Strand 2: Human-Computer Interaction

The foundational question for Strand 2 is: What advances in theories, interaction-paradigms, and frameworks are needed to orchestrate effective student and teacher interactions with AI partners? Building on their accomplishments over the winter and spring, the Strand 2 team focused their attention on three main goals over the summer: 1) measuring and modeling Collaborative Problem Solving, 2) understanding teacher orchestration and scaffolding within the Sensor Immersion curriculum, and 3) conducting experimental studies to help inform the design and development of an AI Partner aimed at supporting Collaborative Problem Solving.

Collaborative Problem Solving

Working towards the goal of measuring and modeling Collaborative Problem Solving, the iSAT Lab Experiment team collected multimodal data from 20 small-groups (dyads and triads) as they completed a series of Collaborative Problem Solving tasks. The team collected a range of data including task performance, surveys on team processes, self-reported social, cognitive, and affective states, as well as speech, webcam video, eye tracking, and Kinect video during the collaborations. They are in the process of conducting initial analyses on the data and are excited to begin writing up the results in the fall.

Analysis Beyond Speech

In efforts to improve understanding of teacher orchestration and scaffolding within the Sensor Immersion curriculum, Strand 2 developed a coding scheme to analyze non-verbal aspects of communication, such as eye gaze, joint attention, working with technology tools, etc. This analysis is aimed at understanding how indicators other than speech will help the team at iSAT understand collaborative processes. Additionally, they are also examining how peers scaffold each other during group work. Together, these two projects will foster greater understanding surrounding how to support students working in collaborative groups.

Cross-strand Collaboration

Strand 2 team members have also begun planning several studies for the fall including a collaboration with Strand 3 that will test lessons from the Games Unit curriculum in the iSAT lab. Additionally, team members will support a trainee-led project on the role of an agent in supporting learning of computer programming and debugging. Furthering the team’s cross-strand collaboration efforts, Strand 2 will continue to collaborate with both Strand 1 and Strand 3 in exploring designs for interactive displays that will be tested in the lab this fall prior to the classrooms. Strand 2 will continue to work closely with the other strands in the coming months to consider implications for design decisions and how to best support teacher and student needs.