Abstract: Individual human tutoring is perhaps the oldest, most effective, and most expensive form of instruction. Cognitive Tutors are a successful educational technology that provide some of the benefits of individual human tutoring and have surpassed the effectiveness of typical human tutors, but not the best human tutors. In the 2004-2005 school year, Cognitive Tutor Algebra and Geometry courses are in use in about 2000 schools around the country by about 200,000 students. This talk will describe the embedded ACT-R cognitive models that enable Cognitive Tutors to provide just-in-time advice to students in complex problem solving and the simple Bayesian modeling that enable Cognitive Tutors to accurately estimate student knowledge and predict performance. The talk will also discuss machine-learned psychometric modeling of inappropriate student metacognitive behaviors and efforts to encourage more effective metacognitive skills. As a final note, Cognitive Tutors are featured in the new NSF-funded Pittsburgh Science of Learning Center that provides infrastructure for researchers around the country to conduct rigorous educational technology-related studies in high school and university classrooms.
Bio: Albert Corbett is an Associate Research Professor in the Human-Computer Interaction Institute at Carnegie Mellon University, Co-Director of the Pittsburgh Advanced Cognitive Tutor (PACT) Center at CMU and a co-founder of Carnegie Learning, Inc. Dr. Corbett received his B.A. in psychology from Brown University and his Ph.D. in experimental psychology from the University of Oregon. His expertise is in the cognitive modeling of problem solving and learning and the design of intelligent tutoring systems. Over the years he has developed cognitive tutors for programming, mathematics and genetics problem solving that have been deployed and evaluated in middle school, high school and university classrooms.
Papers:
Baker, R.S., Corbett, A.T., Koedinger, K.R. (2004) Detecting Student Misuse of Intelligent Tutoring Systems. Proceedings of the 7th International Conference on Intelligent Tutoring Systems, 531-540. http://www-2.cs.cmu.edu/~rsbaker/BCK2004MLFinal.pdf
Corbett, A.T., McLaughlin, M.S. and Scarpinatto, K.C. (2000). Modeling student knowledge: Cognitive tutors in high school and college. User modeling and user-adapted interaction, 10, 81-108.
http://pact.cs.cmu.edu/corbett/CorbettUMUAI2000.pdf