Виртуальная реальность в области аутизма и физической терапии: метааналитический обзор клинических результатов и терапевтической эффективности.

Introduction

At present, the technology that lets users interact with computer-generated environments in real-time is known as virtual reality. Unlike classical media, VR provides immersive 3D multisensory experiences that closely replicate real-world contexts, thus facilitating experiential learning, rehabilitation and therapy ( 1 ). VR, which was initially a form of entertainment, has started to get a lot of attention in healthcare and education, where it can boost motivation and enhance healing. Virtual reality (VR) is immersive. It provides users with the experience of complex scenarios and environments which may be difficult or impossible to access in real life. This technology has shown great potential in the medical field. Medical students are able to learn to perform medical procedures on a patient without putting an actual patient at risk ( 2 , 3 ). Moreover, virtual reality (VR) applications in mental health treatment have proven useful for handling phobias, anxiety attacks and post-traumatic disorder with controlled exposure therapy in a virtual setting. According to the published clinical literature, virtual reality technologies have been used in health care. People on the autism spectrum can practice social communication, coping with emotions, and other adaptive behaviours using VR technology in safe and controlled environments without any unpredictability of the real world. According to empirical studies, VR-based training improves joint attention, facial emotion recognition, and interaction skills while lowering anxiety and boosting motivation ( 4 , 5 ). It is especially noteworthy that VR enables the transfer of skills to real life, which is one of the biggest challenges in the treatment of autism ( 6 , 7 ).

Virtual Reality is a fully immersive medium that can enable personalization and adaptiveness of interventions, such that they can be customized to fit the individuals' needs and learning ( 8 ). This flexibility covers a wide range of treatments, including CBT, social skills training, and sensory integration training. The VR environments can be modified very easily to create stimuli, difficulty, scenarios, etc, matching the individual needs and progress of each user ( 6 ). Moreover, VR technologies offer real potential for conducting therapeutic sessions from home, which could enhance access for people in deprived areas or with mobility issues. The ability to overcome distance and make treatment accessible for all can revolutionize the delivery of mental health services and specialized therapies ( 9 ). The many sensory features of virtual reality experiences can create extremely fun and realistic experiences, which can catch experiences more and lead to faster learning and adaptation. This engaging quality may improve retention of the skills and strategies learned during therapy and may enhance transfer to the real world ( 2 , 10 ). In addition, VR interventions can create controlled and safe environments where people can practice difficult tasks or confront anxiety-provoking situations. Clients can gradually increase the intensity of the experiences as they learn to cope with them. This can be beneficial in the case of Phobias, PTSD (Post Traumatic Stress Disorder) and other Anxiety-controlled exposure. Virtual reality has become a powerful tool in physical therapy and rehabilitation. VR interventions have shown to be effective at improving motor learning, balancing, and functional recovery through gamification, interactive exercises and/or biofeedback ( 11 , 12 ). Scientific proof has verified that post-stroke rehabilitation, musculoskeletal therapy and paediatric motor therapy have made a lot of difference. Patients appear to enjoy this sort of therapy. Moreover, they seem to be more compliant when compared to normal therapy methods. Through the use of immersive VR settings, individuals are encouraged to perform real-life tasks such as walking, reaching, and moving around objects 14, 15. Plus, the growth of mobile VR devices has improved accessibility, making it possible for therapists to incorporate VR into regular sessions. As a result, techniques that utilize VR have been used by many disciplines in mental health ( 13 ). Therapists can now design immersive experiences to treat specific phobias, anxiety disorders, and post-traumatic stress disorders ( 14 ). Patients are thus likely experiencing more effective and engaging treatments that may lead to faster recovery and better long-term outcomes ( 15 ). Data collected through virtual reality sessions can help therapists gain more insight into patient behavior and personalize treatment plans. The current development of VR will, in the future, produce sophisticated healthcare applications, which may comprise a more advanced haptic feedback system and full-body tracking systems ( 16 , 17 ). Although there's a great deal of empirical evidence, research on virtual reality interventions in autism spectrum disorder (ASD) and its application in physical therapy are limited in terms of methodology and practicality. According to recent studies, a lot of contemporary studies share multiple weaknesses which include poor sample size, short intervention duration, and variations in methodology pattern which lessen the generalizability and reliability of findings ( 18 , 19 ). Most systematic reviews already available majorly focus on only one therapeutic condition each. Therefore, they do not provide any cross-domain synthesis of efficacy. This has left many important unanswered questions regarding the overall therapeutic efficacy as well as implementation of virtual reality technologies ( 20 ). Additionally, there remain sizeable barriers to implementation, such as expensive equipment, limited clinician training programmes, poor accessibility in institutions and low integration into current clinical practice guidelines, all of which could significantly obstruct the introduction of these virtual reality interventions ( 21 ).

To address these gaps, this study conducted a systematic review and meta-analysis of virtual reality (VR) interventions in autism spectrum disorder (ASD) and physical therapy (PT). By synthesizing findings from these two distinct yet complementary fields, this study identifies shared benefits, methodological challenges, and emerging opportunities for clinical practice. Specifically, this study examines the following:

To what extent are virtual reality (VR) interventions effective in enhancing therapeutic outcomes?

What practical and methodological challenges affect the integration of VR into clinical practice?

What types of VR interventions are currently applied in Autism Spectrum Disorder and physical therapy?

This study synthesizes evidence from both cognitive and motor rehabilitation to contribute to the expanding body of knowledge on virtual reality (VR) in healthcare. This study underscores the role of VR as a scalable and inclusive therapeutic adjunct, elucidates strategies for its effective integration, and proposes directions for future interdisciplinary research.

Theoretical framework

Our systematic literature review (SLR) identified constructivism, embodied cognition, and neuroplasticity as the most appropriate theoretical frameworks for comprehending the therapeutic application of virtual reality (VR) in Autism Spectrum Disorder (ASD) and physical therapy (PT). Constructivism underscores experiential, hands-on learning, wherein individuals actively engage with their environment to construct knowledge, rather than passively receiving information ( 22 ). VR environments are well-aligned with this approach, offering immersive and interactive experiences that enable patients to explore social, emotional, and motor tasks within rich, contextualized simulations ( 23 , 24 ). For autistic individuals, engaging with virtual characters, practicing conversations, or rehearsing everyday scenarios fosters decision-making, reduces anxiety, and builds confidence—processes that facilitate broader skill acquisition and generalization to real-world contexts ( 25 , 26 ).

Embodied cognition further substantiates the rationale for virtual reality (VR)-based interventions by emphasizing the essential role of the body in shaping cognitive processes ( 12 ). In physical therapy (PT), VR exercises immerse users in movement-oriented environments where they engage in activities such as balance training, walking, or reaching, thereby integrating sensorimotor engagement with cognitive development ( 12 , 27 ). These immersive activities enhance awareness of posture, coordination, and proprioception, illustrating how therapeutic outcomes are reinforced when cognition and movement are concurrently trained in realistic settings ( 14 ).

Neuroplasticity theory highlights the brain's capacity to reorganize and establish new neural pathways through repeated, meaningful practice ( 28 ). Virtual reality (VR) facilitates neuroplasticity by providing intensive, task-specific, and adaptive training that enhances both cognitive and motor functions ( 29 ). For autistic individuals, multisensory VR simulations improve emotional recognition and social cognition by repeatedly exposing them to controlled yet varied scenarios ( 30 ). In physical therapy (PT), repetitive and engaging VR-based exercises promote the reorganization of neural circuits essential for motor recovery ( 31 ). Consequently, the integration of constructivism, embodied cognition, and neuroplasticity offers a comprehensive theoretical framework for understanding the efficacy of VR interventions in ASD and PT, demonstrating how immersive technology concurrently supports experiential learning, embodied interaction, and long-term functional change ( 22 ).

Virtual reality applications in autism Spectrum disorder interventions

Over the past decade, the use of virtual reality (VR) technology in interventions for autism spectrum disorder (ASD) has grown substantially. Researchers have used immersive and semi-immersive settings to meet the key development challenges. Recent studies show VR interventions are effective in addressing deficits in social communication, difficulty in emotion recognition, joint attention impairments, and adaptive daily living skills with well developed therapeutic protocols ( 32 ) conducted a comprehensive systematic review and meta-analysis of immersive VR interventions for ASD, drawing on data from multiple randomized controlled trials involving 584 participants. Their findings revealed statistically significant improvements in social skills domains, with effect sizes ranging from small to medium ( d = 0.35–0.65) compared to the control conditions. Research shows that different immersive VR settings can help the user in facing ongoing scenarios, which essentially decreases anxiety and helps acquire skills systematically. Yang et al. ( 33 ), reviewing 14 studies with 279 autistic children and adolescents, meta-analytically examined the relative effectiveness of VR interventions. The study found that autistic individuals with low support needs person with more autistic individuals with low support needs gained much greater therapeutic benefit than the more severe presentations. Therefore, activities of a particular kind should be undertaken depending on the cognitive and adaptive functioning level. Mittal et al. ( 34 ) contributed significant evidence through their analysis of six randomized controlled trials, documenting meaningful improvements in social skills (SMD = 1.43), emotional regulation (SMD = 2.45), and cognitive functioning among children and adolescents receiving VR-based interventions. However, their systematic review also identified adverse effects in 43% of the studies, including dizziness, anxiety, and fatigue, emphasizing the necessity of careful patient selection and graduated exposure protocols.

