External distraction, visual guidance, and support for return
When should XR help learners stay focused — and how should it support their return?
The practical question is simple: does the external event require the learner to stop the lesson? The answer changes what helpful system behaviour looks like.
Optional visual guidance can emphasize relevant content around a competing event, but the intervention itself must not add unnecessary workload or obstruction.
RQ1 measures how realistic external distractions affect learning and experience. RQ2 maps the available attention-guidance design space. RQ3 combines both foundations to test when guidance should appear, what it should communicate, and whether its benefits outweigh its costs.
First understand the competition for attention. Then understand the available guidance mechanisms. Finally, use both foundations to design and evaluate support in XR learning.
How do external distractions affect attention-related behaviour, learning outcomes, and user experience during individual, non-co-located learning in desktop and XR learning environments?
What visual attention-guidance mechanisms and evaluation dimensions exist and can inform the design of attention support for individual, non-co-located learning in desktop and XR learning environments?
How can visual guidance be designed and timed to mitigate the effects of external distractions and interruptions on individual, non-co-located learning in XR learning environments, and what benefits and costs does it introduce?
A formative student survey informed a shared set of auditory scenarios. Two studies then examined distraction across desktop, immersive VR, and three Apple Vision Pro presentation environments.
Students reported distractions experienced during face-to-face and online learning. Four auditory scenarios were selected for controlled experiments.
A distractor can compete for attention without proving that the learner disengaged or missed content.
The four distractors differ in timing and duration, so episode-level comparisons are descriptive unless the design supports stronger inference.
Eye tracking is useful behavioural evidence, but looking at content does not establish cognitive engagement or learning.
The attention-guidance work maps a fragmented design space, gives it a shared vocabulary, and then compares selected techniques under one common visual-search task.
Arrows, labels, blur, flicker, binocular modulation, highlights, and many other mechanisms had been difficult to compare directly.
107 technique entries are organized by perceptibility, mechanism, platform, visual variables, and Gestalt principles.
18 paper-informed conditions were recreated in the same visual-search task and compared with subjective and gaze-based measures.
Noticeability, preference, findability, fixation success, timing, and replay effort can diverge.
Select a dimension to see what it contributes to the design vocabulary.
The study compares noticeability, preference, perceived findability, target fixation, guidance-relative timing, and replay effort. It is exploratory: the goal is to reveal trade-offs, not declare a universal winner.
With 18 conditions, exhaustive unique pairing would require 153 comparisons per participant. The study used a fixed budget of 76. Early selection prioritised broad exposure, then minimum exposure, followed by more targeted comparisons. Noticeability drove the adaptive schedule; preference was collected on the same displayed pairs.
Choose a profile.
RQ3 is not a single predetermined cue. It is a design problem with two separable choices: what guidance communicates and when or whether it appears.
For example: visual emphasis, a cue, a checkpoint, or a contextual reminder.
For example: around a competing event, when essential content appears, or only on return.
The final study should isolate benefit from the extra demands introduced by assistance.
The distinction is deliberately behavioural rather than technological: the same XR system may need different policies depending on whether the external event is irrelevant or requires the learner's attention.
Select a direction to see the mechanism, timing policy, comparison, and main evaluation question.
Gaze return and perceived helpfulness should be assessed separately from learning outcomes. Comparisons intended to isolate visual guidance should match instructional information and viewing time. Recaps and replay alter the information or exposure available, and visual emphasis alone cannot recover inaudible narration.
40 studies, 107 technique entries, 24 families, five classification dimensions.
18 conditions, 12 participants, adaptive pairwise comparison with subjective and gaze-based outcomes.
Study 1 and Study 2 data collection is complete. Analysis is ongoing, including time/content-aware questions where the data support them.
Use RQ1 and RQ2 evidence, symposium feedback, lesson fit, and XR constraints to select a bounded mechanism + delivery policy.
Evaluate one attention-support intervention and determine whether a further timing or resumption study is warranted.
Do the distraction studies and guidance work form a coherent contribution to attentional control in XR learning?
Which offers the strongest next step: event-linked, content-linked, or coordinated maintenance and resumption?
What evidence demonstrates meaningful learning benefit, and which assistance costs matter most?
What can one focused intervention study establish, and what would justify an additional study?
The current Doctoral Symposium paper positions the thesis across interruption science, XR learning, attention guidance, and resumption support.
This site is grounded in the current ISS 2026 poster and six-page extended abstract for the Doctoral Symposium, with additional detail from the ongoing dissertation work.
Visual overview of RQ1, RQ2, and proposed RQ3 directions.
Open PDF ↗
Detailed thesis framing, supporting evidence, candidate intervention directions, and symposium questions.
Open PDF ↗