Should Computers Know What You Can Do with Them?
by Don Nix - 1988
The use of computers in education can be an opportunity for children to surprise themselves and their teachers. The key is to empower the child with tools of self-expression.
The use of computers in education can be an opportunity for children to surprise them-m selves and their teachers. The key is to empower the child with tools of self-expression.
The use of computers in education has the potential to contribute to the dignity of the child. However, as computers are typically used, this potential is neither met nor explored. In many cases, in fact, the educational use of the computer poses a threat to human dignity and uniqueness. A story by Woody Allen captures the nature of this threat. My father was tired. He was technologically unemployed. My father worked for the same firm for twelve years. They tired him. They replaced him with a tiny gadget that does everything my father does, only it does it much better. The depressing thing is that my mother ran out and bought one.1 This story blurs the distinction between machines and humans. Furthermore, it makes the machine attractive in a way that triumphs over the human. If one can imagine Franz Kafka surviving in the world of computers long enough to make a comment on this situation, he might add, as an extension to Woody Allens story, And my father went out and bought one too.
Dignity, then, for the purposes of my discussion, refers to the childs experience of himself or herself as intrinsically different from the way a computer functions, specifically by being able to actively consider his or her processes and feelings, and to be unpredictable in a creative way. (I am not arguing that this is what or all that dignity is, but merely being specific about my starting point.) Much of the use of computers in education currently poses a threat to this aspect of the dignity of children.
In order to focus clearly on this problem that computers pose, I will consider this question: Should computers know what you can do with them? When we discuss the problem in this context, the notion of creativity in the sense of unpredictability can be highlighted. In current computer applications in education, the computer to a considerable extent knows ahead of time what the student will learn in some content domain. That is, the computer as experienced by the child is structured so that one can predict to a high degree what the child will learn, specifically because of what the computer is programmed to do. This predictability in fact is commonly used as a measure of the success of both the computer and the child. The danger here is one of mediocrity of knowledge. In addition to predictability in a given content domain, such as math, programming, or history, there is the likelihood of predictability in meta-cognitive ways. Based on the structure of the interaction between the computer and the child, the child learns a certain way, for example, of learning, or problem solving, or obedience. The experience contributes to the childs development of a conception of what learning is, and what his or her role is with regard to it, in terms of both cognition and modes of feeling. The danger here is a threat to dignity.
Presumably it is true that computers can have an effect on cognitive processing and feelings. This is a common and strategic assumption made both by people who are skeptical of computers2 and those who see computers as at least potentially having a positive and liberating impact.3 The truth of this assumption, however, has not been empirically supported by the normal experimental paradigms of psychology and educational psychology. Moreover, an assumption of this type, which involves complex issues of human functioning, is not a convenient one to test with such paradigms. Thus, based on evidence that-is more experientially realistic than what can be provided by results in experimental psychology, I will assume that prolonged and/or critically important interaction with a computer can and does, in a somewhat Whorfian way, cause predictable changes in both cognition and affect in children.
In typical current applications of computers to education, then, computers know in an important sense what the child will learn. This predictability applies both to the content domain involved and to the way a child is influenced to think of and feel about learning and his or her role in it. The first poses the problem of mediocrity of knowledge, and the second constitutes a threat to the dignity of the learner.
EXAMPLES OF PREDICTABILITY
The most pervasive use of computers in education is the often derided, scorned, and ridiculed drill-and-practice mode, along with the related tutorial mode. These originated in traditional computer-assisted-instruction (CAI) days, over thirty years ago, and closely resemble programmed instruction technology, as well as the world of print workbooks and ditto sheets. In this mode, to the extent that the program is successful, the degree of predictability is high. The child learns prepackaged information (such as addition, decoding, history facts, science facts, and so on). The information is relatively simplistic, due to the exigencies of teaching in this mode of instruction. The meta-level of learning includes, among other messages, the notion that the concepts of right and wrong are central to thinking about learning, and that these concepts can and should be unambiguously defined. The child also has the experience that he or she as a person is irrelevant to the process of learning. Learning is hunting for and remembering facts, rather than a process whereby the child explores, creates, and owns. The computer knows this.
