Against Multimedia

Brian Harvey
University of California, Berkeley

About 15 years ago, at the dawn of the personal computer era, I attended a talk by one of the early leaders in educational computing. The Votrax voice response unit had just been introduced, and he was very excited about it. To demonstrate the potential of this early multimedia technology, he first showed a standard, boring arithmetic drill program. The program presented a problem on the screen, the student would type in an answer, and the program would either announce that the answer was correct, and go on to the next problem, or announce that the answer was wrong, and repeat the same problem.

The speaker "improved" this program by using the voice response unit to give messages such as "that's still wrong, but you're getting warmer" for incorrect answers. The machine was able to control the tone and syllable emphasis enough to present these hints in an amusing fashion, and most of the audience was quite entranced by this new technology.

Unfortunately, the modified program's standard for "warmer" or "colder" was nothing more sophisticated than the numerical distance between the student's answer and the correct answer. So, for example, if the problem was 6 times 8, and the student first answered 42 (confusing this problem with the nearby 6 times 7), then answered 47 (a ridiculous answer), the program would congratulate the student for getting "warmer." The result is that the program was turned into a "guess my number" game, and the idea of learning arithmetic was lost!

Of course, the voice technology could have been used in more helpful ways (for example, the program could have recognized common errors and given messages such as "No, 42 is 6 times 7, not 6 times 8"), but the point of the example is that this otherwise intelligent designer of educational software, who would never have thought to use "warmer" and "colder" in the program's printed messages, was so mesmerized by the new technology that he wasn't thinking at all about the educational issues.

Now move forward a dozen years. The new technology is laser disc and CD-ROM. A leading multimedia researcher at Apple Computer demonstrates educational applications; one example is a program to teach physics students about levers. On the computer screen is a depiction of a see-saw. Using the mouse, the student can position three children, of different weights, at various positions on the see-saw. The goal for the student is to get it to balance. The see-saw is held in the horizontal position until the student finishes placing the three children; then the student clicks a button to release it. This is where the multimedia part comes in: The developers have videotaped three actual children, with the correct weight ratios, at all possible integer positions along the length of an actual see-saw. If the student has put two children at the same position, the actual children are seen in each other's lap, or on each other's shoulders. It's fun to watch.

Now imagine you're a student using this program. You've positioned two children arbitrarily, and you're trying to figure out where to put the third child so that the see-saw will balance. Most likely, the correct position is not at an integer distance! This wouldn't be a problem for the computer simulation alone, without the video enhancement, but in fact the program had to be written to allow only integer positions, because that's all they videotaped.

What these examples have in common is that in their eagerness to use the latest technology, both developers have actually made their programs worse, from a pedagogic point of view. Multimedia is limiting! You end up tailoring your work to the demands of the medium, rather than the other way around.

Multimedia is limiting in another way, also. Because good video is difficult and expensive to produce, the range of ideas available is restricted. One of the prototypical educational applications of multimedia is to ask students to prepare a report on Martin Luther King in which they use multimedia software to select and arrange excerpts from available videodiscs. But the Martin Luther King presented by Time-Life and by ABC News is the "I have a dream" integrationist, the pacifist, the martyr -- not the socialist, not the angry militant, not the multifaceted King that students could find in books. The videos don't emphasize the FBI's spying on King's sexual activities, nor the recent accusations of plagiarism in King's scholarly works. Similarly, the ABC videos on the Gulf War are based essentially on information provided by the United States government, some of which has been shown to be inaccurate.

Do all of these examples merely reflect the immaturity of the technology, rather than inherent weakness? Perhaps, but only if slick, professional video production will someday be as easy as computer programming is now. As long as it takes dozens of skilled technicians to produce television shows, the literature of multimedia will be limited. Of course there are better examples, such as the "Voyage of the Mimi" project developed at Bank Street College. (This project centered around a weekly broadcast TV show in which the protagonists explored the world by boat; the broadcasts were supplemented with a range of written material, interactive computer programs, and classroom activities.) But that project took years to produce, using federal research funds. And the result only teaches specific lessons; it's not open-ended in the way that programs like Geometric Supposer (which allows learners to perform geometric constructions on the screen; measure the resulting lengths, angles, and areas; hypothesize general rules; and test the hypotheses by repeating the construction with different starting shapes), or programming languages like Logo (a dialect of Lisp designed specifically for children learning mathematics, with simplified syntax and with a variety of mathematical "microworlds" to explore, of which the best known is turtle graphics), are open-ended. Therefore, the payoff for all that money is quite limited, compared to non-multimedia computer software development.

It is possible to use multimedia technology in a more open-ended way. Coco Conn, for example, has done some spectacular work with kids and video, in which the kids themselves plan and carry out the entire project, operating the cameras and the video editors. If there is any promise at all in educational multimedia, I think it's in that style of work. But it hasn't happened on a large scale because it requires both expensive equipment and a lot of skilled adult support. Even then, the kids' work probably won't be as slick as professionally made video, so we may find that students will be tempted to stick with the professional work rather than make their own, just as interest in computer programming has declined because of the wide availability of slick programs with graphical interfaces.

Because multimedia revolves around video, it shares many of the questionable properties of broadcast television: the emphasis on visually strong news such as violence and disasters, the short attention span, the focus on celebrities. People have argued for many years about whether these things are necessary implications of television, but even programs like Nova that attempt to be somewhat scholarly use the quick-cut style, and show the faces of Nobel laureates talking while the soundtrack gives us the simplified platitudes of a narrator. The more honest media enthusiasts don't even deny that a diet of television works against literacy; instead, they're proud of it, describing print literacy as obsolete in the light of the new "media literacy."

Like much of the "information superhighway" that's been in the news recently, multimedia provides the illusion of interactivity, in which the user's control extends merely to a selection among professionally prepared alternatives. Selecting images from an ABC News video is essentially similar to the idea of the "electronic town meeting," in which TV watchers can vote on choices presented by Al Gore and Ross Perot. In a real town meeting, the decision-makers are the same people who carry out the discussion and who invent the choices. The electronic version is deeply antidemocratic.

Why don't the same arguments apply to any educational use of computer technology? My answer is that in some cases they do apply. The kind of computer technology that presents a learner with multiple-choice tests, or with "Computer-Managed Instruction," is indeed just as limiting as multimedia technology. But computers lend themselves to a different style of work, one in which learners are presented with tools rather than with constraints. Word processing, spreadsheets, calculators, and drawing programs can help learners explore and present their own ideas. Most flexible of all is a programming language, in which learners can invent their own tools. So far, at least, multimedia is a much less accessible medium.