The Human Factor in Multimedia Joachim P. Hasebrook Bank Academy & University of Banking, Frankfurt (Germany) 1 Multimedia: The promises With the help of multimedia, learning is like playing – all fun and leisure. This is what publishers promise selling their edutainment products. "Edutainment", the combination of education and entertainment, seems to be the learner's paradise: Important facts and complicated structures are understood at a glance. And everything can remembered easily with the help of sounds, animation, and videos. Visual knowledge maps and new forms of navigating the multimedia knowledge space allows incidentally learn about medieval lyrics, the principles of evolution, and the theory of relativity. Not to mention the cyberspace: You can visit archeological places, museums, opera houses, and the human metabolism without leaving your living room (e.g. Jonassen, Beissner & Yacci , 1993; Unsworth, 1996). All of these breath-taking wonders are based on the general assumption that the more stimulation and the more media is involved, the easier it is to learn. There is hardly a multimedia book or oral presentation that does not refer to charts or other visual aids, because "hearing & seeing" results in better retention rates as compared to "hearing alone". This theory assumes that there is a simple addition of effects of media upon learning (Hasebrook, 1995a). But it is easy to realize that this cannot be the only truth. Consider, for example, a student who swears that he would study Latin best, when listening to the radio, watching TV, and reading a Latin dictionary. Certainly you won't believe him – so why believe the multimedia industry that any combinations of media lead to better retention? Many of these simple theories of multimedia refer to the famous book "Audio-visual Methods in Teaching" by Edgar Dale which was first published in 1946. Of course, Dale was thinking of musical records, motion pictures, etc, rather than multimedia applications when he was writing his book. He stated that "symbolic information", like words and graphics, are far from direct understanding. "Iconic information", like photographs, movies, or demonstrations, are much more intelligible. Finally, "direct information" derived from simulation, experiments, or goal-oriented experience are the best way to learn. This model suggests the same recommendations which Johan Amos Komensky (Comenius, 1592-1670) made in his pioneering work about the use of educational media: "Orbis Sensualium Pictus" ("The visible world", 1658). He advised the teacher to present media with the adequate sense - that is, learning lessons about music by hearing, and learning lessons about art by seeing. The foundations of his pedagogy published in the book "Magna Didactica" ("The big didactics", 1657) influenced generations of teachers. Nobody has ever tested retention rates for all possible combinations of media. Therefore, the relative importance of those media components for the learning process are uncertain. Furthermore, there are a number other problems related to it: • What kind of retention is needed (e.g., recognition like in multiple-choice tests, cued recall like in fill-in- tests, or free recall like in interviews)? Only specific forms of learning correspond to certain forms of tests. • What kind of cognitive resources are needed to understand the information? For example, music does not help very much to understand linear algebra, and colorful pictures are not appropriate to illustrate old Norwegian music. • What "format" of information has to be displayed? For example, realistic photographs make it hard to understand human anatomy but colored drawings make it easy. The crucial factor is not how multi the media is, but how multimedia is used! Multimedia: The shortcommings Multimedia applications shows only little learning effects - or none at all. One big problem is related to the issues of what multimedia exactly is, how it is tested, and what it is compared to. Joliceur and Berger (1986), two researchers from a software corporation, tried to collect studies about the efficacy of commercial multimedia applications and defined three simple conditions: (1) The results should clearly derive from the application and not from a cluster of technical and organizational factors. (2) The learning effects should be measured by a reasonable kind of test. (3) The results should be compared to some kind of control group, such as reading a 1 Bankakademie e.V., Dr. Joachim Hasebrook, Head of Concept/Programme Development, Oeder Weg 16-18, D-60318 Frankfurt/Main (eMail: hasebrook@bankakademie.de)