Key result
VR adaptive training signals reflect user-controlled settings rather than motor skill.
Why the study?
Adaptive training systems require a reliable signal indicating when the user is at the productive edge of their ability, but candidate signals need validation against user-controlled parameters.
Observational (n=32)
Candidate signals for adaptive training systems must be validated against user-controlled parameters before adapting to them.
Answers are grounded in this paper and cite their sources.
Signals for adaptive VR training warrant caution due to user-parameter confounding; leaves open validation of flow metrics against true ability in prospective studies.
Adaptive training systems need a signal that tells them whether the user is at the productive edge of their ability. This poster tests three candidate signals in bimanual polyrhythmic tracing in virtual reality (Meta Quest 3): movement accuracy, ratio fidelity, and autonomic state. Data come from a single-session study (28 healthy adults, 227 runs) and a four-day pilot (4 participants, 144 runs, with heart-rate variability). Each candidate turns out to be driven by a parameter the user controls but the system did not log. Path error mostly reflects movement amplitude. Ratio error reflects how many cycles were observed. Autonomic–coordination links disappear once time-in-session drift is controlled. Self-rated flow tracking self-selected tempo is the only association that survives. The per-hand range-of-motion calibration also acts as a hidden difficulty setting. The poster closes with design requirements: validate candidate signals against user-controlled parameters before adapting to them, and log what the projection discards. The four-day pilot (n = 4) is descriptive only. Presented at the MishMash Opening Conference, Kristiansand, 14–15 September 2026.
No takes yet. Share an insight, caveat, or question.
Adam Bartłomiej Lewartowski (2026) conducted an observational in Healthy adults (n=32). Bimanual polyrhythmic tracing in virtual reality was evaluated on Candidate signals (movement accuracy, ratio fidelity, and autonomic state). Candidate signals for adaptive training in VR were driven by user-controlled parameters rather than true ability, with only self-rated flow tracking self-selected tempo surviving.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: