Scene Design

The design of immersive learning scenes in Classlet follows a structured, room-based layout that aligns spatial navigation with cognitive flow. Each scene is segmented into distinct “rooms” or learning zones, typically containing around five non-player characters (NPCs) who act as embedded learning agents—delivering narrative prompts, questions, or contextual information.

This custom home scene, developed for a European History project at Hong Kong PolyU, functions as an interactive entry hub. Learners can explore the visually rich environment and teleport to sub-zones, such as exhibits on the Roman Empire, enabling non-linear navigation and personalized learning journeys.

Custom virtual scenes can be institutionally branded, as shown here with Lingnan University. This allows educators to reinforce identity, provide contextual relevance, and create a more immersive and recognizable learning environment.

Scene Design

These room-based progressions align with instructional design principles by scaffolding complexity and orienting learners through manageable cognitive units. Whether visually mapped or subtly embedded, each spatial segment is intentionally crafted to anchor content, reduce disorientation, and guide learners toward mastery through exploratory yet structured interaction.

Instructors designing VR scenes should begin with clear learning outcomes and align each room to a specific concept or objective. For example, one room may focus on content familiarization (e.g., visuals or text), the next on applied decision-making (e.g., NPC dilemmas or grab tasks), and a final one on reflection or synthesis. This spatial segmentation supports cognitive load management, allowing learners to process complex topics incrementally while staying immersed.

This image illustrates a structured scene design approach for immersive learning, organized into three thematic "rooms": Room 1 (Temple), Room 2 (Entrance & City View), and Room 3 (Colosseum). Each space is mapped with intentional learning goals, enabling students to navigate historical content progressively while maintaining spatial orientation and cognitive flow.

Each room typically includes 3–5 NPCs, with distinct conversational roles. Instructors can configure NPCs for different task types: delivering prompts, asking questions, presenting dilemmas, or reinforcing concepts through branching conversations. By using either pre-scripted dialogue or GPT-generated responses, educators can scaffold student reasoning while preserving a sense of exploration and agency.

To support learner autonomy, scenes can include navigational cues such as lighting, signs, spatial audio, or visual "milestones" like portals or clear entryways. These cues help learners orient themselves without breaking immersion. In exploratory scenes, optional guidance avatars or visible path indicators can help those who may need more direction, while still allowing more advanced students to roam freely.

This sci-fi city scene is structured with an "invisible" three-room progression, guiding the learner from the opening dialogue through a linear narrative path that culminates at the end of the alley. Each section introduces new clues and characters, supporting scenario-based learning embedded in an exploratory urban environment.

What you can give us

• Suggested Scene Context or Environment: Whether you'd like the scene set in a city, lab, clinic, courtroom, space station, ancient site, etc.

• Content Topics or Curriculum Focus: Key themes, concepts, or units (e.g., "European colonization", "food safety hazards", "AI ethics").

• Scene Flow Suggestions: If you imagine a progression (e.g., start in a lobby, go to 3 sub-rooms), let us know — or we can suggest one for you.

• Media Assets: Images, diagrams, or short videos you’d like embedded in the scene.