Enhancing the efficacy and user experience of virtual reality exposure therapy for social anxiety disorder: A pilot study

Abstract

Background

Social anxiety disorder (SAD) significantly impairs social functioning. Virtual reality exposure therapy (VRET) offers a promising treatment by providing a controlled, customizable environment. This study aimed to develop and evaluate the efficacy and user experience of a VRET program.

Methods

The study was conducted in two phases: Phase I with the general population and Phase II with individuals diagnosed with SAD at a university hospital. Social anxiety, depression, anxiety, and stress were measured using the Social Interaction Anxiety Scale and the Depression Anxiety and Stress Scale at three time points: before, immediately after, and two weeks post-VRET. The Virtual Reality Neuroscience Questionnaire assessed user experience, game mechanics, in-game assistance, and any VR-induced symptoms. Our VRET program utilized graded exposure techniques within culturally relevant social scenarios.

Results

Both groups exhibited significant reductions in social anxiety levels following VRET sessions (ps < 0.01) compared to pre-VRET levels. However, individuals with SAD reported increased social anxiety at the two-week follow-up, while the general population maintained their improvement. The VR software received satisfactory ratings for usability, safety, and acceptability.

Conclusion

This program demonstrates potential for reducing social anxiety and provides a satisfactory VR experience, supporting its feasibility for individuals with SAD in a developing country. Given the pilot nature and limited sample size, these findings should be interpreted cautiously. Future research with larger samples and repeated sessions is needed to enhance efficacy and ensure long-term benefits. Comprehensive treatment protocols, including tutorials, relaxation techniques, and stress monitoring, are recommended for optimal outcomes.

Keywords

Exposure therapy feasibility phobia social treatment outcome virtual reality

1 Introduction

Social anxiety disorder (SAD) is a prevalent mental health condition, affecting approximately 12.4–14.3% of individuals aged 18–59.¹ It is characterized by a fear of negative evaluation and avoidance of social situations,² leading to interpersonal difficulties and impaired relationships.^3,4 These challenges often adversely affect occupational and educational functioning and are sometimes accompanied by comorbid conditions such as depression.^4–6 Despite experiencing a diminished quality of life, only a third of individuals with SAD seek mental health services.⁷

Exposure therapy (ET) is a recognized treatment for SAD, effectively reducing distress in social situations. ET involves techniques where individuals confront fear-inducing situations to elicit and subsequently extinguish fear responses. Common methods include flooding or desensitization, which entail either full or gradual exposure to feared stimuli or scenarios.⁸ In the context of social anxiety, exposure may involve activities like shopping or public speaking, transitioning from imagined to real-life situations (in vivo).⁸ However, many individuals with SAD face obstacles in engaging in exposure exercises due to fear, inconvenience, or lack of relevant opportunities, potentially leading to symptom relapse.⁹

Virtual reality exposure therapy (VRET) has emerged as a promising intervention for anxiety disorders such as agoraphobia, arachnophobia, aviophobia, and social anxiety, offering advantages such as convenience and safety within a controlled environment.^10–15 VRET allows for customizable settings, pacing, and content adjustments during exposure sessions, making it particularly suitable for anxiety treatment.¹⁶ VRET is also recognized as an effective and appropriate alternative for preventive treatment due to its advantages in standardization, accessibility, safety, and patient acceptance.¹⁷ Recent meta-analyses underscore the efficacy of VRET for treating social anxiety.^{10–13,18,19} While VRET shows promise, it still has significant areas for growth. We need to improve its effectiveness and how well patients engage with it. Enhancing the VR experiences themselves is also crucial. Furthermore, we must address socio-cultural differences to overcome challenges in VR development, its phased introduction, and necessary educational programs.^20–22

To improve VRET's effectiveness, it is crucial to design realistic social interaction scenarios that accurately reflect real-life situations pertinent to the therapy. These scenarios must be capable of eliciting the social distress experienced by patients and be embedded within an appropriate cultural context.^10,23 Additionally, it is important to establish the optimal duration for VR exposure and to incorporate a diverse range of scenarios, such as public speaking, attending gatherings, dining, or placing orders in retail environments.

^12,20Beyond boosting effectiveness, enhancing the VR experience itself—focusing on usability and safety—is also suggested to make VRET more impactful.²¹

This study aimed to develop a VRET program that prioritizes both therapeutic efficacy (measured by social anxiety levels) and the overall VR experience (assessed by safety, usability, and acceptability). We hypothesize that after going through our VRET program, individuals with Social Anxiety Disorder (SAD) will experience a decrease in social anxiety and find the VR experience satisfactory.

Given the limited software development capabilities among low- to middle-income countries, such as Southeast Asian countries, this limitation affects testing in developing countries, which may potentially influence the results and generalizability.^13,24 It is recommended that incorporating realistic verbal interactions and culturally relevant stimuli is essential to enhance treatment effectiveness.^10,16,23 Tailoring VRET to specific cultural contexts, such as creating scenarios relevant to local social norms and values, is advised.^25,26 Thus, this study was designed to develop a VRET program in a developing country context—Thailand.

2 Method

2.1 Study design and setting

This pilot study was conducted at a university hospital between October and December 2023. The study tested the VRET program in two phases. Phase I involved recruiting participants from the general population to ensure VRET safety. Phase II was conducted among individuals diagnosed with SAD. The ethical oversight for this study, including the research type, potential risks, the intervention protocol, and adverse event management, was provided through approval from the Institutional Review Board of Prince of Songkla University, Thailand (reference number: REC.65-201-3-1).

2.2 Participants

Participants aged 20 to 40 were included in both phases. Individuals over 40 were excluded due to potential limited experience with digital platforms and an increased risk of falls during VRET sessions. Diagnosis of SAD was confirmed by a psychiatrist at the university hospital using the 10th revision of the International Statistical Classification of Diseases and Related Health Problems (ICD-10). Exclusion criteria included individuals with other psychiatric comorbidities such as major depressive disorder, generalized anxiety disorder, panic disorder, or schizophrenia, as well as those with physical conditions preventing them from engaging in VRET, such as paralysis, muscle weakness, Meniere's disease, dizziness, epilepsy, pregnancy, or light sensitivity.