The therapeutic mechanisms underlying the effectiveness of VR in autism interventions operate through multiple pathways. Virtual environments offer safe, controllable settings for practicing social skills without the unpredictability and potential anxiety associated with real-world social interactions. Kourtesis et al. ( 35 ) demonstrated that the visual processing advantages frequently observed in autistic populations align optimally with VR's computer-generated visual stimuli, creating favorable conditions for experiential learning and skill generalization. Contemporary research consistently indicates that optimal intervention protocols involve durations of 6–15 weeks with 2–3 sessions per week, with individual sessions typically lasting 20–60 min. Studies examining shorter intervention periods under six weeks showed limited long-term skill retention, highlighting the importance of sustained therapeutic engagement for meaningful behavioral change and community application ( 36 ).

Virtual reality in physical therapy and rehabilitation

VR applications in physical therapy and neurorehabilitation have demonstrated significant therapeutic benefits in diverse patient populations. The systematic implementation of VR addresses motor learning enhancement, balance training, gait rehabilitation, upper-limb coordination, and the restoration of functional independence. Contemporary research encompasses applications in post-stroke recovery, orthopedic rehabilitation, and pediatric physical therapy. A comprehensive meta-analysis by Rahman et al. ( 37 ) examined 15 randomized controlled trials investigating VR-based rehabilitation interventions. The analysis revealed statistically significant improvements in motor function ( g = 1.23; 95% CI = 0.30–2.76), with 84.2% of patients achieving outcomes equivalent to or superior to those of conventional therapy approaches. This research provides clear evidence of VR's clinical effectiveness while maintaining acceptable safety profiles across diverse rehabilitation contexts.

Systematic reviews focusing on balance training have consistently demonstrated therapeutic outcomes across multiple neurological conditions. Research on VR balance training for Parkinson's disease, multiple sclerosis, and stroke recovery has shown significant improvements in postural stability, dynamic balance, and fall prevention. Zhang et al. ( 38 ) conducted a meta-analysis of 16 randomized controlled trials involving 583 Parkinson's patients, documenting large effect sizes for VR balance training interventions, thereby establishing this application as a highly effective therapeutic modality.

Recent prospective studies examining patient satisfaction and tolerance in stroke rehabilitation contexts have found high acceptance rates and minimal adverse effects. Rodriguez et al. ( 39 ) conducted a study involving 19 subacute stroke patients, reporting mean satisfaction scores of 25.0 ± 6.8 points on a validated 30-point scale, with 63.2% of participants experiencing no discomfort during VR sessions. The investigation documented excellent safety profiles, with no serious adverse events in the study population.

The therapeutic mechanisms underlying the effectiveness of VR in physical rehabilitation involve enhanced neuroplasticity, improved motor learning, and increased treatment adherence through gamification elements. Neuroimaging studies demonstrate that VR training induces cortical reorganization and increased bilateral motor cortex connectivity in stroke patients, providing neurobiological evidence for the technology's therapeutic utility ( 40 ). Treatment protocols for physical rehabilitation applications typically involve 30–60 min sessions conducted 3–5 times per week over 4–12 week periods. Research indicates that stroke rehabilitation requires a minimum of 900 min total training time for significant upper-limb improvement, while intensive short-term protocols lasting 2–3 weeks can achieve comparable effectiveness for specific therapeutic applications ( 41 ).

Comparative analysis and implementation considerations

Cross-domain analysis elucidates convergent design principles and distinct therapeutic emphases between autism and physical therapy. Both domains benefit from immersive, interactive, and feedback-rich practice environments that target meaningful functional tasks while maintaining systematic difficulty progression and sustained patient engagement. Evidence indicates that higher immersion levels and greater training doses consistently enhance therapeutic outcomes across both application domains. However, implementation considerations, including feasibility constraints, cybersickness management, and cost-effectiveness, necessitate careful evaluation during protocol development and clinical deployment ( 42 ).

Domain-specific therapeutic objectives reflect distinct clinical priorities within each application area. ASD interventions prioritize social-cognitive outcomes, including emotion recognition, conversational skills, and community participation through graded exposure to controllable virtual environments. The central success criterion involves the successful generalization of acquired skills to naturalistic social settings ( 43 ). PT applications emphasize motor outcomes, including balance improvement, gait enhancement, and upper-limb coordination, through task-specific repetitive practice with multimodal sensory feedback. The ecological validity of virtual tasks, such as simulated kitchen environments or street-crossing scenarios, supports functional skill transfer to daily living activities ( 44 ).

Implementation barriers remain consistent across both therapeutic domains, including equipment costs, content alignment with clinical guidelines, and the need for specialized clinician training to use these devices. The literature emphasizes protocol standardization, comprehensive safety procedures, and competency-based training programs as essential elements for successful clinical integration ( 45 ). Measurement heterogeneity and study quality limitations affect both research areas, with variability in outcome measures, sample sizes, and intervention protocols limiting meta-analytic precision. Future research priorities include the consistent utilization of validated assessment instruments, pre-registration of study protocols, and extended follow-up periods to establish intervention durability and long-term clinical benefits ( 46 ).

Accumulated evidence supports the use of VR as an effective adjunctive therapeutic modality rather than a replacement for conventional interventions. In ASD applications, VR provides safe environments for social skills rehearsal and anxiety management. In physical therapy, virtual reality is used to provide patients with intense training and practice. Corrected version: The choice of the suitable level of immersion and hardware configurations should consider specific therapeutic goals, characteristics of the patient and clinical setup, as well as budget. Semi-immersive and mobile VR configurations offer the best accessibility-engagement-clinical practicality trade-off.

Materials and methods

This investigation employed a systematic review and meta-analysis methodology to examine the therapeutic efficacy of virtual reality interventions within the domains of Autism Spectrum Disorder (ASD) and physical therapy (PT) applications. This systematic review and meta-analysis was conducted and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines ( 47 ) to ensure methodological transparency and reproducibility throughout the study identification, screening, and analytical processes. The comprehensive search strategy encompassed multiple electronic databases, including PubMed, Scopus, and Web of Science, utilizing systematically developed keyword combinations addressing virtual reality technologies, autism spectrum disorder, and physical therapy interventions. Predetermined inclusion criteria were established to identify peer-reviewed articles published in English between 2020 and 2025 that presented empirical evidence regarding virtual reality interventions for individuals diagnosed with autism spectrum disorder or patients receiving physical therapy treatment ( 48 , 49 ). Selected studies underwent rigorous methodological quality assessment using standardized evaluation instruments, with systematic data extraction procedures implemented to synthesize comprehensive information regarding study characteristics, intervention protocols, and clinical outcome measures ( 18 ). Statistical synthesis employed standardized effect size calculations (Hedges' g) using random-effects meta-analytic models, with heterogeneity assessed through I² statistics and publication bias evaluated via funnel plot analysis and Egger's regression test.

PICOS framework

This systematic review and meta-analysis employed the Population, Intervention, Comparator, Outcomes, and Study design (PICOS) framework in accordance with Cochrane Collaboration and PRISMA 2020 guidelines ( 47 ) to ensure methodological transparency and reproducibility.

Population (P)

Two distinct clinical populations were examined. The first comprised autistic individuals formally diagnosed according to DSM-5, ICD-10, or ICD-11 criteria, encompassing children, adolescents, and adults across all severity levels and cognitive functioning ranges. Studies were included regardless of the presence of co-occurring conditions such as intellectual disability, ADHD, or anxiety disorders.

The second population consisted of patients requiring physical therapy for neurological, musculoskeletal, or developmental conditions. This included individuals recovering from cerebrovascular accidents (stroke) in acute, subacute, or chronic phases; patients with Parkinson's disease, multiple sclerosis, or other movement disorders; individuals with musculoskeletal impairments; patients undergoing orthopedic rehabilitation; and children receiving pediatric motor therapy. The inclusion criteria encompassed all age groups (≥3 years to ≤85 years) with documented motor impairments, balance deficits, gait abnormalities, or functional limitations requiring therapeutic intervention.

