Using Virtual Reality to Treat Social Anxiety in Autistic Adolescents

Using Virtual Reality Assisted Therapy for Social Anxiety in Adolescents With Autism Spectrum Disorder: A Case Series

South London and Maudsley NHS Foundation Trust, National Institute for Health Research, United Kingdom, Guy's and St Thomas' NHS Foundation Trust

This study will test the feasibility and acceptability of virtual reality assisted cognitive behavioural therapy for the treatment of social anxiety in autistic adolescents. Five adolescents will receive the intervention and a parent/caregiver of each adolescent will be asked to act as informants on some questionnaires and interviews.

Difficulties interacting with others in social situations is a core characteristic of autism spectrum disorder (ASD). These difficulties are intensified by social anxiety - commonly experienced in autistic adolescents. A promising psychological treatment is cognitive behavioural therapy (CBT), but there are constraints in its use for the autistic population. For example, CBT requires patients to imagine being in a social situation that would cause anxiety, but autistic people often experience difficulties with imagination. The use of virtual reality (VR) in CBT may help to lessen such constraints. VR uses 3D computer-generated visual environments displayed through a head-mounted display. The images are synchronised to the movements of the user such that they experience feeling immersed in the virtual scene. The virtual scene can involve social scenarios, making it an ideal tool for eliciting social anxiety in the treatment setting. The investigators can see if these virtual scenarios are indeed eliciting social anxiety by recording physiological responses (e.g. heart rate) in parallel with patients' anxiety ratings during exposure. This is important for CBT to be effective. This study will test the feasibility and acceptability of VR-assisted CBT, combined with the use of physiological measurements, in autistic adolescents experiencing social anxiety. The therapy will typically consist of 8-20 weekly sessions and will be delivered by clinical psychologists in local mental health services. Conventional outcome measures will be used and feedback from youth and their parents/caregivers will be requested. The results of this study may lead to modification of the treatment and research methods. They will inform a later pilot randomised controlled trial.

Between 8 to 20 individual in-person sessions of VR-assisted CBT will be delivered on a weekly basis by NHS therapists who are trained in delivering CBT to this patient group.

The intervention will be delivered in NHS services and will involve using VR to assist the delivery of conventional CBT for the treatment of social anxiety in autistic adolescents. Social situations that are commonly anxiety-evoking for this patient group will be simulated in VR and used for exposure (a typical component of CBT for social anxiety). Therapists will use a modular approach involving optional modules such as psychoeducation on anxiety in the context of ASD, managing expectations of outcome and emotional literacy training. Essential modules include rapport building, introducing the use and purpose of VR, and exposure (incorporating VR). Because CBT is delivered in routine clinical practice, the VR exposure sessions will be the main subject of interest.

This will be indicated by the proportion of the patients who commenced the VR component that completed it. Completion is defined as 75% attendance of the total VR exposure sessions recommended by the therapist. Not all therapy sessions involve VR exposure.

Proportion of eligible patients successfully recruited (i.e. the proportion of patients approached for whom consent was provided to participate).

The proportion of participants completing each stage of the study (i.e. baseline assessments/measures, therapy and therapy measures, and post-treatment assessments/measures).

This is a 25-item self-report questionnaire. Participants are asked to rate the extent they agree with each statement using a 5-point Likert scale (0 = Not at all; 4 = Very much). Statements refer to their experience of the therapy and whether taking part has caused any distress. Higher scores indicate increased adverse effects.

Qualitative feedback regarding the intervention will be sought from therapists and participants using a semi-structured interview. The interview will follow a topic guide aimed at understanding opinions on helpful aspects of the therapy, suggestions for improvement, and the perceived level of social presence (i.e. the subjective experience of being with the non-player characters) in the VR environments. Feedback will be summarised.

This will be measured using Goal Attainment Scaling (GAS). Goals are agreed at the start of therapy. Initially patients rate the importance, difficulty, and how they are functioning at baseline, for each goal. After therapy patients indicate whether the goal has been achieved and rate the outcome (e.g. +2 better than expected, -2 worse than expected).

Measured using the Revised Children's Anxiety and Depression Scale (and Subscales) (RCADS; Chorpita et al. 2000). This is a 47-item self-report questionnaire. Both youth (suitable for ages 8 to 18 years) and parental/caregiver forms will be administered. Participants rate each item using a 4-point Likert scale (0 = Never; 3 = Always). The questionnaires are scored using spreadsheets provided by the developer. A t-score is computed based on the person's school year. T-scores of 70 and higher are classified as clinically significant. Higher t-scores indicate increased symptoms.

