From Mutual Diagnosis to Collaboration Engines:

Some Technical Aspects of Distributed Cognition

P. Dillenbourg TECFA (Educational Technology Unit) School of Education and Psychology University of Geneva 9, Route de Drize, CH1227 Carouge (Switzerland)
Abstract This contribution is based on the development and the evaluation of two learning environments in which the learner had to collaborate with the machine. This experience revealed that existing knowledge-based techniques are not appropriate to support 'real' collaboration, i.e. to cover a range of flexible, opportunistic and robust interactions which enable to agents to build a shared understanding of a problem. I argue for the design of collaboration engines which integrate dialogue models at the rule instantiation level. The role of dialogue models is not simply to improve the interface. The challenge is to develop models which account for the role of dialogues in problem solving and learning. Such models would reflect current theories on 'distributed cognition', one of the approaches placed under the 'situated cognition' umbrella. Most of the implications of these theories to the design of interactive learning environments (ILEs) that have been discussed so far concern the choice of methods (e.g. apprenticeship - Newman, 1989 - or project-oriented group work - Goldman et al, 1994) or software engineering (e.g. participatory design - Clancey, 1993). I address here the implications of these theories at a more technical level. I structured the argument as a discussion with myself, which is quite natural within the distributed cognition approach where interacting agents are viewed as forming a single cognitive system.

Conclusion Viewing cognition as distributed does not simply impact on the design of learning activities, but on the whole architecture of an ILE. Instead of viewing an ILE as a set of agents which exchange messages, the system and the learner are viewed as forming a joint cognitive system. The main functions of an ILE (tutoring, explaining, problem solving, diagnosis,...) are intrinsically collaborative. Cognitive effects are due to the mechanisms (e.g. appropriation) by which, despite the fact that cognition is distributed, agents succeed in forming a joint cognitive system. I illustrated this with the diagnosis process. While ILE researchers traditionally implemented diagnosis as a one-side process, dialogue models treat it as a mutual process. This does not simply imply that we should plug in a grounding dialogue module on top of the existing architecture. The growing importance of dialogue studies in AI&ED is not simply targeted to develop natural language interfaces. These techniques are required to model the key point in the 'distributed cognition' approach: participating into a joint cognitive system impacts on individual's cognition. I argue for a modification of the technology of knowledge-based systems in such a way that dialogues and problem solving are conducted by the same operators, that the very process of instantiation variable is implemented as a negotiation process.
Acknowledgements Thanks to Michael Baker, Patrick Mendelsohn and David Traum for their comments on previous drafts of this extended abstract, to the students who used MEMOLAB or PEOPLE POWER and to those who transcripted the videotapes. Parts of this work were supported by grants 4023-27013 and 1114-040711 from the Swiss National Science Foundation.