The Importance of VR Simulations with Analytics for Instant Feedback

What are VR simulation analytics?

VR simulation analytics are structured records of a learner's actions and performance within a virtual learning environment. They can show what the learner attempted, which decisions they made, how they interacted with equipment, how long they spent on each stage and whether their actions met defined learning or safety criteria.

A conventional learning management system may record that a learner opened a module, completed it and received a score. A VR simulation can provide a more detailed record of performance because the learner is acting within a three-dimensional environment.

Depending on the simulation, the system may record:

• Which actions were completed
• The order in which actions were performed
• How long the learner took to recognise or respond to a situation
• Which equipment settings or clinical parameters were selected
• Positioning, distance, orientation and movement within the environment
• Missed steps, repeated errors and requests for assistance
• Safety-related decisions
• Performance across repeated attempts

These records become educationally useful when they are interpreted against a defined objective, standard, rule or assessment criterion. Raw activity data alone do not explain whether a learner understands a task or can perform it safely in clinical practice.

What types of learning data can VR capture?

Some systems can also collect eye tracking, voice, hand movement or physiological data. These data may provide additional information about attention, communication or stress, but they require careful validation and stronger privacy safeguards.

How do VR analytics become useful feedback?

A useful VR feedback system follows a clear learning cycle:

1. Observe: The simulation records a relevant learner action or decision.
2. Interpret: The action is compared with a defined learning objective, performance rule or expected range.
3. Respond: The learner receives a cue, explanation, warning, result or summary.
4. Act: The learner adjusts their approach or repeats the task.
5. Review: The learner or educator examines performance across the complete activity and over time.

This distinction matters because an analytics dashboard is not automatically a feedback system. A score such as 72% tells the learner how they performed overall, but it may not explain what went wrong, why it matters or what they should do next.

More useful feedback might identify that a learner selected an inappropriate equipment setting, delayed a time-critical action or repeatedly positioned themselves in a higher-risk location. It should then explain the significance of the error and support another attempt.

Why does instant feedback matter in VR learning?

Timely feedback allows learners to connect a consequence with the action that produced it. In a virtual environment, the learner can see how a decision affects the simulated patient, image, procedure, equipment or safety outcome while the context is still clear.

A meta-analysis of computer-based learning found that elaborated feedback, which provides an explanation, produced better learning outcomes than simply telling learners whether an answer was correct or incorrect. The review also found that the most appropriate timing depends partly on the type of learning involved.^[4]

This means that not every mistake requires an immediate interruption.

• Immediate feedback is often appropriate when an action creates a clear safety risk, when the learner cannot progress or when continuing would reinforce an incorrect procedure.
• End-of-stage feedback may be more appropriate when the learner needs to complete a sequence before reflecting on it.
• End-of-session feedback can support review of overall strategy, decision-making and patterns across the complete scenario.
• Educator-led debriefing remains valuable when performance involves uncertainty, professional judgement, teamwork or communication.

Effective systems can therefore combine immediate cues with a fuller report and an opportunity to repeat the activity.

What does the evidence show about VR, analytics, and feedback?

Research supports the use of VR and technology-enhanced simulation in healthcare education, but outcomes depend on the learning design, comparison method, learner group, and capability being assessed.

A systematic review and meta-analysis of VR in health professions education found a modest improvement in post-intervention knowledge compared with traditional learning. The authors also reported variation between studies and called for further research into immersive and interactive VR, cost-effectiveness and changes in clinical practice.^[1]

A larger review of technology-enhanced simulation found benefits for knowledge, skills and behaviours when simulation was compared with no educational intervention. The review examined instructional features including feedback, mastery learning, repetitive practice and curriculum integration.^[2]

An earlier BEME systematic review identified feedback, repetitive practice, increasing difficulty, curriculum integration and clearly defined outcomes among the features associated with effective high-fidelity simulation.^[3]

Evidence of transfer beyond the simulator

One of the best-known studies of VR surgical simulation examined whether simulator training transferred to the operating room. Residents trained to a defined proficiency level completed gallbladder dissection 29% faster and made substantially fewer errors than the conventionally trained group.^[5]

This study involved a particular surgical task and should not be treated as proof that every VR application will produce the same result. It does, however, show that measured, proficiency-based simulation can support transfer when the simulator, training criteria and clinical task are closely aligned.

Evidence for automated feedback

A 2025 randomised crossover study compared medical emergency VR training with automated feedback against video seminars of equivalent duration and content. Short-term knowledge gains were similar, but the VR condition produced higher knowledge retention after 30 days. The VR system provided positive notifications, warnings about potentially harmful actions, prompts for missed time-critical actions and an end-of-session checklist.^[6]

The authors noted that the study involved one institution, relatively short learning sessions and knowledge rather than practical-skill retention. The finding therefore supports further use and investigation of automated VR feedback, rather than proving that automated feedback is always superior.

