Immersive Virtual Reality is More Effective Than Non-Immersive Devices for Developing Real-World Skills in People with Intellectual Disability

People with intellectual disability demonstrate persistent challenges around developing life skills. Immersive virtual reality (IVR) is gaining interest as a tool for training life skills as it enables individuals to engage in hands-on learning in a safe, controlled, and repeatable environment. However, there are concerns about the potential drawbacks of IVR, such as cybersickness and practical challenges with using VR equipment, which may hinder its widespread adoption in educational settings. The current study aimed to compare the effectiveness of training in IVR and a non-immersive virtual environment for improving real-world skills in people with intellectual disability.. In the present study, 36 adults (16f, 20m) with intellectual disability were recruited from a disability organisation. Participants completed a real-world assessment of waste management skills before and after training in either the IVR or non-immersive group. Consistent with our hypotheses, the IVR group scored significantly higher in the real-world assessment after virtual training ( d = 1), and at the one-week follow-up ( d = 1.12), compared to the non-immersive group. Further analyses showed that the IVR group, but not the non-immersive group, significantly improved performance in the real-world assessment across timepoints. Our findings indicate that IVR was more effective for improving and retaining real-world waste management skills. This study supports IVR as a viable tool for professionals and caregivers to develop skills for independent living among people with intellectual disability.


Introduction
People with intellectual disability often require intensive support from caregivers to manage activities of daily living [1].This reliance largely stems from challenges in intellectual and adaptive functioning [2].Difficulties in adaptive functioning impact the ability to perform the social, conceptual and practical skills necessary for everyday life [3].Over 60% of people with intellectual disability experience severe or profound challenges in performing life skills [4].
These are the practical skills required to complete everyday tasks without the need for caregiver assistance such as cooking, showering, and cleaning [5].Difficulties in performing life skills prevent people with intellectual disability from living independently, which is associated with decreased feelings of happiness and poorer quality of life [6].Therefore, there is a need to find effective methods for developing life skills in this population.
People with intellectual disability often face challenges in conceptual understanding and working memory, which can limit their ability to process and retain information that is not visually or externally represented [7].Conversely, people with intellectual disability have unique characteristics that make them well-suited for experiential learning [8,9].Experiential learning (or 'learning-by-doing') describes the cyclical process of gaining knowledge through direct and hands-on experience of a task and then applying this knowledge to future problems to consolidate motor learning [10].Methods that utilise hands-on interactions have largely found to be effective [7,9,11].Research shows that experiential learning is more effective for developing skills among individuals with ID compared to passive and didactic methods of learning [12,13].
For example, Scruggs, Mastropieri [13] compared two groups of high school students with intellectual disability who learned science-based skills from either a practical lab-based activity or reading a textbook.They found that the practical activity group performed significantly better than the textbook group in a post-test skill examination.Therefore, there is promising but limited research demonstrating the benefits of experiential learning for improving skills among individuals with intellectual disability.
Opportunities for people with intellectual disability to participate in hands-on learning are often minimal due to concerns surrounding risk of injury, time restraints, and lack of staff training [14,15].Disability support workers report that minimal time, resources, and professional training limit their ability to provide opportunities for in-vivo experiences for their clients [15].While perhaps optimal for skill development, experiential learning also introduces potential hazards.
Certain activities may lead to injuries such as mastering the safe crossing of bustling streets or handling household waste that contains items such as broken glass.This risk is particularly pronounced for people with intellectual disability, who are more likely to incur preventable injuries in comparison to people without disability [16].In response to these challenges, immersive virtual environments emerge as a strategic solution, offering controlled, hands-on, and realistic learning experiences in a safe and controlled environment [17,18].
Immersive virtual reality (IVR) is a simulated digital environment that replicates aspects of the real world or imagines entirely new worlds, designed to completely immerse users' senses and create the sensation of being physically present in that environment [19].This immersive experience is typically facilitated by technologies such as head-mounted displays (HMD) [19].
One of the pivotal advantages of IVR is its capacity to enable the training of realistic gestures and movements within a safe and repeatable setting.This unique attribute establishes HMDs as an ideal instrument for experiential learning, where participants can navigate and interact with lifelike scenarios without real-world risks [20,21].However, despite its promising application, there is a discernible gap in research exploring its efficacy in enhancing life skills among people with intellectual disability.
Michalski et al. [22] published one of the first studies demonstrating the effectiveness of IVR training for improving life skills among individuals with intellectual disability.In their withingroups study, participants were immersed in an HMD and completed a virtual waste management intervention.Performance in an equivalent real-world assessment indicated that participants significantly improved their real-world waste management skills following IVR training, to a moderate-to-large effect (d = 0.68).Encouragingly, the authors also found that skills were retained at a follow-up assessment one week later (d = 0.21) [22] Thus, Michalski et al. [22] provided preliminary support for the effectiveness of IVR for improving life skills among people with intellectual disability.
While preliminary evidence suggests promising outcomes, there are notable drawbacks associated with IVR in vulnerable populations.For instance, adverse effects such as cybersickness pose significant concerns regarding the suitability of IVR as a learning tool for people with intellectual disability [21].In healthy populations, up to 16% of participants drop out of studies due to symptoms of cybersickness from using IVR, including eye strain, dizziness, and nausea [23].Some research suggests that participants with neurodevelopmental disorders (such as Autism Spectrum Disorder) could be more affected by adverse effects than their neurotypical peers [24].The current understanding of the prevalence of these adverse symptoms and their impact on skill acquisition among individuals with intellectual disability remains limited due to insufficient research.Past studies that have assessed IVR-induced cybersickness in populations with neurodevelopmental disorders have yielded substantial variations in prevalence rates, ranging from 13.8% to 83.3% of participants experiencing symptoms [24,25].Thus, perhaps there is a need to compare the effectiveness of IVR to lower-risk alternatives.
An alternative method for providing controlled and safe learning environments, without the risk of adverse side effects, is training using non-immersive virtual environments [21].Nonimmersive virtual environments such as tablets, desktops, or other flat-screen displays allow users to interact with the learning environment using a touchscreen display or via keyboard and mouse [26,27].Several studies have shown that training using non-immersive virtual environments can significantly improve real-world skills from pre-to post-training among individuals with intellectual disability [28,29].Although these devices are more widely adopted in educational settings and do not elicit adverse symptoms [24], they fail to facilitate the same level of realism and naturalness that IVR training provides [21,30].This raises the question of whether training in IVR or a non-immersive virtual environment would be more effective for developing skills in people with intellectual disability.
The current study aimed to compare the effectiveness of training in IVR and a non-immersive virtual environment for improving real-world skills in people with intellectual disability.. Building upon the methods used by Michalski et al. [22], this study assessed waste management skills.The real-world assessment was completed before (pre-test), after (post-test), and one-week after (delayed-test) virtual training in either immersive (HMD) or non-immersive (tablet) virtual environments.It was hypothesised that: (H1) skill performance, measured by total correct disposals in the real-world assessment at post-test, would be significantly higher in the IVR group compared to the non-immersive tablet group; (H2) skill retention, measured by total correct disposals in the real-world assessment at delayed-test, would be significantly higher in the IVR group compared to the non-immersive tablet group.

