Speaking AnxietySocial Communication

Three-arm RCT (n=60) of stand-alone VR exposure therapy vs in-vivo exposure vs waiting-list for social anxiety disorder: in-vivo exposure was superior to VRET on multiple secondary outcomes - the opposite pattern to Bouchard 2017

Kampmann IL et al. · 2016 · Behaviour Research and Therapy · RCT · n = 60 · Adults with social anxiety disorder · DOI
Evidence certainty: Moderate certainty
How this was rated

Pre-specified three-arm RCT with formal randomisation, manualised treatments, and pre-post-follow-up assessment - a strong design. Sixty participants is modest for a three-arm comparison but consistent with the SAD VRET RCT literature of the period. Peer-reviewed in Behavior Research and Therapy (Elsevier, established high-impact peer-reviewed venue in clinical psychology). Multilevel regression analyzes appropriate to the design. Limitations: stand-alone VRET without cognitive components is a deliberately stripped-down test condition - the result that iVET outperforms VRET applies most directly to this stripped-down configuration. The Bouchard et al. 2017 trial (CBT-embedded VRET) found VRET superior to iVET, indicating the integration with cognitive components matters substantially.

Ratings use a simplified four-tier scheme (High, Moderate, Low, Very Low) informed by the GRADE working group. Learn more about how studies are rated.

Sixty participants diagnosed with social anxiety disorder were randomly assigned to individual virtual reality exposure therapy (VRET), individual in-vivo exposure therapy (iVET), or a waiting-list control. Both treatments improved social anxiety symptoms, speech duration, perceived stress, and avoidant-personality-disorder beliefs versus waitlist. However, iVET (but NOT VRET) also improved fear of negative evaluation, speech performance, general anxiety, depression, and quality of life. iVET was superior to VRET on social anxiety symptoms at post and follow-up and on avoidant-PD beliefs at follow-up. At follow-up, almost all improvements remained significant for iVET; for VRET only the perceived-stress effect held.

Clinical bottom line

A three-arm RCT (n=60) directly comparing stand-alone VRET (without cognitive components) versus in-vivo exposure therapy (iVET) versus waitlist for DSM-diagnosed SAD. Both treatments beat waitlist on core social-anxiety outcomes, BUT in-vivo exposure was MORE effective than VRET on multiple secondary outcomes (fear of negative evaluation, speech performance, general anxiety, depression, quality of life) and was superior to VRET on social-anxiety symptoms at post and follow-up. This is the opposite pattern to Bouchard et al. 2017 (which found VRET superior to in-vivo when both were embedded in CBT). The likely mechanism for the divergence: Kampmann's VRET was stand-alone without cognitive components, while Bouchard's VRET was embedded in a CBT package. Clinically, this argues for INTEGRATING VRET into CBT rather than using it stand-alone.

Key findings

  • 60 participants (mean age 36.9 years, 63.3% women) diagnosed with social anxiety disorder were randomly assigned to individual VRET, individual in-vivo exposure therapy (iVET), or waiting-list
  • VRET in this study was STAND-ALONE - comprising verbal interaction with virtual humans WITHOUT any cognitive components. This is a deliberately stripped-down condition to isolate the exposure mechanism
  • Both VRET and iVET improved social anxiety symptoms, speech duration, perceived stress, and avoidant-personality-disorder-related beliefs from pre to post, relative to waitlist
  • iVET (but NOT VRET) additionally improved fear of negative evaluation, speech performance, general anxiety, depression, and quality of life relative to waitlist
  • iVET was SUPERIOR to VRET on social anxiety symptoms at post-treatment and at follow-up, and on avoidant-PD-related beliefs at follow-up
  • At follow-up, all improvements were significant for iVET; for VRET only the perceived-stress effect remained significant
  • The pattern OPPOSES Bouchard et al. 2017 (BJPsych three-arm RCT), which found VRET superior to in-vivo when both were embedded in CBT. The most plausible mechanism: VRET's gains compound when paired with cognitive components, but stripped of cognitive content the in-vivo modality is more effective

