Wearable Biosignal and Pupillometry Analytics for Stress-Probe Evaluation in Mixed Reality Illegal Checkpoint Training

High-stakes peace mission and first-responder training must expose trainees to threat, ambiguity and time pressure without compromising safety or repeatability. This paper presents a cutting-edge Mixed Reality (MR) training and learning-analytics concept for the novel “SmartSkills” training procedure and reports an exploratory proof-of-concept pilot study conducted at the Johanniter Simulation Centre in Vienna in 2025. The system combines digital twins of operational training sites, live and virtual actors, a scenario editor, eye tracking, cardiovascular and wearable biosignal sensing, and an instructor-facing decision support layer. The methodological core is a stress-probe approach: short, time-locked acoustic and interaction events, especially aggressive shouting and other emotionally aversive auditory cues, are embedded in unavoidable key scenes of an illegal checkpoint scenario. Physiological responses are analysed in standardized windows using heart rate, short-term HRV/RMSSD and pupillometry, with gaze-based indices of situation awareness. The pilot study included 12 participants in six trials, each involving a trainer and two trainees. Results indicate that synchronized wearable and eye-tracking data can identify probe-locked autonomic changes and longer episodes of sustained stress during MR training. Although the study is exploratory and not powered for inference, the approach provides a scalable pathway toward objective after-action review, adaptive scenario difficulty and individualized stress-resilience training.