Method Ethics
This study was granted ethics approval from the University of South Australia Human Research Ethics Committee (Protocol No. 202640).

Participants
Forty-one adults (21 male, 20 female) with an intellectual disability were recruited from a nonprofit disability organisation in South Australia using a convenience sampling method.We recruited participants with any severity of intellectual disability.An a priori power analysis estimated that our sample would need 21 participants in each group for sufficient power (0.80) to detect a large effect (d > 0.8) with α = 0.05 [31].
Participants were eligible to be included in the sample if they successfully passed the virtual tutorial which demonstrated adequate task understanding and motor capability to complete the subsequent virtual training.There were no restrictions on age, comorbid conditions, or motor abilities to be eligible for the study.Exclusion did occur if participants failed the tutorial, scored above 90% in the real-world assessment at pre-test due to insufficient room for improvement, or did not complete all virtual training sessions.
Five of the 41 participants recruited were excluded from the final sample.Reasons included failing the tutorial (n = 1), disinterest (n = 3) and feeling overwhelmed (n = 1).Therefore, 36 participants were included in the sample who were initially assigned to either the IVR group or the non-immersive tablet group using A-B-A-B sequencing.Participant characteristics are reported in Table 1.None or Not Reported 5 5 Note.Descriptive classification of adaptive functioning level included extremely low (<71), low (71-79), below average (80-89), average (90-109), above average (110-119), and high (120 or more) [32].a Adaptive functioning level was measured using the GAC standardised score provided by the ABAS-III [32].b No significant difference (p = .424)in GAC standardised score between both groups.

