[This link-filled essay from Bioethics Today, the website for the American Journal of Bioethics, raises some thought-provoking and important concerns about the use and impact of (presence-evoking) simulations in medical education. –Matthew]

Virtual Reality in Medical Education: Can Empathy Become a Simulation?

By Hana Abbasian, MS, a Research Assistant at Harvard Medical School and Centre for Addiction and Mental Health (CAMH)
October 28, 2025

Medical education has always relied on simulators. From wooden mannequins in the 17th century to the digital cadavers of today, physicians have long used different tools to practice care without consequence. But virtual reality (VR) changes the moral equation. When a headset can recreate the tremor of an elderly hand, the panic of an asthma attack, or the exhaustion of a long shift, the question is no longer just how doctors learn, but what kind of moral world they are learning within.

The practice of medicine constantly navigates the balance between emotion and algorithm, and VR pushes that balance into a new moral territory. Virtual reality promises to revolutionize medical training: immersive anatomy lessons, lifelike emergency simulations, and modules that let students “learn” empathy. What does it mean to learn compassion through code? To treat simulated patients whose suffering is both real and unreal? And who decides which experiences, and whose bodies, are worth simulating? These questions focus on how we come to understand another’s pain when that pain is mediated by data and algorithms.

Advocates of VR in medicine often describe it as an “empathy machine.” By letting trainees embody a different perspective, for instance, through simulations that replicate what it feels like to inhabit an aging body, educators hope to cultivate moral sensitivity. At the MIT AgeLab, researchers developed the Age Gain Now Empathy System (AGNES), a suit and virtual module that constrains movement, blurs vision, and impacts hearing to mimic the sensory and physical limitations of older adults. When integrated with virtual reality, the system allows individuals across multiple fields to feel age rather than simply study it.

However, some scholars believe that empathy is not an algorithm, and embodiment is not the same as understanding. A ten-minute simulation of visual impairment may not be able to reproduce the daily negotiation of identity, dignity, and dependence that real-life individuals experience. VR may risk trivializing suffering by packaging it into an “experience” designed for education and, increasingly, commercial use. The ethical question is whether these empathy simulations lead to sustained moral growth or to moral fatigue, where compassion becomes another module to “complete.” Therefore, the moral risks extend beyond the classroom to healthcare settings, where empathy itself risks becoming a commodity.

Many VR simulations incorporate biometric and behavioral data, including heart rates and facial expressions, to create a sense of “embodied” patient experience. While this data can improve training, it also risks turning empathy into a metric rather than a human moral engagement. Without careful oversight, trainees might be evaluated on how closely their biometric signals match an expected pattern of concern or distress, shifting the goal from understanding suffering to performing it. Therefore, integrating data-driven realism into VR raises an important ethical question: can empathy be taught as a dataset, or does it require the unpredictability of real human experience?

Ethically, this raises the specter of moral quantification, the idea that empathy, attentiveness, or composure can be measured and optimized like surgical speed. Medical educators should ask whether such metrics improve learning or simply reward emotional performativity, or the simulation of compassion rather than actively building it.

Additionally, VR may offer a seductive illusion: control. Inside a simulation, mistakes can be reversed, deaths undone, and ethical dilemmas replayed until the “correct” decision emerges. Yet this reversibility may ignore one of medicine’s most important moral lessons, that some choices have irreversible consequences.

When students practice resuscitation on virtual patients who always come back, the emotional weight of loss may be softened. And when a simulation reduces ethical ambiguity to a series of branching options, moral judgment becomes gamified. The real clinic is not so forgiving. Patients do not always survive and uncertainty cannot be reset.

Ethical medical education depends on building moral resilience, including the ability to live with imperfection, grief, and doubt. VR can assist in that process only if it acknowledges its own limitations, resisting the temptation to replace uncertainty with artificial clarity.

True empathy in medicine often arises from discomfort, from being powerless in the face of pain or ambiguity. By contrast, VR offers an experience of compassion which has reduced unpredictability. When medical education outsources moral experience to technology, it may risk producing practitioners who are capable of describing empathy, but less practiced in living it.

The ethical principle of justice demands that technological progress not worsen educational inequities. Open-source or low-cost VR models, collaborative data-sharing, and equitable partnerships with institutions in the resource-limited settings could help democratize access. Otherwise, VR will serve as another form of technological gatekeeping in medicine.

Virtual reality will undoubtedly reshape medical education. But as we build increasingly sophisticated worlds of simulation, we must remember all the ethical dimensions that shape, and sometimes distort, our sense of empathy and reality. The ethics of virtual medicine is a call for justice, empathy, and awareness of the lives behind the data. As students heal in simulation, they must not forget the bodies, histories, and inequalities those virtual patients stand in for.