2.3 Measurement

Demographic data, including gender, age, monthly income, and highest educational degree, were collected for both groups. For individuals diagnosed with SAD, data on illness duration, current prescribed medications, and previous psychological treatment were recorded. The following instruments were used:

Social Interaction Anxiety Scale (SIAS) comprising 20 items that measure fear of social interaction using a 5-point Likert scale ranging from 0 (not at all) to 4 (very much). Scores range from 0 to 80, with higher scores indicating greater social anxiety.²⁷ This scale has demonstrated excellent psychometric properties, with a test-retest correlation of 0.92 and a Cronbach's alpha of 0.94.²⁷ The Thai version also showed a Cronbach's alpha of 0.94.²⁸ A score exceeding 36 indicates significant severity of social anxiety.²⁹

Depression Anxiety Stress Scale (DASS-21) was designed to evaluate depression (7 items), anxiety (7 items), and stress (7 items), this scale employs a 4-point Likert scale ranging from 0 (not at all) to 3 (very much).³⁰ Higher scores indicate greater severity of symptoms. The scale has shown excellent reliability, with Cronbach's alpha values ranging from 0.84 to 0.91,³⁰ and similar reliability in the Thai version (0.85–0.91).³¹

Virtual Reality Neuroscience Questionnaire (VRNQ) comprised 20 items rated on a 7-point Likert scale from 1 (very low) to 7 (very high).³² This questionnaire assesses VR experiences across four subscales: user experience, game mechanics, in-game assistance, and VR-induced symptoms and effects (VRISE). This questionnaire demonstrated good to excellent internal reliability, with Cronbach's alpha values of 0.89 for user experience, 0.89 for game mechanics, 0.90 for in-game assistance, and 0.89 for VRISE.³² Higher scores indicate more positive experiences. The Thai version was developed through forward-backward translation to ensure linguistic and conceptual equivalence. (see more details of VRNQ in supplementary material)

This questionnaire assesses VR experiences across four subscales: user experience (e.g., intensity of the immersion, the level of enjoyment or the quality of the graphics, sound, and VR technology); game mechanics (e.g., the ease to navigate, physically move, and interact with the virtual environment); in-game assistance (e.g., the quality of the tutorial(s), in-game instructions (such as description of the aim of the task), and prompts (such as arrows showing the direction); and VR-induced symptoms and effects (VRISE) which is the intensity of primary adverse symptoms and effects pertinent to VR (i.e., nausea, disorientation, dizziness, fatigue, and instability). Higher scores indicate more positive experiences. The minimum thresholds imply that if the median score for each sub-category is 25 and the total score is 100 (equivalent to a median of at least 5 for each item), then the results of the VRNQ indicate that the assessed VR software is of sufficient quality to avoid significant VRISE. Moreover, the more conservative thresholds suggest that if the median score for each sub-category is 30 and the total score is 120 (equivalent to a median of at least 6 for each item), then utilizing these stricter thresholds more strongly validates the appropriateness of the VR software.

Since it is recommended that standardized measurement should be used to assess VR experiences in VRET research,²¹ VRNQ was chosen in this study. This questionnaire demonstrated good to excellent internal reliability, with Cronbach's alpha values of 0.89 for user experience, 0.89 for game mechanics, 0.90 for in-game assistance, and 0.89 for VRISE.³² The Thai version of VRNQ was developed through forward-backward translation following the process of cross-cultural adaptation,³³ ensuring linguistic and conceptual equivalence between the English and Thai versions. Forward translation from English to Thai was conducted by WA and another academic professional not specialized in psychiatry, followed by backward translation by another professional who had not seen the tool. Discrepancies between the original and translated versions were resolved, and the Thai version was used in this study. Optional qualitative feedback was collected for each subscale.

2.4 Improvements to VRET scenarios and VR experiences

To enhance VRET effectiveness, social interaction scenarios, exposure duration, and scenario variety were crucial. The VRET program was designed for interactive experiences, requiring participants to navigate virtual scenarios for ET.^12,20 Scenarios were created by a team of experienced psychiatrists to reflect common situations encountered by people with SAD and suitable for VR development. Three scenarios were developed: engaging in small talk in canteens and stores, and delivering speeches in public at events like wedding parties (Figure 1). Each scenario featured at least three tasks such as ordering food, asking for information, waiting in line, facing others in public, and giving a speech. Participants were actively engaged in these tasks to gradually expose them to social stimuli, with exposure timing gently increased using desensitization techniques.⁸

Figure 1.

Designed scenarios with tasks for the virtual reality exposure therapy including canteen, supermarket and wedding party.

Moreover, VR experiences were also considered to improve. Safety was addressed by limiting VRET sessions to 20 min as advise³⁴ with 5-min breaks between scenarios. Usability was improved by providing a tutorial on using VR before starting the VRET session and incorporating physical and verbal interaction into every scenario. Verbal interaction was designed as one-way communication to practice speaking performance. Acceptability was ensured by designing scenarios with appropriate fear stimuli to potentially trigger social anxiety without causing panic attacks or emotional distress. User experience was assessed using standardized measures, with VRNQ for reporting VR experiences, and DASS-21 for monitoring depression, anxiety, and stress.

2.5 VRET protocol

Before commencing the VRET sessions, participants were provided with basic information about the project, instructions on using the VR equipment, and guidance on adjusting the head-mounted display (HMD). Participants underwent a five-minute tutorial to familiarize themselves with the VR scenes, including scenario boundaries, symbols, and the HMD.

