This project is one of four research thrusts in the Pittsburgh Science of Learning Center. The Social and Communicative Factors in Learning Thrust is jointly lead by Carolyn Rosé and Lauren Resnick. This summary focuses mainly on the contribution of my group to this joint work.
In this project, we are partnering with the 9th grade biology team in a nearby urban school district. As part of that partnership, we are working with administrators and teachers across the district to introduce what are referred to as Academically Productive Talk (APT) practices into classrooms. Early studies of APT in classrooms showed success with highly skilled teachers or privileged student populations. The goal we are working towards is increasing the extent to which these APT based practices are used within typical urban classrooms, using professional development with teachers and Computer Supported Collaborative Learning (CSCL) experiences for small groups of students as tools for reshaping the classroom culture.
An important part of the collaborative work between my group and Lauren Resnick's group has been an effort to build bridges between the Classroom Discourse Community and the Computer Supported Collaborative Learning community. We have worked towards this by exploring the connections between important constructs for analysis of discussion for learning that originate from both communities. Our integrated understanding (Sionti et al., 2011; Howley, Mayfield, & Rosé, 2013) forms the foundation for the work we have done since the beginning of the thrust work (Suthers et al., in press; Dyke et al., 2012; Howley et al., 2011; Mayfield & Rosé, 2011).
This effort builds on an earlier history of successful deployment of intelligent conversational agents for support of small group learning. In early studies, these computer agents served the purpose of elevating the conceptual depth of collaborative discussions by leading students in groups through directed lines of reasoning, referred to as knowledge construction dialogues, which were meant to scaffold the process of groups constructing conceptually rich explanations together (Howley et al., 2011; Wang et al., 2011). In this more recent work, we have begun exploring how we might design conversational agents that employ APT practices as scaffolding for on-line collaborative learning discussions (Adamson et al., 2013; Dyke et al., in press).
Classroom studies where we have used APT agents to support learning in groups have demonstrated their significant positive learning impact within the collaborative sessions. We have also found that the positive effect of the online interaction extends into the classroom environment as well, elevating the quality of whole group teacher led discussions in classroom sessions immediately following the ones where the online activities took place(Clarke et al., in press).