Modelling dialogue and beliefs as a basis for generating guidance in a CSCL environment

Kristine Lund (1) , Michael Baker (1, 2, 3, 4, 5) , Monique Baron (6)

1 ICAR - Interactions, Corpus, Apprentissages, Représentations
2 CNRS - Centre National de la Recherche Scientifique
3 LEAPLCE - Laboratoire d'études sur l'acquisition et la pathologie du langage chez l'enfant
4 MoDyCo - Modèles, Dynamiques, Corpus
5 GRIC 5 - Groupe de Recherches sur les Interactions Communicatives
6 LAFORIA-IBP - Laboratoire Formes et Intelligence Artificielle

Existing approaches to integrating ITS and CSCL environments have not attempted to model the collaborative interaction between human learners in order to exploit its pedagogical potential. We present an approach to modelling such interactions and students' underlying beliefs, within the framework of a CSCL environment designed for learning the concept of energy in physics (C-CHENE). The environment is based on flexible structuring of the collaborative interaction using a specially designed communication interface, and a dialogue grammar. We propose three types of automatic guidance that could be provided on this basis.

Mots clés: CSCL

Domaines: Linguistique; Education; Environnements Informatiques pour l'Apprentissage Humain

Kristine Lund, Michael Baker, Monique Baron. Modelling dialogue and beliefs as a basis for generating guidance in a CSCL environment. Intelligent tutoring systems: third International Conference, ITS '96, Jun 1996, Canada. pp.206-214. ⟨halshs-00195338⟩