Much of the evidence evaluates VR, feedback, repetition and simulation design as a combined intervention. It is often impossible to isolate the effect of analytics alone. Institutions should therefore evaluate the complete learning design rather than assuming that collecting more metrics will improve outcomes.

How should instant feedback be designed?

Begin with the learning objective

Every recorded action should have a reason for being collected. If the objective is to improve radiation safety, the system might analyse equipment geometry, shielding, distance and exposure-related decisions. If the objective is patient communication, equipment settings alone will not provide meaningful evidence.

Explain what happened and why it matters

Correct or incorrect messages are rarely enough. Feedback should identify the relevant action, explain its effect and direct the learner towards a safer or more effective approach.

Keep feedback close to the decision

The learner should be able to understand which action generated the feedback. Long lists of undifferentiated errors at the end of a complex scenario can make reflection difficult.

Avoid unnecessary interruption

Constant alerts can disrupt concentration and turn an authentic task into a sequence of prompts. Immediate interruption should be reserved for circumstances in which it supports the learning objective. Other information can be provided at a natural pause or during debriefing.

Allow learners to try again

Feedback is most valuable when learners can act on it. A simulation should normally allow them to repeat a task, adjust their approach or revisit the scenario with less assistance.

Separate practice from assessment

Practice mode may provide cues, explanations and encouragement. Assessment mode may need to limit assistance so that performance can be evaluated independently. Learners should know which mode they are using and how their data will be interpreted.

Do not confuse a metric with competence

Faster completion is not necessarily better. A learner may work quickly while overlooking patient communication, safety checks or clinical reasoning. Performance measures should be interpreted together and linked to the capability being taught.

Validate important judgements

Automated scoring rules should be reviewed by subject-matter experts and tested against the intended learning objectives. High-stakes decisions should not depend on an unvalidated score simply because it was generated automatically.

How do VR analytics support educators?

Learner feedback addresses the immediate question, “What should I do differently?” Educator analytics answer a different set of questions:

• Which learning objectives are causing difficulty across the cohort?
• Which errors persist after repeated practice?
• Are learners improving between sessions?
• Do learners understand the theory but struggle to apply it?
• Which scenarios or instructions may need redesign?
• Which learners may benefit from additional support?

Aggregated results can reveal patterns that are difficult to see through observation of individual sessions. For example, an educator may discover that many learners select the correct answer in a knowledge test but apply the principle inconsistently in a simulated clinical environment.

Analytics should support professional judgement rather than replace it. Educators still need to interpret performance in the context of the curriculum, learner experience, scenario design and available evidence.

How should institutions implement VR learning analytics?

A successful implementation begins with curriculum and assessment requirements rather than hardware specifications.

1. Define the learning objective. Identify the knowledge, skill, behaviour or decision the activity is intended to develop.
2. Identify observable evidence. Decide which actions would reasonably demonstrate progress towards that objective.
3. Choose the feedback point. Determine whether guidance should appear immediately, after a stage, at the end of the scenario or during educator debriefing.
4. Define the explanation. Tell learners what happened, why it matters and what they should do next.
5. Provide opportunities for repetition. Allow learners to apply the feedback rather than simply read it.
6. Test scoring and thresholds. Confirm that the system responds accurately to expected and unexpected learner behaviour.
7. Plan educator reporting. Decide which information educators need and avoid dashboards filled with unused metrics.
8. Establish data governance. Define who can access learner data, why it is collected, how long it is retained and whether it may be used for research.
9. Evaluate learning outcomes. Measure whether the activity improves the intended capability, not only whether learners enjoyed using VR.

A scoping review of VR implementation in health professions education identified recurring considerations including collaboration, availability, cost, guidelines, technology, careful design, evaluation and user training.^[7]

What are the limitations and challenges of VR learning analytics?

Data overload

VR can produce a large volume of data, but collecting everything does not mean that everything is educationally meaningful. Too many measures can obscure the outcomes that matter.

False precision

A detailed score may appear objective even when the underlying rule has not been validated. Analytics should not imply a level of certainty that the simulation cannot support.

Incomplete representation of competence

A simulation records performance within a designed environment. It may not capture every aspect of clinical competence, including communication, adaptability, teamwork, professional judgement and response to real patients.

Technical integration

Institutions may need to connect simulation data with a learning management system, learner record store or reporting platform. Consistent learner identity, event definitions and data formats are required if records are to remain interpretable across systems.

