Mathematical Ability Prediction Model Based on Psychological and Cognitive Characteristics

This study integrates multi-dimensional psychology variables and cognitive ability indicators to construct a predictive model for mathematical ability and explore the underlying mechanisms of individual differences in mathematical performance. By collecting data on core psychological variables (e.g., motivation, anxiety, attention, and cognitive strategies) alongside mathematical achievement scores, we aim to form a theoretical foundation of individual characteristics supportive strategies for educational interventions. We will utilize machine learning methods to compare the predictive efficacy of various feature and feature combinations, identifying the key psychological and cognitive indicators with the highest explanatory power. On this basis, the study aims to build a stable, generalizable predictive model and evaluate its potential for the early identification of students with potential learning difficulties. We hope that through questionnaires, simple behavioral experiments, and functional magnetic resonance imaging (fMRI), this study will provide evidence for the psychological mechanisms of math ability variance, reveal the neural basis of human numerical cognition, provide a scientific basis for the formation mechanisms of the number concept and neural interventions for cognitive dysfunction, and offer a feasible theoretical foundation for educational interventions and personalized support strategies.