Vol. 14, No. 4, December 1998

Special Issue: CAL and AI - a time for rapprochement

Artificial intelligence in education: an exploration

G. Cumming

Email: g.cumming@latrobe.edu.au

School of Psychological Science, La Trobe University

The research field of Artificial Intelligence in Education (AIED) embraces a wide diversity of research interests. Psychology, education and cognitive science are strongly represented, alongside computer science and artificial intelligence. A key interest is in modelling, especially of learning processes and cognition. This paper gives a brief outline of the development of AIED, and examples of current issues and projects. The 'AI' in the title may give a misleading picture of a research field that is in fact dynamic and broad, with many links to the classroom.

Keywords: Artificial intelligence; Cognition; Learning environment; Learner modelling; Situated learning

Invited paper

Has Minsky anything to say for education?

M. Boyle

Faculty of Education, Monash University

Email: martin.boyle@education.monash.edu.au

This paper examines some of the concepts outlined by Artificial Intelligence researcher Marvin Minsky in his seminal work, The Society of Mind. The paper takes an overview of the work of Minsky and some of the criticism directed towards it. It then concentrates on two of the most significant agents suggested as models of aspects of the human mind by Minsky: the Frame and the K-Line. An attempt is made to illustrate the significance of the agents of mind for human learning and organising learning milieu and the significance of computing as a meta-model for the Society.

Keywords: Agents; Artificial intelligence; Frames; K-Lines; Learning theory; Minsky

Invited paper

Adding intelligence to a learning environment: learner-centred design?

P. Brna & R. Cox*

Computer Based Learning Unit, Leeds University

* Human Communication Research Centre, Edinburgh University

Email: p.brna@cbl.leeds.ac.uk

The development of a specific learning environment into an intelligent learning environment (switchER II) is used as the basis of a discussion about the nature of 'learner-centred design' - an approach which is contrasted with user-centred design, and is being advocated as an important move in the development of effective educational computing systems. An analysis of the notion of learner-centred design leads to the need to develop appropriate methodologies to support learner-centred design, to ensure that individual differences are respected, and that Artificial Intelligence techniques are applied appropriately.

Keywords: Intelligent learning environments; Learned-centred design; User-centred design

Invited paper

Developing mathematical problem solving skills

B. Abidin* & J.R. HartleyÈ

* MARA Institute of Technology, Malaysia and È University of Leeds

Email: j.r.hartley@cbl.leeds.ac.uk

Problem solving often requires a representation of given information in a structured form which can stimulate and support the problem solving process. FunctionLab is a computer based learning environment meeting this requirement and designed to assist problem solving and the development of problem solving skills in algebra word problems. A feature of the system is its interface design which incorporates tools for students to represent problem information as generic and dynamic process models which support investigatory learning and illustrate the structural characteristics of solutions. The design of FunctionLab is described together with an initial evaluation study of its effects on problem solving performances and methods of solution.

Keywords: CAL; Interface design; Learning environments; Mathematics education; Problem solving; Validation study

Invited paper

Cognitive development and qualitative modelling

J. Ogborn

Institute of Education, University of Sussex

Email: j.m.ogborn@sussex.ac.uk

This paper argues that in order to encourage the learning of high level cognitive skills through the use of computers, it is necessary to use systems which are adapted to forms of natural human reasoning. This is specifically the case in using modelling programs. Two kinds of such reasoning are identified, both qualitative: reasoning using imagined objects and events, and semi-quantitative reasoning. Modelling systems which support each form are described. Some results of research with them are outlined.

Keywords: Cognitive development; Computer assisted learning; Qualitative modelling; Reasoning processes

Invited paper

Dialogues in support of qualitative reasoning

R. Pilkington

Computer Based Learning Unit, University of Leeds

Email: r.m.pilkington@cbl.leeds.ac.uk

Do students always interact with computers reflectively on tasks and improve qualitative reasoning? It seems that students often manipulate software without changing existing conceptions or exploring the implications of their conceptions to generate deeper explanations. This undermines the educational value of such interactions - though dialogue with a peer or adult can make the interaction more valuable. However, it is uncertain when and how such dialogues work. If high quality educational interactions with computers are to take place, then it is necessary to understand the situations in which particular dialogue forms are effective and to find ways of modelling these. In this paper a framework for the design of such interactions is proposed based on a dialogue analysis that employs transactional analysis and logical dialogue game theory. The framework is applied to a medical (simulation-based) learning context to illustrate how it may enhance interactions with such systems.

Keywords: Dialogue analysis; Human-computer interaction; Qualitative reasoning; Simulation-based learning

Invited paper