Quantifying time perception during virtual reality gameplay using a multimodal biosensor-instrumented headset: a...

Area of Science:

• Neuroscience
• Human-Computer Interaction
• Virtual Reality

Background:

• Time perception in virtual reality (VR) significantly impacts user experience, influencing immersion and engagement.
• Objective measurement of time perception in VR is limited, with most studies relying on subjective questionnaires.
• Understanding how physiological signals relate to time perception is crucial for designing more engaging VR environments.

Purpose of the Study:

• To objectively characterize changes in time perception within a virtual reality environment.
• To explore the utility of multimodal biosensor data for predicting subjective time perception.
• To identify key physiological features correlating with altered time perception in VR.

Main Methods:

• Utilized a multimodal biosensor-embedded VR headset to collect electroencephalography (EEG), electrooculography (EOG), electrocardiography (ECG), and head movement data.
• Recruited eight gamers to play a commercial action game involving puzzle-solving and combat scenarios.
• Applied a two-step feature selection process to biosignals and employed Gaussian process regression to predict time perception ratings.

Main Results:

• Features extracted from EEG, EOG, ECG, and head movement data were successfully used to predict time perception.
• The developed models significantly outperformed chance in predicting subjective ratings of time distortion and time loss.
• Identified key physiological features that correlate with altered temporal experiences in VR.

Conclusions:

• Multimodal biosensor data offers a viable objective method for assessing time perception in virtual reality.
• Insights from physiological features can inform the design of more immersive and engaging VR experiences.
• This approach provides a foundation for developing adaptive VR systems that respond to user's temporal experiences.