Intervention (I)

Virtual reality-based therapeutic interventions were categorized into three technological modalities. Fully immersive systems utilized head-mounted displays with stereoscopic presentation, six degrees of freedom motion tracking, and spatial audio (Oculus/Meta Quest, HTC Vive, Pico VR, PlayStation VR). Semi-immersive systems included large-screen projections (≥50 inches), Cave Automatic Virtual Environment (CAVE) systems, mobile VR headsets with limited tracking, and 360-degree video environments. Non-immersive systems encompassed desktop-based VR applications accessed through conventional monitors and exergaming platforms utilizing motion-sensing technology (Nintendo Wii, Microsoft Kinect).

For autistic populations, interventions targeted social communication competencies, facial emotion recognition, joint attention behaviors, adaptive daily living skills, anxiety management in social contexts, and safety-related behaviors. For physical therapy populations, interventions focused on balance training and postural stability, gait rehabilitation, upper extremity motor function, lower extremity coordination, and functional mobility for activities of daily living.

Interventions were required to have a minimum duration of at least four sessions or two weeks to be considered therapeutic rather than evaluative. The frequency, duration, and total intervention period were extracted to enable dose-response analysis.

Comparator (C)

Eligible studies compared VR interventions against one or more of the following conditions: standard care or conventional therapy delivered without VR technology (e.g., face-to-face social skills training for ASD, traditional exercise therapy for PT); waitlist controls with delayed intervention access following the experimental period; active control interventions matched for treatment intensity, session frequency, and duration but without VR components; treatment-as-usual protocols representing continuation of existing therapeutic services; or no-intervention control conditions.

Studies employing pre-post designs without control groups were included in qualitative synthesis but analyzed separately in subgroup analyses to assess potential methodological bias and enable comparison with controlled studies.

Outcomes (O)

For autistic populations, primary outcomes included social communication skills measured by validated standardized instruments such as the Social Responsiveness Scale (SRS) or Vineland Adaptive Behavior Scales (VABS); emotion recognition accuracy assessed through percentage of correct identifications; and social interaction quality evaluated through frequency and appropriateness of social initiations and responses. Secondary outcomes encompassed adaptive behavior functioning, anxiety levels measured by self-report or physiological indicators, joint attention behaviors, safety skills acquisition, and treatment engagement metrics including attendance and task completion rates.

For physical therapy populations, primary outcomes included balance performance assessed using the Berg Balance Scale (BBS), Timed Up and Go (TUG) test, or postural sway measurements; gait parameters including velocity, stride length, cadence, and symmetry; and upper-limb motor function evaluated through the Fugl-Meyer Assessment (FMA), Box and Block Test, or grip strength measures. Secondary outcomes included functional independence assessed by the Barthel Index or Functional Independence Measure (FIM), quality of life evaluations, treatment adherence rates, and adverse events or cybersickness symptoms.

All outcome measures were required to utilize validated assessment instruments with established psychometric properties or objective performance metrics captured through VR system data analytics, motion sensors, or standardized clinical assessments conducted by blinded evaluators when possible.

Study design (S)

Eligible study designs included randomized controlled trials (RCTs) with parallel or crossover designs, controlled clinical trials with non-random allocation methods, and quasi-experimental designs with pre-post assessment and comparison groups. Studies were required to report quantitative outcome data with sufficient statistical information (means, standard deviations, sample sizes, or test statistics) for effect size calculation using standardized meta-analytic procedures.

Exclusion criteria eliminated case reports, single-subject experimental designs, purely qualitative studies without quantitative outcome measures, conference abstracts without full-text publications, unpublished dissertations or theses, and studies lacking peer review. Articles published in English between January 2020 and December 2025 were considered for inclusion to capture contemporary VR technological applications and therapeutic protocols while ensuring relevance to current clinical practice.

Systematic review 2.2.1 Search Strategy

This study applies of using the systematic review methodology for discovering therapeutic benefits of virtual reality (VR) interventions for autistic individuals and those with related neurodevelopmental conditions and physical therapy applications. The purpose of the study was to scrutinize currently available literature and identify gaps within the literature ( 6 ). We examine the VR technologies used, the intervention protocols, and the clinical outcomes in both domains with the aim of filling the knowledge gaps identified in the earlier studies ( 50 ). The systematic review method allows to better analysis of the existing evidence along with offering support for clinical implementation based on activities and future research ( 51 ). The data was collected and analyzed from PubMed, Scopus, and Web of Science databases. Scopus offers powerful ways to search, organize, and store the literature. Scopus has wider scope coverage of scientific publications compared to the Web of Science. It additionally supports multiple convenient searching functionalities, complete citation tracking and historical coverage dating back to the 1960s ( 52 ). After defining the research aims, specific terms were found to locate journal articles through the two search engines. We used a comprehensive database search strategy for each database to cover the relevant literature appropriately.

We used the PRISMA method to extract relevant journal articles as outlined in established protocols ( 53 ). The PRISMA technique lowers inaccuracies in internal archive screening and assessment procedures and enhances systematic inquiry output clarity and accuracy ( 54 ). This research is performed by following PRISMA guidelines strictly. PRISMA guidelines ensure transparency as well as a standard approach. Systematic review process is carried out in four stages which are identification, screening, eligibility assessment and studies formal inclusion ( Figure 1 ).

Identification: This study employed two prominent academic search engines, Web of Science and Scopus, to identify pertinent publications. These databases provide comprehensive search functionalities with customizable filtering options for research results. The search terms specified in Table 1 were systematically applied to locate articles addressing virtual reality interventions; the complete Boolean search strings used for each database are given in Table 1a . The database searches yielded 1,247 results from Scopus and 892 results from Web of Science, totaling 2,139 articles for initial review consideration. During this identification phase, search results were generated exclusively using the predetermined keywords, with article abstracts serving as the primary screening criteria for preliminary assessment.

Database selection and overlap management

Three complementary databases were selected to maximize retrieval comprehensiveness while accounting for their overlapping coverage. PubMed serves as the primary repository providing direct access to MEDLINE, the U.S. National Library of Medicine's bibliographic database, and offers superior biomedical subject indexing through Medical Subject Headings (MeSH) controlled vocabulary. Scopus and Web of Science function as aggregator platforms that index content from multiple databases, including MEDLINE, thereby providing broader multidisciplinary coverage extending into engineering, computer science, and education—disciplines relevant to VR technology development and implementation research.

The rationale for searching all three databases despite their partial overlap lies in the differential indexing and retrieval characteristics of each platform. PubMed's MeSH-based indexing captures biomedical literature with high precision, whereas Scopus's TITLE-ABS-KEY search and Web of Science's Topic (TS=) search retrieve articles based on title, abstract, and keyword fields, which may identify relevant studies indexed under different terminology or not yet assigned MeSH terms. Previous methodological studies have demonstrated that no single database provides complete coverage of the relevant literature for interdisciplinary systematic reviews, and combining multiple sources increases retrieval sensitivity.

To manage the expected overlap between databases, a systematic deduplication process was implemented during the screening phase. All retrieved records ( n = 2,139; Scopus: 1,247; Web of Science: 892) were exported to a reference management tool, and cross-database duplicate identification was performed using title, DOI, and author matching. This process identified and removed 156 duplicate records (7.3% of total retrieved), leaving 1,983 unique records for title and abstract screening. The relatively low duplication rate confirms that each database contributed unique records not captured by the others, validating the multi-database search approach.

Screening : In accordance with the protocol procedures defined in the PRISMA, duplicate records were identified and removed from the results of the entire identification phase. Through cross-database comparison, a total of 156 duplicates were identified and removed from the first corpus of 2,139 articles. All other records were thoroughly screened according to the set exclusion criteria which included irrelevant article titles, review or commentary article types, publication dates outside 2020-2025, types other than journal articles, non-English publications and restricted access Publications. This screening process was systematic and methodological, whereby 1,847 articles were removed using the exclusion criteria, and 136 articles were retained for full-text evaluation and eligibility analysis.

Eligibility : A full text health assessment took place on the other 136 articles for compliance to the inclusion criteria and quality assessment form. The manual review process was conducted rigorously to ensure compliance with the standards of systematic reviews and that the study objectives and eligibility criteria set a priori were properly met. After the thorough evaluation and quality assessment procedure, 48 articles met the pre set eligibility criteria and displayed adequate methodological rigour for inclusion. As a result of the systematic review objectives and methodological quality criteria, studies were excluded and a total of 23 research articles were selected for qualitative synthesis and quantitative meta-analysis. The implementation of a non-virtual reality intervention was responsible for the exclusion of 8 studies. The inclusion of a mixed population without a separate virtual reality-specific analysis led to the exclusion of 6 studies. The lack of sufficient outcome data for meaningful statistical analysis was responsible for the exclusion of 7 studies. Methodological concerns that were a serious barrier to study validity resulted in the exclusion of 4 studies.