The designed scenarios included shopping in a supermarket, interacting with others in a canteen, and delivering a speech at a wedding party. Each scenario lasted approximately 10–15 min, with a five-minute break between scenarios. Throughout the test, researchers were available to provide assistance and support. Additionally, our protocol for VRET incorporated breathing relaxation techniques before starting the VRET to improve retention at the beginning of the treatment. This coping skill was aimed at managing negative emotions and panic attacks in VR.

2.6 Data collection

The study project was publicly announced in the hospital to recruit the general population for Phase I, while convenience sampling was conducted from the hospital's information technology system for Phase II. Six participants were recruited for each phase, totaling 12 participants. Interested individuals completed informed consent procedures and questionnaires assessing social anxiety, depression, anxiety, and stress before and after the VRET sessions, as well as two weeks post-test. The VRNQ was completed immediately after the test.

2.7 Data analyses

Descriptive statistics, including proportions, means, or medians, were employed to analyze the data collected in this study. Dependent t-tests were utilized to compare scores on the SIAS and the DASS-21 before and after the VRET, as well as two weeks post-treatment. Data analysis was conducted using R program version 3.4.1 (R Foundation for Statistical Computing). A significance level of 0.05 was employed, with 95% confidence intervals utilized to determine significant outcomes.

2.8 Software and hardware

For software, the Unity 3D Game Engine version 2021.3.15f1 LTS was used to develop the scenarios. The hardware included a personal computer with an AMD Ryzen 5 5600G CPU, an Nvidia RTX 3060 (8GB) GPU, and 16 GB of RAM. The HMD utilized was the Meta Quest Pro VR headset by Meta Platforms, Inc.

3 Results

Twelve participants took part in the study, with six in each phase. All participants were Thai and Buddhism. In Phase I, 50% were male, and in Phase II, 66.7% were male. The mean age of participants was 31 years old, with a standard deviation of 5.8. The majority of participants had a monthly income of 10,001–25,000 Baht (41.7%) and held Bachelor's degrees (58.3%).

In Phase II, individuals with SAD had been experiencing the illness for an average of 44 months, and all of them were prescribed medications. Five of them (83.3%) received psychoeducation, followed by Cognitive Behavioral Therapy/Behavioral Therapy (4, 66.7%), and Relaxation technique/Breathing technique (1, 16.7%) (Table 1).

Table 1.
Demographic characteristics (n = 12).

Number (%)

Demographic characteristics and medical history Total (%)(n = 12) Phase I:General population(n = 6) Phase II:Individuals with SAD(n = 6)

Gender

Male 7 (58.3) 3 (50.0) 4 (66.7)

Female 5 (41.7) 3 (50.0) 2 (33.3)

Age (years)

Mean ± S.D. (Min – Max) 31 ± 5.8 (24–39) 30 ± 5.7 (25–37) 32 ± 6.3 (24–39)

20–25 3 (25.0) 2 (33.3) 1 (16.7)

26–35 5 (41.7) 2 (33.3) 3 (50.0)

> 35 4 (33.3) 2 (33.3) 2 (33.3)

Income (Baht/month)

No 1 (8.3) 0 (0.0) 1 (16.7)

< 10,000 2 (16.7) 1 (16.7) 1 (16.7)

10,001–25,000 5 (41.7) 3 (50.0) 2 (33.3)

25,001–50,000 1 (8.3) 1 (16.7) 0 (0.0)

75,001–100,000 3 (25.0) 1 (16.7) 2 (33.3)

Highest level of education

High school 1 (8.3) 0 (0.0) 1 (16.7)

Bachelor's degree 7 (58.3) 4 (66.7) 3 (50.0)

> Bachelor's degree 4 (33.3) 2 (33.3) 2 (33.3)

Duration of illness (months)

Mean ± S.D. (Min, Max) 44 ± 22.3 (12–72)

	Number (%)
Gender
Male	7 (58.3)	3 (50.0)	4 (66.7)
Female	5 (41.7)	3 (50.0)	2 (33.3)
Age (years)
Mean ± S.D. (Min – Max)	31 ± 5.8 (24–39)	30 ± 5.7 (25–37)	32 ± 6.3 (24–39)
20–25	3 (25.0)	2 (33.3)	1 (16.7)
26–35	5 (41.7)	2 (33.3)	3 (50.0)
> 35	4 (33.3)	2 (33.3)	2 (33.3)
Income (Baht/month)
No	1 (8.3)	0 (0.0)	1 (16.7)
< 10,000	2 (16.7)	1 (16.7)	1 (16.7)
10,001–25,000	5 (41.7)	3 (50.0)	2 (33.3)
25,001–50,000	1 (8.3)	1 (16.7)	0 (0.0)
75,001–100,000	3 (25.0)	1 (16.7)	2 (33.3)
Highest level of education
High school	1 (8.3)	0 (0.0)	1 (16.7)
Bachelor's degree	7 (58.3)	4 (66.7)	3 (50.0)
> Bachelor's degree	4 (33.3)	2 (33.3)	2 (33.3)
Duration of illness (months)
Mean ± S.D. (Min, Max)			44 ± 22.3 (12–72)

SAD: social anxiety disorder.

In both Phase I and Phase II, data regarding SIAS scores are presented in Table 2. In Phase I, there was a notable decrease in SIAS scores among the general population following participation in VRET sessions (Median 14.5 (IQR 10.5–185)) and at the two-week post-test (14.0 (14–14)), in comparison to their scores prior to the VRET session (16.0 (12.3–19.85)). Conversely, individuals diagnosed with SAD in Phase II exhibited a significant decrease in SIAS scores following the VRET session (29.0 (23.0–59.0)) compared to pre-session scores (39.5 (30.0–59.5)), and but there was no significant decrease at the two-week post-test (40.0 (39.0–66.0)) (Figure 2).

Figure 2.

Median of pre - posttest and 2-week posttest for social anxiety, depression, anxiety and stress of phase I (n = 6) and II (n = 6). SAD: social anxiety disorder.

Table 2.

