Supporting Student Planning with Open Learner Models in Middle Grades Science

通过中年级科学的开放学习者模型支持学生规划

• 批准号: 1761178
• 负责人: James Lester
• 金额: $ 149.92万
• 依托单位: North Carolina State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-08-15 至 2023-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1761178&HistoricalAwards=false
• 关键词: Supporting; Student; Planning; Open; Learner

This project is supported by the Education and Human Resources Core Research (ECR) program, which supports fundamental research in STEM learning and learning environments. Self-regulated learning requires learners to set learning goals, plan how to achieve them, monitor the success of their plans, and make needed changes. Critical gaps exist in our understanding of how and to what extent middle school students can engage in self-regulated learning during science inquiry activities. Analogous gaps exist in our understanding of how to help students become self-regulated learners. Computer-based learning tools might help fill these gaps, but these tools are usually "closed," meaning that how the software interprets students' knowledge and progress is hidden from the student. In contrast, "open" learning tools provide students with understandable, visual representations of their knowledge and progress. Such "open learner models" may be useful for helping students develop self-regulated learning skills. For example, by providing easy-to-understand representations of student progress, open learner models may support the self-regulated learning processes of goal setting and planning. This project focuses on the design, development, and investigation of open learner models for student goal setting and planning in middle school science. The project will explore how students engage in goal-directed learning behaviors using data from eye tracking, log files, and think-aloud exercises. It is anticipated that the project results can advance the goal of high quality STEM learning experiences for all students. A major project goal is improving students' problem-solving abilities and learning outcomes through a theoretically grounded, data-driven open learner model. The project is designed to integrate an open learner model into the Future Worlds science learning environment for middle school ecosystems education. A culminating between-subjects experiment will be conducted to compare a baseline version of Future Worlds that does not have an open learner model (control condition) with a version of Future Worlds that has an embedded open learner model (experimental condition). The project will use state-of-the-art artificial intelligence computational frameworks to recognize students' goals and plans from observations of their problem-solving activities in an online learning environment. It is hypothesized that the open learner model will yield better student learning outcomes, with improved science problem-solving skills, increased science content knowledge, increased metacognitive awareness, and enhanced science self-efficacy. The project is designed to make significant contributions to both theory and practice of self-regulated learning. The project's aim is to make contributions to foundational knowledge and theory by advancing our understanding of how to improve science learning with open learner models: (1) With a focus on student goal setting and planning in technology-rich learning environments, the project aims to formulate an empirically-based theoretical framework for open learner model-enhanced learning that addresses both cognitive and metacognitive components of middle grades science education. (2) By conceptualizing goal setting and planning to account for self-regulated learning in the context of science problem solving, the project is designed to create a rich framework that connects goal setting and planning to students' problem solving and metacognitive processes. (3) By expanding to fine-grained process data, the project is intended to make methodological contributions that will enable the field to go beyond self-report measures, which have long dominated research on self-regulated learning. (4) The project plans to produce learning analytic techniques that yield predictive models of goal setting and planning in science problem solving, as well as open learner models to effectively support student goal setting and problem solving.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.