Design
A between-group longitudinal design was used to test the research hypotheses.The two experimental groups included the IVR group and the non-immersive tablet group.
Both groups completed the real-world assessment, the virtual tutorial, and 12 virtual training sessions across four data collection phases, as shown in Figure 1.The predictor variables for the main analysis included the virtual environment (IVR or non-immersive tablet) and real-world assessment timepoint (pre, post and delayed).The outcome variable was performance in the Real-World Assessment, measured by total correct disposals.

Questionnaires and Assessments
Adaptive Behaviour Assessment System Third Edition (ABAS-III).The standardised ABAS-III was used to assess and classify participants' level of adaptive functioning (extremely low to high).[32].The adult form of the ABAS-III was completed by organisation staff (n=2) who frequently interacted with the participant being assessed.Staff provided scores of observed ability between zero (not able) to three (always or almost always able) on a list of 11 skill areas, such as verbal communication.Scores were combined to form the General Adaptive Composite (GAC) standarddised score (ranging from 40 to 160) [32].
Cybersickness Questionnaire.Cybersickness was assessed using a questionnaire that read, "I felt dizzy or sick."Participants were required to select one of three potential answers "yes", "not sure", "no" with matching emoticons to indicate level of sickness.This questionnaire with simplified graphics was developed by Michalski, Szpak [33] to assess symptoms of cybersickness from IVR among individuals with intellectual disability.Real-World Assessment.All 24 items listed in Table 2 were randomly spread out on a table placed in front of three labelled bin options, as shown in Figure 2. Participants were instructed to dispose each item one-by-one into the bin that they thought correctly matched the item until all items were disposed.A tally of total correct disposals was recorded and ranged from 0 (no items in correct bins) to 24 (all items in correct bins).

Figure 2
Layout of the Real-World Assessment

Software and Applications
The virtual tutorial and training applications were custom-designed using Unity 3D game engine (Unity Technologies; http://unity3d.com)and Javascript (Oracle; https://developer.oracle.com/languages/javascript.html) for the Quest 2 and Lenovo tablet, respectively.
Virtual Tutorial.The tutorial displayed nine objects of different shapes one-by-one on a table adjacent to three bin options with matching labels (see Figure 3).Participants had to correctly dispose eight of the nine objects in the matching bin to pass the tutorial.In the IVR group, virtual objects could be grabbed and disposed by holding and releasing the trigger button on the controller which required movement of the whole body to transport objects from the table to the bins.In the tablet group, grabbing and disposing the objects required tapping and dragging the touchscreen using a fingertip.

Figure 3
Virtual Tutorial in Both Virtual Environments Note.A screenshot of the tutorial is shown from within the IVR head-mounted display (left) and non-immersive tablet (right).

Virtual Training. The virtual training task used in each training session replicated the
Real-World Assessment in all components besides the number of items.In virtual training, only 18 items were presented (see Table 1).The virtual interactions in the training task were identical to the virtual tutorial, and feedback on performance was provided (see Figure 4).tablet.This feedback appeared after each disposal and indicated to the participant whether the disposal was correct or incorrect.
The frequency of each item appearing varied in each virtual training session and was based on the principles of spaced repetition using the Leitner System [34].The Leitner system is one of the earliest and most widely used models of spaced repetition which illustrates the optimal frequency of repetition using a multi-box arrangement (see Figure 5) (Beursgens, 2022; Leitner, 1972).Past research has shown the enhanced effectiveness of spaced repetition for improving learning outcomes compared to non-spaced learning schedules among individuals with developmental learning disability [35][36][37][38].

Procedure
A simplified informed consent form was signed by each participant and a staff member also signed to confirm the mental capacity of the participant to consent.

Real-World Assessment (Pre-test)
The real-world assessment was set up in a 10 m x 5 m room, shown in Figure 2.Each participant was allowed one disposal attempt per item and no further instruction or feedback was provided until all items in the task were disposed.

Virtual Tutorial
A researcher then assisted the participants to either fit on the Quest 2 HMD, or setup the Lenovo M10 device, depending on group allocation.Participants were assessed using the cybersickness questionnaire after completing the tutorial.For participants who reported feeling cybersickness, further questioning occurred to confirm their desire to continue the study.Upon passing, participants immediately began the first virtual training session.
To successfully pass the VR tutorial, participants were required to correctly place at least eight out of nine items into the designated bins.Those who failed to meet this criterion after a maximum of three attempts were excluded from the study.The tutorial was deemed incomplete if participants remained unresponsive in VR, failed to follow the task instructions, or did not press the necessary button at the correct time, despite several reminders.Throughout the tutorial, a researcher was on hand to offer reminders and assistance.Participants who successfully completed the VR tutorial advanced to the next phase, the VR training.