Pre - posttest and 2-week posttest for social anxiety, depression, anxiety and stress of phase I (n = 6) and II (n = 6).

	Phase I (General population); Median (IQR)
Factors	Pretest	Posttest	2-week Posttest	Pretest vs Posttest	Pretest vs 2-week Posttest
Social anxiety	16.0 (12.3–19.8)	14.5 (10.5–18.5)	14.0 (14.0–14.0)	0.004^a	0.003^a
Depression	1.5 (0–4.5)	0 (0–1.5)	0 (0–1.5)	0.823^a	0.968^a
Anxiety	2.0 (2.0–2.75)	1.5 (0.25–2.0)	1.0 (0.25–1.75)	0.774^a	0.502^a
Stress	2.5 (2.0–3.75)	1.5 (0.25–3.5)	3.0 (0.75–3.0)	0.149^a	0.437^a
	Phase II (Individuals with SAD); Median (IQR)
Social anxiety	39.5 (30.0–59.5)	29.0 (23.0–59.0)	40.0 (39.0–66.0)	0.003^a	0.063^a
Depression	5.0 (2.75–6.5)	3.5 (1.5–7.75)	1.0 (0–7.0)	0.020^a	0.066^a
Anxiety	8.0 (6.25–12.75)	7.0 (7.0–8.5)	4.0 (3.0–8.0)	0.004^a	0.018^a
Stress	9.5 (8.25–12.25)	7.0 (6.0–9.5)	7.0 (6.0–7.0)	0.002^a	0.006^a

IQR: Interquartile range; SAD: social anxiety disorder.

Wilcoxon matched paired sign rank test.

Note: One participant did not report back at two-week posttest of Phase II.

Based on the SIAS cut-off point (greater than 36), in Phase II individuals with SAD reported social anxiety level which met the threshold for social anxiety at Pretest, Posttest, and 2-week Posttest, with percentages of 66.7%, 33.3%, and 20.0%, respectively. However, no statistically significant differences were observed.

Concerning the mental health domains using DASS-21, the overall scores among the general population were lower compared to individuals diagnosed with SAD. In Phase I, there were no statistically significant differences in scores observed between pre- and post-VRET session assessments, as well as at the two-week post-test. However, in Phase II, anxiety and stress scores following the VRET session and at the two-week post-test were significantly lower than before the VRET session, only depression scores did not show a statistically significant difference.

VR experiences using VRNQ were also assessed in both groups (Table 3 and Supplementary Figure 1), with the general population reporting higher scores than individuals with SAD. In both Phase I and II, the lowest score was observed for Game Mechanics, with median scores of 28.0 (IQR 26.3–29.8) and 24.0 (22.3–25.0), respectively, while the highest score was observed for VRISE, with median scores of 32.5 (32.0–34.5) and 28.5 (27.3–32.8), respectively. Other details of VR experiences are described in Supplementary Table 1.

Table 3.

Median with Interquartile of Virtual Reality Neuroscience Questionnaire (VRNQ).

	Median (IQR)
VRNQ	Phase I	Phase II	P-value
User experience	32.5 (26.8–33.0)	24.5 (22.3–27.5)	0.065
Game mechanics	28.0 (26.3–29.8)	24.0 (22.3–25.0)	0.077
In-game assistance	30.5 (28.3–32.8)	27.5 (27.0–28.8)	0.197
VRISE	32.5 (32.0–34.5)	28.5 (27.3–32.8)	0.376
Total	123.5 (108.5–128.8)	105.5 (98.3–112.0)	0.065

IQR: Interquartile range; VRIS: VR-induced symptoms and effects; VRNQ: Virtual reality neuroscience questionnaire.

Note: P-value using Mann-Whitney U-test.

4 Discussion

The primary aim of this study was to test software development while assessing the feasibility, both improving the efficacy and user experience, of VRET in a Southeast Asian context – Thailand. Given this objective, the sample size in this pilot study was limited. Effectiveness was assessed by measuring social anxiety levels before, immediately after, and two weeks following VRET participation. VR experience was evaluated using the standardized VRNQ, which examines user experience, game mechanics, in-game assistance, and VRISE. The findings suggest that VRET is promising in reducing social anxiety in both the general population and individuals with diagnosed social anxiety. Additionally, the developed VR software demonstrated satisfactory user experiences, supported by safety interventions. These findings indicate the feasibility of VRET for individuals with SAD in a developing country. Nonetheless, it is essential to approach these findings with caution due to the small sample size of this pilot study.

4.1 Efficacy of VRET

Social anxiety levels were measured using SIAS before and after the VRET session to assess efficacy. Participants with SAD reported a decrease in SIAS scores after the VRET session, and individuals from the general population also reported lower SIAS scores, consistent with previous research.¹² Moreover, similar reductions in social anxiety have been observed in prior VR-based pilot studies with 6–20 participants in Singapore,³⁵ the UK^36,37 and USA.³⁸ Our findings suggest that VRET initially induces heightened social fears, allowing individuals with SAD and the general population to learn coping strategies through graded exposure tasks. This process enhances social skills, enabling participants to manage social situations within the VRET environment.^39,40

Despite limitations in generalizability, our findings are consistent with prior research, highlighting VRET's potential effectiveness in diverse geographical and cultural settings, including those within Asia (Singapore, South Korea) and Western countries (UK, USA).^35–38,41

At the two-week post-test, the general population reported a sustained decrease in social anxiety levels, while individuals with SAD reported an increase. These results imply that our developed VRET program induces social fear while facilitating the acquisition of effective behavioural skills during exposure. The increase in social anxiety levels among individuals with SAD at the follow-up may be due to the differences in social context from the VRET. These individuals may need more time to gain a sense of control or self-efficacy and become more confident in applying coping strategies to other socially feared situations.^42,43 This supports the notion that a single session of behavioural exposure may not sufficiently reduce anxiety or facilitate the acquisition of new coping mechanisms, suggesting that repeated VRET sessions yield better anxiety outcomes.⁴⁴ Additionally, combining cognitive behavioural therapy (CBT) with follow-up sessions and pharmacologically augmented psychotherapy may further alleviate social anxiety.^38,45 Given that individuals with SAD exhibited higher stress levels than the general population, this elevated stress may have adversely affected their coping strategies, leading to stress-related withdrawal behaviors⁴⁶ and increased social anxiety outcomes.

