Overview
Our project's objective is to combine insights from recent work in computer science, psychology and education to create and study "teachable agent" (TA) environments in mathematics and science that are motivating to students, intuitive to teachers and parents, and lead to high degrees of student learning. The hallmark of these environments is that students learn by instructing "teachable agents" who then venture forth in simulation-based exploratory environments and attempt to solve problems that require knowledge relevant to the disciplines of mathematics or science. If the agents have been taught properly they solve the problems; otherwise they need to be educated further. The simulation-based environments are carefully designed to focus attention on important concepts in science and mathematics, and to make explicit the errors that occur during problem solving. Students "scout" the problem solving requirements of various environments before attempting to teach their agents. Additional help and coaching agents are available to point students in the right direction when they make errors or produce sub-optimal solutions.
One key issue to be studied is how student learning is affected by opportunities to teach agents to prepare for particular challenges, and how this learning is influenced by the design of systems that vary in the degree to which they let students
- "scout" in order to find problems that arise in the agents' environments
- teach the agents with different representations and techniques
- measure the successfulness of their teaching by placing their agents in mini-assessment environments prior to engaging them in full-blown "challenge environments"
- receive different degrees and forms of feedback when their agents encounter difficulties (e.g. feedback can vary from specifically noting what needs to be fixed to simply noting that a problem exists)
- educate the personality, as well as the knowledge variables, relevant to learning, problems solving and collaboration.
The Teachable Agents project requires contributions from, and has important implications for, at least three disciplines: computer science, psychology, and education. It will inform computer science by exploring new approaches to the design of environments that support learning, and new ways to program intelligent agents. An especially important contribution will be research on defining "open" computational architectures where teachable agents created in one environment can be placed in other problem solving environments. A number of research issues dealing with intrinsic and extrinsic characteristics of multi-agent architectures will be addressed.
Our project will inform psychology by creating a rich environment for studying how different designs for "learning by teaching" affect student learning. It will inform education by exploring design principles that are student, teacher and parent friendly, and that help all students achieve. The Teachable Agents project also has the potential to create new forms of assessment (by letting students design and debug agents to solve various problems), and to transform popular video game technologies into environments that help students learn important content (rather than accumulate "power" or "weapons", agents can accumulate relevant knowledge for solving problems). The quality and impact of the project will be enhanced through its association with the NSF-funded Center for Innovative Learning Technologies (CILT), whose mission is to foster collaboration among members of the education and technology community. CILT will provide guidance and feedback to the project.