The Vanderbilt University
Grant: $681,541 - National Science Foundation - Jul. 31, 2009
No votes have been cast for this award yet
Join the conversation: Post a comment about this award
Award Description: This proposal extends our previous work to designing, implementing, and analyzing choice-adaptive systems that will prepare students for future learning in real-world environments. Learning outside of school and in technology-enhanced learning environments of the future includes many different possible learning resources. To achieve the goal of adaptive intelligent systems that will prepare students to learn on their own, we need to perform additional foundational research, include several types of research: (a) designing and implementing systems that guide students when learning in virtual environments that provide a variety of choices; (b) using machine learning techniques to analyze students? choice patterns and their correlations w/ student learning outcomes; & (c) experimental comparisons of a range of choice adaptive and non-adaptive systems that include transfer measures of whether students are prepared to learn on their own when no longer being guided by an adaptive system. Our primary hypothesis is that helping students develop the metacognitive abilities to make learning choices will have strong effects on their subsequent abilities to learn in unstructured and open-ended learning environments. If the proposed work is true, it will lead to the creation of a set of methodologies that can be used to make the increasing number of virtual learning worlds into intelligent adaptive systems. Our goal is to create more dynamic and adaptive virtual agent environments w/ features to observe and analyze students? interactions with the agents. We call this a choice-adaptive framework, because the student can make choices in picking their learning topics & the resources they use for learning. The system monitors the students? choices, their activities on the system, & their performance on the learning tasks, & adapts to help them improve their learning. For example, the system may provide feedback to guide the student away from using trial and error approaches to solve problems to one where the student is encouraged to selectively read the resources on topics that they have not been doing well. Choice adaptive systems differ from most adaptive systems that primarily model student knowledge within a relatively rigid sequence of problem tasks and scripted learning resources. Students? choice patterns of what and how to learn will permit the learning environment to adapt content and choices to improve student learning and metacognition. The goal is to help students develop more effective skills for learning in complex science domains in both formal classrooms, and on their own when they need to make choices about what to learn. Ideally, choice adaptive systems will help students learn content better, and it will prepare them to make better choices for future learning, especially when the students no longer have the benefit of a scripted curriculum, instructor, or intelligent adaptive system. Overall, this research will address a number of significant issues that impact student learning in formal and informal settings: The design and development of adaptive learning systems using extended multi-agent architectures and machine learning techniques that can help students learn in the choice-filled computer worlds that are growing in education. Experimental comparisons of different choice adaptive designs, as well as comparisons of choice adaptive systems versus non-adaptive systems. Key outcome measures include which type of initial instruction prepares students to learn more effectively on their own in the future. A central hypothesis is that to learn metacognitive skills of regulating learning choices, students need opportunities to make choices along with guidance about when those choices are effective or sub-optimal. The analysis of choice behavior as a viable dynamic learning assessment on par with knowledge measurements that only look at speed, accuracy and error.
Project Description: This proposal extends our previous work to designing, implementing, and analyzing choice-adaptive systems that will prepare students for future learning in real-world environments. Learning outside of school and in technology-enhanced learning environments of the future includes many different possible learning resources. Our primary hypothesis is that helping students develop the metacognitive abilities to make learning choices will have strong effects on their subsequent abilities to learn in unstructured and open-ended learning environments. If the proposed work is true, it will lead to the creation of a set of methodologies that can be used to make the increasing number of virtual learning worlds into intelligent adaptive systems. Our goal is to create more dynamic and adaptive virtual agent environments w/ features to observe and analyze students? interactions with the agents. We call this a choice-adaptive framework, because the student can make choices in picking their learning topics & the resources they use for learning. The system monitors the students? choices, their activities on the system, & their performance on the learning tasks, & adapts to help them improve their learning. Overall, this research will address a number of significant issues that impact student learning in formal and informal settings: The design and development of adaptive learning systems using extended multi-agent architectures & machine learning techniques that can help students learn in the choice-filled computer worlds that are growing in education. Experimental comparisons of different choice adaptive designs, as well as comparisons of choice adaptive systems versus non-adaptive systems. The analysis of choice behavior as a viable dynamic learning assessment on par with knowledge measurements that only look at speed, accuracy and error.
Jobs Summary: The recovery act funds for this award helped to create or retain the following types of positions: professor, graduate student research assistant and an undergraduate research assistant. (Total jobs reported: 3)
Project Status: Less Than 50% Completed
This award's data was last updated on Jul. 31, 2009. Help expand these official descriptions using the wiki below.
No comments have been added for this project.