In the ever-evolving landscape of adult education and training, the integration of virtual reality (VR) has emerged as a revolutionary paradigm, offering immersive and interactive experiences that transcend traditional learning methods. In this article, we will review and summarize utilized hardware tools, technical skill acquisition, and technical skill generalization with healthy adult individuals in the context of VR; with a specific focus on medical settings and occupations where these skills are of paramount importance for ensuring safety and optimal performance. The scope of this review is brief and narrow to focus on recent studies that involve fully immersive VR training experiences. As the technology and training applications are rapidly advancing, it is critical to examine the comparison of these new procedures to business-as-usual (BAU) analogue training.

Researchers followed PRISMA checklist guidelines (Page et al., 2021) when conducting this review. Researchers first began by using ProQuest to search for articles on immersive virtual reality training. The search was conducted in November 2022 and included all published articles up until that date. The ProQuest database itself included the following databases in all search attempts conducted: APA PsycINFO, Coronavirus Research Database, Education Collection, Psychology Database, PTSDpubs, and Publicly Available Content Database. Researchers selected ProQuest as it was made available through professional organizations to which researchers belonged, as well as that it was likely to include virtual reality training studies that may have been conducted during the pandemic due to the focus on virtual training during that time. Researchers used the ProQuest search feature and searched the terms “immersive virtual reality,” “medical training procedure,” and “learning outcome.” Researchers limited the results to “peer-reviewed” articles only and the source type to “scholarly journals” only. This search retrieved 1,556 results. From these results, the second author initially screened titles and article abstracts that met all of the following criteria: (a) involved the use of virtual reality (VR)-based training to teach a skill; (b) involved a medical or safety-related training; (c) involved participants who worked or studied in a healthcare-related field; (d) was published in a peer-reviewed journal; (e) was published in English. Any articles that did not meet all the previously mentioned criteria were not included for further review.

This initial screening produced a total of 44 included articles, which were then screened once again to determine which articles met the final inclusion criteria: (a) used immersive virtual reality (IVR) that included a headset and hand controllers; (b) involved the training of non-surgical healthcare-related procedures only; (c) included participants who were medical students, practitioners, emergency responders or life support safety instructors; (d) compared the performance between participants who underwent VR-based training and participants who underwent traditional BAU training. As a result, a total of six articles were included in this review (see Figure 1).

As the field of VR-based training matures, the development of evidence-based guidelines becomes imperative. Drawing on insights from research, this paper reviews general practices in the design, implementation, and assessment of VR programs aimed at enhancing non-specialized skills in medical settings where the skills are critical to safety. The summary of this topic is useful as it provides a baseline of understanding of training outcomes for non-specialized skills, speaking to the generality and effectiveness of VR training when considering the participants engaging in this type of training.

Interestingly, we found several studies reported no statistically significant differences between trainees who underwent VR training versus BAU. While this may appear to be a result of failed effects, in essence, it speaks to the interchangeability and substitution of BAU for VR in terms of effectiveness. It may then follow that other variables (e.g., cost savings, trainee satisfaction) may impact the choice of one method over the other (Young and Greenberg, 2013; Bumbach et al., 2022). As a case in point, a study by Farra et al. (2019) found overall cost savings for a hospital when they compared VR versus BAU training on safety drills for hospital workers. The researchers found initially that VR is more expensive for the live drill (BAU) versus for VR training. When development costs are extrapolated to repeated training over 3 years, however, the virtual exercise becomes less expensive, while the cost of live exercises remains fixed. The authors concluded the larger initial investment in virtual reality can be spread across many trainees and a longer time with little additional cost.

Of note, pedagogy and specific learning theories were seldom mentioned, if at all, in the articles reviewed. Though not recognized in all fields, human service fields frequently incorporate the use of hands-on simulation training with opportunities for the trainee to rehearse and practice. As an evidence-based example, behavioral skills training (BST) is frequently used to build skills for individuals delivering behavior assessment and intervention strategies/techniques for children with disabilities (Shea et al., 2020; Schaefer et al., 2021; Smith et al., 2022). BST entails four separate phases of training: instructions, modeling, rehearsal, and feedback. The first phase involves didactic instruction which seeks to explain the how and why of the processes that underlie the strategy being taught. Though not as important as the modeling or the rehearsal and feedback phases, an explanation of procedures satisfies an element of social validity within teaching a new skill. Next, the expert models the skill for the trainee. This can be done in-vivo, through a confederate, by viewing pre-recorded video, or in a virtual space. The third step involves rehearsal with the trainee in a role-play format. During the process the final step is executed, feedback. The expert gives direct feedback to the trainee on what the trainee did correctly and how their performance could be improved. This process is typically repeated until the trainee demonstrates skill mastery.

