The purpose of this research is to investigate: (1) how students' mathematical dispositions differ depending on whether they were taught using a Problem-Based Learning model or an Ordinary Learning model; (2) how the mathematical dispositions of students are affected by the relationship between the learning style and Previous Students' Ability. This study is a quantitative investigation employing procedures that are semi experimental. The instruments of this research are problem solving ability tests and student disposition questionnaires. The participants in this research project are all 222 students enrolled in class VIII at SMP Negeri 5 Stabat during the academic year 2022/2023. There are a total of 222 individuals. The method of sampling utilized for this investigation was a straightforward random sampling method, and the total number of students comprising the sample was calculated using the Issac and Michael formula. Essay examinations and questionnaires were the research tools that were utilized in this investigation. A Two Way ANOVA was utilized in the analysis of the results. The findings of the research indicate the following: (1) the mathematical disposition of students in the class that uses problem-based learning is higher than that of students in the class that uses ordinary learning; (2) there is no interaction between the learning model and early mathematical abilities (high, medium, and low) on students' mathematical dispositions.Keywords: Problem Based Learning, Quasi Experiment, Mathematical Dispositions