Virtual Training
Participants completed 12 virtual training sessions in total.Each virtual training session ended once all items were correctly disposed into the bins.Participants completed four training sessions per day.After every second training session, participants were assessed using the cybersickness questionnaire.Although the spacing between training sessions varied due to staff availability, the aim was to complete four virtual training sessions per week.

Real-World Assessment (Post and delayed-test)
Immediately following the last virtual training session, participants completed the real-world assessment again (post-test).One week later, participants completed the same real-world assessment (delayed-test).

Statistical Analyses
The data was screened for normality and appropriate diagnostic tests were run before all inferential tests.
A linear mixed-effects regression was run to examine performance differences in the Real-World Assessment between both groups, using the GAMj package in Jamovi (Gallucci, 2019).Total correct disposals in the real-world assessment (counts ranging from 0-24) was entered as the discrete outcome variable; virtual environment group with two levels (IVR and non-immersive tablet), real-world assessment timepoint with three levels (pre, post and delayed), and real-world task timepoint by virtual environment group interaction were entered as fixed effects; and participants were included as the random intercept effect.Significant main and interaction effects (significance cutoff p < 0.05) were followed up with Bonferroni-Holm corrected post hoc comparisons to test H1 and H2.Cohen's d effect sizes and confidence intervals were calculated using raw data and the Esci package in Jamovi [39].

Results
Five of the 41 recruited participants were excluded from the sample, as seen in the adapted CONSORT flow diagram in Figure 6 [40].

Figure 6
Participant

Real-World Assessment Descriptives
The mean duration in days between the real-world assessment at pre-test and post-test was 17 days (SD = 19) for the IVR group and 12.72 days (SD = 10.22) for the non-immersive tablet group.Between post and delayed-test, the mean duration was 7.66 days (SD = 2.96) for the IVR group and 9.5 days (SD = 7.12) for the non-immersive tablet group.A pair of independent samples t-tests revealed no significant between-group differences in days between the real-world assessment at pre to post-test (p = .406,d = -0.27)or post to delayed-test (p = .318,d = 0.33).

Real-World Assessment Analysis
Diagnostic tests did not reveal any assumption violations for the regression.
A linear mixed-effects regression revealed significant main effects of virtual environment group (F1,34.0= 8.93, p = .005),real-world assessment timepoint (F2,67.1 = 12.34, p < .001),and a significant interaction effect between both fixed factors (F2,67.1 = 4.56, p = .014)on total correct disposals in the real-world assessment.Table 3 summarises the between and within-group post hoc comparisons.H1 was supported by the IVR group showing significantly more correct disposals in the realworld assessment at post-test compared to the non-immersive tablet group, as seen in Table 3.
H2 was supported by the IVR group showing significantly more correct disposals in the realworld assessment at delayed-test compared to the non-immersive tablet group.This effect was large as indicated by d = 1.12, 95% CI [0.45, 1.94].There was also no significant between-group difference (p = .288,d = 0.61) at pre-test, as shown in Figure 7. Significance testing completed using Bonferroni-holm corrected post hoc comparisons.Nonsignificant comparisons using an alpha cutoff value of p < .05were labelled by ns.Significant comparisons were labelled *p < .05,**p < .01.

Adaptive Functioning
Multiple Pearson correlations revealed no significant relationship between adaptive functioning level and change in real-world assessment performance in the IVR group.In the tablet group, a significant correlation between adaptive functioning and change in correct disposals in the realworld assessment was found between pre to post-test.No other significant associations were found, as seen in Table 4. Note.Task performance = correct disposals, Adaptive Functioning = General Adaptive Composite, *p < .05.

Cybersickness
Cybersickness was assessed across all participants after each virtual tutorial and every second training session, totalling 252 assessments.Three (16.7%) of the participants in the IVR group answered "not sure" at least once and one participant (5.56%) answered "yes" to feeling cybersickness, and reported feeling "dizzy."However, symptoms subsided within a few minutes and all four participants completed the training sessions.The remaining 14 participants in the IVR group, and all participants in the non-immersive tablet group, reported no symptoms.

