Immersive Virtual Reality as a Scalable Intervention for Persistent Voices in Schizophrenia: findings from the CHALLENGE trial

Overview of VIRTU’s Work

The CHALLENGE trial is the largest randomized controlled trial to date assessing immersive virtual reality-assisted therapy for treatment-resistant auditory verbal hallucinations in individuals with schizophrenia spectrum disorders. Conducted across three Danish regions, the study tested the safety, feasibility, and efficacy of a novel intervention (Challenge-VRT) delivered via a head-mounted display that enables therapist-guided 3D dialogues with avatars representing distressing voices.

Led by the VIRTU Research Group and supported by multiple Danish research foundations, this trial represents a major step in translating virtual reality (VR) therapies into routine psychiatric care.

Target Group and Unmet Need

Persistent auditory verbal hallucinations are among the most debilitating symptoms in schizophrenia, often resistant to pharmacological treatment. Approximately one-third of patients experience little to no symptom relief from antipsychotics, and many face long-term distress and functional impairment. While cognitive-behavioral therapy offers some benefit, effect sizes are modest, and access is limited.

Challenge-VRT targets this gap by offering a brief, relational therapy co-developed with individuals who hear voices. The intervention builds on the AVATAR therapy framework but delivers it in a fully immersive environment, aiming to enhance emotional processing and patient engagement.

Intervention Design and Delivery

Challenge-VRT consists of seven weekly sessions and two booster sessions. Patients create a visual and auditory avatar of their most distressing voice with support from a therapist. Through therapist-voiced avatars in VR, patients engage in progressively more assertive dialogues across three therapeutic phases: (1) Reclaiming power, (2) Building self-worth, and (3) Recovery.

Therapists could switch between their natural and transformed voice in real time and adjust proximity and setting (e.g., beach scenes) to match arousal levels. The software was co-developed with lived-experience experts, and the intervention was delivered by psychologists, psychiatrists, and nurses trained locally and supervised remotely.

AVATAR strict teacher woman 1
AVATAR: Strict Teacher Woman. Image courtesy of Heka-VR

Key Findings

In total, 271 patients were randomized. At 12 weeks, Challenge-VRT significantly reduced the severity of auditory hallucinations compared to enhanced treatment-as-usual:

  • Primary outcome (PSYRATS-AH Total): Adjusted mean difference –2.26 (95% CI –4.26 to –0.25; p = 0.027), Cohen’s d = 0.27

  • Secondary outcome (PSYRATS-AH Frequency): Significant reduction at both 12 and 24 weeks

  • Satisfaction: Significantly higher in the Challenge-VRT group (12-week difference +2.74, p < 0.0001)

Although effects on distress and psychosocial functioning were not significant, exploratory findings suggest long-term impact on voice frequency may be maintained with booster sessions.

Implementation Feasibility and User Experience

The intervention was well tolerated:

  • 79% completed all sessions

  • Mean sessions completed: 6.08 (of 7 core) and 1.35 (of 2 booster)

  • High therapist fidelity and skill ratings

However, ~29% required extra preparation due to anxiety entering VR, and 48% experienced at least one technical issue. While most adverse events were unrelated to the intervention, five cases of voice exacerbation and one self-harm incident were deemed possibly related.

Participant testimonial:

“It felt like facing my voice in real life – and finally standing up to it.” Challenge-VRT participant

Behandler og patient 1
Behandler Org Patient. Image courtesy of Heka-VR

Challenges and Lessons Learned

  1. Initial Symptom Exacerbation: A temporary increase in symptoms during the early sessions was reported by 37% of participants, underscoring the need for pre-exposure preparation and emotional regulation tools.

  2. Software Stability: Technical issues occurred in nearly half of participants. Improving the VR platform’s robustness is essential before broader rollout.

  3. Comparator Heterogeneity: The control group received unstructured supportive counselling, complicating attribution of effects solely to the VR component.

Key Insights for Implementation

  • Scalability: The intervention can be delivered by existing mental health staff following brief training and remote supervision, making it viable for integration into routine care.

  • Short Format, Real Effects: Despite lasting just seven sessions, Challenge-VRT achieved outcomes comparable to longer therapies.

  • Collaborative Design: Involvement of individuals with lived experience enhanced both relevance and acceptability.

Next Steps and Global Relevance

VIRTU is currently developing adapted versions of Challenge-VRT for use in eating disorders and treatment-resistant depression, alongside ongoing efforts to integrate AI for real-time personalization of avatar responses. Additional work is exploring the feasibility of home-based booster sessions using portable VR technology. The findings from the CHALLENGE trial build on and extend the results of the recent AVATAR2 study, reinforcing the potential of immersive VR as a scalable, adaptable, and clinically acceptable modality in contemporary psychiatric care.

Further resources:

https://www.thelancet.com/journals/lanpsy/article/PIIS2215-0366(25)00161-0/abstract

Share this post

About the Author

Louise Birkedal Glenthøj

at VIRTU Research Group

Louise Birkedal Glenthøj is a Senior researcher, Doctor of Medical Science (Dr. Med.), PhD, associate professor, clinical psychologist, and specialist in psychotherapy.

VIRTU Research Group

Authors

Louise Birkedal Glenthøj

VIRTU Research Group

Partners

Virtu Research Group logo
Virtu Research Group
Strategic Partner

ADVERTISEMENT

Disclaimer

The views shared are those of the authors and do not necessarily reflect those of eMHIC.  For more details, see our Privacy Policy & Terms of Service

Our Audience

eMHIC has an audience of 26 member countries (and growing) with thousands of subscribers around the world.

Something to Share?

Contribute quality news and resources to the eMHIC Knowledge Bank. Your submissions will be carefully considered for our global community.

More Reading