Thursday, September 17, 2009

Paradigms of Shared Knowledge

Multiple paradigms for analyzing shared knowledge in collaboration. Collaborative learning is all about sharing knowledge. Without a shared base of knowledge (common ground), discourse itself is impossible, let alone collaborative interaction. Collaborators must share a means of communication (language), a joint focus (object-orientation), and a compatible orientation (perspective). In addition to being dependent upon the presence of existing shared knowledge, successful collaboration or collaborative learning involves the construction of new knowledge, created jointly and thereby shared by the participants. Knowledge can take many forms, not necessarily rational, propositional, explicit, factual knowledge. There is tacit and explicit knowledge, focal and background, propositional and procedural, personal and institutional, individual and group.

Studies of CSCL each tend to focus on a certain form of knowledge and assume a certain way of sharing this knowledge. These choices depend upon the theoretical position implicitly or explicitly adopted by the study. You may find it interesting to figure out which paradigm of shared knowledge corresponds to various articles in CSCL.

The topic of paradigms of shared knowledge may seem abstrusely theoretical and remote from the practical concerns of CSCL. However, it strongly effects whether a given educational intervention—incorporating pedagogical resources, computer technologies, scripting, grouping of students—will foster effective collaborative learning. In fact, it may affect this even more when the teachers, researchers, or other CSCL designers are not explicitly aware of their assumptions about shared knowledge.

The four articles of the December issue of ijCSCL, which deal with computer support for shared knowledge, are all focused on the practical design of technologies to support collaborative learning: wikis, virtual reality, PowerPoint, and group formation software. They each presume a different paradigm of shared knowledge. The following paragraphs define four paradigms spanning a range:

The paradigm of sharing individual mental representations. Perhaps the most commonsensical view of shared knowledge in a small group is that the individual members of the group each possess the same knowledge. This can be elaborated theoretically by hypothesizing that each member has mental representations that are sufficiently similar to specific mental representations of each of the other members. The classic analysis of grounding (Clark & Brennan, 1991) that is often cited in CSCL research describes how two typical collaborators might establish shared knowledge by externalizing their ideas and explicitly comparing the propositional expressions of their mental representations. This paradigm assumes that individuals possess well-formed opinions and can unproblematically express them. Sharing is here taken to be a matter of transferring and comparing ideas in ways that typically do not change the ideas.

The paradigm of sharing an object. A quite different view conceives of shared knowledge as a natural consequence of a group being collaboratively involved with the object of their work together. They are all oriented in common toward the same object (an artifact, a problem, a goal) and thereby come to share knowledge of that object, in particular the knowledge about that object that arises from their work with it. The sharing of knowledge about a common object does not need special coordination, once the object is truly shared. A recent ijCSCL article (Çakır, Zemel & Stahl, 2009) described an example of how students in an online group worked to define and share multiple realizations of a mathematical object; once they could all “see” the same object, the construction of new shared knowledge (such as the formulation of an algebraic expression to solve their problem) proceeded quickly. Another recent article (Dohn, 2009) discussed how the affordances of an object must be enacted; in the collaborative case, this is accomplished interactively as the group comes to know and share the object. In both analyses, the shared knowledge is new knowledge for all the participants, arising out of their interactions with each other, with the shared object, and with other resources for communication and understanding, including available computer supports.

The paradigm of sharing a situation. If we broaden the notion that shared knowledge comes from a joint focus on an object of collaboration, we come to the idea that a group can share knowledge by being situated in a common context—a joint problem space (Roschelle & Teasley, 1995) or an indexical ground of reference (Hanks, 1992). The situation includes the shared object, but it also includes other resources and constraints, such as the affordances of a CSCL environment. Above all, it includes the past discourse of the group, which has created a complex network of shared concepts, interactions and experiences. According to (Duranti & Goodwin, 1992), the situation and the discourse “stand in a mutually constitutive relationship to each other, with talk, and the interpretive work it generates, shaping context as much as context shapes talk” (p. 31). According to this paradigm, the engagement in collaborative discourse can automatically generate shared knowledge as an ongoing process.

The paradigm of sharing a community. The social sciences generally take an even broader view. They argue that the shared knowledge that makes life together possible comes from belonging to the same communities, cultures, and societies. It is the understanding of the same historically accumulated knowledge, values, perspectives, artifacts, and ways of life—largely encapsulated in language—that makes communication aaapossible. In proposing that we broaden our thinking about computer support to include complex technological infrastructures, Jones, Dirckinck-Holmfeld, and Lindström (2006) tried to show how the institutional macro-level could be related to the mezzo-level of collaboration and even to the micro-level of fine-grained interaction analysis.

The existence of multiple effective paradigms for understanding something like shared knowledge is not necessarily problematic. It may be possible to select the most appropriate paradigm for any given study. However, it does raise the question of how the paradigms might fit together—a topic for another time.

ijCSCL Issue for December 2009

International Journal of Computer-Supported Collaborative Learning (ijCSCL)
Volume 4, Number 4, December 2009

Paradigms of shared knowledge
Gerry Stahl * Friedrich Hesse

Wikis to support the "collaborative" part of collaborative learning
Johann A. Larusson * Richard Alterman (USA)

Earth science learning in SMALLab: A design experiment for mixed reality
David Birchfield * Colleen Megowan-Romanowicz (USA)

Contrasting the use of tools for presentation and critique: Some cases from architectural education
Gustav Lymer * Jonas Ivarsson * Oskar Lindwall (Sweden)

An ontology engineering approach to the realization of theory-driven group formation
Seiji Isotani * Akiko Inaba * Mitsuru Ikeda * Riichiro Mizoguchi (Japan)