4.2 VR experience

Individuals from the general population reported all sub-scores and the total score of the VRNQ higher than 25 and 100, respectively, indicating that the VR software meets adequate quality standards in all aspects.³² Conversely, among individuals diagnosed with SAD, the sub-scores for in-game assistance and VRISE, as well as the total score, met the criteria for adequate quality of the VR software, ensuring that significant VRISE was not experienced. User experience and game mechanics showed tendencies to meet the quality standards for VR, though they did not fully satisfy the more rigorous thresholds. Specifically, game mechanics consistently ranked as the lowest-rated subscale across both groups, especially among individuals with SAD, suggesting a need for enhancements in virtual navigation, physical movement, and interaction capabilities. Furthermore, optional qualitative feedback for each subscale highlighted areas for software improvement, particularly in graphic design (e.g., “virtual environment is larger than room lab/real world this make confused sometime”) and user-friendliness (e.g., “white lighting sometimes cause eyes sensitivity during VR”), which corresponded with the lower scores observed for Game Mechanics and User Experience within the SAD group.

The satisfactory ratings received for the VR software's usability, safety, and acceptability can be attributed to several factors. Our meticulously designed tutorials and integrated interactive elements fostered participant engagement and mitigated technological hurdles, thereby ensuring strong usability—a critical component for effective exposure therapy. Furthermore, we implemented deliberate safety measures, such as restricting session durations and improving VR resolution, to minimize any VR-induced symptoms, guaranteeing a comfortable and continuous therapeutic experience for users. The high level of acceptability was achieved by crafting scenarios that, while designed to evoke social anxiety, did so without inducing panic, thus ensuring participants’ willingness to engage fully with the exposure tasks and maximize their therapeutic gains. Collectively, these elements established a supportive environment for graded exposure, empowering participants to effectively acquire coping strategies and contributing to a reduction in their symptoms.

Further studies with larger populations are necessary to confirm the software quality. Improvements in software development, particularly in graphic design and user-friendliness, are warranted. Consideration of individual social concerns and the integration of nonverbal social cues (NVSC) are crucial for refining VRET.^26,47 Thoughtful design should ensure realistic, precise, and safe social anxiety exposure experiences. Additionally, when examining mental health domains using the DASS-21, there was a notable decrease in depression, anxiety, and stress scores at the two-week post-test among individuals diagnosed with SAD. No extreme stress exacerbation was observed, suggesting that our developed VR program supports its acceptability for individuals with SAD.

4.3 Strengths and limitations

This VR software is innovative in simulating diverse social situations with interactive tasks. It incorporates recommendations for effective exposure therapy principles, such as appropriate exposure durations, physical interaction, and scenario varieties.^12,20 Safety, usability, and acceptability were prioritized,²¹ with limited exposure time and enhanced VR resolution to prevent motion sickness.³⁴ Graded exposure techniques, including increasing social interactions and response times, and incorporating breathing relaxation techniques, were used to enhance treatment retention. Tutorials and instructions were provided to address concerns regarding technology use and to streamline exposure sessions, acknowledging varying levels of digital literacy. In addition to focusing on software development, this study highlights the importance of comprehensive treatment protocols that include tutorials, relaxation techniques, and stress monitoring. The VRET was customized to suit specific cultural contexts, creating appropriate stimuli to induce fear among individuals from Southeast Asian countries. While other digital interventions, such as augmented reality and mobile cognitive training, offer flexibility and accessibility, VRET provides a fully immersive experience with precise environmental control, optimizing simulated social scenarios.

Despite these strengths, several limitations were noted. First, data collection for each scenario was limited, as participants experienced all three scenarios simultaneously before completing questionnaires. Future studies should collect data per scenario individually to refine the VRET program. Second, the small sample size affects statistical power and generalizability. However, previous VR studies in Singapore, the UK and the USA with similarly small samples shown significant anxiety reductions.^35–38 While results should be interpreted with caution, the findings indicate that the VRET effectively induced social fears while maintaining safety. Further research with larger samples and individualized scenario testing is needed to validate these outcomes and enhance VRET effectiveness. Third, the absence of follow-up sessions or reinforcement after the initial VRET session indicates that a single session may not be sufficient for lasting effects—a common limitation in pilot studies. A more comprehensive follow-up schedule, such as multiple follow-up or booster sessions, is advised to improve long-term efficacy. Lastly, to capture more granular data related to VRET-specific outcomes, future studies should incorporate additional measures, such as virtual presence and emotional engagement.

4.4 Further implications

The reluctance of individuals with SAD to seek professional help poses a risk of undertreatment, impacting daily functioning. Even those who receive treatment often struggle to practice social skills in real-world situations. Expanding treatment options to include VRET, which offers a controlled, customizable, and accessible alternative, could help bridge this gap.^11,48 Future advancements, such as AI-driven interactive verbal feedback and more dynamic, adaptive nonverbal social cues (e.g., eye gaze, facial expressivity) from virtual characters, or internet-integrated communication, may further enhance the immersive experience, improving engagement and treatment effectiveness.^45,49,50 This could serve as an alternative means of increasing treatment accessibility for individuals with SAD. Moreover, future work should explicitly consider the cost-effectiveness and scalability of VRET implementation in resource-limited settings, and explore simpler, more accessible VR platforms to improve reach and adoption. This approach directly addresses the challenges faced by low- to middle-income countries in accessing mental health services.