The CAI modes are relatively clear examples of outcome predictability. Many objections to traditional and updated CAI have been expressed, different authors finding different features to excoriate. One particularly formidable objection is that of Papert.4 A significant focus of this objection is the passive role the child is constrained to play in the drill-and-practice, tutorial CAI mode. Such a role is undignified. At best it ignores human potential. At worst it represses it.
An alternative plan for computers and education is exemplified by work by Papert and others, and embodied for example in the Logo language for children. In a Logo environment, the intention is for the child to program the computer, rather than being programmed by it. The child creates something with the computer. What is created can be a picture, or animation sequence, or, in principal, any kind of program that can be written with a general programming language. The computer is not pre-programmed to teach a certain set of facts about a certain content area. At this level, the computer cannot know what the child can or is doing with it.
At the processing level, however, the situation is different. Logo is a tendentious language. The language is designed so that, as an ideal, a child will learn a type of problem-solving technique. A significant goal of the Logo orientation is to have an impact on creative, self-expressive, problem-solving strategies. The goal is not for the child to learn unambiguous facts, but to learn processes that will not only be used to solve problems in subsequent Logo and Logo-like environments, but also to generalize to other domains in the life of the child. The problem-solving technique is computeristic. The domain of expression is defined basically in conceptual schemes related to computers, such as algorithmic thinking, procedural thinking, logical debugging, and modularization. Supporters of this type of role for the child at the computer define an important aspect of success in terms of the childs becoming an active problem solver of the type described. An ideal is for the child to master and become proficient at this type of creativity and problem solving in a wide range of domains beyond the specific computer setting that fosters it. It is at this processing level of predictability that the computer knows what the child will do.
Another area of computer use in education, and a recent development, comes from a noneducational field noted for its extravagant claims: artificial intelligence (AI)/intelligent computer-assisted instruction (ICAI). Proponents of artificial intelligence view the computer as knowing something in a more literal sense. In theory, the computer would, for example, understand the content to be learned, the goals of the student, and the ongoing cognitive states of the student, with regard at least to his or her progress in learning what is to be taught. In a sense, then, the computers knowing what the child can do with it is an explicit design goal. Again, success of the ICAI system can be measured in terms of a certain type of predictability. From the standpoint of technology, the ICAI approach and related expert-systems (or, more modestly, paraprofessional-systems) approaches are radically different from the two referred to above, the drill-and-practice and the creative-programming approaches. From the dignity perspective, however, the differences are not that clear, although very few systems have actually been implemented for children, and even fewer have been explored in real-life settings. At the content level, the computer knows what the child will learn if the system is successful, similar to the situation with drill and practice, although the teaching methodology is quite different.
At the level of having an effect on the way a child might learn to think about problems, ICAI is basically an empirically unexplored area. It is easy to consider ways in which, for example, paraprofessional systems could be designed to teach either overtly or implicitly a manner of approaching some domain of problems, rather than or in addition to specific success in solving problems in that domain. However, the basis of such systems so far consists of explicit rules and algorithms for combining them. Given this deterministic type of system, no matter how complex, it is likely that the way in which a child learned to think and solve problems like the expert would be predictable. The computer would know.
THE COMPUTER KNOWS
In the above examples, arguments can be made that what the systems teach is worth teaching. This is not the issue. Any content materials can be taught in a variety of ways. The issue is whether these genres of activity between the computer and the child have side effects that are inimical to the childs exploring ideas and feelings that go beyond those comprehensible to computers, and do so in a manner of exploration that goes beyond the information-processing style of problem solving embodied in computers.
My assumption is that dignity is a desideratum. The above descriptions of several computer uses in education show that those uses do not exploit the computers potential to foster dignity either fully or at all. Whether these uses are a boon, a bust, or Big Brother is not the issue. The point is that they are not conceptualized in terms of the type of creativity that is based on computeristic unpredictability. This does not mean they should be replaced. What it does mean is that there is considerable room in the world of exploring the uses of computers in education for a different type of system with different starting assumptions, and different effects. There is room for the type of computer in education where the computer does not and cannot know what you can do with it.
