Philip Barker and Paul van Schaik

Human-Computer Interaction Laboratory
University of Teesside, UK

As a consequence of the activities in which they become involved, people build a variety of different types of cognitive structure in order to hold the information and knowledge that is important to them. One of the most important types of structure that people build are ‘mental models’. These are often regarded as representations created ‘in the head’ as a result of exposure to various kinds of ‘learning’ event. These may be of either an implicit or an explicit nature. The quality of these learning events strongly effects the richness of the mental models that are produced and, subsequently, the quality of a person’s task performance in a problem solving or goal seeking situation. This idea is embedded in the mental model hypothesis which states that the quality of task performance in a given domain is strongly depended on the quality of an individual’s mental models. This has significant implications for the design of computer-based learning (CBL) resources and the ways in which they are used.

Because of the importance of mental models as a basis for designing effective CBL materials we have conducted a programme of research which identifies appropriate tools and methods of using these tools for studying mental models. For the purpose of this study we have used the following tools: rating, sorting, laddering, concept mapping, GST diagramming and teach-back. These have been used as a basis for two experimental studies. The first involved a qualitative investigation of students’ mental representations of Web browsers. The second involved an in-depth quantitative study of Microsoft’s Word for Windows word-processing system. The latter study also involved measuring students’ performance with MS-Word. This paper describes the methods that were used in the experimental studies, the results obtained and provides supporting evidence for the mental model hypothesis. Finally, the implications of this work for the design of computer-based learning materials are discussed.

Address for correspondence: Professor Philip Barker, School of Computing and Mathematics, University of Teesside, Borough Road, Middlesbrough, TS1 3BA, UK. Tel: 01642-342660; Fax: 01642-342067; Email: Philip.Barker@tees.ac.uk