Exploring User Perceptions of Virtual Reality Scene Design in Metaverse Learning Environments

Rahatara Ferdousi, Mohammed Faisal, Fedwa Laamarti, Chunsheng Yang, Abdulmotaleb El-Saddik

TL;DR

This study investigates how the aesthetic design of Virtual Reality (VR) scenes—ranging from traditional classrooms to minimalist underwater environments—impacts a learner's cognitive load and preference in the Metaverse. The researchers found that minimalist, high-contrast designs significantly reduce mental effort and frustration, especially for users with existing vision issues like myopia.

What Problem Does This Paper Address?

While the "Metaverse" is often touted as the next frontier for immersive education, most research focuses on avatars or interactive tools rather than the environment itself. Many developers default to replicating real-world settings (like a standard wood-paneled lecture hall), assuming familiarity equals comfort.

However, VR presents unique physiological challenges. The authors address a critical gap: how do specific design elements—style, color, texture, and background—actually affect the "brain power" required to learn? Furthermore, they explore how these design choices interact with a user’s physical health, specifically vision-related issues, which are often overlooked in VR UX research.

Methods

The researchers conducted a controlled user study with 16 participants using Meta Quest 2 headsets and the Spatial app. The core of the experiment was a head-to-head comparison of two distinct environments:

  1. Scene-01 (Traditional): Modeled after a standard lecture hall with neutral tones (beige, brown), wooden textures, and familiar objects like desks and projectors, set against a campus background.
  2. Scene-02 (Non-traditional): A minimalist, "aqua reef" environment featuring bright tones (pearl white, light blue), smooth surfaces, and unique objects like shells and pearls.

Participants watched one-minute educational tutorials in each scene. To measure the impact, the team used:

  • Seven-point Likert Scales: To rank scene preference.
  • NASA TLX (Task Load Index): A standard multidimensional scale to assess mental demand, physical demand, temporal demand, performance, effort, and frustration.
  • Mann-Whitney U Test: To statistically validate the differences in user preferences.

Datasets

As this is a primary human-computer interaction (HCI) study, the "dataset" consists of the qualitative and quantitative responses from the 16 recruited participants.

  • Demographics: Average age of 25.6; diverse VR expertise (62.5% somewhat familiar, 32.5% novices).
  • Vision Profile: Crucially, 50% of the group had vision-related issues (37.5% myopia, 6.25% migraines, 6.25% hypermetropia), providing a robust baseline for accessibility analysis.

Key Insights and Ideas

  • Minimalism Beats Realism: Contrary to the urge to make VR look "just like the real world," the study suggests that a minimalist design with plain textures and fewer objects is superior for learning. It reduces visual "noise," allowing the brain to focus on the content.
  • The "Soothing" Factor: Participants specifically noted that the blue/white contrast of the underwater scene was "eye-soothing" and "calm." This suggests that environmental affect (how a space makes you feel) directly influences cognitive bandwidth.
  • The Accessibility Tax: Users with vision issues (like myopia) experienced significantly higher cognitive loads across the board. In the "busy" traditional scene, these users reported feeling overwhelmed, with some experiencing headaches or "zoomed out" vision changes.
  • Design as a Mitigant: The study found that the minimalist Scene-02 helped lower the "accessibility tax," reducing the overall cognitive load for users with vision issues compared to the traditional scene.

Key Results

  • Preference: Scene-02 was the clear winner with a sample average ranking of 5.62 compared to 3.93 for Scene-01.
  • Statistical Significance: The Mann-Whitney U test yielded a p-value of 0.009729, well below the 0.05 threshold, confirming the preference wasn't due to chance.
  • Cognitive Load (NASA TLX):
    • Frustration: Dropped significantly in Scene-02 (from a score of 52.22 down to 25 for those with vision issues).
  • Mental Demand: Users without vision issues reported a score of 30.00 in the traditional scene vs. 28.57 in the minimalist scene.
  • Performance: Users felt they performed better in the minimalist scene (scoring 76.88 vs 71.88).

Limitations and Future Work

The authors acknowledge a few key constraints:

  • Sample Size: With only 16 participants, the results are an exploratory "snapshot" rather than a definitive universal law.
  • Duration: The tutorials were only one minute long. Cognitive load and "VR fatigue" might manifest differently during a full 60-minute university lecture.
  • TLX Version: The study used the Raw NASA TLX rather than the weighted version, which might overlook which specific dimensions (e.g., mental vs. temporal) participants valued most.

Future Directions: The team plans to test longer, real-life class durations and suggests that future Metaverse platforms should allow users to choose their own environment "skins" to match their personal comfort and physiological needs.

Summary updated

Abstract

Metaverse learning environments allow for a seamless and intuitive transition between activities compared to Virtual Reality (VR) learning environments, due to their interconnected design. The design of VR scenes is important for creating effective learning experiences in the Metaverse. However, there is limited research on the impact of different design elements on user’s learning experiences in VR scenes. To address this, a study was conducted with 16 participants who interacted with two VR scenes, each with varying design elements such as style, color, texture, object, and background, while watching a short tutorial. Participant rankings of the scenes for learning were obtained using a seven-point Likert scale, and the Mann-Whitney U test was used to validate differences in preference between the scenes. The results showed a significant difference in preference between the scenes. Further analysis using the NASA TLX questionnaire was conducted to examine the impact of this difference on cognitive load, and participant feedback was also considered. The study emphasizes the importance of careful VR scene design to improve the user’s learning experience.
BibTeX
@inproceedings{ferdousi2024exploring,
  author = {Rahatara Ferdousi and Mohammed Faisal and Fedwa Laamarti and Chunsheng Yang and Abdulmotaleb El-Saddik},
  title = {{Exploring User Perceptions of Virtual Reality Scene Design in Metaverse Learning Environments}},
  booktitle = {ICCE},
  year = {2024},
  url = {https://openreview.net/forum?id=dJh7pQzKK2},
}

Loading related content...

Figure