While further research is needed to establish VRET as a stand-alone treatment, it should be integrated into established therapeutic frameworks, including cognitive behavioural therapy (CBT) and relaxation techniques.¹⁶ Recent reviewed models emphasize cognitive-behavioural approaches in SAD treatment, highlighting factors such as attention and interpretation biases, implicit associations, and emotion regulation strategies.⁵¹ Addressing maintenance mechanisms of social anxiety—such as self-focused attention, safety behaviours, and post-event processing—could enhance treatment outcomes.^51,52 Incorporating more comprehensive treatment protocols, including repeated VRET sessions, is suggested for better social anxiety outcomes.⁴⁴ Future research should explore personalized VRET interventions based on cognitive-behavioural principles, rather than solely relying on virtual exposure.^14,45

Further VRET development should incorporate internal self-appraisal processes, guiding individuals to focus on their self-perception rather than external judgment.^51,53,54 Additionally, avoidance behaviours, such as gaze and head movements including avoidance measurement, should be examined in future studies,^55,56 as understanding these behaviours can inform targeted interventions to enhance exposure tolerability.⁵² Moreover, a comprehensive concern of social contexts for VR development, including cultural, racial, and ethnic backgrounds, remain essential to accurately address social anxiety in VR.²⁶ Nonverbal social cues, including eye gaze, facial expressivity, and vocal characteristics, should also be considered, as they influence social hierarchy and may offer new therapeutic opportunities through embodied cognition.⁴⁷

VRET has potential applications beyond SAD, offering broader benefits in mental health. Adolescents with limited social skills^57,58 and individuals with autism spectrum disorder (ASD) who struggle with social interactions⁵⁹ may find VRET useful in managing social fears. Given the shared cognitive mechanisms between social anxiety and paranoia,^60,61 VRET-based behavioural strategies could also help mitigate fear responses in individuals with paranoia or psychosis.^62,63 Expanding VRET to address various socially feared situations warrants further research to enhance its therapeutic potential.

5 Conclusion

The developed VRET shows promise in reducing social anxiety among the Thai population and provides satisfactory VR experiences regarding safety, usability, and acceptability. These findings support the feasibility of VRET for individuals with SAD in developing countries. However, due to the study's pilot nature and small sample size, the results should be interpreted with caution and further testing with larger samples studies is needed. Additionally, for therapeutic efficacy, VRET protocols should focus on both software improvements and comprehensive treatment designs, including tutorials, breathing relaxation techniques before sessions, and stress monitoring.

Supplemental Material

sj-docx-1-thc-10.1177_09287329251360523 - Supplemental material for Enhancing the efficacy and user experience of virtual reality exposure therapy for social anxiety disorder: A pilot study

Supplemental material, sj-docx-1-thc-10.1177_09287329251360523 for Enhancing the efficacy and user experience of virtual reality exposure therapy for social anxiety disorder: A pilot study by Warut Aunjitsakul, Kanthee Anantapong, Pakawat Wiwattanaworaset, Aimorn Jiraphan, Teerapat Teetharatkul, Katti Sathaporn, Kreuwan Jongbowonwiwat and Sitthichok Chaichulee in Technology and Health Care

Footnotes

Acknowledgements

We would like to express our gratitude to all the volunteers who participated in this research. Their valuable feedback significantly contributed to the final version of the virtual reality program.

ORCID iDs

Warut Aunjitsakul

Kanthee Anantapong

Pakawat Wiwattanaworaset

Aimorn Jiraphan

Teerapat Teetharatkul

Author contributions

WA: Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Resources; Writing - original draft; and Writing - review & editing, KA: Conceptualization; Investigation; Methodology and Writing - review & editing, AJ and KS: Conceptualization; Investigation and Methodology, PW and TT: Conceptualization; and Methodology, KJ: Data curation; Investigation; Methodology; and Project administration, SC: Conceptualization; Funding acquisition; Methodology; Resources. All others have approved the final draft.

Funding

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by National Science, Research and Innovation Fund (NSRF) and Prince of Songkla University (Grant No: MED6601327S).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Supplemental material

Supplemental material for this paper is available online.

References

Kessler

Berglund

Demler

, et al. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the national comorbidity survey replication. Arch Gen Psychiatry 2005; 62: 593–602.

Clark

Wells

. A cognitive model of social phobia. In: Social phobia: diagnosis, assessment, and treatment. New York, NY, US: The Guilford Press, 1995, pp. 69–93.

Tonge

Lim

Piccirillo

, et al. Interpersonal problems in social anxiety disorder across different relational contexts. J Anxiety Disord 2020; 75: 102275.

Davila

Beck

. Is social anxiety associated with impairment in close relationships? A preliminary investigation. Behav Ther. 2002; 33: 427–446.

Wittchen

Fuetsch

Sonntag

, et al. Disability and quality of life in pure and comorbid social phobia. Findings from a controlled study. Eur Psychiatry. 2000; 15: 46–58.

Fehm

Pelissolo

Furmark

, et al. Size and burden of social phobia in Europe. Eur Neuropsychopharmacol 2005; 15: 453–462.

Ruscio

Brown

Chiu

, et al. Social fears and social phobia in the USA: results from the national comorbidity survey replication. Psychol Med 2008; 38: 15–28.

Rodebaugh

Holaway

Heimberg

. The treatment of social anxiety disorder. Clin Psychol Rev 2004; 24: 883–908.

Geraets

CNW

Veling

Witlox

, et al. Virtual reality-based cognitive behavioural therapy for patients with generalized social anxiety disorder: a pilot study. Behav Cogn Psychother 2019; 47: 745–750.

10.

Kampmann

Emmelkamp

Morina

. Meta-analysis of technology-assisted interventions for social anxiety disorder. J Anxiety Disord 2016; 42: 71–84.

11.

Chesham

Malouff

Schutte

. Meta-analysis of the efficacy of virtual reality exposure therapy for social anxiety. Behav Change. 2018; 35: 152–166.

12.

