The evolution of dynamic storytelling in game design has been significantly influenced by advances in procedural generation techniques, allowing for the creation of adaptive and player-driven narratives. While traditional narratives in games follow a linear path, procedural generation enables stories to evolve based on player choices, offering unique and personalized experiences. This research aims to explore the potential of procedural narrative generation in enhancing player engagement and learning outcomes in educational games. The study employs a mixed-methods approach, combining quantitative analysis of player engagement, knowledge retention, and problem-solving abilities with qualitative insights from participant interviews. A total of 240 participants were divided into three groups: those who interacted with games featuring fully procedural narratives, semi-procedural narratives, and traditional static narratives. The results show that players in the fully procedural narrative group demonstrated higher levels of engagement and cognitive performance, particularly in knowledge retention and problem-solving. These findings indicate that procedural narratives can significantly enhance player immersion and educational outcomes, providing a more interactive and personalized gaming experience. The study concludes that the integration of procedural generation in narrative design offers a promising avenue for improving educational game design, fostering deeper engagement and more effective learning.