Collaborative Research: Engaging High School Students in Computer Science with Co-Creative Learning Companions

这项研究调查了最先进的创意和教学代理如何改善学生的学习、态度和对计算机科学的参与。该项目将使用 EarSketch 在高中教室进行，EarSketch 是一种在线计算机科学学习环境，吸引了全球超过 160,000 名学习者使用 JavaScript 或 Python 代码制作音乐。研究人员将构建第一个共同创造的学习伙伴Cai，它将为学生提供教学策略，包括利用学习者代码来说明抽象性和模块化、建议新代码来构建新概念、提供帮助和提示以及解释其决策。这项工作将直接解决国家对培养计算素养作为 STEM 核心技能的需求。拟议的工作汇集了一个经验丰富的跨学科团队来研究这样一个假设：在富有表现力的计算机科学学习环境中添加一个共同创造的学习伙伴将改善学生的计算机科学学习（通过代码复杂程度和概念知识来衡量）、对计算的积极态度（自我效能和动机）和参与度（学习过程中的集中注意力和参与度）。共同创意学习伴侣的迭代设计和开发将基于对 EarSketch 课堂中人类协作的研究、共同创意文献和虚拟代理研究的发现，以及研究人员对 EarSketch 在课堂中使用的观察。这项工作将解决以下研究问题：1）表达性编程的共同创意学习伙伴应该执行哪些基本教学动作？ 2）对于共同创造的学习伙伴来说，什么教育策略最有效地支撑学习、对计算的有利态度和参与？ 3) 在高中课堂的社会文化背景下，EarSketch 中的共同创意学习伙伴如何提高计算机科学学习、参与度和对计算机科学的积极态度？拟议的研究有可能改变我们对如何通过表达计算环境支持学生学习和扩大参与的理解。该奖项反映了 NSF 的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

This research investigates how state-of-the-art creative and pedagogical agents can improve students' learning, attitudes, and engagement with computer science. The project will be conducted in high school classrooms using EarSketch, an online computer science learning environments that engages over 160,000 learners worldwide in making music with JavaScript or Python code. The researchers will build the first co-creative learning companion, Cai, that will scaffold students with pedagogical strategies that include making use of learner code to illustrate abstraction and modularity, suggesting new code to scaffold new concepts, providing help and hints, and explaining its decisions. This work will directly address the national need to develop computing literacy as a core STEM skill.The proposed work brings together an experienced interdisciplinary team to investigate the hypothesis that adding a co-creative learning companion to an expressive computer science learning environment will improve students' computer science learning (as measured by code sophistication and concept knowledge), positive attitudes towards computing (self-efficacy and motivation), and engagement (focused attention and involvement during learning). The iterative design and development of the co-creative learning companion will be based on studies of human collaboration in EarSketch classrooms, the findings in the co-creative literature and virtual agents research, and the researchers' observations of EarSketch use in classrooms. This work will address the following research questions: 1) What are the foundational pedagogical moves that a co-creative learning companion for expressive programming should perform?; 2) What educational strategies for a co-creative learning companion most effectively scaffold learning, favorable attitudes toward computing, and engagement?; and 3) In what ways does a co-creative learning companion in EarSketch increase computer science learning, engagement, and positive attitudes toward computer science when deployed within the sociocultural context of a high school classroom? The proposed research has the potential to transform our understanding of how to support student learning in and broaden participation through expressive computing environments.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.