THE COMPUTER DOES NOT KNOW
A paradigm we are exploring is intended to fit into the computer does not know category, in contradistinction to the ways computers are generally used in education. The term Making a Scene will be used to refer to the general environment and point of view we are implementing. This making-a-scene paradigm has not as yet been explored in detail. It is based on the use of a child-controlled multimedia computer system, including voice, video disc and VCR, audio cassette, animation, graphics, and a language named Handy. The focus is on how the child can use the technology in an expressive way, rather than on the technology itself. The goal is to enable the child to be creative and self-expressive using the computer, but in areas that are not intrinsically related to computeristic concepts, and that cannot be expressed computeristicallyways that, in short, defy computeristic limitations.
This form differs from the traditional drill-and-practice systems, and the apparent aspirations of ICAI and paraprofessional systems, in that the stress is on self-expression. It differs from the programming-oriented systems such as Logo in that the areas of self-expression are more experientially familiar to humans. These areas include, for example, fun, humor, passion, satire, aesthetics, and interpersonal negotiations, in such topics as civil rights, rock video, political campaign advertisements, and television game shows. The making-a-scene computer uses the computer as a decentered participant, in contrast to most computers in the classroom. The computer does not know anything about a specific content area. The computer language does not intentionally embody a specific style of problem solving. The decentered role the computer plays prevents whatever hidden Whorfian determinants there are from being a focal point and a source of significant predictability. This is technology and clutter in the style of Max Headroom. It is an attempt to get closer to Paperts metaphor of the computer as pencil.
MAKING A SCENE
In order to explore ways that children could try out ideas using a computer, we created an experimental computer language. The language, Handy, will be briefly described here, and several case studies of its use will be sketched.
Handy enables a child (or other person as an author) to construct what can be called interactive scenes presented on a computer. The scenes can integrate video disc, video tape, audio tape, synthesized voice, digitized voice, touch panel, animation, and graphics. The child can make the scene interactive according to whatever plan he or she has in mind. Examples of events include animating stories of interest, telling jokes, making a rock video or a play, and constructing an essay consisting of video and voice-over and text and touchable menus of questions that enable the reader-listener-viewer to browse this multimedia event.
Using Handy, the child creates the script and objects that are to be visible to whomever will subsequently interact with the scene or scenes. A script is a program written in the Handy language. The objects are individual elements that can be displayed on the computer screen. Objects can be pictures created by the child, or text, or windows onto video disc or video tape segments. Once the child has created all or part of a script and its associated objects, he or she can try them out in an interactive, iterative manner. Thus, once a scene or part of a scene has been created, it can be run, interrupted, revised, and continued. Scenes can be simple in the case of a small set of objects and a single script, or complex in the case of hundreds of objects and scripts that call other scripts or that run the objects themselves, instead of just showing them.
To take a relatively simple making-a-scene example, suppose a child wanted to comically annotate a few scenes from the movie Ghostbusters, and then show it to the class. The child might use a scenario as follows. A robot would come out onto the stage (the computer screen), and ask the viewer what he or she wanted to seethe school library, or the librarian. The robot would speak the question using the voice synthesizer, and also show two boxes for touching to indicate the choice. If the viewer touches the box indicating the library, the script would move the robot offstage, and then play from the video disc a scene from Ghostbusters showing the front of the New York Public Library. While this scene was playing on the computer monitor, the script could move onstage an object with text reading This is the modest PS 233 School Library. When the scene ended, the text would disappear, and the robot would return and present the two choices again.
If the user touches the box indicating that he or she wants to see the librarian, the robot might be moved offstage, and the script would play the scene from Ghostbusters that shows a ghost drifting through the card catalog area. The label Our Librarian could be superimposed on the video. The video would then be scripted or programmed to freeze on a closeup on the ghosts face. A box would shrink down to frame just the face, blanking out the rest of the video image. The robot would come out. A text balloon would appear above the ghosts face with the words I dont know why the children are afraid to check out books. Then once the user touches the screen, the box framing the ghosts face would close all the way up so that no video was visible, and the robot would ask the choice question again.
