View Proposal

Proposer
Marta Vallejo
Title
Investigating Visual Attention and Cognitive Load During Walking Under Different Environmental and Auditory Conditions Using Eye Tracking
Goal
The project aims to investigate how visual attention and cognitive load are influenced by different environmental and auditory conditions during walking. Using Pupil Labs Pupil Invisible glasses, the study will analyse eye movements, gaze behaviour and physiological indicators such as blink rate while participants navigate realistic environments. The project will extend previous work by developing a richer understanding of how environmental hazards and distractions influence attention, while creating a high-quality dataset for future research in human factors, safety and explainable artificial intelligence.
Description
Human attention is fundamental for safe navigation in everyday environments. Walking requires the continuous integration of visual information while simultaneously responding to environmental hazards and cognitive demands. Understanding how attention changes under different conditions is important for designing safer public spaces, workplaces and assistive technologies. This project builds upon previous MSc research on visual attention and cognitive responses using Pupil Labs Pupil Invisible eye-tracking glasses. Existing datasets and software pipelines provide a solid foundation, allowing the student to focus on designing new experiments and extending previous findings rather than starting from scratch. Participants will wear the eye-tracking glasses while walking through predefined environments under different experimental conditions. Potential investigations include: - quiet versus noisy environments; - different auditory conditions, such as ambient sounds, traffic noise, conversations, alarms or music; - environments with varying levels of visual complexity and potential hazards; - age-related differences in attention and distraction. Eye-tracking data, including gaze behaviour, fixation duration, saccades and blink rate, will be analysed to investigate how environmental and auditory stimuli influence attention and cognitive load. The project may also incorporate objective hazard annotations to compare where participants look with independently identified hazards, providing a more robust understanding of visual attention in safety-related scenarios. Students with an interest in machine learning may also explore computer vision and explainable AI techniques to analyse gaze behaviour, identify patterns associated with increased cognitive demand, and compare human attention with AI-based visual attention models. The project is expected to provide: - A quantitative analysis of how different environmental and auditory conditions influence visual attention during walking. - Insights into the relationship between environmental hazards, cognitive load and gaze behaviour. - Analysis of blink rate and other eye-tracking metrics as indicators of attentional engagement. - Comparison between participant gaze patterns and independently defined hazard locations. - A high-quality eye-tracking dataset to support future research in human factors, ergonomics, safety science and explainable AI. - The opportunity to contribute towards a journal publication extending previous work in this area. Reference: Vallejo, M. and Wang, S., 2026. Comparing Human Gaze and Vision-Language Model Attention in Safety-Relevant Environments. arXiv preprint arXiv:2606.15202.
Resources
- The Pupil Labs "Pupil Invisible" glasses. - Existing datasets and software developed by previous MSc students.
Background
The student should have an interest in machine learning, computer vision, eye tracking and data analysis. The project is suitable for students interested in artificial intelligence, human factors, cognitive science, ergonomics or explainable AI. Since the project builds upon existing research and software, it offers an excellent opportunity to extend previous work and potentially contribute to a journal publication.
Url
External Link
Difficulty Level
Variable
Ethical Approval
Full
Number Of Students
2
Supervisor
Marta Vallejo
Keywords
machine learning, data analysis, computer vision, eye-tracking technology, pupil labs glasses, visual attention.
Degrees
Bachelor of Science in Computer Science
Master of Engineering in Software Engineering
Master of Science in Artificial Intelligence
Master of Science in Computing (2 Years)
Master of Science in Data Science
Master of Science in Human Robot Interaction
Master of Science in Robotics
Master of Science in Software Engineering
Bachelor of Science in Computing Science
Bachelor of Engineering in Robotics
Master of Science in Robotics with Industrial Application
BSc Data Sciences
Back to List