This paper attempts to rehabilitate student models within intelligent tutoring systems. Recently, some researchers have questioned both the need for detailed student models and the practical possibility of building them. We regard it as axiomatic that any intelligent tutoring system needs a student model. This paper suggests some practical guidelines and changes in philosophical approach which may help in building effective student models. 1. Introduction In a review of the 1987 "Artificial Intelligence and Education" Conference, Sandberg (1987) summarised a general opinion that "detailed user models do not necessarily enhance the capability of an intelligent tutoring system ... good teaching can do without a detailed user model, because in good teaching serious misconceptions are avoided, and errors will be repaired on the spot ... it is debatable whether the cost of constructing very 3 detailed, complex user models that are runnable and have to be maintained all the time is worth...

User analysis is a crucial aspect of user-centered systems design, yet Human-Computer Interaction (HCI) has yet to formulate reliable and valid characterizations of users beyond gross distinctions based on task and experience. Individual differences research from mainstream psychology has identified a stable set of characteristics that would appear to offer potential application in the HCI arena. Furthermore, in its evolution over the last 100 years, research on individual differences has faced many of the problems of theoretical status and applicability that is common to HCI. In the present paper the relationship between work in cognitive and differential psychology and current analyses of users in HCI is examined. It is concluded that HCI could gain significant predictive power if individual differences research was related to the analysis of users in contemporary systems design.

The learning behaviour and performance of 65 postgraduate students using a hypermedia-based tutorial were measured. Data were also obtained on cognitive style, levels of prior experience, motivation, age, and gender. A number of statistically significant interactions were found. Field-dependent/independent cognitive styles were linked to strategic differences in navigation. Levels of prior experience were linked to quantitative differences in both navigation behaviour and learning performance. The implications of these findings are discussed. The rapid rise in the use of the World Wide Web (WWW or Web) in teaching and learning has brought hypermedia into prominence as a mode of information accessing. The term “hypermedia ” signifies both mode and media of information presentation. Hypermedia may be distinguished from hypertext insofar as the former may include sound and/or moving images in addition to text. However, the research reported here focuses on the hyper element

andaspectsofresearchers’problem-solvingandrelated information-seeking behavior. One hundred and eleven postdoctoral researchers were assessed for Witkin’s field dependence/independence using Riding’s Cognitive Styles Analysis and for Pask’s holist/serialist biases usingitemsfromFord’sStudyProcessesQuestionnaire. These measures were correlated with the researchers’ perceptions of aspects of their problem-solving and information-seeking behavior, and with those of the search intermediary who performed literature searches on their behalf. Anumber of statistically significant correlations were found. Field-independent researchers weremoreanalyticandactivethantheirfield-dependent counterparts. Holists engaged more in exploratory and serendipitous behavior, and were more idiosyncratic in their communication than serialists.

Abstract. This paper describes a study in which individual learner models were built for students and presented to them with a choice of view. Students found it useful, and not confusing to be shown multiple representations of their knowledge, and individuals exhibited different preferences for which view they favoured. No link was established between these preferences and the students’ learning styles. We describe the implications of these results for intelligent tutoring systems where interaction with the open learner model is individualised. 1

The study described in this article investigated how differences in cognitive style and online search experience influence the search performance and navigational pattern of individuals utilizing a university World Wide Web (WWW) site. Forty-eight undergraduate students with diverse academic backgrounds participated in the study. On the basis of cognitive style and prior experience with online database search, the participants were assigned to 1 of the following groups: (a) field-dependent (FD) with little or no online search experience, (b) FD with substantial online search experience, (c) field-independent (FI) with little or no online search experience, and (d) FI with substantial online search experience. It was found that cognitive style influenced search time, whereas online search experience affected navigational style, such as jumps and layer traversal. Cognitive style and online search experience also interacted to influence search performance and navigational style. FDs with little or no online search experience navigated the WWW in a fairly linear mode, using embedded links frequently. They also tended to visit more nodes and used “Home ” more frequently than the rest. The results imply that as FD searchers gain more online search experience, their navigational style and search performance change and become comparable to that of FDs. Based on these findings, some suggestions are made to improve the WWW interface and WWW user training programs. 1.

Abstract not found

Abstract: Multimedia learning environments combine multiple forms of representations like texts, static and animated pictures or graphs. Knowledge acquisition from multiple representations requires that learner create referential connections between corresponding elements and corresponding structures in different representations. As this process is usually difficult, learners frequently fail to construct coherent mental representations and, thus, do not sufficiently understand the subject matter. This paper analyzes the effects of different kinds of instructional help on the process of coherence formation from multiple representations by learners with different prior knowledge. Three groups of university students with different domain-specific knowledge had to learn a complex subject matter from chemistry using six different forms of representation. In addition, a first group received directive help for coherence formation. A second group received non-directive help, and a third group received no instructional help. Results indicate that directive help is effective for recall performance because of its summarizing and repeating function. Furthermore learners with different levels of prior knowledge show different reactions when help is given. For learners with insufficient prior knowledge help is not helpful or in case of recall performance even harmful. Learners with a medium level of prior knowledge can increase

There has been tremendous growth in Web-based instruction over the past few years. Because the user group of Web-based instruction includes learners from heterogeneous backgrounds, individual differences become an important issue in the development of Web-based instructional programs. Among a variety of individual differences, cognitive style is a particularly important characteristic. This study aims to determine the relationships between learners ' cognitive styles and their perceptions and attitudes toward the features of a Web-based instructional program. The results indicate that cognitive styles influence students ' reactions to nonlinear interaction, independent learning, and navigation tools and the difficulties and problems that they encounter. The findings are applied to develop a learning model that can support the design of Web-based instructional programs. 1.

Hyperproof is a multimodal tool designed for the teaching of first order logic. An experiment was conducted which gathered computerised logs of exam answers by a group of subjects undertaking this course. The current study utilises natural language corpus-based statistical tests to analyse the proof logs. A pre-course test was administered to the subjects on the Hyperproof class, and subjects were classified according to their performance on this test. Interesting differences in rule use and proof style were found between these groups. Considered with reference to theories of the fundamental reasoning mechanism, against the background of the individual difference literature, the current study has implications for several areas of cognitive science. First, the idea that there is just one observable reasoning mechanism underlying human inference is challenged. Second, support is provided for the position that graphical methods ought to be encouraged in reasoning. Third, with regard to hu...