Last modified: Mon Aug 16 18:33:58 1999
held at the
Recent learner modelling research has been exploring the opening of the learner model, in the sense of allowing the learner to inspect and control the model, as well as to participate in its creation and management. Several modifiers have been used in referring to this sort of learner modelling: overt, inspectable, open, participative, cooperative, collaborative, learner-controlled, etc. The aim of every approach has been either to improve the quality of the models or to obtain additional benefits from their construction and maintenance beyond the usual adaptation of the environment.
Given the variety of approaches taken, and the different names given frequently to very similar approaches, the aim of this workshop is to serve as a communication medium for people involved in these areas of research to share experiences, results, and promising research directions. Hopefully, a more coherent picture of the broad spectrum of current investigations will emerge as a result.
A general overview of the workshop content, that is also the introduction to the workshop proceedings, can be found in overview.html
A frequent reason for developing open learner models has been that learner collaboration in the modelling task would result in more accurate models of some sort. Other research has suggested, however, that learners know very little about themselves.
Another common reason for opening learner models to learner inspection has been to encourage learners to reflect upon, and be aware of, their own knowledge. Reflection and awareness would then lead to improving the quality of the knowledge acquired by learners.
Even if overt learner models were not more accurate, nor they resulted in an immediate improvement on learners' knowledge, they could still be useful tools for a variety of purposes to improving the learning process (for example, facilitating collaboration among learners).
Another indirect advantage of overt learner models could be that the ratio cost/benefit were better for them than for more traditional (covert) models.
Finally, there could be fundamental ethical reasons for allowing learners' control of learner models, or even of the whole learner modelling process, despite all its possible drawbacks.
In the first half of the workshop each speaker will make a short presentation on their work. They will be asked to address at least one of these five questions in their presentation. The second half of the workshop will be devoted to a discussion focused on the five questions listed above.
Well, the rush is over. The workshop was an enjoyable experience, thanks to all the participants who provided an intellectually rich and socially friendly environment. Hopefully, however, the energy is not over and we will be soon able to collect here conclusions from the workshop, as well as follow up discussions.