I am a Post-Doctoral Fellow at the Human Computer Interaction Institute and Pittsburgh Science of Learning Center, at
Carnegie Mellon University.
Previously, I was a Research Fellow at the Learning Sciences Research Institute
at the University of
Nottingham. I completed my doctorate in Human Computer Interaction, at Carnegie Mellon University, in December 2005.
I am associate editor of the Journal of Educational Data Mining, and was program chair (with Joseph Beck) of the First International Conference on Educational Data Mining, which was held June 20-21, 2008 in Montreal, Canada.
My goal is to develop interactive learning environments that can adapt effectively and sensitively to differences between students. Towards this end, I study how differences in student affect, attitudes, and motivation impact students' choices of how to use learning environments, and how these choices in turn impact their learning. I study these issues with a variety of techniques, drawn from education, computer science, and psychology, including quantitative and qualitative field observations [Read More] , labeling text-based replays of log data [Read More] , data mining and machine learning [Read More] , and analyses of questionnaire responses. I use machine learning to to develop detectors of the student behaviors associated with poorer learning, and to differentiate between behavioral categories that seem unitary at a surface level [Read More].
My research has shown that "gaming the system", attempting to succeed in an interactive learning environment by exploiting properties of the system rather than by learning the material, leads to significantly worse learning in intelligent tutoring systems. I have studied the prevalence of gaming behavior and what affective states and attitudes are associated with the choice to game the system, both in intelligent tutoring systems and educational games.
I have developed a system that can detect when students are gaming the system, and an interactive software agent, Scooter the Tutor, who responds to when students game. Scooter significantly improves gaming students' learning. [Read More].
Other recent research I have conducted has involved
- Modeling student learning, affect, and gaming the system within educational games
- Modeling off-task behavior in intelligent tutoring systems
- Generalization of machine-learned models between different lessons in intelligent tutoring systems (e.g. transfer learning)
These recent research projects do not yet have web pages, but please check out my publications web page for recent papers.
Past projects I was involved in included
- Developing cognitive tutors to teach data representation and interpretation
- Visualizing students' implementations of data structures, at a conceptual level [Read More].
