A simulator-based Intelligent Tutoring System (ITS) is a computer system that is made to provide students with a learning experience that is both customizable to a student’s needs (e.g., level of expertise, pace) and includes simulation, e.g., demonstrate certain domain concepts or allow problem-solving while replicating a real-life situation. ITSs offers convenient and low-cost studying. We aim to explore the recent trends and identify limitations and opportunities in recent work on STEM (Science, Technology, Engineering, and Mathematics) self-study simulator based ITSs, we conducted a systematic literature review investigating 47 papers from four different databases. The research encloses ITSs from various educational sectors, ranging from elementary, middle, secondary, tertiary, and after school training. The majority of the systems targeted tertiary education. As a result, there are many research opportunities in introducing a more generalizable approach to simulator-based ITSs which will make it easier to address many STEM-related subjects. There are also opportunities in utilizing help methods that emphasize encouragement and self-reflection. We noticed that the number of STEM-related simulator-based ITSs is relatively low. Another finding was that most simulator-based ITSs are domain-dependent, and therefore they are not reusable for other subjects. Finally, we found that the traits of feedback in simulator-based systems that result in positive learning outcomes are ones that combined immediate and delayed feedback, used procedural information in their feedback, of formative feedback type, and detailed feedback.

基于模拟器的智能辅导系统（ITS）是一种计算机系统，旨在为学生提供既可以根据学生的需求（例如专业水平，步伐）进行定制的学习体验，又可以进行模拟，例如演示特定领域概念，或者在复制现实情况时解决问题。ITSs提供了便捷且低成本的学习方式。我们旨在探索基于ITS的STEM（科学，技术，工程和数学）自学模拟器的最新趋势，并确定近期工作中的局限和机遇，我们对来自四个不同数据库的47篇论文进行了系统的文献综述。该研究包括来自各个教育部门的ITS，范围从小学，初中，中学，大专和课后培训。大多数系统都以高等教育为目标。结果是，为基于模拟器的ITS引入更具通用性的方法有很多研究机会，这将使解决许多STEM相关主题变得更加容易。也有机会利用强调鼓励和自我反省的帮助方法。我们注意到，与STEM相关的基于模拟器的ITS的数量相对较少。另一个发现是，大多数基于模拟器的ITS都是依赖于域的，因此它们不可用于其他主题。最后，我们发现，在基于模拟器的系统中，能够带来积极学习成果的反馈特征是：结合了即时和延迟反馈，在反馈中使用了程序信息，形成了反馈类型以及详细的反馈。