Teaching requirements elicitation to students who do not have practical experience is challenging, as they usually cannot understand the difficulty. Several recent studies have reported their experience of teaching requirements elicitation with a serious game. However, in these games, the fictitious characters have not been carefully designed to reflect real scenarios. For example, they always respond the same no matter how many times a learner interacts with them. Moreover, most existing serious games contain only one specific scenario and cannot be easily extended to cover various cases. In this paper, we design and implement a dynamic state-based serious game (BARA) for teaching requirements elicitation, which can realistically simulate real-world scenarios and automatically record learners’ actions for assessment. Specifically, we model fictitious characters’ behaviors using finite-state machines in order to precisely characterize the dynamic states of stakeholders. We also developed an easy-to-use editor for non-programmers to design fictitious characters and thus construct various simulated scenarios. Finally, BARA records learners’ actions during the game, based on which we can gain an in-depth understanding of learners’ performance and our teaching effectiveness. We evaluated BARA with 60 participants using a simulated scenario. The result shows that most participants are immersed in BARA and can reasonably complete the requirements elicitation task within the simulated scenario.

@inproceedings{LiuLHY23,
  title = {BARA: A Dynamic State-based Serious Game for Teaching Requirements Elicitation},
  author = {Yu Liu and Tong Li and Zheqing Huang and Zhen Yang},
  booktitle = {ICSE},
  year = {2023},
  doi = {10.1109/icse-seet58685.2023.00020}
}