Similar to BST, training in the medical field involving simulation involves the technique of rapid cycle deliberate practice (RCDP). In RCDP learners rapidly cycle between deliberate practice and directed feedback within the simulation scenario until mastery is achieved (Taras and Everett, 2017). RCDP has been shown to be effective in teaching a spectrum of resuscitation skills, such as airway management skills (e.g., Gross et al., 2016), pediatric resuscitation (Hunt et al., 2014), and resuscitation for cardiac arrest (e.g., Kutzin and Janicke, 2015). Future research should further investigate VR models of BST (e.g., Clay et al., 2021) and RCDP.

Ethical considerations surrounding the use of VR in medical training warrant careful examination. Issues such as informed consent, psychological wellbeing, and potential desensitization to critical situations necessitate a thoughtful and ethical approach to the development and implementation of VR programs. Recent studies by Zechner et al. (2023), Gasteiger et al. (2022), and Slater et al. (2020) contribute valuable perspectives on ethical considerations in the use of VR for psychological skill development. As Zechner et al. (2023) mention, for police officers to be able to properly make decisions in high-risk scenarios, they need to feel a certain amount of stress. Stressful situations could be visually simulated within AR/VR, and as technology improves these simulations will only become more immersive as resolution increases in each eye of the headset (Zhan et al., 2020) and haptic feedback (van Wegen et al., 2023) becomes easier to program into the software through ancillary hardware. This stress is necessary to elicit since most closely resembles the state of the participant in the natural environment and would, theoretically, greatly help with the generalization component of some training. Clearly, this also comes at a potential price as practitioners must delicately balance the introduction of stressful situations as they correspond to the expected application of the strategies by the trainees. If not carefully controlled, undue stress would be presented to a trainee. This evaluation of modulated stressful situations is not present within some trainings in VR/AR literature, and it deserves a closer look.

Related to the potentially aversive experiences in VR, positive experiences reported via social validity measures were found in all studies we reviewed. Of note, all studies reported high satisfaction with the VR training program. Part of the experience of VR training is how immersive the experience is and may impact how a user reports positive experiences in the simulation. Data on the degree of immersion as it relates to positive or negative social experiences would be valuable for future researchers to collect.

Some limitations of this current review include a narrow range in the scope of the variables identified, reviewed, and discussed. We specifically attempted to provide a review of VR programs for discrete skills that may be used in medical settings but excluded specialized skills (e.g., surgery), dynamic skills (e.g., teamwork communication; cf. Bracq et al., 2019), and disease diagnosis as multiple previous reviews of VR skills training exist for these skills. Another limitation related to scope of the review and recent research that has since been published. We conducted the review in late 2022 and it’s likely more studies have been published since then, which creates an excellent opportunity for future research. Another limitation is that we did not combine and calculate the overall effect of the treatments for comparisons between studies analyzed as might be the case in larger meta-analyses. Future researchers may attempt effect size calculations when summarizing VR training research for a more complete analysis of the extent to which VR training contributes to the acquisition of technical skills. A third limitation is that we only used one database (i.e., ProQuest). For a more thorough and systematic review, researchers should include comparison with at least one other database.

In summary, our review provides information on VR non-specialized training to practitioners and researchers to inform evidence-based practice and future research directions. Importantly, we provided measures of efficacy that involved comparison to business-as-usual or control training, while also including descriptions of technology that was used as well as identification of social validity and generalization. Despite efficacy being significant for all studies reviewed we would suggest taking caution in interpretation as only one-third of the studies reviewed reported generalization measures. Future directions include assessing both IVR and augmented reality as training tools as well as comparisons of these formats in skills training. We would encourage future researchers to include some measure of generalization outside the virtual environment to increase the validity of training. Future researchers might also investigate which variables lead to the generalization of the skill outside the virtual environment to increase the validity of the training.