Doctoral Research Explores the Future of Personalised Virtual Reality Exposure Therapy

How can virtual reality and artificial intelligence support more precise, ethical and effective approaches to exposure therapy? A doctoral research project by Kamilla Bergsnev at UiT The Arctic University of Norway investigates this question by bringing together fear-learning science, immersive virtual reality (VR), psychophysiology and interpretable AI.

The project, Next-Generation Exposure Therapy: Integrating Extinction Science, Personalisation and Artificial Intelligence in Virtual Reality, examines how virtual reality exposure therapy (VRET) can be designed around the mechanisms through which people learn that previously feared situations are safe.

Exposure therapy is one of the most effective treatments for fear- and anxiety-related disorders. Yet relapse and the return of fear remain significant challenges. VR offers an important opportunity to address this problem: it can create realistic but controlled environments where cues, contexts and intensity can be adjusted carefully. At the same time, AI may eventually help clinicians tailor these environments to individual needs—but only if its use is grounded in sound theory, robust evidence and meaningful human oversight.

The research has been developed in collaboration with Professor Ana Luisa Sánchez Laws, whose expertise in virtual reality, immersive learning and interdisciplinary methodologies contributes to the project’s conceptual, technical and methodological development.

From technology to learning mechanisms

A central contribution of the research is its shift away from treating VR or AI as solutions in themselves. Instead, it asks how these technologies can support established principles from learning science.

The work draws particularly on inhibitory learning theory, which understands exposure not as erasing a fear memory, but as creating new learning: a feared cue can be encountered without the anticipated negative outcome. This new learning must be strong enough to be retrieved across contexts and over time.

The research examines how elements such as timing, predictability, context and individual responses may shape this process in immersive environments. It also highlights the importance of distinguishing between promising laboratory findings and interventions that are ready for clinical use.

“The aim is not to automate therapy or replace clinical judgement,” says Bergsnev. “It is to understand which features of exposure can be meaningfully personalised, and how VR and AI might support therapists and patients in that process.”

Mapping personalisation and AI in VRET

The doctoral work includes two literature reviews co-authored by Bergsnev and Sánchez Laws.

The first, published in Frontiers in Virtual Reality, examined how researchers have approached personalisation in VR research and treatment for fear-related disorders. The review found that VR has considerable potential to tailor contexts and interactions, but that studies often use highly varied measures and make limited use of VR’s interactive possibilities.

The review called for a more theory-driven approach to personalisation—one in which measurements are directly connected to clearly defined hypotheses and mechanisms, rather than collected without a specific role in the intervention.

A second systematic review, “Artificial Intelligence (AI) for Virtual Reality Exposure Therapy (VRET): A Systematic Review,” was published in Translational Psychiatry. The review maps 23 studies and identifies three broad uses of AI in VRET:

• Machine learning for classifying states or predicting outcomes;

• Conversational AI for engagement, guidance and interaction within virtual environments;

• Knowledge-based and hybrid AI systems for therapist decision support and adaptive scenario design.

The review finds that AI-enabled VRET is still an emerging field. While some systems show promise, many studies lack strong theoretical integration, external validation and meaningful involvement of therapists and patients in the design process.

The authors argue that future systems should be human-in-the-loop: transparent, explainable and designed to support—not replace—clinical expertise.

A roadmap for responsible innovation

The research programme argues for a cautious and mechanism-first approach to innovation in mental health technology. It highlights the need for:

• exposure protocols grounded in established learning theory;

• transparent and interpretable AI tools;

• therapist oversight and the possibility to override automated recommendations;

• fairness and bias assessments;

• stakeholder involvement from patients, clinicians, engineers and ethicists;

• robust reporting, preregistration and open-science practices;

• clinical studies that compare VR-based and in vivo exposure while holding personalisation targets constant.

As the work develops, Bergsnev plans to further investigate what should be personalised in exposure therapy and to compare personalised VR exposure with personalised in vivo exposure. This future work will focus on whether VR adds value when the same therapeutic mechanisms, decision rules and therapist training are applied across both formats.

For EDGE Lab, the project contributes to a growing body of research on immersive technologies that is both technically innovative and critically grounded. It demonstrates how VR can serve not only as a delivery platform for interventions, but also as a controlled environment for investigating fundamental questions about learning, emotion and human–technology interaction.

Publications

• Bergsnev, K., & Sánchez Laws, A. L. (2022). Personalizing Virtual Reality for the Research and Treatment of Fear-Related Disorders: A Mini Review.Frontiers in Virtual Reality, 3, 834004.
  https://doi.org/10.3389/frvir.2022.834004

• Bergsnev, K., & Sánchez Laws, A. L. Artificial Intelligence (AI) for Virtual Reality Exposure Therapy (VRET): A Systematic Review. Translational Psychiatry. https://www.nature.com/articles/s41398-026-03936-4

HerStory in VR Receives Erasmus+ Funding: EDGE Researchers to Advance Immersive Storytelling, Gender and Conflict Memory

The Erasmus+ project HerStory in VR: Empowering Female Humanities Students through Digital Storytelling has received funding under the 2026 Cooperation Partnerships in Higher Education call.