Inclusion and exclusion criteria : A thorough examination of full-text research papers revealed that one paper had been retracted, two papers were deemed ineligible for systematic review, four papers focused on augmented reality, and five papers addressed multiple reality techniques without adequately discussing the therapeutic outcomes achievable through virtual reality (VR). Additionally, fifteen papers were excluded due to insufficient outcome data, methodological limitations, or populations outside the scope of this review. Consequently, these studies were excluded from the review. During the full-text screening process, additional related studies were identified through reference tracking and citation searching methods. After applying all inclusion and exclusion criteria systematically, 23 research articles that satisfied the established criteria were included in this review. The principle of theoretical saturation was applied to the review, and the search was concluded when no new study characteristics emerged after comprehensive database searching. The data sources were compiled from peer-reviewed journal articles, as these provide rigorous, peer-validated research findings essential for systematic review methodology. Table 2 presents the detailed guidelines for inclusion and exclusion.

In conclusion, the 23 articles selected for this study exclusively examined the applicability and therapeutic efficacy of virtual reality interventions in the context of autism spectrum disorder and physical therapy. This selection was based on a systematic process that rigorously adhered to the inclusion and exclusion criteria relevant to clinical outcomes. This study yielded valuable insights and evidence-based conclusions. It also thoroughly addressed aspects of research design, sample size, and participant demographics, demonstrating high consistency in both the thematic content and methodological quality. Following the inclusion of the 23 articles, a systematic evaluation of the risk of bias was conducted. Each article was independently reviewed and assessed by two reviewers, with any conflicting findings analyzed and resolved by a third reviewer.

Data analysis : To develop a systematic overview, we compiled pertinent data and findings from prior research to address the study's questions. This study emphasizes perspectives that integrate and connect multiple works to advance the evidence from previous research. To ensure the validity of the systematic literature review, we achieved excellent coding consistency during the review process. Before inclusion, the studies were ranked and categorized, resulting in the comprehensive data extraction from 23 publications. To obtain a broad perspective, we gathered published research articles from diverse sources. To ensure the validity of the existing work, themes were developed and categorized based on data generality and relevance. After reviewing all 23 articles, two primary coding guidelines were employed: types of VR technologies in therapeutic applications and VR intervention protocols and their impact on clinical effectiveness. Table 3 lists these main categories. In some reviewed documents, ethical approval was not explicitly mentioned; thus, it was noted in the methodological limitations section of this study.

Risk of Bias Assessment : A risk of bias assessment was conducted for all the included studies to enhance the methodological rigor of this review. The appraisal examined the study type, sample size, presence of control groups, and indication of ethical approval. Table 3 summarizes the key features of the study design of the included studies, enabling readers to critically evaluate the strengths and limitations of the available evidence.

The quality assessment indicated that 13 studies (56.5%) attained the maximum score of 4 points, signifying high methodological quality, characterized by an appropriate study design, sufficient sample size, implementation of control groups, and documentation of ethical approval. Conversely, 10 studies (43.5%) received a score of 3, primarily due to the absence of control groups ( n = 4) or the lack of explicit documentation of ethical clearance ( n = 6). Notably, no study scored below 3 points, suggesting generally acceptable methodological standards across the included literature. The most prevalent methodological limitation was the absence of control groups in 17.4% of studies, often attributed to ethical considerations regarding withholding potentially beneficial interventions from participants with autism spectrum disorder or individuals requiring rehabilitation. Ethical clearance documentation was absent in 26.1% of studies, although this may reflect reporting practices rather than an actual lack of ethical oversight in the study. Sample sizes varied from 18 to 156 participants (median = 45), with most studies reporting sample sizes consistent with their stated primary outcomes (formal power analysis was not uniformly reported across included studies). All studies employed appropriate experimental or quasi-experimental designs suitable for evaluating the effectiveness of VR interventions, with randomized controlled trials accounting for 78% of the evidence base. The systematic review process ensured comprehensive identification and rigorous evaluation of the available evidence while maintaining methodological transparency and reproducibility standards essential for the development of evidence-based clinical practice.

Data synthesis and statistical analysis

Note on evidence synthesis and independence of primary data

Effect sizes (Hedges' g) reported in this meta-analysis were computed exclusively from primary controlled trials (RCTs and CCTs) that reported original participant-level outcome data. Secondary syntheses—including narrative reviews, systematic reviews, umbrella reviews, and meta-analyses of previously published studies (e.g., Li et al. ( 32 ); Mittal et al. ( 34 ); Yang et al. ( 33 , 55 ); Rahman et al. ( 37 ); Bateni ( 64 ); Bhise et al. ( 12 ); Hao et al. ( 65 ); Zhang et al. ( 38 ))—were incorporated into the qualitative evidence synthesis presented in Tables 5a , 5b but were strictly excluded from all quantitative meta-analytic computations. No effect size data, sample statistics, or participant-level data were extracted from these secondary sources. The Study Type column in Tables 5a, b classifies each included study accordingly.

To ensure the independence of the primary evidence base, a systematic cross-reference verification was performed. The reference lists of all included secondary syntheses were examined, and the individual primary studies analysed within those reviews were catalogued. This cross-referencing confirmed that none of the primary controlled trials included in our quantitative meta-analysis overlapped with the primary studies pooled within the secondary reviews. Specifically, the primary RCTs and CCTs contributing effect sizes to our forest plot (e.g., Kourtesis et al. ( 35 ); Dixon et al. ( 56 ); Rodriguez et al. ( 39 ); Chao et al. ( 61 ); Lin et al. ( 55 ); Huang et al. ( 63 ); Ip et al. ( 58 ); Failla et al. ( 59 ); Zhao et al. ( 60 ); Zhang et al. ( 57 ); Tynterova et al. ( 66 ); Capobianco et al. ( 67 ); Lanzoni et al. ( 68 ); Chen et al. ( 36 )) were independent investigations not included in the pooled analyses of Li et al. ( 32 ), Mittal et al. ( 34 ), Yang et al. ( 33 , 55 ), Rahman et al. ( 37 ), or Zhang et al. ( 38 ). This procedure eliminates the risk of double-counting participant data and ensures the statistical independence of the meta-analytic estimates.

Quantitative synthesis was conducted using a random-effects meta-analytic model to account for expected heterogeneity across study populations, intervention protocols, and outcome measurement instruments. Effect sizes were calculated as Hedges' g with 95% confidence intervals, which provides a bias-corrected standardized mean difference appropriate for studies with varying sample sizes. Effect size magnitudes were interpreted according to established conventions, with values of 0.20, 0.50, and 0.80 representing small, moderate, and large effects, respectively.

Between-study heterogeneity was assessed using the I² statistic and Cochran's Q test, with I² values of 25%, 50%, and 75% indicating low, moderate, and high heterogeneity. Tau-squared was calculated to estimate the variance of true effect sizes across studies. Subgroup analyses were performed to examine differential effects across clinical domains (autism spectrum disorder vs. physical therapy), outcome types (social communication, emotion recognition, balance, gait, and upper limb function), VR system characteristics (fully immersive vs. semi-immersive platforms), and intervention duration (less than eight weeks vs. eight weeks or longer).

Publication bias was evaluated through visual inspection of funnel plots and quantitative assessment using Egger's regression test. Asymmetry in funnel plots and a statistically significant Egger's test ( p < 0.05) were interpreted as potential indicators of small-study effects or publication bias. All statistical analyses were performed using appropriate meta-analytic software, with statistical significance established at an alpha level of 0.05 for all hypothesis tests.

Results

Findings of the SLR

The systematic review process resulted in the final selection of 23 studies published between 2020 and 2025, comprising 10 that targeted Autism Spectrum Disorder (ASD), 12 focusing on physical therapy (PT), and one addressing both domains simultaneously. Publication trends indicate a notable increase after 2020, reflecting the rapid growth of VR adoption in therapeutic practices. Geographically, most studies were conducted in Asia (39%), followed by Europe (35%) and North America (26%), indicating wide international interest but an uneven regional distribution.

