From Boring Tests to Engaging Adventures: A Game-Based Learning Approach in LMS

This paper describes a game-based approach to language placement testing within the ILIAS Learning Management System at Wismar University of Applied Sciences. Its central aim is to address two persistent problems: students being enrolled in courses that do not match their language skills, and their disengagement from assessment. The project, "Journey to Mars," converts the language center's existing English placement tests into a science fiction point-and-click adventure, using learning modules, interactive pictures, embedded audio and video, and automated tests with immediate feedback. The author grounds the design in player type models (Bartle, Yee), Self-Determination Theory, and Aarseth's and Deterding's frameworks for game analysis, and reports a three-step development process completed within a two-week prototype timeline.

The development relied heavily on generative AI tools, including ChatGPT for dialogue and narrative, DALL-E for imagery, and ElevenLabs for voiceovers, which the author describes as significantly reducing content production effort. The game is structured as three chapters, each containing four tests aligned with the language center's three competence levels, set respectively at a spaceport, aboard a spaceship, and on Mars. Additional features such as performance-based and exploration-based badges, ironic narration, and an alternative ending were added to support autonomy and replayability. The paper emphasizes that the LMS itself was the most restrictive factor in development, shaping the choice toward simple, tool-supported mechanics.

Evidence of effectiveness is limited. The author reports that initial feedback from a small group of students who tested the first version was largely positive, with students finding the setting relatable and engaging, but no quantitative data, sample details, or formal evaluation of learning outcomes or placement accuracy are presented. Technical issues with varying screen sizes and ILIAS login barriers were identified as shortcomings. Planned future work includes broader rollout for quantitative analysis, adaptive design and accessibility improvements, dynamic narrative paths, personalized learning routes, and integration with ILIAS's competency management system.