In this example, which is based on a project done by fourth graders, the video scenes would be from the Ghostbusters movie video disc. The objects (robot, text balloons, frames, and so on) would be created by the child, and superimposed on the video material. An example of a script written to perform the above is shown in Figure 1. What each object shows is as follows (uppercase words are the names the child has given to the objects created and appear in Figure 1 in lower case): ROBOT is the robot; LIBRARY contains the text I want to see the library; LIBRARIAN contains the text I want to see the librarian; LIB contains the text This is the modest PS 233 School Library; FRAME is a box for outlining the ghosts face; and WHY contains the text I dont know why the children are afraid to check out books. Much of the script is self-explanatory. The statement ~touch, however, needs explanation. This statement is a variable that is automatically replaced by the word library or librarian once the person interacting with the scene has made a choice and touched one of the objects, LIBRARY or LIBRARIAN. If the LIBRARY object is touched, for example, then the script determines that ~touch matches the word library and so the script runs the subscript labeled playlibrary via the statement then playlibrary. The playlibrarian script is run if the LIBRARIAN object is touched.
The variables ~scene1 and ~scene2 contain the video disc frame numbers for the two scenes the script will play. These are set by the child by viewing the video, and then telling Handy what sections of the video are to go by those names. Any names could be chosen.
The example in Figure 1 is relatively simple. Handy itself has hundreds of additional commands and facilities, and can be used for quite sophisticated programming. The application of Handy in Figure 1 uses only a small subset of the system. It is presented here to indicate the type of language Handy is. However, even in this example, the child has to study the movie; analyze it; reconceptualize it; create a mode of presentation, including a mode of interaction, for the intended audience; plan and create the objects; and write, try out, and revise the script. The creativity involved in doing this is not circumscribed by computeristic concepts. It involves a wider range of issues, many of which are real world issues that are unrelated to computers, such as humor, satire, and a challenge to authority. On the other hand, the child does deal with computer concepts quite directly in order to create his or her production. The script in Figure 1 is a computer program. It is not English. It is written in an artificial language that has to be learned, using where possible what one knows about English as a guide. Once the script is written, or some of it is written, it has to be debugged.
The design goal of Handy was to create a language that would facilitate a type of noncomputeristic self-expression. The intention was to make the language itself as decentered as possible. As children use Handy, their experience will provide information needed to revise it. We have begun a series of exploratory studies in which children create events with Handy. Although there are many questions that can be experimentally studied using this environment, at this early stage the focus is on providing children with the facility, and seeing what they do and where they have problems. Several examples will be briefly outlined here.5
One class of Handy use is that of creating interactive video essays. In general, a child creates an essay that, instead of being written on paper, consists of video segments chosen by the child to make his or her point, combined with superimposed annotation or extended text, and voice-over from the child. The person who reads the essay actually reads and listens to and watches the essay, and can make choices about what he or she wants to see next, or see again. The Ghostbusters example above is a simple example of this type of essay. A more elaborate example is a project in which tenth graders created mock political advertisements for a presidential candidate, using video-tape footage from The Video Encyclopedia of the 20th Century6 consisting of seventy-live hours of newsreel footage containing news clips from 1893 to 1985. In this project, students picked presidential candidates (Ford and Reagan) and used film clips from their campaigns as raw material for the video component of the essay. In addition to the footage from the actual film clips, the students shot additional footage themselves, and then edited it all together.
In creating the essays, a number of decisions had to be made by the students. Examples include: demographics of intended audience; relevance of the ad for the overall campaign; emphasis (record of achievement, personal qualities, platform, positive versus negative campaigning, and others); emotional tone; thematic integration; use of media (camera angles, based on what is available, sound, voice-over, annotation, graphics, and others). The students implement their decisions by creating objects consisting of text and pictures and other types of information, and writing a script to put it all together. For example, at one point in the Ford essay, the script was written to play a video segment showing a Fourth of July celebration with fireworks displayed on a dark sky. The script also showed a computer-generated American flag superimposed on the video, and at the same time played an excerpt from Beethovens First Symphony. The computer did not know what the students had done with it.
A separate but related essay project used a video disc containing civil rights footage from the 1960s. The video was excerpted from several of the Video Encyclopedia video tapes. Eighth graders used shots of speakers (for example, Martin Luther King, Jr., George Wallace, and Julian Bond), as well as police and crowd scenes (for example, federal troops sent to Little Rock, Arkansas), to express their ideas about civil rights issues. For example, for video footage that has no sound with it, a student can record narration on a computer-controlled audio tape, and then write a script to play the video with their narration at the same time. As another possibility, the script could be written to allow the person who is listening to the essay to record his or her own voice over the video.
