psychotic symptoms early in therapy after confronting the voices they feared, underscoring the need for clinicians who are experienced with psychosis and can judge when to push forward and when to pause.

Ready to Roll

Commenting on the findings for Medscape Medical News, Tom K.J. Craig, MD, one of the pioneers of avatar therapy for this indication and professor emeritus of social psychiatry at the Institute of Psychiatry, Psychology and Neuroscience in London, England, said the CHALLENGE results are impressive and “add to my conviction that we’ve got an effective therapy that should be rolled out.”

The effect size from Craig’s pioneering AVATAR1 study (d=src.8) — one of the highest observed for any psychological therapy targeting auditory hallucinations — was much larger than that of CHALLENGE (d=src.3).

Still, he noted that the CHALLENGE intervention achieved an effect size comparable to cognitive-behavioral therapy (CBT), but in just six sessions rather than the typical 9 months to a year of treatment, representing a meaningful therapeutic gain.

He also expects that once the Danish group resolves the technical issues, their results could match those achieved by his team.

“If you look at medication, working with a population, which are people with long-term psychosis who have heard voices for years, the best that medication can do is src.src3. We can double that.” He added that when given the choice, most patients prefer “talking therapy” over another medication trial.

Although Craig believes the therapy is ready for broader implementation, he acknowledges several obstacles. “You have healthcare organizations with fixed budgets, and you’ve got some pushback from therapists who are used to standard CBT and don’t mind that it takes a year to get there — and they’re also skeptical about playing the ‘nasty voice,’” he said.

Technical challenges add another layer of complexity: “What you put in place today will be quickly outdated. So, I’m sure in 1src years’ time this will be obsolete, and something better will come along — but at the moment, we’re there.

Glenthøj reported consultancies for Lundbeck Pharma, Boehringer Ingelheim, and Khora VR. Craig reported research support from Wellcome Trust and the National Institute for Health and Care Research. He is also an unpaid scientific advisor to AVATAR therapy.com.

Kate Johnson is a Montreal-based journalist with more than 3src years of experie