Open Process Models Optimizing Self Regulated Learning in the Classroom

当今许多计算技术的进步，包括人工智能（AI）、机器学习和数据科学，都依赖于分布式和并行计算。因此，开发分布式和并行系统技能对于计算机科学教育和为劳动力做准备至关重要。然而，学习并行编程是困难的，因为学生经常发现并行思考而不是顺序思考是具有挑战性的（在入门课程中是这样教的）。基于游戏的学习方法可以提供一个吸引人的环境，促进元认知和自我调节。这项工作建立在nsf资助的游戏《Parallel》（教育IIS 1523116）的基础上，该游戏吸引学生解决难度增加的谜题，帮助学生使用并行编程的抽象原理来解决谜题。然而，对游戏的研究发现，仅靠游戏本身并不能激发学生通过计划、反思、重新评估和重新计划来管理自己的学习。该项目将开发基于学习科学原理的社区工具，以支持学生的自我调节学习。学生的学习过程将通过界面可视化，允许学生相互学习，教师指导和支撑学习过程。该项目将重点研究如何让学生参与自我调节学习的元认知过程，从而更好地学习并行编程。该项目将通过基于设计的研究来解决这个研究问题。该项目将开展一系列研究，以了解学生目前如何在并行编程课程中学习和自我调节学习，然后利用这些研究结果帮助开发具有增强社区互动机制的人工智能系统。该系统将由一个过程可视化系统组成，该系统扩展了并行游戏，允许学生通过开放的可视化方式查看和反思他们自己和其他学生的数据，揭示解决问题的策略和决策。社区互动机制将允许学生为彼此留下评论，并允许教师通过对学生学习过程的反馈来支撑学习过程。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Many of today’s computing advancements, including artificial intelligence (AI), machine learning and data science, rely on distributed and parallel computation. Thus, developing distributed and parallel systems skills is essential for computer science education and preparation for the workforce. Learning parallel programming, however, is difficult because students often find it challenging to think in parallel rather than sequentially (how they are taught in introductory courses). Game-based learning approaches can provide an engaging environment that fosters metacognition and self-regulation. This work builds on an NSF-funded game called Parallel (educational IIS 1523116), which engages students in solving puzzles that increase in difficulty, helping students to solve puzzles using abstract principles from parallel programming. Studies of the game, however, found that it alone cannot stimulate students to manage their learning through planning, reflection, re-assessment, and re-planning. This project will develop community-based tools based on learning sciences principles to scaffold students' self-regulated learning. Student learning processes will be visualized using interfaces that allow students to learn from each other and instructors to coach and scaffold the learning process. The project will focus on the research question of how to engage students in metacognitive processes of self-regulated learning toward better learning of parallel programming. The project will address this research question through design-based research. The project will develop a set of studies to understand how students currently learn and self-regulate their learning within parallel programming classes and then use the results of these studies to help develop an AI system with augmented community interaction mechanisms. This system will be composed of a process visualization system that extends the Parallel game to allow students to view and reflect on their process through an open visualization of their own and other students’ data revealing problem-solving strategies and decisions. The community interaction mechanisms will allow students to leave comments for one another and for teachers to scaffold the learning process through feedback on the student’s learning process.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.