Analysis of Attention Based on Gaze and Head Movements under Task Difficulty in an Extended Reality Diagnostic Simulation

Abstract

This iLEAD type’s study examines how task difficulty shapes learners’ visual and embodied exploration during strabismus diagnosis in an extended reality (XR) simulation. 21 medical students completed three diagnostic subtasks (ocular motility test, cover–uncover test, and prism cover test) while eye gaze and head movement data were continuously recorded (gaze hit position; head position and rotation). Visual exploration was quantified using gaze path rate (scanpath length per second) and gaze speed variability (SD of gaze speed), and embodied exploration was quantified using head translation path rate (head movement rate) and head angular speed variability (SD of head angular velocity). Behavioral indicators were compared across the three tasks (ocular motility test, cover–uncover test, prism cover test), accounting for repeated observations within individuals. Results showed that higher task difficulty was associated with increased gaze path rate, increased gaze speed variability, and increased head movement rate. Head rotation variability showed an increasing tendency with task difficulty but did not reach statistical significance. By integrating continuous eye and head movement indicators, this research provides quantitative evidence that task difficulty systematically alters how learners explore and physically navigate XR diagnostic simulations, offering a behavioral basis for understanding performance processes in immersive clinical training environments.