Running from September 2026 to August 2029, the project brings together partners in Cyprus, Norway, Greece and Serbia to address the digital confidence gap among female humanities students through virtual reality, 360° video production and oral-history practice.

At UiT The Arctic University of Norway, the project will be supported by researchers from the Centre for Geopolitics, Peace and Security and the EDGE Lab. Dr Antigone Heraclidou will play a central role in UiT’s contribution, bringing her expertise in oral history, gender, conflict, difficult heritage and digital cultural heritage. Professor Ana Luisa Sanchez Laws will support Heraclidou’s work, contributing her longstanding research on immersive media, virtual reality, conflict transformation, presence and learning.

Heraclidou’s research sits at the intersection of historical testimony, contested memory and emerging technologies. In HerStory in VR, she will contribute to the project’s research and curriculum-design work, helping to develop ethical and pedagogically grounded approaches to documenting and presenting testimonies from women affected by conflict and displacement. She will also support UiT’s pilot activities in Norway.

The project is coordinated by CYENS Centre of Excellence in Cyprus and includes UiT, the University of Belgrade, the Greek Council for Refugees and ProjectSquared Ltd. UiT leads the project’s research and curriculum-design work package, drawing on the Centre for Peace Studies’ interdisciplinary expertise in peace and conflict studies, oral history, gender, and immersive technologies.

Building digital confidence through meaningful storytelling

HerStory in VR will develop an openly licensed four-module curriculum for female humanities students with no prior experience in virtual reality or digital production. Participants will gain practical skills in oral-history methodology, ethical and trauma-informed interviewing, 360° video production, digital archiving and immersive storytelling.

Rather than treating digital skills as separate from humanities education, the project positions technology as a tool for humanistic inquiry and public engagement. Students will use immersive media to document stories of women with lived experience of conflict, displacement or post-conflict transition.

The project will pilot the curriculum with at least 60 students across Cyprus, Norway, Greece and Serbia. Through a carefully governed process of testimony collection and production, the initiative will also create a collection of immersive oral histories for the HerStory VR application.

An immersive archive of women’s testimonies

Developed by CYENS, the HerStory VR application will use a “Matrix of Voices” format to present curated 360° testimonies in an immersive environment. The final application is expected to host 20 testimonies, selected from student-produced material across the four pilot countries.

A dedicated ethics and safeguarding protocol will guide all testimony work. The project places particular emphasis on informed consent, participant control, multilingual access, trauma-informed practice and psychological safety when working with sensitive narratives.

For EDGE Lab, the project offers an opportunity to further explore how VR can support engagement with difficult heritage and conflict-related experiences—while also critically considering the ethical responsibilities involved in designing immersive encounters with testimony.

Open resources and long-term impact

The project will produce a freely available curriculum and Facilitator Guide in English, Greek, Serbian and Norwegian, alongside an evaluation report and policy brief on gender-sensitive digital education in higher education.

Its outputs will be published open access, enabling universities, cultural institutions and civil-society organisations to adapt the model beyond the partner institutions. The project will also include public VR exhibitions in Cyprus, Norway, Greece and Serbia, culminating in a final dissemination event in Cyprus.

By combining digital skills development with oral history and women’s conflict narratives, HerStory in VR aims to expand access to immersive technologies while ensuring that underrepresented experiences become part of Europe’s evolving digital cultural record.

AI meets virtual reality therapy: new review maps what works, what’s missing, and what’s next

A systematic review in Nature’s Translational Psychiatry charts how artificial intelligence is being woven into virtual reality exposure therapy (VRET) to treat fear- and anxiety-related disorders—and where the science needs to go next.

The open‑access paper, “Artificial intelligence (AI) for virtual reality exposure therapy (VRET): A systematic review,” by Kamilla Bergsnev and Ana Luisa Sánchez Laws (UiT The Arctic University of Norway), evaluates how AI is being used to personalize VR-based exposure, support clinicians, and engage patients. The bottom line: early signals are promising, but much of the work remains proof‑of‑concept and too often excludes the very people it aims to help.

Why this matters

• Exposure therapy is a gold‑standard treatment for phobias, PTSD, OCD, and related conditions, but it’s underused in routine care due to logistical hurdles and clinician/patient concerns.

• VRET can make exposure safer and more practical by simulating feared situations. AI could take it further—tailoring intensity, pacing, and context in real time, and scaling support to more patients.

• Getting that promise right requires tight alignment with exposure/extinction theory and strong “human‑in‑the‑loop” safeguards.

How the review was done

• Scope: Peer‑reviewed studies through November 14, 2025; 377 records screened; 23 studies included.

