Chapter 7
Educational Technology and Multimedia From a Cognitive Perspective: Knowledge From Inside the Computer, Onto the Screen, and Into Our Heads?
Erica de Vries
Laboratory for Educational Sciences, University of Grenoble II, France
Computer programs specifically designed for educational settings form a large share of the multiple uses of multimedia technology for human information processing. In an educational setting, multimedia, or more generally, the way information is presented, is the essential passageway from information stored in computer memory to knowledge constructed by the learner.
A computer program for learning embodies a specific view on the knowledge in a domain, and on the way in which learners acquire knowledge. This chapter claims that the way in which multimedia are used in effect is a result of theoretical views on how learning takes place and on what role the computer should fulfill in the ideal learning context. After an organization of existing research into three subfields, the chapter examines the consequences of multiple perspectives on learning and multiple educational approaches for multimedia research.
MEDIA, MULTIMEDIA, AND HYPERMEDIA RESEARCH
Ever since teaching and learning can be mediated, that is, may take place with an intermediary such as a book or a blackboard between teacher and learner, the question can be asked as to which is the optimal way of transferring knowledge. This is a rather delicate question because it needs specifying what is meant by the optimal way and by transfer of knowledge. For a long time, the question has been treated in terms of establishing the instructional method that is most cost-effective and that produces high outcomes expressed in learning results. With the development of each new technology, the efficiency question can be asked again, giving rise to a new line of research comparing existing methods with an approach involving new technology. This section briefly structures the field into three different subfields corresponding to the three major debates that have been taking place in this area: the fields of media, multimedia, and hypermedia research.
Media
The first debate centers around the question of whether or not using different media for teaching has an effect on learning. In the context of this question, a medium denotes the channel for passing on information. Possible media are a human, a book, a radio, a television, or a computer. According to Clark (1983), one of the participants in the debate, a particular medium does not influence learning because it can be considered as a mere vehicle for delivering instruction. Clark describes two confusions that, he argues, are at the origin of studies that report media effects on learning. The first confusion is between a medium and the instructional method selected, for example, when comparing a human and a computer (media) delivering respectively a lecture and large numbers of exercises (instructional methods). The learning effects measured then can be imputed to the instructional method rather than to the medium. The second confusion is between a medium and its media attributes, for example when a learning effect is attributed to video (medium) whereas in fact the effect is caused by the possibility of zooming in on details (media attribute). Because an attribute may not be exclusive to a medium, zooming in on details can be done on video but also on the computer, Clark (1983) argued that it is unjustified to impute the effect to the medium. According to Clark, the two sources of confusion have led to a large number of unjustified claims about the appropriateness of particular media for teaching and learning.
In reaction to Clark's (1983) reasoning, one could argue that it is a rather artificial operation to want to separate a medium from its defining characteristics. This argument can be found in Kozma (1991). In Kozma's interpretation, a medium consists of a technology, a symbol system, and a number of processing characteristics. By technology, Kozma means the physical, mechanical, and electronic aspects that determine the function of the medium, and the surface characteristics that make something be an instance of a specific type of medium. For example, a television is something that emits sound and pictures and that has a screen. But more important are the symbol systems (text, pictures) a medium can employ and the processes that can be performed with it. For example, a radio cannot be used to display pictures (symbol system) and a book cannot easily be used to search for all the occurrences of a certain word (processing characteristic). Following Kozma's (1991) reasoning, it is more fruitful to focus on the influence of different symbol systems and processing capabilities on learning. He makes a strong case for this type of research by stating that the choice of a medium in fact does not enforce exploiting all of its possible symbol systems and processing capabilities (a television that becomes radio, a video-disc player that becomes broadcast television). It is only the capabilities actually used that can be expected to have an effect on learning processes and outcomes. Consequently, Kozma's (1991) review of the media question focuses on the specific characteristics of symbol systems that can be used in a particular medium. Effects on learning are studied by looking at how learners construct, structure, and modify their representations of the information, and the quality of the information processes that act on these representations. The main measures used involve free or cued recall of the presented information and tests involving comprehension and solving new problems (see Kozma, 1991, for an overview).
