The role of multisensory feedback in the objective and subjective evaluations of fidelity in virtual reality environments.

摘要

The use of virtual reality in academic and industrial research has been rapidly expanding in recent years therefore evaluations of the quality and effectiveness of virtual environments are required. The assessment process is usually done through user evaluation that is being measured whilst the user engages with the system. The limitations of this method in terms of its variability and user bias of pre and post-experience have been recognised in the research literature. Therefore, there is a need to design more objective measures of system effectiveness that could complement subjective measures and provide a conceptual framework for the fidelity assessment in VR. There are many technological and perceptual factors that can influence the overall experience in virtual environments. The focus of this thesis was to investigate how multisensory feedback, provided during VR exposure, can modulate a user’s qualitative and quantitative experience in the virtual environment. In a series of experimental studies, the role of visual, audio, haptic and motion cues on objective and subjective evaluations of fidelity in VR was investigated. In all studies, objective measures of performance were collected and compared to the subjective measures of user perception. The results showed that the explicit evaluation of environmental and perceptual factors available within VR environments modulated user experience. In particular, the results shown that a user’s postural responses can be used as a basis for the objective measure of fidelity. Additionally, the role of augmented sensory cues was investigated during a manual assembly task. By recording and analysing the objective and subjective measures it was shown that augmented multisensory feedback modulated the user’s acceptability of the virtual environment in a positive manner and increased overall task performance. Furthermore, the presence of augmented cues mitigated the negative effects of inaccurate motion tracking and simulation sickness. In the follow up study, the beneficial effects of virtual training with augmented sensory cues were observed in the transfer of learning when the same task was performed in a real environment. Similarly, when the effects of 6 degrees of freedom motion cuing on user experience were investigated in a high fidelity flight simulator, the consistent findings between objective and subjective data were recorded. By measuring the pilot’s accuracy to follow the desired path during a slalom manoeuvre while perceived task demand was increased, it was shown that motion cuing is related to effective task performance and modulates the levels of workload, sickness and presence. The overall findings revealed that multisensory feedback plays an important role in the overall perception and fidelity evaluations of VR systems and as such user experience needs to be included when investigating the effectiveness of sensory feedback signals. Throughout this thesis it was consistently shown that subjective measures of user perception in VR are directly comparable to the objective measures of performance and therefore both should be used in order to obtain a robust results when investigating the effectiveness of VR systems. This conceptual framework can provide an effective method to study human perception, which can in turn provide a deeper understanding of the environmental and cognitive factors that can influence the overall user experience, in terms of fidelity requirements, in virtual reality environments.