Measured using the Clinical Global Impression-Improvement Scale (CGI-I; Guy, 1976). At baseline, the researcher will rate their "impression" of illness severity using baseline qualitative data from the Anxiety Disorders Interview Schedule (ADIS-IV; Albano & Silverman, 1996) on a 7-point Likert scale (1 = normal, not ill at all; 7 = among the most severely ill patients). The ADIS-IV is a semi-structured clinical interview which will be conducted separately with patients and their parents/caregivers (informants). The same researcher will compare baseline ADIS-IV data with data obtained from a follow-up (qualitative) interview to rate "improvement". The follow-up interview will be semi-structured, follow a topic guide, and be conducted with the patient and their parent/caregiver (informants) separately. Topics will cover functioning across activities of daily living. "Improvement" will be rated on an 8-point Likert scale (1 = completely recovered; 5 = no change; 8 = very much worse).

Obtained from a galvanic skin response sensor on an Empatica E4 wristband taking continuous measurements. Measured in μS. Sampled at 4Hz. Recordings will only be made during the first and final VR exposure sessions. In both sessions, this will involve a 5-minute baseline recording while participants experience a relaxing VR scene. This will be followed by VR exposure for <1 hour during which recordings are also made. A button on the device will be used to timestamp the start and end of VR relaxation/exposure tasks.

First VR exposure session (up to 19 weeks); Final VR exposure session (up to 20 weeks). Timepoints of the first and final VR exposure session will be decided by the therapist case-by-case, see Assigned Intervention for details.

Inter-beat intervals (RR intervals) can be computed from blood volume pulse which will be measured using a photoplethysmography sensor on an Empatica E4 wristband. RR intervals are the successive time difference between heartbeats in ms which can be transformed into beats per minute. Sampled at 64Hz. Recordings will only be made during the first and final VR exposure sessions. In both sessions, this will involve a 5-minute baseline recording while participants experience a relaxing VR scene. This will be followed by VR exposure for <1 hour during which recordings are also made. A button on the device will be used to timestamp the start and end of VR relaxation/exposure tasks.

First VR exposure session (up to 19 weeks); Final VR exposure session (up to 20 weeks). Timepoints of the first and final VR exposure session will be decided by the therapist case-by-case, see Assigned Intervention for details.

Computed from heart rate measured using a photoplethysmography sensor on an Empatica E4 wristband. Sampled at 64Hz. HRV will be measured by calculating the root mean square of successive RR interval differences (RMSSD). Units: ms. Recordings will only be made during the first and final VR exposure sessions. In both sessions, this will involve a 5-minute baseline recording while participants experience a relaxing VR scene. This will be followed by VR exposure for <1 hour during which recordings are also made. A button on the device will be used to timestamp the start and end of VR relaxation/exposure tasks.

First VR exposure session (up to 19 weeks); Final VR exposure session (up to 20 weeks). Timepoints of the first and final VR exposure session will be decided by the therapist case-by-case, see Assigned Intervention for details.

A composite index will be computed for overall body movement during the sampling period using a 3-axis accelerometer. Sampled at 32Hz. Range [-2g, 2g]. An index for overall movement will be extracted using the standard Euclidean metric for each axis sensor at each sampling point and summed for the total recording sample length. Recordings will only be made during the first and final VR exposure sessions. In both sessions, this will involve a 5-minute baseline recording while participants experience a relaxing VR scene. This will be followed by VR exposure for <1 hour during which recordings are also made. A button on the device will be used to timestamp the start and end of VR relaxation/exposure tasks.

First VR exposure session (up to 19 weeks); Final VR exposure session (up to 20 weeks). Timepoints of the first and final VR exposure session will be decided by the therapist case-by-case, see Assigned Intervention for details.

Measured using the Subjective Units of Distress Scale (SUDs; Kim et al., 2008; Wolpe, 1973). This is a self-report scale with ratings ranging from 0 (no distress) to 10 (highest distress/anxiety ever). Ratings acquired every 120 seconds using a visual analogue scale (fear thermometer) in the VR environment. Ratings will only be collected during the first and final VR exposure sessions. In both sessions, this will involve 5-minute baseline ratings while participants experience a relaxing VR scene. This will be followed by VR exposure for <1 hour during which ratings are collected. A button on the device will be used to timestamp the start and end of VR relaxation/exposure tasks.

