Medical education within Pediatric Cardiology is increasingly recognizing that extrapolation of 3 dimensional structures from 2 D models, while possible and trainable, may not be the optimal way to teach trainees about the heart. Given the close relationship between spatial orientation and physiology that characterizes Pediatric Cardiology, there is a pedagogical imperative to develop approaches that make fields such as congenital heart disease more visually accessible. 3DP and VR have arisen as potential tools in this effort and the preponderance of evidence to date suggests benefits in regard to learner engagement (1, 3, 10–12). Less robust evidence exists to compare the relative value of these modalities. This question is important to address as the two approaches vary significantly in regard to questions of cost, preparation time and portability which are key factors in wider adoption of these approaches in curricula. It was this gap that our study was designed to begin to address. By systematically comparing trainee experience with each modality side by side, a meaningful assessment was obtained to help guide further training efforts and studies.
The results were somewhat surprising in how definitively they skewed toward VR versus 3DP (87% vs 13% summative preference). We hypothesized that there would be a significant number of participants for whom the tactile and haptic qualities of 3DP models would make them preferable to VR models. For the few who in fact had this preference, these factors were mentioned. Also noted, however, was the limitation of the pre-determined cutting planes. In contrast, VR, had a slicing tool which offered multiple planes in virtually any orientation and was repeatedly cited as an appealing factor in narrative comments. This ability to direct the learning experience more precisely is what a previous study on virtual skills learning identified as “presence” and “agency” (13). However, the grey scale of the Virtual Reality model was mentioned as a limitation compared to the two colored 3DP models, suggesting there is a clarifying role for color differentiation regardless of modality. Other VR platforms have color options which likely will improve satisfaction even more.
An additional consideration in comparing these modalities is the effect of prior exposure, whether direct as in VR/3DP in CHD education or indirect as in other settings such as video game usage. Related to this exposure question is Roger’s discussion of generational learning where he advocates for a framework that is best suited to the current generation of learners to include technology (14). In our study, participants generally had limited prior exposure to either VR (1.1 ± 0.4) or 3DP (2.1 ± 1.5) but felt generally comfortable with modern technology (7.6 ± 2.1). It is unlikely, therefore, that prior VR/3DP exposure played a significant role in the preference of VR over 3DP. It should not be ignored, however, that there were novel features like pen wand navigation that may have contributed to the appeal of VR. What may be a more relevant question is whether general comfort with the technology that permeates modern life shortens the learning curve for modalities like VR.
Digging deeper into what it means to “like” or “prefer” a modality raises additional questions: did intrinsic parts of the educational process such as learning to use the VR wand offer an internal reward that made preference more likely? Less ambiguous is that experienced raters were essentially unanimous in their VR preference. This preference could reflect less of a need or desire on the part or early learners for dimensional data than expert learners.
The educational potential of Virtual Reality is certainly being explored in a number of other fields as well with potentially transferable principles. In nursing, for example, an intervention group who were taught a procedure using VR were able to perform more of these procedures in an hour compared to the control group (15). However, these gains were not sustained two weeks after the initial study suggesting that some of the benefits that VR imparts may require tech “boosters” to be sustained. VR is also being employed in pharmacy education where dynamic applications are being explored such as tracking a drug as it proceeds through the body observing visually how it is changed at each stage (16). Such dynamic 4D tracking can be applied to real time analysis of cardiac structures as described in a recent technological innovation review (17). In the orthopedic domain, a study examining the impact of VR and 3D models on preoperative planning for humeral fracture repair found the use of these modalities led to shorter operative time and less blood loss than conventional methods (18).
Of critical importance in the ongoing evaluation of these modalities is to consider both objective effectiveness and feasibility. In a recent study examining the impact of VR on participant understanding of atrioventricular canals, no difference was found in post-test scores between the control group (desk-top computer) and intervention group (VR). However, the VR group did report a better learning experience and engagement level. Almost counterintuitively, the VR group also had a stronger correlation between their perceived strength of knowledge and their actual performance suggesting that this modality may have role in bridging the gap between perceived knowledge and actual knowledge. In a counterexample, a study looking at the relationship between participant confidence of correctness and actual correctness in the virtual environment of a pre-surgical planning session found the correlation was low (19). This finding may be related to the challenges of measuring depth and features in VR. There continues to be a need for rigorous studies that evaluate objective improvement in knowledge acquisition and spatial conceptualization which can be difficult to capture. Su’s controlled study examining the impact of 3D models in a medical student curriculum is a promising example (12). By asking both subjective questions as well as fact based and spatial conceptualization questions in the post-test, this study was able to demonstrate improvement in knowledge acquisition more rigorously. In regard to 3DP, a recent review highlights the need to systematically examine if there are certain groups who may benefit more from such modalities (3).
Having demonstrated effectiveness, the final hurdle for the wider use of such modalities is feasibility. 3DP models are expensive and time consuming to prepare (11). VR, depending on the interface, can also involve significant cost but lower technology iterations exist. If such factors as cost can be addressed, VR holds further promise given shareability. Such technological nimbleness and ability to share remotely is critical in an age where we witnessed a physical interaction standstill with the novel coronavirus (COVID-19). Further nuanced work can reveal where modalities like Augmented Reality (AR), which retains the capacity to still see the physical world, may be more optimal (20). VR promises to not only make CHD education more effective, but may also have important global pediatric cardiology applications including the capacity to remotely train others in low and middle income countries (LMIC) where such work could be an important part of capacity building (21). Such work would also form a robust response to the charge issued by the Lancet Independent Global Commission for the Education of Health Professionals for the 21st Century calling for “transformative learning” through the harnessing of technological innovations (22) .