Faculty preparation

Educators need to understand what each measure represents, how it was produced and what conclusions can reasonably be drawn from it. A dashboard is of limited value when its users do not trust or understand the data.

Accessibility and user comfort

VR activities should account for users who experience discomfort, have limited mobility, require seated use or need alternative ways to complete an interaction. Analytics should not penalise a learner for using an approved accessibility pathway.

Privacy and data governance

VR systems may collect detailed movement and interaction data. Research has shown that head and hand motion can be sufficiently distinctive to identify individuals in large datasets. This reinforces the need to treat VR telemetry as potentially sensitive rather than assuming that removing a name makes the data anonymous.^[10]

Institutions should collect only the information needed for the educational purpose, restrict access, define retention periods and tell learners how their data will be used.

The future of analytics-supported VR learning

Future VR learning systems are likely to become more adaptive. A simulation may change the difficulty, reduce prompts, introduce new complications or recommend another activity based on the learner's previous performance.

Artificial intelligence may also help interpret complex patterns involving decisions, communication, movement and repeated attempts. These capabilities could support more specific feedback, but they also increase the need for validation, transparency and human oversight.

The strongest systems will not use analytics simply to generate more scores. They will connect each learning experience to clear evidence, explain what that evidence means and help learners decide what to do next.

Conclusion

VR simulation analytics can make healthcare training more observable, repeatable and responsive. They allow software to record learner actions, identify relevant errors and provide feedback while the learner still understands the context in which a decision was made.

Their value depends on design. Feedback must be linked to a defined learning objective, explain why an action matters and give learners an opportunity to improve. Educator dashboards must present interpretable evidence rather than a collection of disconnected metrics.

VR analytics should also be used carefully. Simulator performance is not identical to clinical competence, automated scores require validation and detailed movement data require appropriate privacy protection.

When immersive practice, meaningful feedback, repetition and educator judgement are combined, VR can do more than recreate a clinical environment. It can help learners understand their performance and make deliberate progress towards safer practice.

Frequently asked questions

References

1. Kyaw BM, Saxena N, Posadzki P, et al. Virtual Reality for Health Professions Education: Systematic Review and Meta-Analysis by the Digital Health Education Collaboration. Journal of Medical Internet Research. 2019;21(1):e12959. doi:10.2196/12959. Read the full article.
2. Cook DA, Hatala R, Brydges R, et al. Technology-enhanced simulation for health professions education: a systematic review and meta-analysis. JAMA. 2011;306(9):978-988. doi:10.1001/jama.2011.1234. View on PubMed.
3. Issenberg SB, McGaghie WC, Petrusa ER, Gordon DL, Scalese RJ. Features and uses of high-fidelity medical simulations that lead to effective learning: a BEME systematic review. Medical Teacher. 2005;27(1):10-28. doi:10.1080/01421590500046924. View on PubMed.
4. Van der Kleij FM, Feskens RCW, Eggen TJHM. Effects of Feedback in a Computer-Based Learning Environment on Students' Learning Outcomes: A Meta-Analysis. Review of Educational Research. 2015;85(4):475-511. doi:10.3102/0034654314564881. View the article.
5. Seymour NE, Gallagher AG, Roman SA, et al. Virtual Reality Training Improves Operating Room Performance: Results of a Randomized, Double-Blinded Study. Annals of Surgery. 2002;236(4):458-464. doi:10.1097/00000658-200210000-00008. Read the full article.
6. Lindner M, Leutritz T, Backhaus J, König S, Mühling T. Knowledge Gain and the Impact of Stress in a Fully Immersive Virtual Reality-Based Medical Emergencies Training With Automated Feedback: Randomized Controlled Trial. Journal of Medical Internet Research. 2025;27:e67412. doi:10.2196/67412. Read the full article.
7. Lie SS, Helle N, Sletteland NV, Vikman MD, Bonsaksen T. Implementation of Virtual Reality in Health Professions Education: Scoping Review. JMIR Medical Education. 2023;9:e41589. doi:10.2196/41589. View on PubMed.
8. Sakr A, Abdullah T. Virtual, augmented reality and learning analytics impact on learners, and educators: A systematic review. Education and Information Technologies. 2024;29:19913-19962. doi:10.1007/s10639-024-12602-5. Read the full article.
9. Liao M, Zhu K, Wang G. Can human-machine feedback in a smart learning environment enhance learners' learning performance? A meta-analysis. Frontiers in Psychology. 2024;14:1288503. doi:10.3389/fpsyg.2023.1288503. Read the full article.
10. Nair V, Guo W, Mattern J, et al. Unique Identification of 50,000+ Virtual Reality Users from Head and Hand Motion Data. In: 32nd USENIX Security Symposium. 2023:895-910. Read the conference paper.