Vol. 16, No. 2, June 2000

Special Issue: Learning from hypermedia systems: cognitive approaches

Guest Editor: Jean-Francois Rouet, Laboratoire Langage et Cognition, University of Poitiers and Centre National de la Recherche Scientifique

Guest editorial: hypermedia and learning – cognitive perspectives

J-F. Rouet, Laboratoire Langage et Cognition, CNRS et Université de Poitiers

This Special Issue of JCAL focuses on the potential of hypermedia systems as learning tools, considered from cognitive perspectives. The term hypermedia results from the combination of ‘hypertext’ and ‘multimedia’. Both terms are now widely used to designate a variety of computer applications. From a more technical point of view, however, they are not synonymous. The notion of ‘hypertext’, coined in the sixties by Ted Nelson, was introduced by Conklin (1987) in the following terms:

"Mechanisms are being devised which allow direct machine-supported references from one textual chunk to another; new interfaces provide the user with the ability to interact directly with these chunks and to establish new relationships between them. These extensions of the traditional text fall under the category of hypertext (also known as non-linear text)." (Conklin, 1987, p. 17)

According to Conklin, a hypertext consists of (a) a database in which nodes are text chunks, and links are references from one chunk to another, and (b) a computer interface which allows the user to display and browse through the database. In other words, hypertext refers to both the contents (a set of interconnected texts) and computer-based tools to access those contents through a user-interface.

Kommers defines multimedia as "those computer-based applications that allow the user to see and hear different types of information via one screen with audio support" (Kommers, 1996, p. 4). Multimedia may include texts, pictures and possibly other types of information. According to Kommers, "hypermedia are computer-based applications for consulting multimedia resources" (ibid., p. 6). Thus hypermedia systems may be broadly defined as a family of computer applications intended to store and retrieve vast amounts of multimedia documents interconnected in nonlinear ways (the process of creating links is in itself part of the concept of hypermedia). ‘Nonlinearity’ means that each unit or page in a hypermedia document may be connected through several others by means of ‘hyperlinks’. This enables a new type of interaction between learners and documents called ‘browsing’ which is believed to have a positive effect on learning (Spiro & Jehng, 1990).

Each page in a hypermedia system may feature texts, pictures, sounds presented in static or dynamic ways, and according to visual or auditory sensory modalities. Again, multimedia presentation of information is considered beneficial to learning, at least in certain conditions (Mayer, 1997). Finally, in addition to content information, hypermedia documents usually include a variety of ‘navigation tools’, ranging from keyword search, explicit or embedded menus, history trees and so forth. The multiplicity of means available for information retrieval may also promote active learning strategies in students.

It is now acknowledged that the effectiveness of hypermedia systems as learning tools depends to a large extent on their compatibility with the psychological processes by which students perceive, understand and learn from complex information sources. As a result, the psychology of hypermedia learning is emerging as a new area of research (McKnight et al., 1993; Rouet et al., 1996; van Oostendorp & de Mul, 1996). Cognitive models of information processing, multimedia integration and memory development are used to generate hypotheses about the effectiveness of specific design decisions. This line of research may contribute to a better understanding of the potential of hypermedia technologies as learning tools. However there is a need to bridge the gap between laboratory research and the concrete practice of educational multimedia design and application. This is the core purpose of the present Special Issue.

The papers featured in this issue of JCAL originated in a conference held in Poitiers (France) on the topic of of ‘hypermedia and learning’ (Rouet & de la Passardière, 1999). The conference aimed at promoting communication between researchers and practitioners of hypermedia technologies in education. Indeed, several contributions illustrated where and how research and practice can meet. Some authors were invited to submit revised versions of their papers for this Special Issue. In addition to popularising the cognitive approach of hypermedia design among the JCAL readership, this issue is also meant to bring some work conducted in French speaking institutions (mostly in France and Switzerland) to the attention of English speaking colleagues.

The paper by Tricot and colleagues discusses the issue of learner support in hypermedia systems as a function of the type of learning activity considered. Clearly hypermedia may support a wide range of learning activities which differ with respect to the cognitive processes involved. Hence the need to think of hypermedia devices in terms of the specific learning processes which they may support. Tricot et al. argue for a distinction between three main types of learning activities: learning-by-doing (which focuses on problem solving), learning-by-instruction (which focuses on comprehension, i.e. the building of mental representations from external representations), and learning-by-exploration (which, according to them, shares some characteristics of the two previous categories). For each type of learning considered, they review and discuss the devices provided by hypermedia systems.