Horigome

Kurokawa

Sawada

, et al. Virtual reality exposure therapy for social anxiety disorder: a systematic review and meta-analysis. Psychol Med 2020; 50: 2487–2497.

13.

Rejbrand

Fure

Sonnby

. Stand-alone virtual reality exposure therapy as a treatment for social anxiety symptoms: a systematic review and meta-analysis. Upsala J Med Sci. 2023; 128.

14.

Bouchard

Dumoulin

Robillard

, et al. Virtual reality compared with in vivo exposure in the treatment of social anxiety disorder: a three-arm randomised controlled trial. Br J Psychiatry 2017; 210: 276–283.

15.

Parsons

Rizzo

. Affective outcomes of virtual reality exposure therapy for anxiety and specific phobias: a meta-analysis. J Behav Ther Exp Psychiatry 2008; 39: 250–261.

16.

Emmelkamp

PMG

Meyerbröker

Morina

. Virtual reality therapy in social anxiety disorder. Curr Psychiatry Rep 2020; 22: 32.

17.

Seuling

Czernin

Schiele

. Virtual reality exposure therapy in the treatment of public speaking anxiety and social anxiety disorder. Neurosci Appl 2024; 3: 104074.

18.

Tan

Xiu

CVY

Darryl

AWH

, et al. Virtual reality exposure therapy for social anxiety disorders: a meta-analysis and meta-regression of randomized controlled trials. Anxiety Stress Coping 2025; 38: 141–160.

19.

Kuleli

Tyson

Davies

, et al. Examining the comparative effectiveness of virtual reality and in-vivo exposure therapy on social anxiety and specific phobia: a systematic review & meta-analysis. J Behav Cognit Ther 2025; 35: 100524.

20.

Wong

Lai

CYY

Qin

. Systematic review and meta-analysis of randomised controlled trials for evaluating the effectiveness of virtual reality therapy for social anxiety disorder. J Affect Disord 2023; 333: 353–364.

21.

Shahid

Kelson

Saliba

. Effectiveness and user experience of virtual reality for social anxiety disorder: systematic review. JMIR Ment Health 2024; 11: e48916.

22.

Glegg

SMN

Levac

. Barriers, facilitators and interventions to support virtual reality implementation in rehabilitation: a scoping review. Pm r 2018; 10: 1237–51.e1.

23.

National Collaborating Centre for Mental H. National Institute for Health and Care Excellence: Guidelines. Social anxiety disorder: Recognition, assessment and treatment. Leicester (UK): British Psychological Society (UK), 2013, Copyright © The British Psychological Society & The Royal College of Psychiatrists, 2013.

24.

Jingili

Oyelere

Nyström

MBT

, et al. A systematic review on the efficacy of virtual reality and gamification interventions for managing anxiety and depression. Front Digit Health 2023; 5: 1239435.

25.

Naeem

Phiri

Rathod

, et al. Cultural adaptation of cognitive–behavioural therapy. BJPsych Adv 2019; 25: 387–395.

26.

Hofmann

Anu Asnaani

Hinton

. Cultural aspects in social anxiety and social anxiety disorder. Depress Anxiety. 2010; 27: 1117–1127.

27.

Mattick

Clarke

. Development and validation of measures of social phobia scrutiny fear and social interaction anxiety. Behav Res Ther 1998; 36: 455–470.

28.

Aunjitsakul

McLeod

Gumley

. Understanding psychological mechanisms linking social anxiety and paranoia: a cross-cultural general population survey in Thailand and the United Kingdom. Psychiatry Res Commun 2022; 2: 100079.

29.

Peters

. Discriminant validity of the social phobia and anxiety inventory (SPAI), the social phobia scale (SPS) and the social interaction anxiety scale (SIAS). Behav Res Ther 2000; 38: 943–950.

30.

Lovibond

. The structure of negative emotional states: comparison of the depression anxiety stress scales (DASS) with the beck depression and anxiety inventories. Behav Res Ther. 1995; 33: 335–343.

31.

Oei

Sawang

Goh

, et al. Using the depression anxiety stress scale 21 (DASS-21) across cultures. Int J Psychol 2013; 48: 1018–1029.

32.

Kourtesis

Collina

Doumas

LAA

, et al. Validation of the virtual reality neuroscience questionnaire: maximum duration of immersive virtual reality sessions without the presence of pertinent adverse symptomatology. Front Hum Neurosci. 2019; 13.

33.

Beaton

Bombardier

Guillemin

, et al. Guidelines for the process of cross-cultural adaptation of self-report measures. Spine (Phila Pa 1976) 2000; 25: 3186–3191.

34.

Taubert

Webber

Hamilton

, et al. Virtual reality videos used in undergraduate palliative and oncology medical teaching: results of a pilot study. BMJ Support Palliat Care 2019; 9: 281–285.

35.

Lin

Lee

T-S

Cheung

, et al. Exposure therapy with personalized real-time arousal detection and feedback to alleviate social anxiety symptoms in an analogue adult sample: pilot proof-of-concept randomized controlled trial. JMIR Ment Health 2019; 6: e13869.

36.

James

Lin

Steed

, et al. Social anxiety in virtual environments: results of a pilot study. Cyberpsychol Behav 2003; 6: 237–243.

37.

Hinrichsen

Clark

. Anticipatory processing in social anxiety: two pilot studies. J Behav Ther Exp Psychiatry. 2003; 34: 205–218.

38.

Masia

Klein

Storch

, et al. School-based behavioral treatment for social anxiety disorder in adolescents: results of a pilot study. J Am Acad Child Adolesc Psychiatry 2001; 40: 780–786.

39.

Hildebrand

Roesmann

Planert

, et al. Self-guided virtual reality therapy for social anxiety disorder: a study protocol for a randomized controlled trial. Trials 2022; 23: 395.

40.

Caponnetto

Triscari

Maglia

, et al. The simulation game—virtual reality therapy for the treatment of social anxiety disorder: a systematic review. Int J Environ Res Public Health. 2021; 18: 13209.