• Databases: PsycINFO, Web of Science, Google Scholar, EMBASE, CINAHL, MEDLINE.

• Method: PRISMA‑guided narrative synthesis clustered studies into:

  1. machine learning (ML),

  2. conversational AI, and

  3. knowledge‑based/hybrid AI.

• Quality: Risk‑of‑bias (RoB) appraisal classified 9 studies as Low RoB, 9 as Medium, 5 as High (all High RoB were ML). Primary conclusions emphasize Low/Medium RoB; High RoB results are treated as sensitivity analyses.

What the authors found

1. Machine learning: modest gains, best with multimodal data

• Use‑cases: outcome prediction (who responds, symptom severity), state classification for biofeedback (e.g., fear/arousal), and neurofeedback frameworks.

• Signal: ML can classify anxiety states from physiological and behavioral signals and sometimes predict severe symptom subgroups better than chance, especially with multimodal inputs (e.g., combining voice, heart rate, movement).

• Limits: Generalizability is weak; many models depend on small, homogeneous samples and don’t transfer well across sites. fMRI‑based approaches show promise but are resource‑intensive.

• Process gap: Over half of ML studies set goals without clinician/patient input; few anchored AI logic in contemporary exposure/extinction theory.

2. Conversational AI: engagement and coaching, not replacement

• Use‑cases: virtual agents that coach patients and scaffold sessions inside VR.

• Signal: Early prototypes show high patient acceptance as a supplement to human‑led therapy; automation can reduce therapist workload and improve session flow.

• Limits: No clear superiority over non‑VR comparators in small studies; speech naturalness, empathy, and continuity across sessions remain challenges; unsupervised use raises safety concerns.

• Best practice emerging: Keep therapists “in the loop”—AI can propose, but humans should decide, especially when safety thresholds are involved.

3. Knowledge‑based/hybrid AI: decision support and adaptive scenarios

• Use‑cases: systems that encode exposure rules and adapt VR scenarios to patient responses, sometimes blending rules with learned models.

• Signal: Case reports (e.g., PTSD simulations with haptics) suggest symptom reductions and high stress‑classification accuracy in controlled settings; therapist decision support often aligned with experienced clinicians’ choices.

• Limits: Small samples; maintenance burden for rule‑based systems; risk of over‑automation if tools don’t respect therapist styles and clinical judgment.

Cross‑cutting concerns

• Theory integration is thin: Few applied systems embed modern exposure/extinction principles (e.g., inhibitory learning, expectancy violation, variability, context shifts) into AI decision-making.

• Stakeholder involvement lags: Higher‑quality studies more often involve clinicians and patients; many ML efforts remain “out‑of‑the‑loop,” risking clinically misaligned goals.

• Ethics and equity are underaddressed: Few studies assess model explainability, calibration, subgroup error rates, or fairness; privacy and data‑governance plans are rarely detailed.

• Implementation barriers: Sensor cost and complexity, external validation needs, and clinician training for interpreting probabilistic outputs slow adoption.

What the authors recommend

• Build larger, multi‑site datasets and prioritize external validation to improve generalizability.

• Make AI theory‑driven: encode expectancy violation, variability, spacing, and context shifts; use AI to optimize exposure schedules within safety bounds.

• Compare head‑to‑head: run trials pitting AI‑enabled VRET against traditional VRET to identify who benefits, when, and why.

• Keep humans in the loop: design for therapist oversight, with clear override controls and session memory. Align with emerging guidance (EU AI Act; APA ethics).

• Demand explainability and accountability: report calibration, uncertainty, subgroup errors; publish model cards/data sheets; audit decisions informed by AI.

• Co‑design from day one: include patients, clinicians, engineers, ethicists, and equity experts; build AI literacy for therapists and VR literacy for developers.

The takeaway
AI is starting to make VRET smarter—personalizing stimuli, enhancing engagement, and supporting clinician decisions. But the field is early. The most credible path forward keeps therapists at the helm, bakes exposure theory into code, and involves patients and clinicians at every step. Done well, AI‑enabled VRET could expand access to high‑quality exposure therapy without sacrificing safety or the therapeutic alliance.

Paper details

• Title: Artificial intelligence (AI) for virtual reality exposure therapy (VRET): A systematic review

• Authors: Kamilla Bergsnev; Ana Luisa Sánchez Laws

• Journal: Translational Psychiatry (Nature), Volume 16, Article 208 (2026)

• Published: 26 March 2026

• Open access: Creative Commons Attribution 4.0

• DOI: https://doi.org/10.1038/s41398-026-03936-4

Ana Luisa Sanchez Laws  –  Professor

Kamilla Bergsnev

Antigone Heraclidou  –  Postdoctoral Fellow

Ania Malinowska  –  Associate Professor

Bernhard Spanlang

Gunhild Hoogensen Gjørv  –  Professor

Laurence Kroese

Marcela Douglas Aranibar  –  Associate Professor