In these projects, the students actively used a computer and a specialized environment to produce essays that could not feasibly be produced without computer support. The computers were an integral part of one aspect of the expression, but not a focal point in the overall experience (similar to the pencil analogy). In this interactive video type of essay, the computer does not know what the child will create, or think, or learn. The most significant aspect of what the child does is not related to concepts about computers. The civil rights essay was a moving multimedia interactive presentation that was a revelation to youngsters for whom the 1960s did not previously exist. The Ford campaign was for some of its audience a persuasive advertisement, and for others a source of satirical hilarity.
Another form of self-expression is a project in which a group of fifth-grade children created a disc jockey program, where a viewer could request a song. They created an object representing a disc jockey, including mustache, Mohawk hair cut, and sweater with WPLJ, their favorite radio stations call letters. They also created a menu that had four songs, Private Dancer, West End Girls, Working on the Highway, and I Call Her Darlin, but She Calls Me Collect. They wrote a Handy script for a scenario in which the disc jockey bounced onstage, and the menu appeared, and the disc jockey said, using the voice synthesizer, Pick a song. The children had created their own video tape. On the tape they were singing and dancing to each of the songs. They had made costumes, picked the songs, done the choreography, and handled the production details. Then they viewed the tape through Handy, in order to assign the segments they wanted to show for the different song choices their viewers would make. Once the user made a song choice, the disc jockey would bounce offstage, the script would locate the beginning of the song on the video tape, and an object would open up on the screen, showing the singing and dancing. The script was written so that, at any point in time, the user could interrupt the song and dance, by pressing a certain key, and then return to the main menu.
This project involved a considerable amount of group interaction and planning, from picking the songs to scheduling time when all four could get together after school to shoot the video. Again, the point is that the focus is not on math drill or computer programming, but on using a computer environment to create (in this case) a rock video, where the most important elements involved in the creation were social interaction, aesthetics, music, humor, popular culture, and production.
Two more examples will be mentioned, which are more ambitious in scope. The first is a soap-opera project done with eighth graders from the Dalton School, and appropriately called Dalton Crest. The students wrote a screenplay for the story that involved a complex set of personal relationships among six people, and in which a murder occurred. The screenplay was written so that each character had a plausible motive for the murder, but so that one character was implicated in subtle ways. We videotaped the soap opera, and each of the creators played a role in it. A viewer of the soap opera encountered the following situation when he or she interacted with it on the computer. A short initial video segment was shown, with computer-generated text, to introduce the characters and establish that a murder had occurred. Then it was up to the viewer to solve the mystery by requesting more information. The additional information was presented in terms of video segments. Background and other information was superimposed on the video. At any point the viewer could offer a solution. The Handy script kept a record of how the viewer traversed through Dalton Crest.
This project involved a wide range of cognitive, affective, and interpersonal activities. The computer played a part in enabling the students to think about a range of issues, and then to be able to creatively explore those issues in a highly motivational, personally meaningful, and publicly entertaining way.
The final example was done by a group of children, grades 3 through 5. The project began simply as an effort to teach a version of Handy to the children, in order to determine where there were problems in Handy that had to be corrected before the language could be used on a wider scale. This project evolved into an elaborate satire of a popular television program, Family Feud. On this show, two families vie against each other by trying to guess how people on a previously conducted survey answered certain questions. The live or so most frequent answers are on a board, but covered up. If a family member guesses one of the answers on the board, the answer is uncovered. If the guess made is not one of the answers on the board, an X appears and a buzzer is sounded. Key elements in the television program are the fawning attitude of the moderator and the way in which members of the two families overreact to the moderator, the questions, the answers, and everything else. The name of the project the children did was Family Fools.
