CTRL (Collaborative Tutoring Research Lab)

Collaborative activities have been shown to increase student learning, but only when students engage in productive behaviors as they collaborate. Providing collaborating students with assistance that adapts to their behaviors may be an effective way of improving the quality of collaboration. Intelligent systems that provide adaptive assistance to student collaboration have been developed, but most have not been evaluated in classrooms in order to answer learning sciences research questions. The goal of this project is to develop an architecture for adaptive collaborative learning support that makes it easier to conduct controlled classroom experiments. The collaborative tutoring research lab (CTRL) facilitates data analysis by integrating multiple streams of process data in a single log. It facilitates development of complex tutoring by allowing custom-generated tutors to use information from existing domain-specific tutors. Finally, CTRL facilitates ablation studies by allowing the efficient removal of components to enable the creation of natural control conditions. Currently, CTRL has one instantiation: an adaptive peer tutoring assistant (described in the ADAPT project). The next steps in the development of CTRL are to demonstrate its generality by instantiating different scenarios and to increase the complexity of its integration mechanism for different types of assistance.

Walker, E., Koedinger, K. R., McLaren, B. M. and Rummel, N. Cognitive Tutors as Research Platforms: Extending an Established Tutoring System for Collaborative and Metacognitive Experimentation. In the Proceedings of the 8th International Conference on Intelligent Tutoring Systems, Jhongli, Taiwan, June 26-30, 2006. [pdf]