Effects of body visualization on performance in head-mounted display virtual reality

• Published in: PloS one 2020-09, Vol.15 (9), p.e0239226-e0239226
• Main Authors: Pastel, Stefan, Chen, Chien-Hsi, Petri, Katharina, Witte, Kerstin
• Format: Article
• Language: English
• Subjects: Behavior; Biology and Life Sciences; Body parts; Computer and Information Sciences; Computer applications; Engineering; Engineering and Technology; Feet; Grasping; Helmet mounted displays; Human factors; Human performance; Medicine and Health Sciences; Motor task performance; Movement (Physiology); Occlusion; Parameters; Physiological research; Realism; Social Sciences; Sport science; Sports; Supervision; Throwing; Virtual environments; Virtual reality; Visualization; Visualization (Computer)
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0239226

Although there are many virtual reality (VR) applications in sports, only a handful of studies visualized the whole body. There is still a lack of understanding, how much of the own body must be visualized in the head-mounted display (HMD) based VR, to ensure fidelity and similar performance outcome as in the real-world. In the current study, 20 young and healthy participants completed three tasks in a real and virtual environment: balance task, grasping task, and throwing task with a ball. The aim was to find out the meaning of the visualization of different body parts for the quality of movement execution and to derive future guidelines for virtual body presentation. In addition, a comparison of human performance between reality and VR, with whole-body visualization was made. Focusing on the main goal of the current study, there were differences within the measured parameters due to the visualization of different body parts. In the balance task, the differences within the VR body visualization consisted mainly through no-body visualization (NB) compared to the other visualization types defined as whole-body (WB), WB except feet (NF), as well as WB except feet and legs (NLF). In the grasping task, the different body visualization seemed to have no impact on the participants' performances. In the throwing task, the whole-body visualization led to higher accuracy compared to the other visualization types. Regarding the comparison between the conditions, we found significant differences between reality and VR, which had a large effect on the parameters time for completion in the balance and grasping task, the number of foot strikes on the beam in the balance task, as well as the subjective estimation of the difficulty for all tasks. However, the number of errors and the quality of the performances did not differ significantly. The current study was the first study comparing sports-related tasks in VR and reality with further manipulations (occlusions of body parts) of the virtual body. For studies analyzing perception and sports performance or for VR sports interventions, we recommend the visualization of the whole body in real-time.