Clinical training in extended reality becomes far more powerful when the scenario can sense how the trainee is coping and respond to it. Instead of a fixed script that plays out identically regardless of who is in the headset, a neuroadaptive scenario continuously infers cognitive load and stress from physiological signals — heart-rate variability, electrodermal activity, and pupillometry — and adjusts what the trainee encounters next.
The demo below makes that closed loop visible. A simulated clinical scene runs against a scrolling readout of synthetic biosensor traces. A transparent, inspectable rule combines those signals into a single load index and sorts it into a regulation band. When the loop is adaptive, the scenario responds: easing difficulty and offering scaffolding when the trainee is overloaded, and raising the challenge when they are calmly in control. Toggle adaptation off to watch the same run play out as a fixed script, inject a complication to perturb the system, and switch between emergency-medicine, pediatric-readiness, and surgical-education domains.
All data shown is synthetically generated for illustration. This is a conceptual demonstration of the adaptive-loop mechanism, not a validated clinical or performance-assessment tool.