Aggregate Representations to Support Scientific Inquiry: A Case Study with Embedded Phenomena Rebecca Cober, University of Toronto, Ontario Institute for Studies in Education Email: rebecca.cober@mail.utoronto.ca Colin McCann, University of Toronto, Ontario Institute for Studies in Education Email: colindmccann@gmail.com Tom Moher, University of Illinois at Chicago, Department of Computer Science and Learning Science Research Institute Email: moher@uic.edu Jim Slotta, University of Toronto, Ontario Institute for Studies in Education Email: jslotta@oise.utoronto.ca Abstract: This study explores how aggregate representations of student-contributed content were used in whole-class discussions to scaffold scientific inquiry in a middle school science classroom. Using tablet computers, students contributed observations concerning HelioRoom, a room-sized “Embedded Phenomena” simulation where students observe colored circles corresponding to planets in our Solar system, as they circle the classroom on display monitors. Students observe the occlusion relationships of the colored circles, working collectively at all four monitors, with the ultimate goal of deducing which colored circles correspond to which planets. Their observations were collected and represented in aggregate form, and were the focus of teacher-led whole- class discussions. This paper identifies four interaction patterns that emerged through the use of these aggregate representations. Objective Over the past two decades there has been a growing interest in the use of socially aggregated content as a resource for learning and instruction (see Rochelle et al., 2007a). These collective representations are generated from the contributions made by individual students or small groups of students using networked devices. A prominent application of this approach was provided by Crouch and Mazur (e.g., 2000), who employed “clickers” to help make their physics lectures more engaging for students, and more responsive to their specific conceptualizations. Another example of a classroom response system is ClassTalk, where teachers pose a multiple-choice question, students enter a response into a networked device, and responses are instantly aggregated and presented in a histogram at the front of the room (Dufresne & Gerace, 1996). Using this visualization as a shared point of reference, teachers can highlight areas of agreement or disagreement in student responses during individual, small group, or whole class discussion (Roschelle, 2003). Other powerful aggregation tools have been studied, including the Active Class Project (Ratto et al., 2003), where students pose and answer questions concerning a curriculum topic, graphs generated from probes and sensors (Tinker, 2000), and collaborative writing (Roschelle et al., 2007b). The emergence of Web 2.0 tools and the ubiquitous nature of mobile devices have advanced a wide range of new opportunities for collecting and re/presenting student observations, votes, tags or other contributed content in aggregate form, often in “real time” (i.e., instantly updating and reflecting any new inputs). However, there remains a significant challenge of matching visualization designs to specific curriculum goals. The objective of the present study is to describe and analyze the interaction patterns that emerge when a classroom of elementary students engages in scientific inquiry using