Towards Effective Human Performance in XR Space Framework based on Real-time Eye Tracking Biofeedback
Abstract
This paper proposes an eye tracking module for the XR Space Framework aimed at enhancing human performance in XR-based applications, specifically in training, screening, and teleoperation. This framework provides a methodology and components that streamline the development of adaptive real-time virtual immersive systems. It contains multimodal measurements - declarative in the form of in-VR questionnaires and objective, including eye tracking, body movement, and psychophysiological data (e.g., ECG, GSR, PPG). A key focus of this paper is the integration of real-time eye tracking data into XR environments to facilitate a biofeedback loop, providing insight into user attention, cognitive load, and engagement. Given the relatively high measurement frequency of eye tracking - recognized as a noninvasive yet robust psychophysiological measure - this technology is particularly well suited for real-time adjustments in task difficulty and feedback to enhance learning and operational effectiveness. Despite its established role in cognitive and attentional studies, implementing eye tracking metrics within dynamic, real-time XR environments poses unique challenges, particularly given the complex moving visuals presented in head-mounted displays (HMDs). This paper addresses these challenges by focusing on the essential aspects of integrating eye tracking in immersive systems based on real-time engines, ultimately facilitating more efficient, adaptive XR applications.
Cite
@article{arxiv.2507.21000,
title = {Towards Effective Human Performance in XR Space Framework based on Real-time Eye Tracking Biofeedback},
author = {Barbara Karpowicz and Tomasz Kowalewski and Pavlo Zinevych and Adam Kuzdraliński and Grzegorz Marcin Wójcik and Wiesław Kopeć},
journal= {arXiv preprint arXiv:2507.21000},
year = {2025}
}