This study examines how students interact with a Gamified Online Judge through cognitive psychology strategies, repeated and distributed practice, and how these patterns contribute to performance improvements. Using a data-driven approach, we analyze students’ usage patterns while considering their prior knowledge and HEXAD profiles to understand how different patterns relate to performance gains. Feature selection resulted in the choice of Number of Attempts and Grouped Hours Between Attempts for further analysis, as these were significantly related to performance gains. Clustering analysis revealed six student usage patterns with varied practice and spacing behaviors, highlighting the complex interplay between repeated and distributed practice strategies by demonstrating that usage patterns influence performance gains differently depending on students’ prior knowledge but not HEXAD user profiles. Thus, this work offers insights into how adaptive learning environments can better support diverse student needs, emphasizing the importance of aligning gamified assessment design with cognitive strategies to personalize learning experiences effectively.