Pedaco-Gen: Scaffolding Pedagogical Agency In Human-AI Collaborative Video Authoring

Injun Baek, Seoul National University

Yearim Kim, Seoul National University

Nojun Kwak, Seoul National University

Date

02/2026

AI Overview

Why it matters

Text-to-video AI models are often optimized for visual fidelity rather than instructional efficacy. Without granular educator control, these black-box pipelines risk generating videos that violate established instructional principles and increase extraneous cognitive load.

Study at a glance

Question: Can a human-AI collaborative authoring system improve the pedagogical quality of AI-generated educational videos?
Setting and data: Online web-based study with 23 Korean education professionals (teachers and education majors).
Method: Within-subject user study comparing baseline generation to collaborative authoring using quantitative ratings and qualitative thematic analysis.
AI tools: PedaCo-Gen, Gemini 2.5 Flash, Veo 3.1

Key findings

Videos created with PedaCo-Gen showed an average quality score increase from 3.07 to 3.86 out of 5 across multimedia learning principles.
The largest pedagogical improvements occurred in the pre-training principle (+0.86 points), the coherence principle (+0.84 points), and for abstract concept explanations (+1.17 points).
Educators rated the collaborative system highly for production efficiency (4.26 out of 5) and guide validity (4.04 out of 5), valuing the reduction in trial-and-error.

Summary drafted with AI from the full paper and may contain errors. See the full paper for details.

While advancements in Text-to-Video (T2V) generative AI offer a promising path toward democratizing content creation, current models are often optimized for visual fidelity rather than instructional efficacy. This study introduces PedaCo-Gen, a pedagogically-informed human-AI collaborative video generating system for authoring instructional videos based on Mayer's Cognitive Theory of Multimedia Learning (CTML). Moving away from traditional "one-shot" generation, PedaCo-Gen introduces an Intermediate Representation (IR) phase, enabling educators to interactively review and refine video blueprints-comprising scripts and visual descriptions-with an AI reviewer. Our study with 23 education experts demonstrates that PedaCo-Gen significantly enhances video quality across various topics and CTML principles compared to baselines. Participants perceived the AI-driven guidance not merely as a set of instructions but as a metacognitive scaffold that augmented their instructional design expertise, reporting high production efficiency (M=4.26) and guide validity (M=4.04). These findings highlight the importance of reclaiming pedagogical agency through principled co-creation, providing a foundation for future AI authoring tools that harmonize generative power with human professional expertise.