CAREER: Intelligent Representations: How to Blend Physical and Virtual Representations by Adapting to the Individual Student's Needs in Real Time

适应性教育技术可以大大提高学生在科学、技术、工程和数学（STEM）领域的学习。这些技术的特别优势在于它们可以提供可视化复杂概念的交互式模型。教育技术通常使用虚拟模型，学生可以通过鼠标或键盘操作。然而，学生用手操作的物理模型可能更直观，因为它们将抽象概念与学生联系起来。现实世界中的身体体验。这个CAREER项目研究了如何最好地将物理模型整合到适应性教育技术中。该项目将重点关注模型在教学中发挥关键作用的领域：本科化学。在两年制和四年制大学进行的一系列实验将测试物理模型或虚拟模型对特定概念是否最有效，以何种顺序呈现它们，以及如何帮助学生在它们之间建立联系。此外，该团队将开发一种技术，可以评估学生如何操作物理模型。结果将在物理和虚拟模型如何影响学生学习的综合理论中得到巩固。结果将帮助教师为他们的学生选择最好的模式。该团队将在研究结果的基础上开发一种教育技术，根据学生的学习进度，自适应地选择对学生个人最有帮助的物理和虚拟模型。这项研究将产生一种新型的教育技术，它融合了物理和虚拟模型，可以适应学生个体的身体互动。这些技术可以让不同背景的学生更容易理解STEM概念。此外，通过让经常缺乏技术创新机会的两年制大学的学生和教师参与进来，该项目将扩大STEM领域的参与度，并增强STEM领域的社会经济平等。学生在视觉表征方面的思考困难会危及他们在科学、技术、工程和数学（STEM）领域（如化学）的学习。物理（有形）和虚拟的表现方式对学生的学习有互补的好处。然而，当学生学习抽象内容时，表征模式是如何相互补充的，目前还没有一个全面的理论。这个CAREER项目的目标就是发展这样的理论。由于物理和虚拟表示的有效组合是概念、行动和学生特定的，因此对于教师来说，如果没有支持，它们可能太复杂而无法实现。教育技术可以提供这样的支持，但它们不能与物理表征相结合。因此，这个CAREER项目的另一个目标是将表征模式的理论转化为适应性教育技术，智能地融合物理和虚拟表征。在二年级和四年级的大学化学中，一系列的实验将研究物理表征和虚拟表征是如何相互补充的，如何最好地对它们进行排序，以及如何最好地帮助学生在它们之间建立联系。与教师和学生一起进行的用户研究将调查教育技术设计如何在真实的学习环境中解决教育需求和系统约束，以及如何支持教师有效地结合表现模式。进一步的项目活动将把智能混合框架扩展到K-12背景下的STEM领域。该项目将提供关于物理和虚拟表示如何相互补充的新理论。它将为教育技术的设计提供实用的建议，以适应个别学生基于身体的互动。一个具体的成果将是一种适用于2年制和4年制大学化学的适应性教育技术，免费提供并传播给不同的人群。通过纳入两年制大学，并使本研究适用于K-16背景下的STEM教育，该项目可能会扩大STEM许多领域的参与并增强社会经济平等。

Adaptive educational technologies can much improve students' learning in science, technology, engineering, and mathematics (STEM) domains. A particular strength of these technologies is that they can provide interactive models that visualize complex concepts. Educational technologies typically use virtual models that students manipulate via mouse or keyboard. Yet, physical models that students manipulate with their hands can be more intuitive because they relate abstract concepts to students? bodily experiences in the real world. This CAREER project examines how best to integrate physical models into adaptive educational technologies. The project will focus on a domain where models play a crucial role in instruction: undergraduate chemistry. A series of experiments at 2-year and 4-year colleges will test whether physical or virtual models are most effective for particular concepts, in which order to present them, and how to help students make connections among them. Further, the team will develop a technology that can assess how students manipulate physical models. Results will be consolidated in a comprehensive theory of how physical and virtual models affect student learning. The results will help instructors select the best model for their students. The team will build on the results to develop an educational technology that adaptively selects physical and virtual models that is most helpful to the individual student given his/her learning progress. This research will yield a new type of educational technologies that blend physical and virtual models and that can adapt to individual students' bodily interactions. Such technologies can make STEM concepts more accessible to students with diverse backgrounds. Further, by involving students and instructors from 2-year colleges who often lack access to technology innovations, the project will broaden participation and enhance socioeconomic equality in STEM.Students' difficulties in thinking in terms of visual representations jeopardize their learning in science, technology, engineering, and math (STEM) domains such as chemistry. Physical (tangible) and virtual representation modes have complementary benefits for students' learning. Yet, there is no comprehensive theory of how representation modes complement one another when students learn abstract content. The goal of this CAREER project is to develop such theory. Because effective combinations of physical and virtual representations are concept-, action-, and student-specific, they are likely too complex for instructors to achieve without support. Educational technologies can offer such support, but they cannot interface with physical representations. Hence, another goal of this CAREER project is to translate theory about representation modes into adaptive educational technologies that intelligently blend physical and virtual representations. A series of experiments in 2- and 4-year college chemistry will investigate how physical and virtual representations complement one another, how best to sequence them, and how best to help students make connections among them. User studies with instructors and students will investigate how educational technology design can address educational needs and systemic constraints in real learning contexts and how to support instructors in combining representation modes effectively. Further project activities will expand the intelligent blending framework to STEM domains in K-12 contexts. The project will contribute new theory about how physical and virtual representations complement one another. It will yield practical recommendations for the design of educational technologies that adapt to individual students' body-based interactions. A concrete deliverable will be an adaptive educational technology for 2- and 4-year college chemistry, available for free and disseminated to diverse populations. By including 2-year-colleges and by making this research applicable to STEM education in K-16 contexts, this project may broaden participation and enhance socioeconomic equality in many STEM domains.