该项目得到了教育和人力资源核心研究（ECR）计划的支持，该计划支持STEM学习和学习环境中的基本研究。 自我调节的学习需要学习者设定学习目标，计划如何实现目标，监控计划的成功并进行所需的更改。 我们对中学学生如何以及在科学探究活动期间可以进行自我调节的学习的理解中存在关键差距。 我们对如何帮助学生成为自我调节的学习者的理解存在类似的差距。 基于计算机的学习工具可能有助于填补这些空白，但是这些工具通常是“封闭的”，这意味着软件如何解释学生的知识和进度对学生隐藏。 相比之下，“开放”学习工具为学生提供了可以理解的知识和进步的视觉表示。 这种“开放学习者”模型可能有助于帮助学生发展自我调节的学习技能。 例如，通过提供易于理解的学生进步表示，开放的学习者模型可以支持目标设定和计划的自我调节学习过程。该项目侧重于开放学习者模型的设计，开发和调查，用于中学科学的学生目标和计划。该项目将使用眼睛跟踪，日志文件和Think-Aloud练习中的数据探讨学生如何从事目标指导的学习行为。 预计项目结果可以推动所有学生的高质量STEM学习体验的目标。一个主要的项目目标是通过理论上的，数据驱动的开放学习者模型来改善学生的解决问题的能力和学习成果。该项目旨在将开放的学习者模型纳入中学生态系统教育的未来世界科学学习环境。将进行最终的受试者间实验，以比较未来世界的基线版本，该世界没有开放的学习者模型（控制条件）与具有嵌入式开放的学习者模型（实验条件）的未来世界版本。该项目将使用最先进的人工智能计算框架来从在线学习环境中观察到解决问题的活动来识别学生的目标和计划。 假设开放的学习者模型将通过提高科学解决问题的技能，提高科学内容知识，提高元认知意识并增强科学自我效能感，从而产生更好的学生学习成果。 该项目旨在为自我调节学习的理论和实践做出重大贡献。该项目的目的是通过促进我们如何通过开放学习者模型来改善科学学习的理解来为基础知识和理论做出贡献：（1）重点是在技术丰富的学习环境中的学生目标设定和计划，该项目旨在建立一个基于经验的理论框架，用于开放的学习者模型学习者，以解决认知能力和元认知组成的中等程度科学教育。 （2）通过概念化目标设定并计划在科学问题解决的背景下考虑自我调节的学习，该项目旨在创建一个丰富的框架，将目标设定和计划与学生的问题解决和元认知过程联系起来。 （3）通过扩展到细粒度的过程数据，该项目旨在做出方法论贡献，使该领域能够超越自我报告措施，这些措施长期以来一直占据了对自我调节学习的研究。 （4）该项目计划生产学习分析技术，从而产生科学问题解决目标和计划的预测模型，以及开放的学习者模型，以有效地支持学生的目标设定和问题解决。该奖项反映了NSF的法定任务，并被认为是值得通过基金会的知识分子优点和更广泛的影响来通过评估来支持的，审查了审查的标准。

项目成果

Autonomy and types of informational text presentations in game-based learning environments

基于游戏的学习环境中信息文本呈现的自主性和类型

• 发表时间: 2019
• 期刊: Proceedings of the Twentieth International Conference on Artificial Intelligence in Education
• 作者: Dever, D. A.;Azevedo, R.
• 通讯作者: Azevedo, R.

Leveraging Student Goal Setting for Real-Time Plan Recognition in Game-Based Learning

在基于游戏的学习中利用学生目标设定进行实时计划识别

• DOI: 10.1007/978-3-031-11644-5_7
• 发表时间: 2022
• 期刊: Proceedings of the 23rd International Conference on Artificial Intelligence in Education
• 作者: Goslen, A.;Carpenter, D.;Rowe, J.;Henderson, N.;Azevedo, R.;Lester, J.
• 通讯作者: Lester, J.

Examining Gaze Behaviors and Metacognitive Judgments of Informational Text Within Game-Based Learning Environments

在基于游戏的学习环境中检查信息文本的注视行为和元认知判断

• 发表时间: 2019
• 期刊: International Conference on Artificial Intelligence in Education
• 作者: Dever, Daryn;Azevedo, Roger
• 通讯作者: Azevedo, Roger

“The role of achievement goal orientation on metacognitive processes in game-based learning.”

– 成就目标导向对基于游戏的学习中元认知过程的作用。 –

• 发表时间: 2019
• 期刊: International Conference on Artificial Intelligence in Education
• 作者: Cloude, Elizabeth;Taub, Michelle;Lester, James;Azevedo, Roger
• 通讯作者: Azevedo, Roger

Language Modeling for Plan Generation in Game-Based Learning Environments

基于游戏的学习环境中计划生成的语言建模

• 发表时间: 2023
• 期刊: AIED’23 : Empowering Education with LLMs - the Next-Gen Interface and Content Generation
• 作者: Goslen, A.;Kim, Y.;Rowe, J.;Lester, J.
• 通讯作者: Lester, J.