User-Centered Quality of Experience Assessment for Extended Reality: From Passive Perception to Interactive Engagement

Extended Reality is transforming how users perceive and interact with digital environments, yet evaluating the resulting Quality of Experience remains difficult due to the interdependence of technological, perceptual, cognitive, and social factors. This thesis adopts a user-centered approach to investigate how these factors shape immersive experiences across a progression of scenarios: passive perception, active exploration, multi-user interaction, and applied cultural-heritage contexts. The first study examines passive experiences, assessing whether head-movement signals captured by head-mounted displays can serve as unobtrusive indicators of cognitive stress during non-interactive Augmented Reality tasks. Stress-inducing protocols, a public dataset, and a machine-learning model were developed to explore the feasibility of lightweight behavioral monitoring. The analysis then considers active experiences, where users’ actions shape their multisensory engagement. Experiments on the visual quality of 3D meshes presented on desktop and in Virtual Reality, together with a study on multisensory integration, investigate how exploration behavior, viewpoint control, and sensory cues influence perceptual judgments. Interactive scenarios are addressed through a methodological framework for evaluating interaction quality, social presence, and participation patterns in asymmetric collaborative environments. This framework supported the development of ITU-T Recommendation P.1321 and was validated in a 5G-based tele-rehabilitation study examining how network variability affects communication and interaction fluency. Finally, the methodological framework was translated into practice through a collaboration with the Roman National Museum, guiding the user-centered development of a Virtual Reality experience for the site of Palazzo Altemps (Roman National Museum) in Rome. Summarizing, the thesis shows that Quality of Experience in Extended Reality arises from a complex interplay of system behavior, content properties, and user strategies. Across all studies, behavioural cues emerge as a promising foundation for developing adaptive systems that better reflect users’ needs and behavior.