Integrating NotebookLM, Source-Grounded Artificial Intelligence, and Storytelling for Enhanced Physiology Education: A Perspective

Physiology education is often faced with the challenge of engaging first-year medical students since the subject matter is abstract and mechanistic. Traditional teaching may not be able to make physiological mechanisms applicable to daily life, leading to the challenge of retaining and applying such information for the students. Storytelling is quickly gaining recognition as a powerful pedagogical tool that can make physiological principles personal, promote wonder, and make learning easier by placing information in everyday contexts. Nevertheless, it is difficult for teachers to compose scientifically accurate and engaging stories within curricular and time constraints. Google's NotebookLM offers a novel solution to this difficulty. Unlike general-purpose Large Language Models (LLMs) such as ChatGPT, NotebookLM enables users to upload specific documents and generate content based solely on those sources. This minimizes hallucinations and ensures created stories are also compatible with trusted academic sources. In this paper, we report on how the open-source “Anatomy and Physiology 2e” textbook was integrated into NotebookLM to create personalized, context-aware storytelling content for physiology pedagogy. This would enable teachers to create accurate and interesting stories instantaneously, making storytelling integration with lectures or tutorials seamless. Through quoting a trustworthy source, teachers can also be open and scientifically trustworthy. With the improvement of LLMs, it is feasible for tools like NotebookLM to transform physiology teaching by rendering complex mechanisms accessible, memorable, and directly usable for students with the power of narrative, aided by trustworthy sources.