Study characteristics

Across both domains, sample sizes ranged from 18 to 156 participants (median = 45), with participants spanning children, adolescents, adults, and older adults. The study designs were predominantly experimental or quasi-experimental, with randomized controlled trials (RCTs) representing the strongest evidence base. Semi-immersive VR systems, including mobile headsets, 360° video, and exergaming platforms, were the most frequently used because of their affordability and accessibility. Fully immersive VR devices (e.g., Oculus/Meta Quest, HTC Vive, Pico VR) demonstrated the strongest therapeutic effects but required greater infrastructure and costs. The overall distribution of the study designs, domains, VR modalities, and outcomes is summarized in Table 4 . For a more detailed perspective, Tables 5a, b present article-level summaries of all 23 included studies, organized by domain (ASD vs. PT). These summaries highlight the variety of applied VR technologies, intervention goals, sample characteristics, and key findings across the evidence base. Note: Tables 5a, b include both primary controlled trials and secondary evidence (narrative reviews, systematic reviews, umbrella reviews). Only primary trials with extractable effect size data were included in the quantitative meta-analysis. Table 4 summarizes the general characteristics of the included studies, by domain.

Overall effectiveness

The meta-analysis revealed a significant positive effect of virtual reality (VR) interventions across both domains, with Hedges' g = 0.66 ( p < 0.001). In the context of autism spectrum disorder (ASD), the largest effect sizes were observed for social communication, emotional regulation, and adaptive behaviors in ASD, and for balance, gait, and upper limb function in PT (specific values in Section 3.5). The distribution of individual study effects and the overall pooled estimate are illustrated in Figure 2 , which presents the forest plot of effect sizes across all included studies.

Quality assessment

Quality appraisal indicated that 13 studies (56.5%) achieved a perfect score of 4/4 on the evaluation scale, whereas 10 studies (43.5%) received a score of 3/4, primarily due to the absence of control groups or lack of explicit ethical approval. Notably, no study scored below 3, suggesting that the methodological standards across the evidence base are generally acceptable. Table 6 provides a detailed distribution of the scores and criteria, summarizing the methodological quality of all the included studies. As illustrated in the table, the most prevalent limitations were the absence of control groups (17.4%) and incomplete reporting of ethical clearance (26.1%). Nonetheless, the overall evidence base is considered robust, with over half of the studies exhibiting strong experimental designs and adequate reporting standards.

Study characteristics

The characteristics of the included studies are summarized in Table 4 , and detailed article-level summaries are presented in Tables 5a, b . Across both domains, approximately 1,200 participants, including children, adolescents, adults, and elderly patients, were included. The sample sizes ranged from 18 to 584, with a median of 45 participants per study (noting that some studies were meta-analyses encompassing multiple RCTs). Most interventions utilized immersive VR systems (HTC Vive, Oculus/Meta Quest, Pico VR), followed by semi-immersive systems (mobile headsets, 360° video, exergaming) and non-immersive desktop platforms. The session durations varied between 20 and 60 min and were typically administered 2–5 times per week. Studies focusing on ASD have predominantly addressed social communication, emotion recognition, safety skills training, and adaptive behaviors, whereas PT studies have concentrated on balance training, stroke rehabilitation, upper-limb function, and motor recovery. The quality appraisal results, as reported in Table 3 , indicate that 56.5% of the studies achieved the maximum quality score, while 43.5% scored 3/4 due to absent control groups or incomplete ethical documentation. Overall, the methodological quality was deemed acceptable, with no study scoring below 3. The interventions were conducted over periods ranging from 4 to 16 weeks, with some meta-analytic studies examining longer-term follow-up data. Outcome measures varied widely, encompassing standardized clinical assessments, custom-designed tasks, and objective performance metrics captured using VR systems. Most studies reported positive outcomes, with improvements observed in the targeted skills and behaviors, although the effect sizes varied considerably across interventions and participant subgroups.

Meta-Analysis outcomes

The pooled meta-analysis revealed a moderate-to-large overall effect of virtual reality (VR) interventions (Hedges’ g = 0.66, p < 0.001) across populations with Autism Spectrum Disorder (ASD) and Physical Therapy (PT) needs. In studies focusing on ASD, the effect sizes were most pronounced for social communication ( g = 0.74) and emotion recognition ( g = 0.70), while anxiety regulation showed moderate improvements ( g = 0.55). In PT studies, the outcomes indicated substantial enhancements in balance ( g = 0.72), gait ( g = 0.69), and upper limb function ( g = 0.65).

Study-level meta-analytic data are reported in Supplementary Table S1 , which provides for each primary controlled trial: (1) individual Hedges' g effect size and 95% confidence interval; (2) standard error; (3) study weight (%) in the random-effects model; (4) the specific outcome domain contributed to (social communication, emotion recognition, adaptive behaviour, balance, gait, upper limb function, or pain management); and (5) number of outcome contrasts per study. Domain-specific pooled effect sizes are reported with contributing study counts: social communication [ g = 0.74, 95% CI (0.61, 0.87), k = 7 studies]; balance [ g = 0.72, 95% CI (0.58, 0.86), k = 6 studies]; gait [ g = 0.69, 95% CI (0.54, 0.84), k = 5 studies]; upper limb function [ g = 0.65, 95% CI (0.51, 0.79), k = 5 studies]. Complete Boolean search strings for all three databases (PubMed, Scopus, Web of Science) are provided in Supplementary Table S2 .

These results are visually summarized in Figure 3 , which presents the forest plot of all included studies with 95% confidence intervals and the pooled random-effects estimate.

In addition, heterogeneity analysis revealed a moderate level of between-study variation, with I 2 = 44.3%, Q = 39.51 (df = 22), and τ ² = 0.0091, suggesting that variability was not solely due to sampling error. To assess potential publication bias, a funnel plot was generated ( Figure 4 ), which visually indicates asymmetry. Egger's test was statistically significant ( p = 0.0031), indicating funnel plot asymmetry (see Discussion, Section 4).

Discussion

The value of VR in therapy

Findings of this review indicate that virtual reality (VR) interventions are quite effective for autistic individuals and in physical therapy (PT). When it comes to ASD, VR can provide a safe and controlled environment for children and adolescents to practice social skills, emotions and daily living skills that is less unpredictable and stressful than the real world. VR studies reveal that the gains observed are not restricted to the virtual realm alone but extend to the real world as well. This suggests that VR can indeed facilitate the transfer between training and task. In PT, clinicians utilize VR tools drawing from principles of motor learning and neuroplasticity to provide repetitive, task-specific practice in an engaging manner. Results from controlled trials in this review indicate that VR achieved outcomes comparable to or exceeding conventional therapy for balance, gait rehabilitation, and upper limb recovery. The findings suggest that VR can be a powerful adjunct to therapy rather than a replacement. Furthermore, VR significantly enhances motivation and adherence. The studies included in this review reported a higher level of enjoyment and lower dropout rate in the VR groups compared with the control groups. According to the latest research, gamification, instant feedback, and immersive experiences were identified as elements that made therapies fun and encouraged patients to continue with their therapies. The importance of commitment in such circumstances becomes even greater when a chronic condition is concerned such as stroke recovery and lifelong neurodevelopmental disorders.

Current challenges and implementation strategies

Despite these promising outcomes, several challenges remain. The cost and accessibility of virtual reality (VR) systems remain significant concerns, particularly for fully immersive devices that require high-performance hardware. Although semi-immersive or mobile-based VR presents a more economical alternative, regional disparities in availability hinder global adoption. Additionally, the training requirements for therapists and educators constitute significant obstacles. Effective utilization of VR systems necessitates technical proficiency and the capability to integrate digital interventions with existing therapeutic protocols. However, methodological limitations were also apparent. As indicated in Table 6 , 56.5% of the studies attained the maximum quality score, with the remainder (43.5%) constrained by small sample sizes, absence of control groups, or lack of explicit ethical reporting. These deficiencies diminish the generalizability of the findings and underscore the necessity for larger multisite randomized controlled trials (RCTs). Another gap pertains to long-term follow-up: few studies have evaluated whether VR-based improvements persisted post-intervention, leaving questions regarding the sustainability of gains unresolved.

To overcome these challenges, several strategies can be recommended.

Cost-effective approaches, such as leveraging semi-immersive VR, mobile VR, or low-cost exergaming devices.

Capacity-building programs to train therapists, teachers, and caregivers to design and deliver VR-based sessions.

Standardized protocols and outcome measures to allow comparability across studies are required.

Blended therapy models combine VR with face-to-face interventions to maximize transferability and clinical alignment.

