Cross-Device Validation of Eye-Tracking Metrics in Visualization Comprehension

Aufgabenstellung / Topic

Description

This thesis project builds on prior research on visualization comprehension by reimplementing and validating a completed study using alternative eye-tracking hardware. The original experiment investigated human comprehension of grouped bar charts and line charts using a laboratory-grade eye tracker (EyeLink 1000 Plus). In this project, the experimental paradigm will be recreated in PsychoPy while integrating a more accessible eye-tracking system (Pupil Labs Core).

While systems such as the EyeLink 1000 Plus offer excellent data quality, their cost and infrastructure requirements limit accessibility for many researchers and practical applications. The Pupil Labs Core is more portable and affordable, opening opportunities for visualization comprehension research in less resource-intensive environments. This project addresses a central question: Can the indicators of visualization comprehension identified with high-end lab equipment be reliably replicated using more accessible eye-tracking technology?

By validating grouped bar and line chart comprehension using Pupil Labs Core, this work aims to assess whether gaze-based indicators (e.g., fixations, scanpaths, temporal measures) and performance patterns are robust across different eye-tracking systems. Successful validation would support broader adoption of eye-tracking-based visualization assessment in educational settings, industry applications, and accessibility research.

Research Phases

The research consists of the following phases:

1. System review and planning: Review the capabilities and API of Pupil Labs Core and identify integration options with PsychoPy for visualization experiments.
2. Implementation and integration: Adapt the existing PsychoPy experimental framework to interface with the Pupil Labs Core eye tracker, including calibration, validation, and reliable data recording for grouped bar and line chart tasks.
3. Validation study: Design and conduct a user study that replicates the original experimental conditions, collecting performance data, presentation times, and gaze metrics using Pupil Labs Core.
4. Comparative analysis: Analyze gaze patterns, fixations, and temporal measures in relation to comprehension performance and compare the results to the original EyeLink-based study to assess replicability, data quality, and measurement consistency.

You Will

• Conduct a technical review of Pupil Labs Core capabilities and integration options with PsychoPy.
• Adapt the existing experimental code to interface with the Pupil Labs Core eye tracker.
• Implement and test calibration and validation procedures appropriate for the new hardware.
• Design and conduct a user study with grouped bar and line chart visualizations.
• Analyze gaze patterns, fixations, and temporal measures in relation to comprehension performance.
• Compare findings with the original study to assess replicability across eye-tracking systems.
• Document your work in a thesis and present your findings to an audience.
• (Optional) Co-write a research paper on the validation of accessible eye-tracking for visualization research.

You Need

• Strong written and verbal communication skills in English.
• Programming experience with Python and PsychoPy.
• Basic knowledge of eye-tracking principles and data analysis.
• Interest in data visualization and human–computer interaction.