Prompted self-explanation is a successful intervention for many domains. However, in our previous work within the domain of second language grammar learning, we found no advantage for self-explanation over practice alone. Here, we continue testing the generality of self-explanation through the development of an adaptive self-explanation tutor and report on results of a classroom evaluation (N=92) in which we compare the adaptive tutor to a practice-only tutor. We investigate both procedural and declarative knowledge acquisition as well as long-term retention. Results show that while self-explanation takes more time than practice alone, it leads to greater learning of declarative knowledge. However, there are no differences between conditions on immediate or long-term retention measures of procedural knowledge.

Authors and Affiliations

1. Human-Computer Interaction Institute, Carnegie Mellon University, USA

   Ruth Wylie & Kenneth R. Koedinger

2. Rice University, USA

   Melissa Sheng

3. Language Technologies Institute, Carnegie Mellon University, USA

   Teruko Mitamura

Editors and Affiliations

1. EECS Department/ISIS, Vanderbilt University, TN 37235, Nashville, USA

   Gautam Biswas

2. Electronic, Electrical and Computer Engineering, University of Birmingham, U.K.

   Susan Bull

3. School of Information Technologies, University of Sydney, 1 Cleveland Street, 2006, Sydney, Australia

   Judy Kay

4. College of Engineering, Department of Computer Science and Software Engineering, University of Canterbury, Private Bag 4800, 8140, Christchurch, New Zealand

   Antonija Mitrovic

Reprints and permissions

© 2011 Springer-Verlag Berlin Heidelberg

Policies and ethics