Notes

* Highlight, page 1

In a previous study [16], I found that online collaborators treated a graphical representation of evidence as a medium through which collaboration took place as well as its object, proposing new ideas by entering them directly in the graph before engaging in (usually brief) confirmation dialogues in a textual chat tool. In contrast to face-to-face users of shared representations, gesture was almost never used online and verbal deixis almost always referenced temporally recent information; instead, direct manipulation of the graph was used to reintroduce old information. In general, actions in the graph appeared to be an important part of participants’ conversations with each other, and in fact was at times the sole means of interaction. It was clear that all mutable media become part of the communication medium in virtual interaction, and therefore should be designed together with this use in mind. These observations led to the questions of the quality of interactions through the graph. Are participants merely operating in parallel on the same workspace, or are they in some sense having a conversation through the graph? If so, is this conversation at all significant from a learning standpoint: can true knowledge construction take place through shared representations? To answer these questions, I identified interactions from our corpus that appeared to constitute collaboration through the nonverbal as well as verbal media, and am engaged in a qualitative analysis of these examples. The purpose of this analysis is to understand how participants made use of the graph representation to mediate meaning making activity, by examining how participants use actions on the representations to build on each others’ ideas. The larger goal is to identify affordances of shared representations for online collaboration and their implications for the design of representational support for collaborative knowledge construction.

* Highlight, page 1

This paper is concerned with the question of how activity mediated by shared representations–notations that are manipulated by more than one person during a collaborative task–might constitute knowledge construction activity. The paper begins with a brief review of theoretical perspectives on how representations mediate collaborative knowledge construction, to identify the kinds of events we would look for as evidence of knowledge construction in via a representational medium. Then the paper draws on data from a prior study in which participants collaborated via a graphical representation as well as a verbal “chat” tool, to identify instances of such events and illustrate ways in which the activity of two individuals can be coupled and joined into a larger cognitive (and sometimes knowledge construction) activity distributed across the persons and representations they are manipulating.