Background

By the mid-2010s, VRET for social anxiety disorder had been established as effective relative to waitlist (Powers & Emmelkamp 2008 meta-analysis, multiple individual RCTs), but the comparison against the gold-standard in-vivo exposure remained underexplored. Most prior VRET-vs-iVET comparisons confounded the exposure-delivery method with the cognitive-behavioral treatment surrounding it. The authors set out to isolate the exposure mechanism by comparing stand-alone VRET (verbal interaction with virtual humans, no cognitive components) against stand-alone iVET, both versus waitlist.

What the researchers did

Sixty participants diagnosed with SAD were randomly assigned to one of three arms:

  1. VRET - individual virtual reality exposure therapy, stripped to stand-alone exposure (no cognitive components)
  2. iVET - individual in-vivo exposure therapy, similarly stand-alone
  3. Waiting-list control

Both active treatments followed manualised protocols. Outcomes were measured at pre-treatment, post-treatment, and follow-up. The primary statistical approach was multilevel regression.

What they found

Versus waitlist (both VRET and iVET):

iVET (but NOT VRET) versus waitlist - additional outcomes:

iVET versus VRET head-to-head:

Durability at follow-up:

Why this matters

The result opposes Bouchard et al. 2017 (three-arm BJPsych RCT, in this Hub as bouchard-2017), which found VRET superior to in-vivo when both were embedded in CBT. The most plausible mechanism for the divergence: Kampmann’s VRET was deliberately stand-alone, stripped of cognitive components. Bouchard’s VRET was embedded in a full CBT package. The clinical message: VRET’s gains appear to compound when paired with cognitive work; stripped of cognitive content, in-vivo exposure is the more effective modality.

For clinicians considering VRET for SAD or PSA - including for PWS with social-anxiety comorbidity - the takeaway is to integrate VRET with cognitive components and CBT structure, not to use it as a pure exposure-delivery mode and expect it to match in-vivo exposure.

Limitations

Implications for practice

For clinicians considering VRET for SAD/PSA, this RCT delivers an important message: stand-alone VRET (verbal interaction with virtual humans, no cognitive components) is less effective than in-vivo exposure on multiple outcomes. This DOES NOT mean VRET is inferior to in-vivo in general - the contrasting Bouchard 2017 result demonstrates that VRET integrated into CBT outperforms in-vivo exposure integrated into CBT. The clinical takeaway is: INTEGRATE VRET with cognitive components and CBT structure. Do not use VRET as a pure exposure-delivery method stripped of cognitive work and expect it to match in-vivo exposure. For PWS, where social anxiety is often comorbid and benefits from integrated cognitive-behavioral work, this argues for VRET embedded in a stuttering-and-social-anxiety CBT framework rather than stand-alone.

Related Studies

Social CommunicationSpeaking AnxietyGeneralization

VR exposure is as effective as in vivo exposure group therapy for public-speaking-focused social anxiety, with lasting results

In a randomized controlled trial of 97 adults with social anxiety disorder whose primary fear was public speaking, eight sessions of virtual reality exposure were as effective as eight sessions of manualized in vivo exposure delivered in groups. Both active treatments outperformed waitlist on self-report and on a behavioral speech task, and improvements were maintained at 12-month follow-up.

Anderson PL et al. · 2013 Read Summary
Speaking AnxietySocial CommunicationGeneralization

First long-term (4-6 year) follow-up of VR exposure therapy and exposure group therapy for social anxiety disorder: durable gains, with 54% no longer meeting diagnostic criteria

Twenty-eight participants from the Anderson et al. 2013 RCT of VR exposure therapy (VRET) vs exposure group therapy for DSM-diagnosed social anxiety disorder were re-assessed a mean of 6 years (range 4-6) after treatment completion. Self-report measures, behavioral speech tasks, and diagnostic interviews all showed statistically significant improvement from pre-treatment to long-term follow-up. The majority (54%) no longer met diagnostic criteria for SAD; 68% rated themselves 'very much' or 'much' improved. With one exception, there were no differences between VRET and exposure group therapy at follow-up - both produce durable long-term benefits, consistent with the broader CBT-for-SAD evidence base.