41.

Kim

M-K

Eom

Kwon

, et al. Neural effects of a short-term virtual reality self-training program to reduce social anxiety. Psychol Med. 2022; 52: 1296–1305.

42.

Bandura

. Social foundations of thought and action: a social cognitive theory. Englewood Cliffs, NJ, US: Prentice-Hall, Inc, 1986, xiii, 617-xiii, p.

43.

Mineka

Thomas

. Mechanisms of change in exposure therapy for anxiety disorders. In: Handbook of cognition and emotion. New York City, United States: John Wiley & Sons Ltd, 1999, pp. 747–764.

44.

Powers

Emmelkamp

. Virtual reality exposure therapy for anxiety disorders: a meta-analysis. J Anxiety Disord 2008; 22: 561–569.

45.

Pelissolo

Abou Kassm

Delhay

. Therapeutic strategies for social anxiety disorder: where are we now? Expert Rev Neurother. 2019; 19: 1179–1189.

46.

Milas

Martinović Klarić

Malnar

, et al. The impact of stress and coping strategies on life satisfaction in a national sample of adolescents: a structural equation modelling approach. Stress Health 2021; 37: 1026–1034.

47.

Gilboa-Schechtman

Shachar-Lavie

. More than a face: a unified theoretical perspective on nonverbal social cue processing in social anxiety. Front Hum Neurosci. 2013; 7.

48.

Ling

Nefs

Morina

, et al. A meta-analysis on the relationship between self-reported presence and anxiety in virtual reality exposure therapy for anxiety disorders. PLoS One 2014; 9: e96144.

49.

Liaw

Tan

Lim

, et al. Artificial intelligence in virtual reality simulation for interprofessional communication training: mixed method study. Nurse Educ Today. 2023; 122: 105718.

50.

Gaikwad

Mathur

Mehta

, et al. Aging in the digital era: analysing the mental health effects of internet use among geriatrics – a narrative review. Eastern Ukrainian Med J 2024; 12: 721–729.

51.

Morrison

Heimberg

. Social anxiety and social anxiety disorder. Annu Rev Clin Psychol. 2013; 9: 249–274.

52.

Piccirillo

Taylor Dryman

Heimberg

. Safety behaviors in adults with social anxiety: review and future directions. Behav Ther. 2016; 47: 675–687.

53.

Schultz

Heimberg

. Attentional focus in social anxiety disorder: potential for interactive processes. Clin Psychol Rev. 2008; 28: 1206–1221.

54.

Mansell

Clark

Ehlers

, et al. Social anxiety and attention away from emotional faces. Cognit Emotion 1999; 13: 673–690.

55.

Anderson

Rothbaum

Hodges

. Virtual reality exposure in the treatment of social anxiety. Cogn Behav Pract. 2003; 10: 240–247.

56.

Wieser

Pauli

Grosseibl

, et al. Virtual social interactions in social anxiety—the impact of sex, gaze, and interpersonal distance. Cyberpsychol Behav Soc Netw 2010; 13: 547–554.

57.

Hugh-Jones

Ulor

Nugent

, et al. The potential of virtual reality to support adolescent mental well-being in schools: a UK co-design and proof-of-concept study. Ment Health Prev. 2023; 30: 200265.

58.

Liu

Zeng

, et al. Virtual reality in adolescent mental health management under the new media communication environment. Humanit Soc Sci Commun 2025; 12: 201.

59.

Adams

Simonoff

Tierney

, et al. Developing a user-informed intervention study of a virtual reality therapy for social anxiety in autistic adolescents. Des Health 2022; 6: 114–133.

60.

Freeman

Garety

Bebbington

, et al. Psychological investigation of the structure of paranoia in a non-clinical population. Br J Psychiatry 2005; 186: 427–435.

61.

Kourtesis

Kouklari

Roussos

, et al. Virtual reality training of social skills in adults with autism spectrum disorder: an examination of acceptability, usability, user experience, social skills, and executive functions. Behav Sci (Basel) 2023; 13.

62.

Freeman

Pugh

Antley

, et al. Virtual reality study of paranoid thinking in the general population. Br J Psychiatry 2008; 192: 258–263.

63.

Aunjitsakul

McGuire

McLeod

, et al. Candidate factors maintaining social anxiety in the context of psychotic experiences: a systematic review. Schizophr Bull 2021; 47: 1218–1242.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.04 MB

	Number (%)
Demographic characteristics and medical history	Total (%)(n = 12)	Phase I:General population(n = 6)	Phase II:Individuals with SAD(n = 6)
Gender
Male	7 (58.3)	3 (50.0)	4 (66.7)
Female	5 (41.7)	3 (50.0)	2 (33.3)
Age (years)
Mean ± S.D. (Min – Max)	31 ± 5.8 (24–39)	30 ± 5.7 (25–37)	32 ± 6.3 (24–39)
20–25	3 (25.0)	2 (33.3)	1 (16.7)
26–35	5 (41.7)	2 (33.3)	3 (50.0)
> 35	4 (33.3)	2 (33.3)	2 (33.3)
Income (Baht/month)
No	1 (8.3)	0 (0.0)	1 (16.7)
< 10,000	2 (16.7)	1 (16.7)	1 (16.7)
10,001–25,000	5 (41.7)	3 (50.0)	2 (33.3)
25,001–50,000	1 (8.3)	1 (16.7)	0 (0.0)
75,001–100,000	3 (25.0)	1 (16.7)	2 (33.3)
Highest level of education
High school	1 (8.3)	0 (0.0)	1 (16.7)
Bachelor's degree	7 (58.3)	4 (66.7)	3 (50.0)
> Bachelor's degree	4 (33.3)	2 (33.3)	2 (33.3)
Duration of illness (months)
Mean ± S.D. (Min, Max)			44 ± 22.3 (12–72)