<![CDATA[This study synthesizes recent empirical research (2015–2025) on the relationships among cognitive load, working memory, metacognitive regulation, self-regulated learning (SRL), and mathematics anxiety in university-level mathematics. Although Cognitive Load Theory (CLT), Working Memory frameworks, and Metacognitive Regulation models have substantially advanced understanding of mathematical information processing, their application to advanced university mathematics remains conceptually fragmented. Through a systematic review of indexed literature, this paper consolidates evidence on how metacognitive strategies—planning, monitoring, and evaluation—interact with cognitive resources and affective factors across mathematical subdomains such as calculus, abstract algebra, statistics, and geometry. The findings indicate that metacognitive regulation supports cognitive load management, that self-regulated learning mediates the relationship between motivational beliefs and performance, and that mathematics anxiety disrupts both working memory efficiency and metacognitive monitoring. Empirical evidence further suggests that cognitive-metacognitive strategy training, particularly when embedded within instructional design or digital learning environments, is more consistently associated with performance gains than purely affective interventions. Rather than proposing a wholly new theoretical framework, this study integrates and clarifies existing perspectives to outline directions for a more holistic model of university mathematics learning that accounts for cognitive, metacognitive, motivational, and emotional dimensions. The review concludes by identifying implications for instructional design and priorities for future empirical research.]]>
