Cybersickness and Its Severity Arising from Virtual Reality Content: A Comprehensive Study

Virtual reality (VR) experiences often elicit a negative effect, cybersickness, which results in nausea, disorientation, and visual discomfort. To quantitatively analyze the degree of cybersickness depending on various attributes of VR content (i.e., camera movement, field of view, path length, fram...

Full description

Bibliographic Details
Main Authors: Heeseok Oh, Wookho Son
Format: Article
Language:English
Published: MDPI AG 2022-02-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/22/4/1314
_version_ 1797476782527479808
author Heeseok Oh
Wookho Son
author_facet Heeseok Oh
Wookho Son
author_sort Heeseok Oh
collection DOAJ
description Virtual reality (VR) experiences often elicit a negative effect, cybersickness, which results in nausea, disorientation, and visual discomfort. To quantitatively analyze the degree of cybersickness depending on various attributes of VR content (i.e., camera movement, field of view, path length, frame reference, and controllability), we generated cybersickness reference (CYRE) content with 52 VR scenes that represent different content attributes. A protocol for cybersickness evaluation was designed to collect subjective opinions from 154 participants as reliably as possible in conjunction with objective data such as rendered VR scenes and biological signals. By investigating the data obtained through the experiment, the statistically significant relationships—the degree that the cybersickness varies with each isolated content factor—are separately identified. We showed that the cybersickness severity was highly correlated with six biological features reflecting brain activities (i.e., relative power spectral densities of Fp1 delta, Fp 1 beta, Fp2 delta, Fp2 gamma, T4 delta, and T4 beta waves) with a coefficient of determination greater than 0.9. Moreover, our experimental results show that individual characteristics (age and susceptibility) are also quantitatively associated with cybersickness level. Notably, the constructed dataset contains a number of labels (i.e., subjective cybersickness scores) that correspond to each VR scene. We used these labels to build cybersickness prediction models and obtain a reliable predictive performance. Hence, the proposed dataset is supposed to be widely applicable in general-purpose scenarios regarding cybersickness quantification.
first_indexed 2024-03-09T21:07:32Z
format Article
id doaj.art-d3f0791e179e499aa8c86d9f6102a39d
institution Directory Open Access Journal
issn 1424-8220
language English
last_indexed 2024-03-09T21:07:32Z
publishDate 2022-02-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj.art-d3f0791e179e499aa8c86d9f6102a39d2023-11-23T21:57:36ZengMDPI AGSensors1424-82202022-02-01224131410.3390/s22041314Cybersickness and Its Severity Arising from Virtual Reality Content: A Comprehensive StudyHeeseok Oh0Wookho Son1Department of Applied AI, Hansung University, Seoul 02876, KoreaSW&Content Research Lab., ETRI, Daejeon 34129, KoreaVirtual reality (VR) experiences often elicit a negative effect, cybersickness, which results in nausea, disorientation, and visual discomfort. To quantitatively analyze the degree of cybersickness depending on various attributes of VR content (i.e., camera movement, field of view, path length, frame reference, and controllability), we generated cybersickness reference (CYRE) content with 52 VR scenes that represent different content attributes. A protocol for cybersickness evaluation was designed to collect subjective opinions from 154 participants as reliably as possible in conjunction with objective data such as rendered VR scenes and biological signals. By investigating the data obtained through the experiment, the statistically significant relationships—the degree that the cybersickness varies with each isolated content factor—are separately identified. We showed that the cybersickness severity was highly correlated with six biological features reflecting brain activities (i.e., relative power spectral densities of Fp1 delta, Fp 1 beta, Fp2 delta, Fp2 gamma, T4 delta, and T4 beta waves) with a coefficient of determination greater than 0.9. Moreover, our experimental results show that individual characteristics (age and susceptibility) are also quantitatively associated with cybersickness level. Notably, the constructed dataset contains a number of labels (i.e., subjective cybersickness scores) that correspond to each VR scene. We used these labels to build cybersickness prediction models and obtain a reliable predictive performance. Hence, the proposed dataset is supposed to be widely applicable in general-purpose scenarios regarding cybersickness quantification.https://www.mdpi.com/1424-8220/22/4/1314virtual reality (VR)VR human factorcybersickness analysisVR cybersickness dataset
spellingShingle Heeseok Oh
Wookho Son
Cybersickness and Its Severity Arising from Virtual Reality Content: A Comprehensive Study
Sensors
virtual reality (VR)
VR human factor
cybersickness analysis
VR cybersickness dataset
title Cybersickness and Its Severity Arising from Virtual Reality Content: A Comprehensive Study
title_full Cybersickness and Its Severity Arising from Virtual Reality Content: A Comprehensive Study
title_fullStr Cybersickness and Its Severity Arising from Virtual Reality Content: A Comprehensive Study
title_full_unstemmed Cybersickness and Its Severity Arising from Virtual Reality Content: A Comprehensive Study
title_short Cybersickness and Its Severity Arising from Virtual Reality Content: A Comprehensive Study
title_sort cybersickness and its severity arising from virtual reality content a comprehensive study
topic virtual reality (VR)
VR human factor
cybersickness analysis
VR cybersickness dataset
url https://www.mdpi.com/1424-8220/22/4/1314
work_keys_str_mv AT heeseokoh cybersicknessanditsseverityarisingfromvirtualrealitycontentacomprehensivestudy
AT wookhoson cybersicknessanditsseverityarisingfromvirtualrealitycontentacomprehensivestudy