Beyond personality traits: a motivation–self-regulation model of mathematical problem-solving through self-efficacy and mathematical thinking

Purpose: This study aims to understand how learning-related traits and process variables collectively influence high school students' mathematical problem-solving ability; and to verify the mediating roles of mathematical self-efficacy and mathematical thinking. Methods: Based on questionnaire and mathematical performance test data from 1,183 high school students in Shanghai, the study employs partial least squares structural equation modeling (PLS-SEM) for analysis. Results: The results show that the model has high explanatory power for mathematical self-efficacy, mathematical thinking, and mathematical problem-solving ability; motivation to math and self-regulated learning are core upstream predictors of problem-solving performance, while mathematical self-efficacy and mathematical thinking serve as the closest direct predictors of problem-solving and significantly mediate the effects of motivation and self-regulated learning on problem-solving ability. In contrast, personality traits exhibit relatively weak direct and mediating effects after controlling for prior academic performance and the aforementioned learning variables. Implications: The contribution of this study lies not only in integrating motivation, self-regulation, affective beliefs, and higher-order mathematical thinking into a unified structural framework, but also in clarifying the relative explanatory roles of distal personality traits and proximal learning processes, showing that mathematical problem-solving is shaped through both affective-belief and cognitive pathways, and demonstrating that the explanatory advantage within the present model lies primarily in domain-specific, modifiable learning processes rather than broad personality dispositions.