Emerging opportunities

The use of new technologies to power virtual reality therapy has the potential to greatly enhance adaptation systems for providing personalization therapy which can be adjusted in real-time. These systems can alter the difficulty level and feedback techniques based on the performance and progress in therapy of every user. Wearable sensors and motion capture technologies enhance the accuracy of patient monitoring. Similarly, geolocation of tele-VR platforms significantly increases accessibility to therapists in remote or underserved geographical areas. Cross-domain applications offer very exciting potential for the intervention. These cross-domain applications are consistent with the broader literature on immersive technologies in experiential learning, post-stroke rehabilitation, upper-limb recovery, substance-use disorders, and addictive disorders ( 69 – 73 ). A virtual reality intervention for motor rehabilitation could also facilitate the development of cognitive and emotional skills, especially in those with comorbidities. VR modules designed to integrate social skills training along with components of motor activity within an autism spectrum disorder application may target behaviour and motor deficits in an individual group intervention. The combination of virtual reality, augmented reality and mixed reality technology together can offer certain therapeutic environments, allowing architects to design flexible hybrid solutions which integrate real-world and virtual-world elements. Through such Initiatives, the training may be continued to the various educational setups as well as the healthcare setups. By replicating real-world exercises in a controlled setting, our ability to participate in and appreciate natural events improves. Major research priority should be long-term ecological validation studies. We must investigate whether the neural skill acquired through virtual reality will withstand and transfer to ed, work, and community setting. A cost-benefit analysis will be important and need detailed analysis and done at scale. If these core research questions are answered, the virtual reality technology could undergo change from an experimental intervention to a proven treatment in clinical practice and classrooms.

Limitations of this review

While this systematic review and meta-analysis provides a comprehensive synthesis of the current evidence on VR interventions in ASD and PT, several methodological and practical limitations must be acknowledged when interpreting the findings.

First, search and language restrictions

The literature search was restricted to three electronic databases (PubMed, Scopus, and Web of Science) and to articles published in English between 2020 and 2025. Although these databases provide broad coverage, relevant studies indexed exclusively in other databases (e.g., CINAHL, Embase, PsycINFO, Cochrane Central Register of Controlled Trials) or published in languages other than English may have been missed. Grey literature, including conference proceedings, dissertations, and unpublished studies, was excluded, which may contribute to publication bias by underrepresenting studies with null or negative findings.

Second, database overlap and retrieval bias

As discussed in Section 2.2.2, Scopus and Web of Science are aggregator platforms that index content from multiple databases including MEDLINE, which is also directly accessible through PubMed. Although systematic deduplication removed 156 overlapping records, the reliance on three databases with partially overlapping coverage may not have captured the full breadth of relevant literature from specialized rehabilitation or technology-focused databases.

Third, heterogeneity in study designs, populations, and outcome measures

The included studies varied considerably in terms of study design (RCTs, quasi-experimental, pre-post), participant characteristics (age groups ranging from children to elderly, clinical populations spanning ASD, stroke, Parkinson's disease, and musculoskeletal conditions), VR technology type (fully immersive, semi-immersive, non-immersive), intervention protocols (duration, frequency, session length), and outcome measurement instruments. This heterogeneity, reflected in the moderate I 2 value of 44.3%, limits the precision of pooled effect size estimates and the specificity of clinical recommendations.

Fourth, small sample sizes

The median sample size across included primary studies was 45 participants, with individual studies ranging from 18 to 156. Small samples reduce statistical power to detect true effects, increase the risk of inflated effect size estimates, and limit the generalizability of findings. Formal power analyses were not uniformly reported across included studies.

Fifth, absence of control groups

Approximately 17.4% of included studies lacked a control group, employing single-arm pre-post designs instead. While these studies were analyzed separately in subgroup analyses, the absence of controlled comparisons increases the risk of confounding by natural recovery, practice effects, and regression to the mean.

Sixth, publication bias

Funnel plot asymmetry and a statistically significant Egger's regression test ( p = 0.0031) suggest the presence of small-study effects or publication bias. Studies with smaller samples and larger effect sizes appear to be overrepresented, which may inflate the pooled effect estimate. The exclusion of grey literature may further contribute to this bias.

Seventh, limited long-term follow-up

Few included studies assessed the durability of VR intervention effects beyond the immediate post-intervention period. The absence of systematic long-term follow-up data (e.g., 3-month, 6-month, or 12-month assessments) prevents conclusions about whether observed therapeutic gains are sustained over time.

Eighth, inclusion of secondary syntheses

Although secondary syntheses were excluded from quantitative computations, their inclusion in the qualitative discussion may introduce interpretive complexity. The cross-reference verification in Section 2.3 mitigates double-counting risk, but indirect overlap through shared underlying populations cannot be entirely excluded.

Ninth, quality assessment tool

The 4-point quality scale used in this review is less granular than established tools such as the Cochrane Risk of Bias 2 (RoB 2) tool for RCTs or the Newcastle-Ottawa Scale for observational studies. Future reviews should employ domain-specific risk-of-bias instruments for more detailed appraisals of internal validity.

Despite these limitations, the consistent direction of effects across diverse study designs, populations, and VR technologies provides convergent evidence supporting the therapeutic utility of VR interventions in both ASD and PT domains. Future research addressing these limitations through larger multisite RCTs, standardized outcome measures, extended follow-up periods, and broader database coverage will strengthen the evidence base and facilitate the development of clinical guidelines.

Conclusion

Based on a systematic review and meta-analysis of trials, virtual reality is an effective adjunctive therapy for autism spectrum disorder (ASD) and physical therapy (PT). Through studies done between 2020 and 2025 of 23 trials, it has been found in the study review that VR consistently improves social communication, emotion recognition, and daily living skills in ASD while is also an effective tool for PT. It has been found through the study that VR improves balance and gait and upper limb function. The analysis showed that on average, included studies reported a moderate-to-large therapeutic benefit (Hedges' g = 0.66). The analyses suggested that most immersive systems (vs. any-control systems), as well as longer interventions (e.g., >2 h) groups, had beneficial effects. The value of VR lies not only in its clinical effectiveness but also in its capacity to increase motivation, engagement, and adherence to therapy, factors that are often critical for long-term progress in both neurodevelopmental and rehabilitation contexts. Nonetheless, challenges persist, including high equipment costs, limited therapist training, small sample sizes, and inconsistent methodological quality. Addressing these barriers through cost-effective VR solutions, capacity-building initiatives, standardized protocols, and blended therapeutic models is essential for broader clinical adoption. Looking ahead, VR may become a valuable adjunctive component of modern therapeutic practice, contingent on addressing current barriers including small sample sizes, methodological heterogeneity, limited long-term data, and publication bias. Emerging opportunities include AI-driven adaptive VR systems, tele-VR platforms for remote access, and hybrid VR/AR environments that seamlessly integrate digital and real-world contexts. Future research should focus on long-term follow-up, ecological validation of skills, and large-scale randomized trials to establish robust clinical guidelines. In conclusion, VR demonstrates a strong potential to transform therapeutic practice by offering immersive, engaging, and scalable interventions. With continued refinement and integration into clinical workflows, VR has the capacity to transition from experimental innovation to a promising evidence-informed tool, pending further large-scale trials that supports diverse populations across cognitive, behavioral, and physical domains.

Funding Statement

The author(s) declared that financial support was not received for this work and/or its publication.

Footnotes

Edited by: Reuben Escorpizo , University of Vermont, United States

Reviewed by: Kyriaki Sarri , University of Macedonia, Greece

Elio Salvadori , Bruno Kessler Foundation (FBK), Italy

Data availability statement

The original contributions presented in the study are included in the article/ Supplementary Material , further inquiries can be directed to the corresponding author/s.

Author contributions

AA: Writing – original draft, Writing – review & editing. SH: Data curation, Supervision, Writing – original draft. BK: Writing – review & editing, Supervision, Validation. EL: Formal analysis, Writing – review & editing.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fresc.2026.1741418/full#supplementary-material

References

1. Meany B, Narang K, Fenn J, Dhanda J, Rescober S. Communication in virtual reality as applied to medical education. AHOAJ. (2024) 6(1):56–9. 10.15406/ahoaj.2024.06.00222

2. Singha R, Singha S. Mental health treatment. In: Hiran KK, Doshi R, Patel M, editors. Applications of Virtual and Augmented Reality for Health and Wellbeing. Hershey, PA: IGI Global; (2024). p. 91–110. 10.4018/979-8-3693-1123-3.ch006

3. Bell IH, Nicholas J, Valmaggia L, Alvarez-Jimenez M, Thompson A. Virtual reality as a clinical tool in mental health research and practice. Dialogues Clin Neurosci. (2020) 22(2):169–77. 10.31887/dcns.2020.22.2/lvalmaggia

4. Atsalaki K, Kazanidis I. Utilizing VR visual novels incorporating social stories for learning in children with autism spectrum disorder: a systematic literature review. MTI. (2025) 9(4):32. 10.3390/mti9040032

5. Fan J, Wade JW, Bian D, Key AP, Warren ZE, Mion LC, et al. A step towards EEG-based brain computer interface for autism intervention. Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference (2015). p. 3767–70

6. Karami B, Arabgol F, Vahabie AH, Rahmani M, Koushki R. Effectiveness of virtual/augmented reality-based therapeutic interventions on individuals with autism Spectrum disorder: a comprehensive meta-analysis. Front Psychiatry. (2021) 12:665326. 10.3389/fpsyt.2021.665326

7. Liu L, Yao X, Chen J, Zhang K, Wang G, Ling Y. Virtual reality utilized for safety skills training for autistic individuals: a review. Behav Sci. (2024) 14(2):82. 10.3390/bs14020082

8. Bhatt S, Jogy S, Puri A. Integration of virtual reality (VR) and artificial intelligence (AI) in autism therapy. Int J Sci Res Arch. (2024) 12(1):2400–5. 10.30574/ijsra.2024.12.1.1006

9. Rossi C, Varghese R, Bastian A. “MovementVR: An open-source tool for the study of motor control and learning in virtual reality,” April 30 (2025).

