My research thus far has benefited from intense involvement both in the language technologies community and in the human-computer interaction community. Because what drives my research is the goal of developing technology capable of both shaping conversation and supporting conversation to achieve a positive impact on human learning, my long term plan is to remain active in both of these communities.
Basic Research on Discourse Analysis and Automatic Conversation Analysis. The primary focus of my pre-tenure work on interaction analysis has been interactions within pairs and small groups. However, as I look forward to supporting collaborative interactions within thriving online learning communities, I am becoming aware of the need to understand how local interactions within pairs or small groups may lead to emergent behavior at the community level, which may then exert downward causality on behavior at the individual and small group level, a phenomenon known in sociology as the macro-micro link. The most recent computational work modeling such patterns is in the multi-agent community. The limitation of this work is that it is based on simulation experiments rather than analysis of data. As large longitudinal datasets from online behavior are becoming easier to obtain, a new wave of work modeling social emergence has the potential to yield new insights, grounded in analysis of data from real communities as they grow and change over time. I am already beginning to engage in this work in a new partnership with Eric Xing where we are exploring how insights from sociolinguistics and techniques from graphical models can be integrated to address questions in this exciting area for inquiry.
Online Interventions Enabled by Conversational Agent Technology and Summarization. Looking to the future building on my pre-tenure research, one major goal is to greatly expand the impact my team is having in the area of online learning through broader dissemination. One current limitation to the impact my team’s work in dynamic support for collaborative learning is able to achieve is that a separate content development effort must be dedicated for each new collaborative unit we can offer to students. A major theme for my post-tenure work in this area will be achieving sustainability through adaptations of the technology that will offer students beneficial collaborative experiences on demand, such as in supported online homework sessions, a direction we’re beginning to work towards collaboratively with Jack Beuth in CMU’s Mechanical Engineering Department. Greater availability of supported collaborative learning experiences can also be offered as more instructors are interested in using them in their courses. In order to form new partnerships with CMU faculty, I am collaborating with CMU’s Eberly Center for Teaching Excellence to offer a workshop to CMU faculty interested in exploring this as an opportunity. I am also building a partnership with Marcela Borge at Penn State who is responsible for introducing online activities into courses throughout their Information Sciences Department.
Interest in dynamic support for collaborative learning has reached such a state of maturity that it is now birthing another new area of Computer Supported Collaborative Learning, namely Robotic Support for Collaborative Learning, spear-headed by Naomi Miyake at the University of Tokyo in collaboration with my group and Sandra Okita from Teacher’s College at Columbia. Building on the paradigm of agent based support for group learning, and bringing it into the three dimensional face-to-face world, Naomi Miyake’s team has been collaborating with Takayuki Kanda and others at Advanced Telecommunications Research (ATR) to develop the concept of robotic support for group learning for use in Japanese elementary school classrooms in 20 school districts across Japan. Their work has so far utilized the robots in wizard-of-Oz mode, but the desire is to eventually fully automate the robotic support for collaborative learning, building on the foundation of dynamic support for online collaboration. As part of this collaboration, PhD student Iris Howley is doing an internship with Takayuki Kanda at ATR in Spring of 2013. I was invited to give a talk on dynamic support for collaborative learning at the first symposium on Robot Facilitation as Dynamic Support for Collaborative Learning organized by Dr. Miyaki at the International Conference of the Learning Sciences in Summer of 2012. We are currently collaborating on what will be a co-edited book on this topic entitled “Collaborative Knowledge Construction: Learning from Virtual and Robot-Supported Environments”.