Supporting Metacognition in Intelligent Tutoring Systems: Self-assessment
Event details
| Date | 09.10.2013 |
| Hour | 14:15 › 15:00 |
| Speaker | Vincent Aleven, Associate Professor in the Human-Computer Interaction Institute at Carnegie Mellon University, Pittsburgh, USA |
| Location | |
| Category | Conferences - Seminars |
Interactive educational technologies, such as intelligent tutoring systems, have shown to be very effective in helping students learn, for example when used in middle-school and high-school mathematics courses. These types of systems provide detailed guidance with problem solving, based on fine-grained modeling and real-time tracking of learners' skill acquisition.
Can these types of technologies also support learners in effectively regulating their learning processes? Theories of self-regulated learning stress the importance of accurate self-assessment - the more learners are aware of how well they master targeted skills, the better their decisions can be as to where to focus their learning efforts. Further, the process of self-assessing by itself may facilitate reflection on the learning materials and deep learning. At the same time, prior research shows that accurate self-assessment is challenging for learners, raising the question of how best to support it.
This talk presents two classroom studies from the work of PhD student Yanjin Long. These studies test the broad hypothesis that intelligent tutoring systems might leverage their learner modeling technologies to support learners' self-assessment and to help them learn more effectively at the domain level (e.g., mathematics problem solving). In both studies, we provided support for learners to reflect on their skill mastery with the help of the system's open learner model. This model presents the system's up-to-date view of an individual learner's mastery of targeted skills. The studies showed that support for self-assessment helps students learn better at the domain level. They confirm the value of open learner models and suggest that advanced learning technologies may help learners become better learners.
Can these types of technologies also support learners in effectively regulating their learning processes? Theories of self-regulated learning stress the importance of accurate self-assessment - the more learners are aware of how well they master targeted skills, the better their decisions can be as to where to focus their learning efforts. Further, the process of self-assessing by itself may facilitate reflection on the learning materials and deep learning. At the same time, prior research shows that accurate self-assessment is challenging for learners, raising the question of how best to support it.
This talk presents two classroom studies from the work of PhD student Yanjin Long. These studies test the broad hypothesis that intelligent tutoring systems might leverage their learner modeling technologies to support learners' self-assessment and to help them learn more effectively at the domain level (e.g., mathematics problem solving). In both studies, we provided support for learners to reflect on their skill mastery with the help of the system's open learner model. This model presents the system's up-to-date view of an individual learner's mastery of targeted skills. The studies showed that support for self-assessment helps students learn better at the domain level. They confirm the value of open learner models and suggest that advanced learning technologies may help learners become better learners.
Practical information
- Informed public
- Free
- This event is internal
Organizer
Contact
- Florence Colomb