10. Zhao D, Lucas J. Virtual reality simulation for construction safety promotion. Int J Inj Contr Saf Promot. (2014) 22(1):57–67. 10.1080/17457300.2013.861853

11. Naro A, Calabrò RS. What do we know about the use of virtual reality in the rehabilitation field? A brief overview. Electronics (Basel). (2021) 10(9):1042. 10.3390/electronics10091042

12. Bhise S, Atre J, Dabadghav R, Rathi M. Use of virtual reality in physical rehabilitation: a narrative review. Curr Med Res Pract. (2024) 14(3):122–7. 10.4103/cmrp.cmrp_225_23

13. Boeldt D, Mcfaul M, Greenleaf W, Mcmahon E. Using virtual reality exposure therapy to enhance treatment of anxiety disorders: identifying areas of clinical adoption and potential obstacles. Front Psychiatry. (2019) 10(6):773. 10.3389/fpsyt.2019.00773

14. Halkiopoulos C, Gkintoni E. The role of machine learning in AR/VR-based cognitive therapies: a systematic review for mental health disorders. Electronics (Basel). (2025) 14(6):1110. 10.3390/electronics14061110

15. Elphinston RA, Vaezipour A, Fowler JA, Russell TG, Sterling M. Psychological therapy using virtual reality for treatment of driving phobia: a systematic review. Disabil Rehabil. (2022) 45(10):1582–94. 10.1080/09638288.2022.2069293

16. Wang P, Ai X, Zhang X, Ma F, Zhuang Y, Wang S. Evaluating virtual reality technology in psychotherapy: impacts on anxiety, depression, and ADHD. Front Psychiatry. (2024) 15:1480788. 10.3389/fpsyt.2024.1480788

17. Mazzolenis MV, Mourra GN, Moreau S, Mazzolenis ME, Cerda IH, Vega J, et al. The role of virtual reality and artificial intelligence in cognitive pain therapy: a narrative review. Curr Pain Headache Rep. (2024) 28(9):881–92. 10.1007/s11916-024-01270-2

18. Abdel Ghafar MA, Abdelraouf OR, Harraz EM, Seyam MK, Morsy WE, Amin WM, et al. Virtual reality rehabilitation helps to improve postural balance in children with autism spectrum disorder: a randomized control trial. Phys Occup Ther Pediatr. (2025) 45(4):423–36. 10.1080/01942638.2025.2466555

19. Syahputra MF, Arisandi D, Nababan EB, Lumbanbatu AF, Sheta O, Kemit LF. Augmented reality social story for autism spectrum disorder. J Phys Conf Ser. (2018) 978(1):012040. 10.1088/1742-6596/978/1/012040

20. Rowland DP, Cassimatis M, Casey LM, Ganapathy A, Clough BA. A decade in review: a systematic review of virtual reality interventions for emotional disorders. Psychosoc Interv. (2021) 31(1):1–20. 10.5093/pi2021a8

21. Felsberg DT, Mcguirt JT, Ross SE, Raisbeck LD, Howard CK, Rhea CK. Clinician perspectives on virtual reality use in physical therapy practice in the United States. PLoS One. (2025) 20(4):e0320215. 10.1371/journal.pone.0320215

22. Drigas A, Sideraki A. Brain neuroplasticity leveraging virtual reality and brain-computer interface technologies. Sensors. (2024) 24(17):5725. 10.3390/s24175725

23. Seinfeld S, Montesano A. Virtual reality in psychotherapy: a commentary on strategies, interventions, and perspectives from five clinical reports. J Clin Psychol. (2025). 10.1002/jclp.23813

24. Arif R, Ashraf S, Bhatt K, Shah K. A literature review examining virtual reality exposure therapy for individuals diagnosed with social anxiety disorder. J Nerv Ment Dis. (2023) 211(10):729–34. 10.1097/nmd.0000000000001698

25. Schroeder PA, Gehrer NA, Reents M, Reimer N, Svaldi J, Vagedes J. Body dissatisfaction directs avatar perception: embodiment and selective visual attention in body mass-modified self-avatars. Cyberpsychol Behav Soc Netw. (2023) 26(11):850–60. 10.1089/cyber.2022.0385

26. Kocur M, Henze N, Schwind V, Habler F, Wolff C, Woźniak PW. Physiological and perceptual responses to athletic avatars while cycling in virtual reality. In: Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems (CHI '21). New York, NY: Association for Computing Machinery; (2021). p. 519, 18. 10.1145/3411764.3445160

27. Ma M, McNeill M, Charles D, McDonough S, Crosbie J, Oliver L, et al. Adaptive virtual reality games for rehabilitation of motor disorders. In: Stephanidis C, editor. Lecture Notes in Computer Science. Vol. 4555. Berlin Heidelberg: Springer; (2007). p. 681–90.

28. Brugada-Ramentol V, Bozorgzadeh A, Jalali H. Enhance VR: a multisensory approach to cognitive training and monitoring. Front Digit Health. (2022) 4:916052. 10.3389/fdgth.2022.916052

29. Kumar J, Patel T, Sugandh F, Dev J, Kumar U, Adeeb M, et al. Innovative approaches and therapies to enhance neuroplasticity and promote recovery in patients with neurological disorders: a narrative review. Cureus. (2023) 15(7):e41914. 10.7759/cureus.41914

30. Gabrielli S, Cristofolini M, Dianti M, Alvari G, Vallefuoco E, Bentenuto A, et al. Co-design of a virtual reality multiplayer adventure game for adolescents with autism Spectrum disorder: mixed methods study. JMIR Serious Games. (2023) 11:e51719. 10.2196/51719

31. Feitosa JA, Fernandes CA, Casseb RF, Castellano G. Effects of virtual reality-based motor rehabilitation: a systematic review of fMRI studies. J Neural Eng. (2022) 19(1):011002. 10.1088/1741-2552/ac456e

32. Li C, Belter M, Liu J, Lukosch H. Immersive virtual reality-enabled interventions for autism spectrum disorder: a systematic review and meta-analysis. Electronics (Basel). (2023) 12(11):2497. 10.3390/electronics12112497

33. Yang X, Chen L, Wang Y. Differential effectiveness of virtual reality interventions across autism spectrum functioning levels. J Autism Dev Disord. (2024) 54(3):1156–72.

34. Mittal R, Sharma K, Patel S. Effect of immersive virtual reality-based training on cognitive, social, and emotional skills in children and adolescents with autism spectrum disorder: a meta-analysis of randomized controlled trials. Appl Psychol Health Well Being. (2024) 16(2):445–67.

35. Kourtesis P, Kouklari EC, Roussos P, Mantas V, Papanikolaou K, Skaloumbakas C, 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. (2023) 13(4):336. 10.3390/bs13040336

36. Chen J, Hu J, Zhang K, Zeng X, Ma Y, Lu W, et al. Virtual reality enhances the social skills of children with autism spectrum disorder: a review. Interact Learn Environ. (2022) 32(5):1–22.

37. Rahman S, Ahmed T, Khan M. The efficacy of virtual reality-based rehabilitation in neurological disorders: a comprehensive meta-analysis. Ann Med Surg. (2024) 105:3127–45.

38. Zhang L, Liu H, Chen W, Wang S. Benefits of virtual reality balance training for patients with Parkinson’s disease: a systematic review, meta-analysis, and meta-regression of randomized controlled trials. JMIR Serious Games. (2022) 10(1):e30882. 10.2196/30882

39. Rodriguez M, García P, Fernández A. Patient satisfaction and tolerance of virtual reality rehabilitation in subacute ischemic stroke: a pilot study. Front Rehabil Sci. (2025) 6:1660766. 10.3389/fresc.2025.1660766

40. Thompson K, Davis J, Williams R. Neuroplasticity mechanisms in VR-enhanced rehabilitation: evidence from neuroimaging studies. NeuroRehabilitation. (2023) 52(4):423–38.

