Interactive Cross Reality-based Data Analysis Tool for User-Studies

Abstract

Cross reality systems become increasingly relevant in fields such as education, medicine and engineering. As these systems grow evermore complex, the analysis of user data within them presents itself as a new challenge. Traditional ex-situ analysis methods often lack contextual information, while in-situ analysis methods are rarely implemented. This thesis explores the combination of both analysis methods to improve the overall understanding of user data in cross reality environments. A prototype analysis tool was developed using the Unity game engine, allowing researchers to record and replay user studies. The system supports immersive in-situ as well as traditional ex-situ analysis modes, features such as synchronized data streams, logging of virtual objects and the generation of heatmaps from the user’s navigational data. A collaborative director mode, that allows two researchers to analyze the data at the same time, is also included. An expert study involving experts in the cross reality research field was conducted and evaluated. The tool’s usability, usefulness and the clarity of information were reviewed, and the results indicate that the integration of the two methods supports a more comprehensive understanding of user data, although some areas still require significant refinement to be usable effectively. Overall, the prototype demonstrates the viability and potential of the two analysis methods into one combined tool, that enhances the evaluation of cross reality-based user studies.

Date of Award	2025
Original language	English
Supervisor	Christoph Anthes (Supervisor)