Why eye tracking seems like a natural fit for VR learning
A VR headset already sits on the face, so adding eye-tracking cameras is technically convenient, and the headset can record gaze continuously while a learner moves through a virtual scene. That makes it tempting to treat gaze as a window into whether someone is paying attention and how hard their brain is working. A review of headset-based learning research frames the goal this way: eye tracking is investigated as a way to assess both visual fatigue and cognitive load during VR learning [3].
The appeal is practical. If gaze patterns could reliably signal overload, a virtual lesson could in principle adapt in real time. Work on eLearning platforms pursues exactly that idea, pairing eye-tracking data with machine-learning models intended to predict cognitive load and support a real-time adaptive response during learning [5]. That is a design ambition, though, not proof that any particular gaze measure is a valid stand-in for mental effort.
What the measures actually capture
The measures fall into a few families. Gaze allocation — where a learner looks and for how long — is used as an indicator of visual attention, and studies in immersive 360-degree video have used eye-tracking equipment built into a VR head-mounted display to evaluate learning behaviors and attention [2]. In that work, cues embedded in the immersive video were examined for their effect on learning and attention, with the underlying concern that poorly managed attention could raise cognitive load toward overload [2].
Physiological signals such as blinking, pupil behavior, and eye-movement patterns are the other main family, and the headset review treats these as the tools for assessing visual fatigue and cognitive load together [3]. The important caveat is that these are indirect. A change in blinking or gaze dispersion is consistent with a change in mental effort, but it can also reflect fatigue, lighting, or the demands of the virtual scene itself — which is why the review pairs fatigue and load rather than treating one signal as pure load [3].
How far the evidence travels beyond the headset
A lot of eye-tracking work on attention and cognitive load was not done in immersive VR at all. One study set out to establish eye-tracking metrics for cognitive load and to analyze how visual attention is allocated in multimedia learning, but it did not involve immersive virtual environments [1]. Similarly, work on AI avatars versus human instructors used eye tracking to capture learners' visual attention patterns and cognitive load in video-based learning, not in VR [4]. These studies support the general logic that gaze and physiological signals track attention and effort, but their specific measures should not be assumed to hold unchanged inside a headset.
The VR-specific evidence is narrower and mostly centers on attention as one component of load rather than on load as a whole. The 360-degree video study is the clearest case of attention being examined directly in an immersive headset setting [2], while the review is the main source that speaks to cognitive load and visual fatigue in headset-based learning specifically [3]. So the honest reading is that attention has been studied in immersive VR, but the mapping from eye signals to cognitive load in VR rests more on review-level synthesis than on a large body of direct VR experiments.
What this means for judging a VR lesson
Treat any single eye-tracking number with suspicion. The evidence supports using gaze allocation to characterize visual attention [2] and using physiological eye measures to assess load and fatigue [3], but it does not establish one measure as a validated index of cognitive load in VR. Combining attention and physiological signals, and interpreting them alongside the learning context, is more defensible than reading one metric as a verdict.
It also helps to know what kind of VR content the evidence covers. The direct immersive work concerns 360-degree video with cues [2], and the review concerns headset-based learning broadly [3]; findings from flat-screen multimedia and video learning [1][4] and from eLearning prediction models [5] are useful background but should not be transplanted into VR without checking. If a claim about VR attention or load cites only non-VR studies, that is a sign the claim is extrapolated rather than demonstrated.
Sources used in this answer
Cognitive load and visual attention assessment using physiological eye tracking measures in multimedia learning
The study uses eye-tracking to build metrics for cognitive load and to analyze visual attention allocation in multimedia learning, but it is not set in immersive VR.
Impacts of cues on learning and attention in immersive 360-degree video: An eye-tracking study
Using eye-tracking built into a VR head-mounted display, the study examined how cues in immersive 360-degree video affect learning and attention, with cognitive overload as a stated concern.
Measuring visual fatigue and cognitive load via eye tracking while learning with virtual reality head-mounted displays: A review
The review investigates how eye tracking can be used to assess visual fatigue and cognitive load during learning with VR head-mounted displays.
The role of AI avatars and human instructors in video-based learning: an eye-tracking study on cognitive load and social presence
In video-based learning, eye-tracking analysis captured learners' visual attention patterns and cognitive load in relation to AI avatar versus human instructors.
Machine Learning Methods for Predicting Cognitive Load in eLearning Environments using Eye Tracking Data.
The work explores combining eye-tracking data with machine-learning methods to predict cognitive load in eLearning platforms for real-time adaptive responses.