Abstract

This review proposes an integrative theoretical framework that explains the development of stable musical score memory as a function of both cognitive architecture and self-regulatory processes. Grounded in the multiple memory systems theory and the three-phase self-regulated learning (SRL) model, the framework conceptualizes musical memorization as a dynamic, recursive process involving the selective activation and coordination of five long-term memory systems—semantic, emotional, perceptual, procedural, and narrative—across the SRL phases of forethought, performance, and reflection. Drawing upon recent empirical findings in cognitive neuroscience, affective memory research, and metacognitive theory (2021–2025), this review introduces a matrix-based model that maps phase-specific memory activation and regulation mechanisms. The model elucidates how SRL functions as a meta-regulatory system that governs memory system engagement to support goal setting, skill acquisition, expressive performance, and adaptive evaluation. Pedagogical implications include the design of SRL-informed instructional strategies that target memory-specific learning functions, enhance learner autonomy, and improve long-term retention. The review concludes by outlining future research directions involving longitudinal tracking of memory dynamics, neurocognitive measurement of learning phases, and AI-enhanced scaffolding of SRL in music education. By reframing musical memorization as a cognitively distributed and strategically regulated process, this work offers a novel contribution to the interdisciplinary understanding of memory-based performance learning.