Scaffolding Student Learning with Interactive Adaptive Technologies: Challenges and Directions
Intelligent Tutoring Systems (ITSs) are interactive computer applications that employ Artificial Intelligence and Human Computer Interaction techniques to instruct students in an “intelligent” way. Although there does not exist an accepted definition of the term “intelligent,” a characteristic shared by many ITSs is that they possess various reasoning capabilities that allow them to adapt their instruction to individual students’ needs. This functionality is motivated by research demonstrating that when students receive personalized one-on-one instruction, they learn significantly better than with standard classroom instruction. Although ITS research has made significant progress in the last decade, many challenges remain, including (1) how to extend ITSs’ capabilities to improve their pedagogical utility, (2) how to unobtrusively and seamlessly assess students’ states of interest during their interaction with an ITS, and (3) how an ITS should use this information to guide student activities.
In this talk, I will describe how my work aims to address these challenges. In general, I believe that educational technologies need to move beyond providing purely cognitive support to also scaffolding students’ meta-cognitive behaviors, affect, and/or creativity. Accordingly, some of my work has focused on designing, implementing, and/or evaluating ITSs that support various meta-cognitive skills, including ones related to analogical problem solving, “gaming”, and exploration. Another area of my work I will describe has involved devising computational models that automatically recognize students' affective states, as well corresponding ITS support and interventions. I will also touch on my work exploring the utility of novel contexts, including learning from tangible environments and from observing others learn.
Kasia Muldner received her Ph.D. from the Department of Computer Science at the University of British Columbia, where she designed and evaluated a computational tutor that supported students during analogical problem solving. She is currently a post-doctoral researcher in the Department of Computing, Informatics, and Decision Systems Engineering at Arizona State University. Her work falls into the intersection of Human-Computer-Interaction and Artificial Intelligence, dealing with the design and evaluation of interactive educational technologies that aim to help students learn effectively though personalized support. She is particularly interested in technologies that support high level student states related to meta-cognition, affect, and creativity. Her work was recently recognized through the annual James Chen User Modeling and User Adapted Interaction (UMUAI) award given to the best journal paper in this venue.