The other papers focus on various issues related to information integration in hypermedia. In its technological sense, integration is the process by which several tools or components are assembled in a given hypermedia system. The term integration, however, also refers to the way different pieces of information are related in a person's mind during comprehension or learning. Understanding the relationships between technological integration and cognitive integration is essential in the design of effective hypermedia systems. The papers presented in this issue address the issue at two distinct levels: a global level and a local level.

Global integration refers to the top-level organisation of documents in the system, as well as to the nature of the information management tools provided for the user. The main obstacle to profitable uses of hypermedia in secondary education lies in their very defining characteristic: the capacity to display vast amounts of ill-structured documentary information in the form of texts, pictures, sounds intertwined in complex networks. Empirical studies have consistently found that most students tend to get lost and overwhelmed in such environments, which results in their inability to retrieve useful information or to retain it for further use (Wright, 1991). Hence the need to focus on learners' strategies in order to get some insight about which tools they actually need.

Beaufils examines the strategies of teenage students exploring a multimedia database on Ancient Greece in order to answer broad cultural questions. His data illustrate the difficulty, for inexpert students, to handle the search, collection and structuring of pieces of information disseminated throughout a large database. Interestingly, students seem aware of the difficulty of the task, and they tend to use rather conservative strategies. For instance, they use familiar and cognitively ’light’ search tools rather than more powerful ones. The reliance on familiar, well understood tools seem to be a constant in novices' behaviour with complex information technologies. Some years ago, Weyer (1982) mentioned that his high-school first-time users of a NoteCards-based electronic history textbook tended to ignore the new and powerful search tools featured in the system. This type of result has been found in other studies of educational hypertext usage. For instance, Britt et al. (1996) observed that students preferred to go back to the table of contents of a hypertext system rather than to use direct document-to-document links, a slower but safer navigation strategy. Beaufils' fine-grained analysis of students' strategies also demonstrates that any feature of the interface has the potential to influence — in either way — the learner's strategy. The study provides a concrete example of how multimedia systems can give way to novel information-based activities. These activities can undoubtedly develop students' informational skills (e.g. document awareness, planning and evaluation abilities). The issue of whether such skills can transfer to lower-tech environments is open for further investigation.

Scherly and colleagues compared the effectiveness of two computer environments for learning virology. One environment (VIROLAB) comprises both a simulation environment and a 96-page hypertext. The other environment comprises only the hypertext section. Twenty eight medical students participated in a 2-hour learning session, using either environment to answer a series of questions about virology. Even though the average gain between pre- and post-test was about the same in the two groups, they found that VIROLAB students were better able to answer those questions which required an understanding of some critical processes, e.g. how a virus develops. This is interpreted as evidence of the value of learning-by-doing as opposed to learning-by-reading a hypertext. In addition Scherly et al. observed that some ‘active search’ tools present in the hypertext (e.g. hyperlinks, glossary, search engine) were seldom used, even though all the participants were categorised as ‘knowledge seekers’, that is, students with an active knowledge construction attitude.

Collaud and colleagues present the results of a research and development study into the pedagogical uses of a multimedia database at the university level. They designed a CD-ROM which was integrated in the pedagogical organisation of an undergraduate psychology course. In designing their system, they attempted to take advantage of the new medium while avoiding some well known problems, e.g. student disorientation, cognitive overhead and task management problems. The design decisions that address those issues may be seen at two levels. First, the use of the CD-ROM was carefully introduced and embedded in a series of pedagogical practices (tutoring, discussion seminars etc.) so that the new tools did not appear as disconnected from students' previous learning experience. Second, the design of the database reflected a concern for explicit structuring, various levels of guidance, and assistance in forming and managing study goals, thus making use of the recent psychological research literature in this area. The preliminary results of the course evaluation showed encouraging results, in that students seemed to cope with the tasks rather well and also performed well in the final examination. Moreover student performance appeared to relate to their actual use of the system, as evidenced by their memory for system commands and interaction procedures.

Local integration has to do with the way information sources are integrated at the level of a single display, a problem close to that of page layout in traditional publishing. The problem is however more complex in hypermedia because many more parameters can vary due to multimodal, dynamic, interactive capacity. One central issue is how to integrate verbal and pictorial information in displays.

There is a general belief that multimedia systems are effective learning tools simply because they include illustrations combined with other sources of information. Merlet reports previous findings which challenge this general and simplistic view. She stresses that comprehension is a complex, multilayered activity. Each source of information present on one display takes some processing resources in order to be perceived and understood. In the case of foreign language learning, the synchronous processing of multiple sources can be overwhelming. As a consequence, pictorial information, if not properly integrated with other sources, can disrupt rather than facilitate comprehension.

