MusicScaffold: Bridging Machine Efficiency and Human Growth in Adolescent Creative Education through Generative AI

Zhejing Hu, Hong Kong Polytechnic University

Yan Liu, Hong Kong Polytechnic University

Zhi Zhang, Hong Kong Polytechnic University

Gong Chen

Bruce X.B. Yu, Zhejiang University

Junxian Li

Jiannong Cao, Hong Kong Polytechnic University

Date

09/2025

AI Overview

Why it matters

Adolescents often lack strategies to turn creative impulses into structured expressions. While generative AI enables rapid music creation, efficiency-driven tools risk fostering student dependency rather than developing lasting expressive skills and autonomy.

Study at a glance

Question: Can an AI scaffolding framework help middle school students develop structured music expression strategies and creative autonomy?
Setting and data: Middle school music classes, 330 students aged 12–14 across a 60-student exploratory study and 270-student comparative study over 4 weeks.
Method: Mixed-methods design with an exploratory study and a three-group comparative classroom experiment with surveys and interviews.
AI tools: Suno, Udio, GPT-4o, MusicScaffold

Key findings

Students using MusicScaffold achieved higher prompt specificity (average score 3.00 versus 1.53 and 1.80) and incorporated significantly broader musical dimensions than control groups.
MusicScaffold shifted student behavior from passive regeneration toward strategic prompt refinement, accounting for 51.1% of adjustments compared to only 1.1% to 2.2% in control groups.
Adolescents using MusicScaffold demonstrated significantly higher self-efficacy in musical expression (mean 3.60 versus 2.01 and 2.37) and voluntary after-class participation that doubled over four weeks.

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

Adolescence is marked by strong creative impulses but limited strategies for structured expression, often leading to frustration or disengagement. While generative AI lowers technical barriers and delivers efficient outputs, its role in fostering adolescents' expressive growth has been overlooked. We propose MusicScaffold, the first adolescent-centered framework that repositions AI as a guide, coach, and partner, making expressive strategies transparent and learnable, and supporting autonomy. In a four-week study with middle school students (ages 12--14), MusicScaffold enhanced cognitive specificity, behavioral self-regulation, and affective confidence in music creation. By reframing generative AI as a scaffold rather than a generator, this work bridges the machine efficiency of generative systems with human growth in adolescent creative education.