First VR exposure session (up to 19 weeks); Final VR exposure session (up to 20 weeks). Timepoints of the first and final VR exposure session will be decided by the therapist case-by-case, see Assigned Intervention for details

Measured using the ITC-Sense of Presence Inventory (ITC-SOPI short version; Lessiter et al., 2001). This is a 12-item self-report questionnaire that measures experiences of media. It will be used to measure sense of presence, engagement, ecological validity, and negative effects in the VR environment. For each item, respondents are asked to rate how much they agree or disagree on a 5-point Likert scale (1 = strongly disagree; 5 = strongly agree). The ratings of the three items corresponding to negative effects will be averaged when scored. Higher scores indicated increased negative effects. This questionnaire will be administered immediately after the first and last VR exposure sessions.

First VR exposure session (up to 19 weeks); Final VR exposure session (up to 20 weeks). Timepoints of the first and final VR exposure session will be decided by the therapist case-by-case, see Assigned Intervention for details

Measured using the ITC-Sense of Presence Inventory (ITC-SOPI short version; Lessiter et al., 2001). This is a 12-item self-report questionnaire that measures experiences of media. It will be used to measure sense of presence, engagement, ecological validity, and negative effects in the VR environment. For each item, respondents are asked to rate how much they agree or disagree on a 5-point Likert scale (1 = strongly disagree; 5 = strongly agree). The ratings of the three items corresponding to sense of presence will be averaged when scored. Higher scores indicated increased sense of presence. This questionnaire will be administered immediately after the first and last VR exposure sessions.

First VR exposure session (up to 19 weeks); Final VR exposure session (up to 20 weeks). Timepoints of the first and final VR exposure session will be decided by the therapist case-by-case, see Assigned Intervention for details

Measured using the ITC-Sense of Presence Inventory (ITC-SOPI short version; Lessiter et al., 2001). This is a 12-item self-report questionnaire that measures experiences of media. It will be used to measure sense of presence, engagement, ecological validity, and negative effects in the VR environment. For each item, respondents are asked to rate how much they agree or disagree on a 5-point Likert scale (1 = strongly disagree; 5 = strongly agree). The ratings of the three items corresponding to engagement will be averaged when scored. Higher scores indicated increased engagement. This questionnaire will be administered immediately after the first and last VR exposure sessions.

First VR exposure session (up to 19 weeks); Final VR exposure session (up to 20 weeks). Timepoints of the first and final VR exposure session will be decided by the therapist case-by-case, see Assigned Intervention for details

Measured using the ITC-Sense of Presence Inventory (ITC-SOPI short version; Lessiter et al., 2001). This is a 12-item self-report questionnaire that measures experiences of media. It will be used to measure sense of presence, engagement, ecological validity, and negative effects in the VR environment. For each item, respondents are asked to rate how much they agree or disagree on a 5-point Likert scale (1 = strongly disagree; 5 = strongly agree). The ratings of the three items corresponding to ecological validity will be averaged when scored. Higher scores indicated increased ecological validity. This questionnaire will be administered immediately after the first and last VR exposure sessions.

First VR exposure session (up to 19 weeks); Final VR exposure session (up to 20 weeks). Timepoints of the first and final VR exposure session will be decided by the therapist case-by-case, see Assigned Intervention for details

Using a tick-box approach, the therapists indicates which VR environments/scenarios from the selection provided were used in each session.

Inclusion Criteria: Participants will be included at screening if: Aged between 13 and 18 years Clinically diagnosed with ASD by a neurodevelopmental clinician Experiencing anxiety in social situations that is impairing daily functioning as determined by clinicians Deemed suitable by the clinical team for the intervention based on factors such as sufficient verbal ability, desire to participate, willingness to undertake exposure-based activities, and no current/active suicidal plans. Able to speak and understand English fluently Patient and parents/caregivers agree to CBT Exclusion Criteria: Participants will be excluded at screening if: Diagnosed with photosensitive epilepsy Receiving concurrent psychological therapy Immediate plans are in-place to change their psychopharmacological medication regimen