In this case, the computer was used to make an answer board like the one on the show. The children created the objects for the answers to questions, and objects to cover them up with decorative patterns. A script was written to check for an answer to be typed in, and, if there was a match (incorrect spellings and other types of approximate answers were accepted as correct), to uncover the answer, and, if not, to show one, two, or three Xs, and sound a buzzer. The production of Family Fools included the children, the sychophantic moderator, and the computer. The children divided themselves into two different families. I was given the role of moderator. Two video cameras were used, one for the moderator and families, and one focused on the computer. The two video sources could be mixed, or one could be inserted into a corner of the other. The show was taped before a live and motley audience consisting of children waiting for their turn on camera, researchers, and passers-by who wandered in because of the noise.
The Family Fools project, although begun as a computer activity and based in large part on a computer program (script), was considerably different in content and feeling from a computeristic activity. The computer was a catalyst and tool for a set of creative efforts that were experientially familiar to the children, that were expressed creatively, and that reached into diverse content areas.
The reason for sketching these representative projects in some detail is to give as tangible as possible an indication of how this kind of use of computers in education differs from current uses, and from most proposed uses, and as suggestions for further exploration. On consideration of these descriptions, or observations of or participation in these and other related projects, it is obvious that something quite different is going on, when compared with what one would normally observe in schools or research laboratories. The computer is a decentered enabling technology for experiences whose significance transcends computeristic ideas.
There are two burning and unresolved issues that are particularly relevant to the type of computer environment described here. I want to place the making-a-scene environment in perspective with regard to these issues, without direct concern for the resolution of the issues themselves.
First, do or can computers have any effect at all, other than simply teaching a mediocre level of content knowledge in some specific domain? Writers who have considered the good and the evil of computers in society, and for children more specifically, including Papert, Weizenbaum, Turkle, Dreyfus and Dreyfus, and most contributers in Sloan, as well as Woody Allen and comic denizens of late night television, assume that computers do have such effects as a result of direct interaction with them over a period of time.7 Little if any empirical evidence is given of the type usually considered necessary to establish a cause-and-effect relationship. This has not been empirically tested in a fair manner to a satisfactory extent, and perhaps cannot be, given the limitations of the test paradigms used, which are usually based on experimental psychology. The relationship is assumed, and the arguments start there, in terms of whether the results are good or bad.
This type of question as applied to the making-a-scene genre of computer activities is shifted somewhat. The focus is not directly on whether the computer makes a difference. It is on whether the nexus within which the computer is decentered, and in which activities involving aesthetics, social interaction, humor, and so on, makes a difference. In this environment, it is easier to assume that a difference is made, because the young people are more totally involved, and they more directly own what they are doing, in terms of both the cognitive and the affective elements. It has in general been shown that the deeper the processing of information, the wider the range of types of processing, and the greater the motivation and sense of ownership, the greater the impact. The working assumption, then, is that computers in education, a la making-a-scene, can have a significant impact.
The second issue to be considered is: Are computers intrinsically limited and different, compared with humans, in terms of mentation and affect? From a theoretical point of view, it is difficult to see how this question can be resolved in the near future, despite the increasing amount of discussion and the strength of conviction many of these discussions demonstrate. If it seems that humans have an element of freedom, then the intrinsic differences are clear. If, on the other hand, it makes more sense to believe that humans are deterministic creatures, then the differences between computers and humans are less intrinsic and more a matter of complexity. However,-despite the theoretical difficulty in resolving the issue, from a more immediate and more practical viewpoint it is clear that computers and children are cognitively and emotionally different. Computers cannot create soap operas that elaborate complex interpersonal relationships and act them out with conviction, and cannot experience strong feelings when confronted graphically with the realities of the racism of the 1960s. Computers represent a cognition that if evidenced in a human would be classified as clinically retarded.
My concern is not that computers may ultimately be as smart as or no different from the brain of a child. My concern rather is that computers as they now exist in education are either not conducive to, or limit, dignity. The making-a-scene genre of computers and children is a means of dealing with these concerns. In a meaningful way, the computer does not know and cannot know what the child will learn. The type of computer-child interaction using the environment described is an additional way to study the impact of computers on education. With a computer that does not know what the child learns we avoid some of the concerns about inimical effects of computers on dignity. More positively, this is a way of considering a wider range of potential enhancements of education by computers. The use of the computer is creative; the creativity is not predictable; and the child in his or her interaction cannot be replaced with a tiny gadget.