Dubois and Vial present an experiment which investigated the effects of different presentation modes on learning Russian words and their French translation. They found that a picture illustrating the meaning of the word is useful only to the extent that it promotes the integration of the word with both its meaning and its phonological representation. An oral comment added to the image promoted learning better than a written comment. However, the results were also function of the type of cue the students received at the time of recall. The integrated image with oral commentary was most effective when the test word was presented orally. The latter result may be interpreted as an instance of a retrieval specificity effect (retrieval is most effective when it occurs in the same context as learning). The practical lesson of this study is that designers should be careful to include only those illustrations that will actually promote the construction of integrated mental representations.

Gyselinck and colleagues examine the effects of illustrations as a means of improving the comprehension of simple explanations in physics. Their experiment shows the facilitative effect of illustrations, especially for answering inferential questions. Moreover, illustrations are most beneficial to students with a higher visuo-spatial capacity. They suggest that multimedia presentation may have different impacts depending on students' characteristics. Presentations that involve too complex text-picture compounds may be detrimental to students with a poorer cognitive capacity.

The papers presented in this Special Issue illustrate the complexity and the diversity of the cognitive processes involved in learning with complex information technology. To put it simply, whereas many believe that powerful computer technologies will make learning more simple, research suggests that it only allows for more complex activities to take place. Whether or not these activities will prove productive in terms of learning depends on a complex interaction between the properties of the device and the learner's cognitive characteristics.

Learning in autonomy with a hypermedia system is a complex activity. Beaufils showed that students benefit a lot from some kind of pre-structuring of the activities, as well as from notetaking, information selection or marking tools that let them draw external representations of intermediate goals, and hence reduce the cognitive load. Scherly and colleagues’ observation that students postpone hypertext search until they have solved the problem also suggests that they try to keep the burden at a manageable level. Both studies also suggest that students are quite reluctant to use novel tools, probably due to the fact that they do not possess adequate ’procedural schemata’ for those tools. The same problems occur at a more local level, when students are confronted by displays that incorporate verbal and graphical information. Students' ability to learn from multimedia information varies as a function of cognitive dimensions, e.g. their level of proficiency in the domain (Merlet) or their working memory capacity (Gyselinck et al.). Modal comple-mentary, or the fact that several sources of information converge towards a single mental representation is a key aspects of multimedia displays (Dubois & Vial).

Together these studies show that the incorporation of information technology in current teaching systems calls for a deep reflection about pedagogical engineering. At the micro-level, pedagogical engineering consists of incorporating psychological and ergonomical considerations into the very design of the system. At an inter-mediate level, pedagogical tasks and activities have to be carefully tailored to the properties of the medium, the characteristics of student thinking and study strategies, and the pedagogical requirements. Finally, at the global level, one has to consider the use of multimedia technology in the context of pedagogical practice, e.g. lecturing, tutoring, and other forms of student-teacher interaction. Becoming a good engineer certainly takes a good deal of practice in real teaching contexts. It may nevertheless benefit from advances in the psychology of complex information processing.

References

Britt, M.A., Rouet, J.-F. & Perfetti, C.A. (1996) Using hypertext to study and reason about historical evidence. In Hypertext and Cognition (eds. J-F. Rouet, J.J. Levonen, A.P. Dillon & R.J. Spiro). pp. 43-72. Lawrence Erlbaum Associates, Mahwah, NJ.

Conklin, J. (1987) Hypertext: an introduction and survey. Computer, 20, 17-41.

Kommers, P.A.M. (1996) Definitions. In Hypermedia Learning Environments (eds. P.A.M. Kommers, S. Grabinger & J.C. Dunlap) pp. 1-11. Lawrence Erlbaum, Mahwah, NJ.

Mayer, R.E. (1997) Multimedia learning: Are we asking the right questions? Educational Psychologist, 32, 1, 1-19.

McKnight, C., Dillon, A. & Richardson, J. (1993) Hypertext: A psychological perspective. Ellis Horwood, Chichester.

Rouet, J-F., Levonen, J., Dillon, A.P. & Spiro, R.J. (eds.) (1996). Hypertext and Cognition. Lawrence Erlbaum Associates, Mahwah, NJ.

Rouet, J-F. & de la Passardière, B. (eds.) (1999) Hypermédia et Apprentisages 4, Actes du quatrième colloque (Université de Poitiers). INRP et EPI, Paris.

Spiro, R.J. & Jehng, J.C. (1990) Cognitive flexibility and hypertext: Theory and technology for the nonlinear and multidimensional traversal of complex subject matter. In Cognition, education and mulrtimedia: Exploring ideas in high technology (eds. D. Nix and R.J. Spiro) pp. 163-205. Lawrence Erlbaum Associates, Hillsdale, NJ.

van Oostendorp, H. & de Mul, S. (eds.) (1996) Cognitive aspects of electronic text processing. Ablex Publishing Co, Norwood, NJ.

