Forecasting Learning Transfer in History Education Games with Narrative Knowledge Structures

Abstract

The integration of serious games into history education has garnered substantial attention as a mechanism to foster deep cognitive engagement and structural understanding of complex past events. However, the extent to which these digital environments facilitate actual learning transfer remains a subject of intense academic debate. This paper investigates the predictive capacity of narrative knowledge structures in determining learning transfer outcomes through a comprehensive simulation study embedded within history education games. By designing a controlled, simulated environment that modulates the narrative coherence, agency, and structural complexity of historical events, this research seeks to isolate the cognitive pathways that enable students to apply learned historical concepts to novel, unencountered scenarios. The study contrasts high-narrative structures, which contextualize facts within human-centric, causally linked storylines, against low-narrative, encyclopedic presentations of the same historical data. Through the meticulous analysis of pre-intervention, in-game telemetry, and post-intervention transfer assessments, the simulation reveals that narrative knowledge structures significantly outperform traditional declarative models in promoting far transfer. The findings indicate that when learners encode historical variables as part of a coherent narrative schema, they are better equipped to abstract underlying principles, such as socio-economic causality and geopolitical dynamics, and map them onto contemporary or divergent historical contexts. This paper provides a robust evidentiary foundation for the design of future educational games, emphasizing that narrative architecture is not merely an engagement tool but a fundamental cognitive scaffold for actionable learning transfer.