Discussion
This study contributes to the existing literature by examining the effectiveness of immersive versus non-immersive virtual environments [41], specifically in enhancing real-world life skills among people with intellectual disability.Identifying effective learning methods for safely developing life skills is crucial for enabling more individuals with intellectual disability to progress toward independent living, which is associated with better psychosocial outcomes [6].
Consistent with our hypotheses, our findings showed that the IVR group scored significantly more correct disposals in the real-world assessment at post-test and delayed-test, compared to the non-immersive group.The effect sizes for the between-group comparisons at both timepoints were large.Further analyses also showed that the IVR group, but not the non-immersive group, significantly increased correct disposals in the real-world assessment across timepoints.These findings indicate that training in IVR was more effective than the non-immersive virtual environment for improving and retaining waste management skills among people with intellectual disability.
Our novel findings support and extend past literature that has shown the benefits of experiential learning for developing real-world skills among individuals with intellectual disability [7,9,[11][12][13].The IVR training used in the present study closely replicated the three-dimensional environment and physical movements required in the real-world assessment.In contrast, training in the non-immersive virtual environment required users to touch and drag a fingertip on the tablet.Our findings indicate that the opportunity for realistic and hands-on practice in IVR training facilitated better skill learning among people with intellectual disability, suggesting that the benefits of experiential learning observed in real-world environments also apply in virtual environments.However, it is also possible that extraneous factors such as engagement, view of environmental distractions, sense of presence (feeling of 'being there'), and motivation may have impacted the observed differences in real-world performance [42,43].For example, IVR is reported to be more engaging and motivating than non-immersive virtual environments for individuals with learning difficulties [43,44], and this is associated with better skill learning [43].Although research has shown the benefits of IVR while controlling for factors such as motivation [20], future studies are encouraged to replicate these findings in people with intellectual disability to more accurately assess the role of experiential learning in improving skills using IVR.
Further analyses revealed that the IVR group significantly increased their performance in the real-world assessment across timepoints, whereas the non-immersive tablet group showed no significant changes.This is consistent with the singular past study that investigated the effectiveness of IVR for improving real-world waste management skills among people with intellectual disability [22].However, Michalski et al. [22] reported medium sized effects of IVR training in their study, whereas the current study identified large effects.One interesting aspect of the current study's findings relate to observed differences across the post-test and delayed-test assessments.The IVR group exhibited not only an immediate improvement in real-world waste management skills from pre-to post-training but also retained these skills effectively at the oneweek follow-up.This contrasts with the non-immersive tablet group, which showed minimal improvement over time.The lack of change in the non-immersive group could be attributed to the less engaging nature of the training medium for learning.These observations are pivotal as they underscore the importance of the immersive quality of the training environment in enhancing not only the acquisition but also the persistence of new skills in people with intellectual disability.
One major difference between the two studies was the type of learning paradigm implemented in the design of the virtual training.Michalski et al. [22] used a simple paradigm of learning via repetition, whereby the time-period between correctly disposed items appearing in the virtual training was fixed (one-session intervals).In contrast, the current study implemented the principles of spaced repetition by designing the virtual training using the Leitner system [34].
This meant that correctly disposed items were shown at increasing intervals across the virtual training sessions in our Foundational research in psychology states that repetitions spaced across larger periods of time result in enhanced memory and learning compared to repetitions massed closer together, known as the spacing effect [45,46].Research also shows that this spacing effect is observed in populations with learning difficulties, despite deficits in intellectual functioning [35][36][37].Therefore, the larger effects found in our study may suggest that spaced repetition is effective for enhancing real-world skill development from IVR training in people with intellectual disability.
This study is one of the first to compare the effectiveness of an immersive and non-immersive virtual environment for improving real-world life skills among individuals with intellectual disability.Our findings support past literature that has demonstrated the benefits of experiential learning for developing skills among individuals with intellectual disability.The benefits identified in this study align with the task-technology fit model, which posits that the effectiveness of the technology is maximised when the capabilities of the technology align closely with the demands of the task it is intended to support [47,48].For learners with intellectual disability, the immersive and interactive features of IVR are particularly suited to enhance experiential learning, which is crucial for engagement and retention of information.The IVR application engaged users through spatial awareness and motor coordination that is more challenging to implement when using traditional, non-immersive technologies.In our study, the alignment between the IVR's capabilities and the demands of the task demonstrated a good tasktechnology fit compared to tablet-based learning.The importance of such an alignment was convincingly highlighted in a meta-analysis by Howard et al. [49], which identified tasktechnology fit as a significant moderator influencing the effectiveness of VR training programs.responses during the assessments.Finally, we were unable to control for length of the training sessions given the spaced repetition approach utilised, and interval between assessments.
A key consideration for future research is the number and length of training sessions required for effective skill learning in IVR [25,55].Several factors may impact the optimal amount of training including task complexity, capability of the population, and adverse symptoms such as cybersickness.To date, past literature shows a clear lack of consistency and explicit reporting of training duration in IVR research [55].In a recent review of 330 studies using typically developed samples, almost half did not provide information on the duration of IVR training (including number and length of sessions) [55].Their review also showed that there was a very desktops are widely used for their accessibility and lower risk of cybersickness, IVR's distinct advantages in enhancing engagement and providing a safe, controlled learning environment make it a promising tool.The efficacy of IVR hinges on overcoming challenges related to cybersickness and improving accessibility.However, it is important to recognise that IVR is not a universal solution.The appropriateness of technology must be carefully matched to the specific needs and contexts of the users.In some cases, simpler or more traditional methods may be equally effective and less burdensome for implementation.

