Embodying the Disney Method: A Multi-Avatar Conversational Interface for Structured Brainstorming
Open Access
- 21 July 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Abstract
Conversational AI systems increasingly support creative collaboration, yet most interfaces rely on a single conversational agent. This paper presents a multi-avatar conversational system that structures brainstorming through distinct reasoning roles. Inspired by the Disney Method, the system represents the Dreamer, Realist, and Critic as embodied conversational agents. Users engage with these roles sequentially to explore ideas, develop implementation strategies, and critically evaluate proposals within a shared problem-solving process. The prototype enables voice-driven interaction with embodied MetaHuman avatars in an Unreal Engine environment and is implemented using a fully open-source conversational pipeline. By representing reasoning perspectives as distinct conversational avatars, the system illustrates how multi-avatar interaction can support structured role-based brainstorming.Keywords
Funding Information
- JSPS KAKENHI (JP25KF0113)
This publication has 7 references indexed in Scilit:
- Conversational Historical Avatars in Virtual Reality Powered by Local Large Language Models2025
- A Latency-Optimized LLM-based Multimodal Dialogue System for Embodied Conversational Agents in VR2025
- Bridging Industrial Expertise and XR with LLM-Powered Conversational Agents2025
- Embodied Conversational Agents in Extended Reality: A Systematic ReviewIEEE Access, 2025
- Exploring LLM-Powered Role and Action-Switching Pedagogical Agents for History Education in Virtual Reality2025
- RAGatar: Enhancing LLM-driven Avatars with RAG for Knowledge-Adaptive Conversations in Virtual Reality2025
- Character-LLM: A Trainable Agent for Role-Playing2023