Both aspects of a medium, the symbol system and the processirig capabilities, are at the heart of the second and third subfields dealt with in this section, respectively the multimedia and hypermedia research fields.
Multimedia
The second debate focuses on the benefits of multimedia for learning. Multimedia in this context designates concurrent or consecutive presentation of information using text, pictures, sound, animations, and so on. Studies in this field involve comparisons of two or more ways of presenting information such as text and pictures, animations and narrations, or printed and spoken text. In addition, some comparisons focus specifically on whether the information is presented concurrently or consecutively and in which order. At the outset, studies in the field aimed at establishing the value of adding illustrations to instructional texts. Numerous overviews to the literature demonstrate the importance of the issue (e.g. Levie, 1987, Mayer, 1993). With the introduction of the computer, the field has exploded due to the possibilities of using animations and narrations. Multimedia research now embodies all research into presenting information and literature overviews on this enlarged field have started to emerge (e.g. Najjar, 1996). In this line of research, the technology for presenting information is not specifically considered as part of the setting. For example, studies focusing on learning with text and pictures may use pieces of paper or a computer screen as a display device. However, studies may focus on the particular characteristics of the symbol system employed. For example, texts and pictures provide stability, whereas sound or animations are transient. Examples of factors studied are visual versus auditory, verbal versus nonverbal, transient versus stable, symbolic versus nonsymbolic, and unimodal versus multimodal presentations of information. Whereas in media research, learning benefits are measured in terms of memory and comprehension of the presented information, multimedia research in addition involves the question of the cognitive factors that can account for learning benefits, that is, information processing advantages of particular symbol systems. In other words, multimedia research deals with why one way of information presentation works better than another rather than only proving that it does. It stresses explanation of the results in terms of cognitive processes rather than only establishing media differences in terms of learning products. Such explanations are sought in a number of directions. For example, augmenting text with pictures, animations, narrations, and so on, could represent advantages through effects related to motivation and repetition (Glenberg & Langston, 1992), and dual coding and mental model construction (Mayer, 1997). Conversely, difficulties might arise due to limitations of working memory and the high cognitive load associated with the integration of multiple sources of information (Chandler & Sweller, 1991; Mayer & Moreno, 1998).
Hypermedia
The third debate, largely instigated by the development of hypermedia technology, concerns the type of access to information. Traditional media involve decisions regarding the order in which information is presented. The development of hypermedia, more specifically, the introduction of electronic linking, has made it possible to leave part of these decisions to the learner. Whereas multimedia techniques are used to present information in different formats, hypermedia techniques allow manipulating access to the information. This feature introduces the idea of interaction of the learner with the information presented. The learner is thought to profit from hypermedia because it enables self-paced, in-depth, and nonlinear access to vast amounts of information structured in a way that is compatible with the workings of the human mind. Detailed overviews of these claims regarding educational hypermedia can be found in Dillon and Gabbard (1998) and in Tergan (1997a, 1997b).
With the development of hypermedia, a number of studies focused on the effects of differences in the structure of presentations on learning. Types of access are keyword, index, linear, hierarchical and network structures of information. Learners typically are asked to study the information at their own pace and are allowed to navigate, crisscross, or browse the information guided by their own curiosity. Learning outcomes in these studies are often measured in terms of comprehension and quality of writ-ten essays (Dillon & Gabbard, 1998). Individual choices of learners are considered to be an asset of hypermedia, but at the same time constitute a problem in hypermedia research. If different hypermedia structures lead to different consultation patterns, should effects on learning be attributed to the structure of information or to the fact that the learners have seen a different subset of information? In any case, there are many issues in the field of hypermedia for learning that remain to be investigated.
Summary of the Conditions and Factors Studied
Table 7.1 summarizes conditions and factors studied in media, multimedia, and hypermedia research. Combining these conditions and factors gives rise to an infinite number of possible ways of...