New Framework Published for Human-AI Collaboration in Virtual Reality Design Using Large Language Models

Aug 4, 2026

A study published on August 4, 2026, introduces a new framework using the VR Dimensionality Framework (8Df) to structure human-AI co-ideation in Virtual Reality design. The research integrates the 8Df across eight dimensions—including spatial, sensory, and psychological—into a dialogic workflow. Testing ChatGPT, Gemini, and DeepSeek on a shared VR narrative design task, the study reveals that ChatGPT is implementation-oriented, Gemini is facilitative, and DeepSeek relies on explicit reasoning.

VR Dimensionality Framework (8Df)

▪The Virtual Reality Dimensionality Framework (8Df) serves as a theoretical scaffold to structure human-AI co-ideation in Virtual Reality design.
▪The Virtual Reality Dimensionality Framework (8Df) guides experience design across eight interrelated dimensions: spatial, sensory, placeness, temporal, cultural, social, cognitive, and psychological.

Human-AI co-ideation workflow

▪Designers often lack structured frameworks that support both creativity and theoretical grounding in complex media workflows like Virtual Reality.
▪The study integrates the Virtual Reality Dimensionality Framework (8Df) into a dialogic co-ideation workflow to guide human-AI design processes.

LLM comparative performance analysis

▪DeepSeek displays explicit reasoning and a principled articulation of design within the framework-guided Virtual Reality design process.
▪Gemini demonstrates a facilitative and alignment-focused approach during the framework-guided Virtual Reality design process.
▪The comparative case study evaluates how ChatGPT, Gemini, and DeepSeek respond to the same framework-guided design process.
▪ChatGPT exhibits an implementation-oriented tendency when responding to the Virtual Reality Dimensionality Framework (8Df) design process.

VR narrative design application

▪The three Large Language Models ChatGPT, Gemini, and DeepSeek were evaluated and compared using a shared Virtual Reality narrative design task.
▪The research on human-AI co-ideation was supported by the Guangdong Higher Education Upgrading Plan and the Guangdong and Hong Kong Universities Joint Research Collaboration Scheme.