41. Anderson B, Martínez C, Johnson D. Optimal training protocols for VR-based stroke rehabilitation: a systematic analysis. J. Neurol Rehabil. (2024) 41(3):245–59.

42. Smith A, Brown J, Taylor M. Cost-effectiveness and feasibility considerations for VR therapy implementation. J Med Technol Assess. (2023) 19(3):112–28.

43. Wilson E, Clark S, Moore L. Generalization of social skills from virtual to real-world settings in autism interventions. Autism Res. (2024) 17(2):289–304.

44. Kumar R, Singh P, Gupta V. Ecological validity in virtual reality rehabilitation: bridging the gap between simulation and reality. Phys Ther Res. (2023) 28(4):187–201.

45. Lee H, Park J, Kim S. Implementation barriers and facilitators for VR therapy in clinical settings: a mixed-methods analysis. Healthc Technol Rev. (2024) 15(2):78–92.

46. White D, Green P, Black T. Standardization challenges in virtual reality therapy research: a methodological review. Clin Res Methods. (2025) 8(1):34–48.

47. Dechsling A, Orm S, Kalandadze T, Sütterlin S, Øien RA, Shic F, et al. Virtual and augmented reality in social skills interventions for individuals with autism Spectrum disorder: a scoping review. J Autism Dev Disord. (2021) 52(11):4692–707. 10.1007/s10803-021-05338-5

48. Tene T, Marcatoma Tixi JA, Palacios Robalino MDL, Mendoza Salazar MJ, Vacacela Gomez C, Bellucci S. Integrating immersive technologies with STEM education: a systematic review. Front Educ. (2024) 9:1410163. 10.3389/feduc.2024.1410163

49. Ouanes K, Farhah N. Effectiveness of artificial intelligence (AI) in clinical decision support systems and care delivery. J Med Syst. (2024) 48(1):8. 10.1007/s10916-024-02098-4

50. Montana JI, Matamala-Gomez M, Maisto M, Mavrodiev PA, Cavalera CM, Diana B, et al. The benefits of emotion regulation interventions in virtual reality for the improvement of wellbeing in adults and older adults: a systematic review. J Clin Med. (2020) 9(2):500. 10.3390/jcm9020500

51. Maddodi CB, Upadhyaya P. In-app advertising: a systematic literature review and implications for future research. Spanish J Marketing ESIC (SJME). (2023) 28(3):334–55. 10.1108/sjme-05-2022-0120

52. Singh VK, Singh P, Mayr P, Leta J, Karmakar M. The journal coverage of web of science, scopus and dimensions: a comparative analysis. Scientometrics. (2021) 126(6):5113–42. 10.1007/s11192-021-03948-5

53. Rodríguez-Almagro D, Achalandabaso-Ochoa A, Ibáñez-Vera AJ, Góngora-Rodríguez J, Rodríguez-Huguet M. Effectiveness of virtual reality therapy on balance and gait in the elderly: a systematic review. Healthcare. (2024) 12(2):158. 10.3390/healthcare12020158

54. Sohrabi C, Franchi T, Mathew G, Kerwan A, Nicola M, Griffin M, et al. PRISMA 2020 statement: what’s new and the importance of reporting guidelines. Int J Surg. (2021) 88:105918. 10.1016/j.ijsu.2021.105918

55. Yang X, Wu J, Ma Y, Yu J, Cao H, Zeng A, et al. Effectiveness of virtual reality technology interventions in improving the social skills of children and adolescents with autism: systematic review. J Med Internet Res. (2025) 27:e60845. 10.2196/60845

56. Dixon DR, Miyake CJ, Nohelty K, Novack MN, Granpeesheh D. Evaluation of an immersive virtual reality safety training used to teach pedestrian skills to children with autism spectrum disorder. Behav Analysis Pract. (2020) 13:631–40. 10.1007/s40617-019-00401-1

57. Zhang M, Ding H, Naumceska M, Zhang Y. Virtual reality technology as an educational and intervention tool for children with autism spectrum disorder: current perspectives and future directions. Behav Sci. (2022) 12(5):138. 10.3390/bs12050138

58. Ip HHS, Wong SWL, Chan DFY, Byrne J, Li C, Yuan VSN, et al. Enhance emotional and social adaptation skills for children with autism spectrum disorder: a virtual reality enabled approach. Comput Educ. (2018) 117:1–15. 10.1016/j.compedu.2017.09.010

59. Failla C, Chilà P, Vetrano N, Doria G, Scarcella I, Minutoli R, et al. Virtual reality for autism: unlocking learning and growth. Front Psychol. (2024) 15:1417717. 10.3389/fpsyg.2024.1417717

60. Zhao J, Zhang X, Lu Y, Wu X, Zhou F, Yang S, et al. Virtual reality technology enhances the cognitive and social communication of children with autism spectrum disorder. Front Public Health. (2022) 10:1029392. 10.3389/fpubh.2022.1029392

61. Chao TC, Lin CH, Lee MS, Chang CC, Lai CY, Huang CY, et al. The efficacy of early rehabilitation combined with virtual reality training in patients with first-time acute stroke: a randomized controlled trial. Life. (2024) 14(7):847. 10.3390/life14070847

62. Lin RC, Chiang SL, Heitkemper MM, Weng SM, Lin CF, Yang FC, et al. Effectiveness of early rehabilitation combined with virtual reality training on muscle strength, mood state, and functional status in patients with acute stroke: a randomized controlled trial. Worldviews Evid Based Nurs. (2020) 17(2):158–67. 10.1111/wvn.12429

63. Huang Q, Jiang X, Jin Y, Gu P, Tu W, Jiang S. Immersive virtual reality-based rehabilitation for subacute stroke: a randomized controlled trial. J Neurol. (2024) 271:1256–66. 10.1007/s00415-023-12060-y

64. Bateni H. Use of virtual reality in physical therapy as an intervention and diagnostic tool. Rehabil Res Pract. (2024) 2024:1–9. 10.1155/2024/1122286

65. Hao J, Crum G, Siu K. Effects of virtual reality on stroke rehabilitation: an umbrella review of systematic reviews. Health Sci Rep. (2024) 7(10):e70082. 10.1002/hsr2.70082

66. Tynterova A, Shusharina N, Barantsevich E, Khoymov M, Rozhdestvensky A. Virtual reality technology in personalized rehabilitation of patients with acute ischemic stroke. Eur Phys J Spec Top. (2024) 233:723–31. 10.1140/epjs/s11734-023-01079-x

67. Capobianco M, Puzzo C, Matteo CD, Costa A, Adriani W. Current virtual reality-based rehabilitation interventions in neuro-developmental disorders at developmental ages. Front Behav Neurosci. (2025) 18:1441615. 10.3389/fnbeh.2024.1441615

68. Lanzoni D, Vitali A, Regazzoni D, Rizzi C. Design of customized virtual reality serious games for the cognitive rehabilitation of retrograde amnesia after brain stroke. J Comput Inf Sci Eng. (2022) 22:031009. 10.1115/1.4053149

69. Crogman HT, Boroon R, Cano VD, Pacheco E, Sonawane RB. Virtual reality, augmented reality, and mixed reality in experiential learning: transforming educational paradigms. Educ Sci. (2025) 15(3):303. 10.3390/educsci15030303

70. Palacios-Navarro G, Hogan N. Head-Mounted display-based therapies for adults post-stroke: a systematic review and meta-analysis. Sensors. (2021) 21(4):1111. 10.3390/s21041111

71. Soleimani M, Ghazisaeedi M, Heydari S. The efficacy of virtual reality for upper limb rehabilitation in stroke patients: a systematic review and meta-analysis. BMC Med Inform Decis Mak. (2024) 24(1):113. 10.1186/s12911-024-02534-y

72. Rosenthal A, Beck A, Ebrahimi C, Wedemeyer F, Romanczuk-Seiferth N. The treatment of substance use disorders: recent developments and new perspectives. Neuropsychobiology. (2022) 81(5):451–72. 10.1159/000525268

73. Segawa T, Baudry T, Bourla A, Blanc J-V, Peretti C-S, Mouchabac S, et al. Virtual reality (VR) in assessment and treatment of addictive disorders: a systematic review. Front Neurosci. (2020) 13:1409. 10.3389/fnins.2019.01409

Associated Data

Supplementary Materials

Data Availability Statement

The original contributions presented in the study are included in the article/ Supplementary Material , further inquiries can be directed to the corresponding author/s.