Anderson PL et al. · 2017 Read Summary
Speaking AnxietySocial CommunicationGeneralization

Pre-registered three-arm RCT: CBT with VR exposure was superior to in-vivo exposure on the primary social-anxiety outcome - and more practical for therapists

A pre-registered three-arm RCT randomized 59 adults with DSM-5 social anxiety disorder to 14 weekly sessions of CBT with VR exposure (n=17), CBT with in-vivo exposure (n=22), or waitlist (n=20). The pre-specified superiority hypothesis was confirmed: VR exposure was significantly more effective than in-vivo exposure on the LSAS-SR primary outcome at post-treatment and at 6-month follow-up, and significantly less cumbersome for therapists (SWEAT: 15.24 vs 24.46). Reliable change: 76.5% VR, 68.3% in vivo, 30.0% waitlist.

Bouchard S et al. · 2017 Read Summary
Speaking AnxietySocial Communication

Preliminary controlled study comparing VR therapy versus group-CBT for social phobia (n=36): both treatments produced statistically and clinically significant improvement, with trivial differences between the two

Thirty-six participants diagnosed with social phobia were allocated to either virtual reality therapy (VRT) using four virtual environments (performance, intimacy, scrutiny, assertiveness situations) OR a control condition of group cognitive-behavioral therapy (CBT). Both treatments lasted 12 weeks and were delivered according to a treatment manual. Both produced statistically and clinically significant improvement. Effect-size comparisons revealed that the differences between VRT and the active group-CBT comparator were trivial - establishing VRT as a non-inferior alternative to evidence-based group-CBT for social phobia.

Klinger E et al. · 2005 Read Summary
Speaking AnxietyImplementation

First RCT (n=25+25) showing that consumer VR hardware and software can deliver effective one-session VRET for public speaking anxiety - both therapist-led (d=1.67) and self-led at home (d=1.35), with gains maintained at 6 and 12 months

Twenty-five participants were randomised to one-session therapist-led VR exposure therapy for public speaking anxiety using consumer VR hardware and software, followed by a 4-week internet-administered VR-to-in-vivo transition program; another 25 served as a waiting-list. Therapist-led VRET produced a very large effect on self-reported PSA (within Cohen's d = 1.67). The waiting-list then received internet-administered, SELF-LED VRET at home, followed by the same transition program - producing a large effect (d = 1.35). Results were maintained or improved at 6-month and 12-month follow-ups. This is the first published RCT demonstrating that off-the-shelf consumer VR hardware and software can deliver effective PSA exposure therapy in both clinician-supervised and home-based formats.

Lindner P et al. · 2019 Read Summary
Speaking AnxietyGeneralizationImplementation

Meta-analysis: VR exposure works as well as evidence-based alternatives, with a dose-response pattern

A quantitative meta-analysis of VR exposure for anxiety disorders found that VR-based treatment produced large gains over waitlist, equivalent gains to established evidence-based alternatives, and a clear dose-response relationship - more sessions produced more benefit.

Opriş D et al. · 2012 Read Summary
Speaking AnxietyGeneralization

Meta-analysis: VR exposure therapy works as well as in-person exposure for anxiety

An early quantitative meta-analysis pooled effect sizes across studies of VR exposure therapy for anxiety disorders, finding VR exposure equally effective as in-vivo (real-world) exposure and significantly more effective than control conditions.

Powers MB, Emmelkamp PMG · 2008 Read Summary
Speaking AnxietyGeneralization

VR exposure equals traditional CBT for public-speaking anxiety, with lower dropout

In an 88-person RCT, VR-based cognitive-behavioral therapy for public-speaking anxiety produced anxiety reductions equivalent to traditional CBT and significantly larger than waitlist, with notably fewer participants dropping out of the VR arm.

