Shadaqnas Dewarif Tri Anggoro

Geometry problem-solving requires learners to coordinate conceptual knowledge, spatial representation, strategy use and the regulation of cognition, motivation and effort. This systematic review examined how self-regulated learning (SRL) operates in geometry problem-solving and which learner, task, social, instructional and developmental conditions shape it. A Scopus search reported according to PRISMA 2020 identified 1,298 records; database limits left 529, and eligibility screening retained 32 empirical studies published from 2021 to 2026. Twelve studies involved undergraduates or pre-service teachers, seven involved school pupils, one involved kindergarten children and 12 did not state the educational level in the extracted records. Study quality was appraised with the Mixed Methods Appraisal Tool, and thematic synthesis identified four themes: planning and strategic regulation; monitoring, control and strategy adaptation; learner characteristics and outcomes; and instructional scaffolding and technology. Successful performance was associated with problem representation, strategy selection, monitoring, error detection, adaptation and evaluation. Diagnostic and elaborated feedback, content-focused peer feedback and supportive scaffolding were more informative than the presence of technology. Evidence on learner characteristics was mainly correlational, and evidence on young learners was thin. The review proposes a Process-Oriented Framework of Self-Regulated Geometry Problem-Solving that links input conditions, forethought, performance regulation, reflection, and outcomes and it is best supported for adolescent and adult learners. Future research should combine self-report with behavioural, longitudinal and process-tracing evidence and develop age-appropriate SRL measures for young children.