Novobo: Supporting Teachers' Peer Learning of Instructional Gestures by Teaching a Mentee AI-Agent Together

2025-05-23

AI Summary

Current teacher gesture-training approaches are often time-intensive, isolated, and overly structured, hindering peer sharing and co-construction of tacit, embodied knowledge. To address this, we propose Novobo—a teachable AI apprentice that pioneers a collaborative pedagogy wherein teachers instruct the AI through natural bodily actions and verbal directives, thereby externalizing, exchanging, and internalizing gesture knowledge. Novobo integrates multimodal understanding (speech + pose), generative motion modeling, human-AI feedback loops, and collaborative dialogue guidance. In 10 co-design workshops involving 30 in-service teachers, Novobo significantly deepened gesture knowledge sharing and fostered localized pedagogical consensus. This work transcends conventional one-way training paradigms, empirically validating the pivotal role of teachable AI in supporting embodied professional development for educators.

Technology Category

• Intelligent Robots: Embodied AI
• Humans and AI: Human-AI Collaboration / Human-AI Teaming
• Machine Learning: Neuro-Symbolic Learning

Application Category

• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

Instructional gestures are essential for teaching, as they enhance communication and support student comprehension. However, existing training methods for developing these embodied skills can be time-consuming, isolating, or overly prescriptive. Research suggests that developing these tacit, experiential skills requires teachers' peer learning, where they learn from each other and build shared knowledge. This paper introduces Novobo, an apprentice AI-agent stimulating teachers' peer learning of instructional gestures through verbal and bodily inputs. Positioning the AI as a mentee employs the learning-by-teaching paradigm, aiming to promote deliberate reflection and active learning. Novobo encourages teachers to evaluate its generated gestures and invite them to provide demonstrations. An evaluation with 30 teachers in 10 collaborative sessions showed Novobo prompted teachers to share tacit knowledge through conversation and movement. This process helped teachers externalize, exchange, and internalize their embodied knowledge, promoting collaborative learning and building a shared understanding of instructional gestures within the local teaching community. This work advances understanding of how teachable AI agents can enhance collaborative learning in teacher professional development, offering valuable design insights for leveraging AI to promote the sharing and construction of embodied and practical knowledge.

Problem

• Enhancing teacher peer learning of instructional gestures
• Addressing limitations in current gesture training methods
• Using AI to facilitate embodied knowledge sharing

Innovation

• AI mentee stimulates peer learning via gestures
• Learning-by-teaching paradigm enhances reflection
• Teachers evaluate and demonstrate AI-generated gestures