Machine Learning-Guided High-Definition Transcranial Direct Current Stimulation Prevents Cybersickness
Date
Authors
Supervisor
Item type
Journal Article
Degree name
Journal Title
Journal ISSN
Volume Title
Publisher
Springer Science and Business Media LLC
Abstract
Extended reality (XR) environments, such as simulators, augmented reality, and virtual reality are major techniques in contemporary AI and entertainment systems. Cybersickness (CS) is a motion-sickness experienced by many users of XR. CS causes debilitating nausea, disorientation, and oculomotor issues. Treatment and prevention for motion-sickness typically involves drugs with sedative properties that impair task performance. These drugs are non-specific to CS and counter intuitive for enabling activity within XR. Our paper finds that there are specific spatiotemporal patterns of brain activity in certain functional networks related to CS and offers a method for the analysis of these patterns. The method can predict CS ahead of its onset and most importantly it suggests what intervention to apply in order to prevent CS in individuals. We apply a novel approach to CS prevention by using our previously developed spiking neural network (SNN) method, which can predict CS using electroencephalogram (EEG) pre-VR usage, before applying neuromodulation to disrupt CS-related functional networks in the brain. This approach provides an additional layer of screening before intervention with high-definition transcranial direct current stimulation (HD-tDCS). The study recruited healthy CS susceptible participants (9 male, 10 female, n = 19, 18–36 years old) and used a within-subjects design. EEG (32-channel, 10–10-configuration) was monitored at seated-rest and processed through the SNN for CS prediction. Immediately following a positive prediction, either sham, anodal or cathodal HD-tDCS was applied at the Cz area (5-min, 1.5 mA, 30 s-ramp-up/down) with subsequent 10-min VR immersion to record CS events. Main results: Cathodal stimulation yielded a significantly higher number of successful preventions compared to anodal (*p = 0.01) and sham (***p = 0.00056), achieving a large effect size (> 0.8) with a 47% reduction in CS likelihood. Significance: The treatment was hypothesized to work through disruption of activity at the motor processing and planning regions under Cz. The area appears to be a marker of ongoing CS susceptibility, and also a contributor towards the condition.Description
Keywords
46 Information and Computing Sciences, 4608 Human-Centred Computing, Networking and Information Technology R&D (NITRD), Bioengineering, Neurosciences, Prevention, Clinical Research, 0801 Artificial Intelligence and Image Processing, 0909 Geomatic Engineering, 1702 Cognitive Sciences, Human Factors, 4607 Graphics, augmented reality and games
Source
Virtual Reality, ISSN: 1359-4338 (Print); 1434-9957 (Online), Springer Science and Business Media LLC, 29(3), 94-. doi: 10.1007/s10055-025-01160-x
Publisher's version
Rights statement
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.