2026-10-22 イースト・アングリア大学(UEA)
Scientists found that tiny changes in pupil size and blinking patterns can reveal how hard the brain is working during virtual reality tasks.
＜関連情報＞
- https://www.uea.ac.uk/about/news/article/how-the-eyes-could-unlock-personalised-pain-relief-with-virtual-reality
- https://www.nature.com/articles/s41598-026-74975-6
VR-FOCUS：慢性疼痛における認知負荷の指標として、仮想現実Nバック課題中の視線追跡を評価する VR-FOCUS: evaluating eye tracking during a virtual reality N-back task as a marker of cognitive load in chronic pain
Jordan Tsigarides, Jennifer O. Bowler, Jack Dainty, Anthony Bagnall & Alexander Macgregor
Scientific Reports Published:14 October 2026
DOI:https://doi.org/10.1038/s41598-026-74975-6 Early provide
Abstract
Immersive virtual reality (VR) is a promising medium for adaptive pain therapeutics, but objective markers of cognitive load suitable for real-time adaptation remain insufficiently characterised in chronic pain. This study evaluated whether eye tracking embedded within a consumer-available VR system provides signatures of cognitive load during a VR N-back task. A total of 84 participants (42 healthy controls, 42 chronic musculoskeletal pain) completed five levels (Baseline to 4-Back) while ocular responses were recorded. Performance declined and subjective workload increased with task level, confirming successful manipulation of cognitive demand, although performance plateaued beyond 3-Back. In mixed-effects models, larger pupil diameter and higher blink rate were associated with task level and workload, without group differences. Logistic regression using summary features discriminated 1-Back from 4-Back at near-chance levels (0.51–0.60). Time-series classifiers exploiting temporal structure performed substantially better in healthy controls (0.74–0.88) but were attenuated in chronic pain (0.58–0.65). Age alone did not appear to explain this difference, although a contribution cannot be excluded. These findings support the feasibility of VR-embedded eye tracking for cognitive load estimation in healthy adults, but performance in chronic pain is currently insufficient for reliable individual-level inference; closed-loop application in this population will require participant-calibrated models and prospective validation.