Weyer, S.A. (1982) The design of a dynamic book for information search. International Journal of Man-Machine Studies, 17, 87-107.

Wright, P. (1991) Cognitive overheads and prostheses: Some issues in evaluating hypertexts. In Proceedings of the Third ACM Conference on Hypertext (eds. R. Furuta & D. Stotts) pp. 1-12. ACM Press, New York, NY.

Specific help devices for educational hypermedia

A. Tricot, C. Pierre-Demarcy & R. El Boussarghini, Brittany University for Teachers' Education and the University of High Brittany, France

Email: andre.tricot@bretagne.iufm.fr

This paper examines the issue of help devices for students in hypermedia environments according to the learning context and, more precisely, according to the cognitive processes involved in that situation. The argument is based on a criticism of the level of generality of traditional help systems which are mostly concerned with user-system interaction. Three main types of learning situations are described: learning-by-doing, learning-by-instruction, learning-by-exploring and the main cognitive processes involved in these three situations are explored. From such descriptions, specific types of help devices are discussed and some perspectives relating to the design and evaluation of educational hypermedia applications are suggested.

Keywords: Courseware; Ergonomics; Guided discovery; Help; Hypermedia; Instruction; Secondary; Student-centred

Journal of Computer Assisted Learning (2000) 16, 2, 102-113

A translated version of a paper based on one selected from Hypermédia et Apprentisages 4, Actes du quatrième colloque (eds. J-F. Rouet et B. de la Passardière). INRP et EPI, Paris.

Understanding multimedia dialogues in a foreign language

S. Merlet, Laboratoire Langage & Cognition, CNRS, Université de Poitiers, France

E-mail: Sylvie.Merlet@mshs.univ-poitiers.fr

The effects of lexical and semantic previews on the comprehension of a computer-presented illustrated dialogue were examined. Based on a multilevel cognitive theory of comprehension, different types of preview were expected to influence specific aspects of the learners’ activity. Thirty French undergraduate students of English as a foreign language were asked to study a computerised illustrated dialogue. A lexical preview decreased the frequency of control actions while listening (e.g. pauses, replays) while the semantic preview improved the quality of information recalled. It is concluded that a fine-grain analysis of language comprehension processes can improve the design of educational multimedia products

Keywords: Control group; Foreign language; Language comprehension; Multimedia; Student-centred; Undergraduate

Journal of Computer Assisted Learning (2000) 16, 2, 148-156

A translated version of a paper based on one selected from Hypermédia et Apprentisages 4, Actes du quatrième colloque (eds. J-F. Rouet et B. de la Passardière). INRP et EPI, Paris.

------------------------------(back to Issue 2 index)

Evaluation of hypertext in an activity learning environment

D. Scherly, L. Roux & P. Dillenbourg, Centre Médical Universitaire and TECFA-FPSE, University of Geneva, Switzerland

Email: Daniel.Scherly@medecine.unige.ch

This paper presents a study which evaluates the use and efficiency of two computer learning tools for basic virology. The first, VIROLAB, is a simulation of a biology laboratory in which the learner uses virtual lab facilities to help defective viruses to multiply. The system also includes a hypertext that provides learners with some knowledge on virology. The second tool is the hypertext which has been extracted from VIROLAB and which is now considered as an independent learning environment. This study addressed two specific questions: (1) which of the two tools was the most efficient for knowledge acquisition, and (2) were there any differences in the use of the two hypertexts? Comparison of pre- and post-test scores showed that neither of the two learning tools is better than the other. However, analysis of the navigation paths showed that VIROLAB users dissociated the reading of the integrated hypertext from the problem resolution activities. The implications of this finding for the design of the learning environment are discussed.

Keywords: Empirical; Hypertext; Medical education; Navigation; Problem solving; Simulation; Undergraduate; Virology

Journal of Computer Assisted Learning (2000) 16, 2, 125-136

A translated version of a paper based on one selected from Hypermédia et Apprentisages 4, Actes du quatrième colloque (eds. J-F. Rouet et B. de la Passardière). INRP et EPI, Paris.