Figure 1 Study
Figure 1

Figure 4 Virtual
Figure 4

Figure 5 Presentation
Figure 5

Figure 7 Correct
Figure 7

Table 1
Participant Demographics in Both Virtual Environment Groups

Table 2
Quest 2. The Quest 2 (developed by Meta; https://www.meta.com/au/quest/)HMD was used to display the virtual environment in the IVR group.The Quest 2 has a refresh rate of Items in the Real-World Assessment and Virtual Training Split by Bin Type 90 Hz and a resolution of 1832x1920 per eye.This device uses inside-out motion tracking to detect direction and positioning of the user's body.One wireless Quest 2 controller held in the dominant hand was used to enhance tracking for movements of the arm and hand.Lenovo M10 Tablet.The Lenovo M10 (developed by Lenovo; https://www.lenovo.com/au/en/p/tablets/android-tablets/tab-series/lenovo-tab-m10/) was used to display the virtual environment in the non-immersive tablet group.The M10 has a 10.1" LED touchscreen display and Android 10 operating system.Items.Twenty-four household items were included in the real-world assessment.The items were classified as 'recycling,' 'general' or 'garden and food organics,' based on South Australian guidelines.Eighteen (six per bin) of those items were digitally replicated and included in the virtual training sessions, as summarised in Table2.Note.Waste classification was based on South Australian Council guidelines.
Flow DiagramAll 36 included participants completed the real-world assessment at pre-test and post-test, and completed all virtual training sessions.One participant in the IVR group did not complete the real-world assessment at delayed-test due to lost contact.No other missing entries or extreme outliers were identified.
group.The mean difference in days to complete training between both groups (M = 12.45, SD = 10.5) was due to random variation in participants' and staff availability.An independent samples t-test revealed that this difference was statistically significant (t = 2.06 , p = .046).

Table 3
Differences in Correct Disposals in the Real-World Assessment Between and Within-Groups

Table 4
Correlations Between Adaptive Functioning and Change in Real-World Assessment large range in reported IVR training session lengths, ranging from 2 minutes up to 10 hours.A similar lack of consistency has been observed among studies in this population.More recently, however, Smith, Van Ryzin [56] explored the dosing efficiency and efficacy of virtual interview training for transition-age youth with disabilities, finding that a targeted number of completed training sessions could significantly enhance job interview skills and subsequent employment outcomes.Further, in Michalski et al. (2023), VR training sessions varied in number until participants reached a learning target or hit a maximum of ten sessions, with a median of 8.5 sessions.This flexible approach tailored the training dosage based on individual progress, effectively accommodating each participant's unique learning pace and needs.This method prevents premature termination or unnecessary prolonging of training.Specific attention to training dose is essential when designing IVR interventions, particularly for people with intellectual disability.IVR shows considerable potential for broad application beyond its current use in building life skills among people with intellectual disability.Its ability to simulate realistic, interactive environments makes it adaptable for training across various contexts, such as vocational tasks, social interactions, and emergency responses.This adaptability could extend its benefits to broader populations, including those with other neurodevelopmental disabilities or in professional training settings.Although non-immersive virtual environments like tablets and