Wallach HS et al. · 2009 Read Summary
Speaking AnxietySocial CommunicationImplementation

Pilot RCT (n=44) of brief self-guided virtual reality exposure therapy for social anxiety disorder: moderate-to-large effects on SAD severity, job interview fear, and trait worry, maintained at 3 and 6 months

Forty-four community-dwelling or undergraduate adults diagnosed with social anxiety disorder (SAD) using the Mini International Neuropsychiatric Interview were randomly assigned to a self-directed VR exposure intervention (designed to last four sessions or more; n=26) or a waitlist control (n=18). Outcomes measured at baseline, post-treatment, 3-month follow-up, and 6-month follow-up. VR exposure produced moderate-to-large reductions in SAD symptom severity, job interview fear, and trait worry (Hedges' g = 0.54 to 1.11). Although between-group differences in depression were not significant, the VR arm reduced depression while waitlist did not. Gains were maintained at 3- and 6-month follow-up. Self-reported presence increased during treatment (g = 0.36 to 0.45); cybersickness decreased (g = 0.43).

Zainal NH et al. · 2021 Read Summary

Cite this study

If you reference this study in your work, the canonical citation formats are:

APA 7th
Kampmann, I. L., Emmelkamp, P. M. G., Hartanto, D., Brinkman, W. P., Zijlstra, B. J. H., & Morina, N. (2016). Exposure to virtual social interactions in the treatment of social anxiety disorder: A randomized controlled trial. Behaviour Research and Therapy. https://doi.org/10.1016/j.brat.2015.12.016.
AMA 11th
Kampmann IL, Emmelkamp PMG, Hartanto D, Brinkman WP, Zijlstra BJH, Morina N. Exposure to virtual social interactions in the treatment of social anxiety disorder: A randomized controlled trial. Behaviour Research and Therapy. 2016. doi:10.1016/j.brat.2015.12.016.
BibTeX
@article{kampmann2016,
  author = {Kampmann, I. L. and Emmelkamp, P. M. G. and Hartanto, D. and Brinkman, W. P. and Zijlstra, B. J. H. and Morina, N.},
  title = {Exposure to virtual social interactions in the treatment of social anxiety disorder: A randomized controlled trial},
  journal = {Behaviour Research and Therapy},
  year = {2016},
  doi = {10.1016/j.brat.2015.12.016},
  url = {https://withvr.app/evidence/studies/kampmann-2016}
}
RIS
TY  - JOUR
AU  - Kampmann, I. L.
AU  - Emmelkamp, P. M. G.
AU  - Hartanto, D.
AU  - Brinkman, W. P.
AU  - Zijlstra, B. J. H.
AU  - Morina, N.
TI  - Exposure to virtual social interactions in the treatment of social anxiety disorder: A randomized controlled trial
JO  - Behaviour Research and Therapy
PY  - 2016
DO  - 10.1016/j.brat.2015.12.016
UR  - https://withvr.app/evidence/studies/kampmann-2016
ER  -

Know of research that should be in this hub? If a relevant peer-reviewed study is not listed here, send the reference to hello@withvr.app. The hub is kept up to date as the literature grows.

Funding & independence

Affiliations: Department of Clinical Psychology, University of Amsterdam; Interactive Intelligence Group, Delft University of Technology; Netherlands Institute for Advanced Study; King Abdulaziz University. Funding sources not extracted in detail from the abstract sections used for this summary. Peer-reviewed in Behavior Research and Therapy (Elsevier). No withVR BV involvement in funding, study design, or authorship. Summary prepared independently by withVR using the published peer-reviewed paper. The VR system used was an era-appropriate research configuration developed in collaboration with Delft's Interactive Intelligence Group, NOT Therapy withVR or Research withVR.

Last reviewed: 2026-05-17 Next review due: 2027-05-17 Reviewed by: Gareth Walkom