Visuospatial working memory in learning from multimedia systems

V. Gyselinck, M-F. Ehrlich, C. Cornoldi, R. de Beni & V. Dubois

Laboratoire de Psychologie Expèrimentale, Universitè Renè Descartes Boulaogne, France and Dipartimento di Psicologia, Universita di Padova, Italy

Email: gyselinck@psycho.univ-paris5.fr

Multimedia systems involve the association of various types of information: verbal information presented visually or auditorily, static or dynamic pictorial information, and sound information. In a cognitive approach, integrating this information involves complex processes constrained by properties of the learner's cognitive system, and especially by the capacity of working memory. This paper reports on an experiment focused on the integration of verbal and pictorial information when students learn a series of physics concepts. The involvement of the visuo-spatial working memory was investigated by means of a dual-task paradigm. Results show that pictorial information enhances the learning process. They also suggest that the visual and the spatial components of visuospatial working memory should be considered. They emphasise the need to consider the limitations in cognitive resources available to the learner.

Keywords: Control group; Multimedia; Physics; Undergraduate

Journal of Computer Assisted Learning (2000) 16, 2, 166-176

A translated version of a paper based on one selected from Hypermédia et Apprentisages 4, Actes du quatrième colloque (eds. J-F. Rouet et B. de la Passardière). INRP et EPI, Paris.

Multimedia design: the effects of relating multimode information

M. Dubois & I. Vial

Universitè Pierre Mendës France, Grenoble, France

Email: michel.dubois@upmf-grenoble.fr

Few models describe learner behaviour during the simultaneous processing of several types of information, yet this is the defining characteristic of the use of multimedia tools, which bring together media in different informational formats (fixed or moving images, sound, text). Following studies in cognitive psychology concerning the increase in the ability to form mental images of words, this study was aimed at defining how different multimedia presentation modes affect the learning of foreign language vocabulary (Russian). A statistically significant effect was observed on word memorisation in the different information presentation modes, suggesting better processing when there is co-referencing of the different sources, especially when the encoding and tests modes are the same. In addition to these experimental results, some principles for the design of multimodal learning tools are discussed.

Keywords: Cognitive load; Experiment: Language learning; Multimedia, Process; Russian; Undergraduate

Journal of Computer Assisted Learning (2000) 16, 2, 157-165

A translated version of a paper based on one selected from Hypermédia et Apprentisages 4, Actes du quatrième colloque (eds. J-F. Rouet et B. de la Passardière). INRP et EPI, Paris.

Design and use of a hypermedia system at the University level

G. Collaud, J-L. Gurtner & P-F. Coen, University of Fribourg, Switzerland

E-mail: gerald.collaud@unifr.ch

This paper .reports on an experiment in which a whole semester course in psychology was replaced by a mixed formula consisting of a CD-ROM complemented by a series of seminars and workshops. The CD-ROM was conceived as a collection of hyper-media documents (hypertexts, research data, references, videos and activities) linked together with genuine Netscape facilities. Students were invited to search through these documents for information to answer a questions (called challenges) on the topic. A multiple-choice questionnaire accompanies each challenge in order to foster students' self-evaluation. The seminars, held every other week, served both as forums to discuss each of the topics under the guidance of an expert,. Careful analysis of students' answers to two questionnaires at the beginning and at the end of the course and during interviews, showed that such a formula was generally favourably accepted by a large majority of students, although it elicited anxiety and lead to greater work load than a traditional course. Positive effects were also observed on learning.

Keywords: CD-ROM; Hypermedia; Problem solving; Psychology; Undergraduate; Illuminative evaluation.

Journal of Computer Assisted Learning (2000) 16, 2, 137-147

A translated version of a paper based on one selected from Hypermédia et Apprentisages 4, Actes du quatrième colloque (eds. J-F. Rouet et B. de la Passardière). INRP et EPI, Paris.

Tools and strategies for searching in hypermedia environments

A. Beaufils, Institut National de la Recherche Pédagogique, Montrouge, France

Email: abeaufls@inrp.fr

This paper presents the results of several investigations into the use made by students of a prototype system of help for the exploitation of hypermedia documents. This prototype was derived from a common type of educational CD-ROM in which the presentation of the contents was improved and certain research and note-taking tools added. The evolving, personalised note-taking system interacts constantly with the environment being explored. Beyond the storage of personal notes (clues, personal knowledge of the area), it allows the user to keep the explored sections in mind and ultimately to plan further explorations. It also enables the extraction and structuring of the data retrieved from the body of the database. The role of this system is to facilitate more thorough searches for information which might sometimes require several sessions stretched out over a long time. It could be profitably used in the context of independent student activities.

Keywords: CD-ROM; Empirical; Guided discovery; Hypermedia; Navigation; Secondary

Journal of Computer Assisted Learning (2000) 16, 2, 114-124

A translated version of a paper based on one selected from Hypermédia et Apprentisages 4, Actes du quatrième colloque (eds. J-F. Rouet et B. de la Passardière). INRP et EPI, Paris.