VR Social Skills Training for Autism: A Safe Rehearsal Space

The problem: practicing social skills without paying the price of failure

Social skills training is a classic intervention for people with autism spectrum disorder, especially adolescents and adults without associated intellectual disability. It works on reading social cues, initiating and sustaining conversations, regulating eye contact, interpreting others' intentions, and managing specific interpersonal situations (interviews, encounters with strangers, group settings).

The practical obstacle is well known. Practicing social skills in session through role-play has a ceiling: the clinician cannot replicate the speed, unpredictability, and real social pressure of an everyday situation. Practicing out in the world exposes the person to a high emotional cost if things go wrong: alongside the cognitive feedback, they may face real shame, frustration, or rejection. This generates avoidance, which perpetuates the condition.

Virtual reality offers a middle ground: social situations realistic enough to activate the target skills, yet controlled, repeatable, and free of the emotional cost of real-world failure. The relevant clinical question is whether that practice generalizes to the natural context.

What the research says: Didehbani, Kandalaft, Maskey

Three lines of research are worth knowing.

Didehbani and colleagues at the University of Texas at Dallas have published studies on social cognition training and conversation skills using VR in children and adolescents with ASD without intellectual disability. Their protocols combine virtual social situations (a first day of school, a job interview, meeting a peer) with structured feedback from the clinician. Available results show improvements in self-reported social cognition and social self-efficacy, with partial generalization to the natural context within a few months. About us About us

Kandalaft and colleagues, from the same team, specifically explored training for job interviews and workplace situations in young adults with ASD. The pilot studies report effects on subjective confidence and self-reported performance in subsequent real interviews, with small samples and, in some cases, designs lacking an active control group.

Maskey and colleagues at Newcastle University have studied the use of VR for phobias and specific anxiety in children and young people with ASD, with consistent results suggesting good clinical tolerability and effects on phobic symptoms comparable to those described in the general population.

The conservative reading of this body of literature is that VR is a useful clinical tool for this population profile, with partial generalization to the natural context and good tolerability when sensory conditions are properly calibrated. It is not a standardized treatment, and it does not replace multimodal intervention programs.

Sensory considerations: the risk of excluding those who need it most

Here, in our reading, is the critical factor that separates responsible clinical use from naive use. People with ASD frequently show differences in sensory processing: hyperreactivity or hyporeactivity to visual, auditory, and vestibular stimuli. A VR headset is, by design, a high sensory-load device: stereoscopic imagery close to the eyes, spatial audio, and the possibility of simulated motion.

If the clinician introduces VR without assessing the patient's sensory profile, two things can happen. The patient may tolerate the immersion well and benefit from the training (the usual case for much of the ASD population without intellectual disability). Or the patient may experience sensory overload, marked distress, visual fatigue, or nausea, abandon the tool, and add one more negative experience to their history.

A reasonable protocol includes a prior assessment of the sensory profile (inventories already available in ASD clinical practice), gradual introduction with low-stimulation scenarios at the start, active control of virtual lighting and sound, short initial sessions (10-15 minutes), active review of tolerance after each session, and a willingness to discontinue the tool if tolerance does not improve with adjustments.

A headset poorly calibrated to a specific sensory profile is not neutral. It is a poor intervention. This holds for the general population too, but in ASD the likelihood of that outcome is higher.

Concrete clinical applications in practice

Within a broader intervention program (psychoeducation, family work where applicable, intervention for comorbid anxiety or depression, coordination with educational or occupational services), VR can contribute to several specific components.

Social cognition training: conversation scenarios where the patient practices starting, sustaining, and closing dialogues with avatars. The clinician can pause the scene, comment on what is happening, rehearse alternative responses, and repeat.

Preparation for specific situations: an upcoming job interview, an oral exam, a significant family gathering. Practicing the specific situation in VR, with calibrated scenarios, reduces the burden of anticipatory anxiety and allows the patient to rehearse responses.

Exposure to specific phobias or social anxiety, which frequently co-occur with ASD. The standard VR exposure protocol transfers well to this profile, with the sensory precautions mentioned above (Maskey and colleagues).

Emotional regulation work in calm, predictable environments: nature scenarios, immersive mindfulness exercises. Useful both as an in-session tool and as an anchor for skills to be used outside the practice.

What VR is not

Some expectations are worth defusing to avoid clinical misuse of the tool.

VR is not social training that replaces real social contact. It is preparatory practice. The therapeutic goal is always generalization to the natural context, and that requires the patient to practice out in the world at some point. If VR becomes a comfortable substitute for real contact, it is being misused.

VR is not an intervention specific to autism. It is an immersive tool applied to diverse populations. Clinical protocols for ASD remain the clinical protocols for ASD, with everything that implies (specific clinician training, multidisciplinary coordination, individualized approach).

And VR does not resolve the underlying question of what is and is not desirable to work on in people with ASD. The neurodivergent movement raises objections, with reasons that deserve to be heard, to certain historical approaches to autism that sought to make a neurodivergent person pass as neurotypical. The responsible clinician works with the patient toward their own life goals, not toward imposing a normative model of social functioning. VR is a neutral tool; the clinical use of that tool is not.

How we integrate this into VRET

VRET's current scenarios that can support work with ASD profiles are the immersive mindfulness scenarios (emotional regulation, sustained attention training, predictable and low-stimulation environments) and the exposure scenarios for specific phobias and enclosed spaces when there is comorbid phobic or anxious symptomatology.

We do not currently offer a dedicated social cognition training module with avatars. It is a complex area of clinical development (requiring specific validation, fine-tuned adjustments to the sensory profile, and specialized ASD clinical supervision) that we are pursuing in collaboration with clinicians who work with this population.

For licensed clinicians specializing in ASD who are integrating VR into their practice, we offer demos with the VRET clinical team to review the suitability of current scenarios, sensory parameter calibration, and operational precautions.

This article is for informational purposes for psychology professionals. It is not clinical advice for any individual case and does not replace the judgment of the licensed psychologist in charge. VRET is professional clinical-support software, not a CE-marked medical device.