53 Not the Computer but Human Interaction is the Basis for Cognitive Development and Education Janni NIELSEN The Royal Danish School of Educational Studies, Institute of Education, Emdrupvej 101, DK 2400 - Copenhagen NV, Denmark A process description of two pupils working with a computer demonstrates that the pupils have great difficulties in learning programming. However, much research leaves the general impression that children do not encounter many problems when working with computers. Programming challenges higher mental functions, and ensures a cognitive development which has general value: a procedural thinking which may be transferred to other areas in the child's life. The implication is that the learning process is of "osmotic" character, "like learning French by living in France . . . . . . without being taught" (Papert, 1980, p. 4). Thus children may as well stay at home by the computer, instead of going to school. With the work process described as point of departure, a cognitive model is presented which is founded in human interaction. The model operates with three fundamental ways of cognition: the sensori-motor, the emotive and the symbolic ± the latter seen here as language. Though they are presented independently, they must be considered a whole, the interaction between them being the basis for the development of human cognition throughout life. On the basis of this model, the theoretical frame of cognition worked within in much research is being questioned. It is argued that working with computers, learning programming demands much teaching, hence human interaction; and this is not a challenge which the computer in the home can meet. Janni Nielsen is a psychologist from the University of 1. "Finally we succeeded" Susan and Jane held their breath, mouths open. Susan touched the return key, and they both stared fascinated at the screen. Slowly, smiles grew upon their faces. " D i d we make that?", asked Susan. "Apparently ..... we ordered it to do it", said Jane with surprise in her voice. "Finally we succeeded", stated Susan, and they laughed. What the girls were looking at was Fig. 1. But the instruction on the exercise sheet was to use their program TURN (illustrated in Fig. 2) and change the right turn order of the triangle so the six triangles would form a secant (Fig. 3). During the introductory work with computers (12 hours) and " L O G O " in a Danish variation, the girls (15 years old) had been introduced gradually to length, angles, "repeat" and subroutines in mathematical education. Beginning with working with the graphics, they moved on to programming, initially by copying programs from the exercise sheet, finally being led on to programming by themselves, but always building on the programs hitherto introduced. When the girls had to change the right turn order, they started out by typing the system order: REMEMBER (this initializes a new program). Realizing this was wrong, they then keyed END. As this did not work either, they entered CATA- Copenhagen. She has carried out research in Office automation, psychological consequences, with special reference to women. At present she holds a research fellowship at the Royal Danish School of Educational Studies; the title of her project is: Computers and Cognitive Development, with special reference to gender-specific differences. Also of particular interest to her is the development of science and technology, and the question of the way to acquire knowledge - this viewed in a historical and gender specific frame. She is a member of the Forum for Women Studies and Research at the Royal School. North-Holland Education & Computing 2 (1986) 53-61 I ?,. ' 5 ¢ y ~." .. DREJ~ TRYK Fig. 1. 54 J. Nielsen / Human Interaction: the Basis of Development Fig. 2. LOGUE, got confused when they saw the screen text, then F O R G E T , then GRAPHIC, and finally again REMEMBER, and got lost. With help from a teacher they eventually got around to making corrections. They had worked with this earlier, but apparently had problems in reading the screen text. By now they had forgotten the instruction and tried instead with REPEAT 25, then REPEAT 31, starting fascinated at t h e screen. Exploring the medium one could say [1,2], but video-taping the work process showed s?mething else. One week later, after 4 more hours of programming, the girls were taped again. They still worked with the program T U R N . This time they made five attempts to solve the problem, at times leaving it and hiding behind the screen, and all one could hear were whispers and screams of laughter. Giving up... " D o we have to make that again? ..... Let's not bother", but returning because "if they (the teacher and the researcher) ask to see it, it is going to be a bit difficult". They were trying to change the subroutine Triangle with Square, as suggested on the exercise sheet, but to start with they had not asked the computer to get their Square program from the disc. Besides, they did not understand the use of subroutines. A program which calls another program was explained by Susan to Jane, after long discussions. " T h e machine then gives it (the program) the name of the new program (the subroutine)". At the fourth attempt the girls had entered this program: TURN, repeat 31, forward 50, right turn 90, to here, END. As the turn is 90 ° and it is repeated, the execution will show a Square. Jane's reaction to this was: Oh - then it just walks around 31 times .... Of course, because it is a Square". Susan: " W h a t is so clear about that?" Jane: " I don't know ..... something must be wrong". Now what appears to be wrong is that Susan forgot to key the order Square. But it was just the word which was missing in the program they had written, not, it seems, the subroutine, and the girls did not reflect on the fact that it drew a Square without this. However, the girls did have some understanding of angles, because later when they had entered the line Square, Susan happened to key a right turn 120, and of course a triangle was drawn. After some discussion, Jane stated: " O f course, because we asked it to turn 120", completely ignoring the subroutine Square, however. By the fifth attempt the solution for the program was: T U R N , repeat 31, Square, forward 50, right turn 95, to here, END. They still had not asked the computer to get their program Square from the disc, and they had changed the angle to 95 (they did not comment on this). The outcome was Fig. 4, and the reaction was: Jane: " H o w cute". Susan: " W e bloody well succeeded". The girls "solved" two more problems the same : ( ............. I Fig. 3. Fig. 4. J. Nielsen / Human Interaction: the Basis of Development way, and when something was drawn on the screen, when there was a product, the problem was considered solved. During the programming work the girls gave up completely trying to correct a program, even if they only happened to touch a wrong key while writing. They simply cleared the screen, and started all over again, by entering the following commands in the following sequence: CATALOGUE, FORGET, END, GRAPHIC, CLEAR and REMEMBER. Whenever a program was entered copying directly from the exercise sheet, they expressed surprise when it was executed. Even if they had worked with the graphics, drawing a figure, and painstakingly noted down the moves with pencil and paper, afterwards entering it as a program, they expressed surprise when seeing it executed. Though they had just drawn it themselves, and keyed the program, they did not seem to perceive the relation between the two activities. The reaction: " D i d we do that?" and whether it was right or wrong, "we succeeded" was most often the case. They were fascinated by the movements on the screen, the unexpected product, but the programming language did not seem to carry any meaning to them. Yet, when they worked directly with the graphics, they used part of the language. They worked in a way which may be termed homing-in [2]. Moving step by step, e.g. forward 30, forward 10, forward 20, back 5 - until "it looks right", and they did the same when turning. They worked mainly with turns of 45, 60, 90, 120 and length 20, 30, 40 etc., these being the numbers the exercise sheet had dealt with. They appeared so much at home with these numbers, to the extent that when Jane suggested, after a lot of shuffle forwards and backwards, that the length needed was 31, Susan's reaction was, "Oh, what a provoking number". She is quite fight of course. One year later, after approximately 40 hours of computerwork, the girls were taped again. During the year, variables had also been introduced. Different programs with variables had been presented and tried out, and the instruction for the exercise was: Use your programs to draw a man. The girls did not use any programs. They went straight to the graphics and worked by homing-in. Shuffle forwards and backwards, and their man was drawn. But they did not return to the actual 55 writing of programs. The two girls are used as an example, but they embody many of the problems that the average pupils - boys and girls - encountered during the year, and did not move beyond. Problems with the programming language, the relation between-the different program orders, the relation between the program and the figure being executed, the system orders and the systematic procedure. Yet much research on children and computers leaves one with the impression that children do not encounter many problems when learning to program [1-8]. 2. Programming and Cognition Programming challenges higher mental functions, and the children acquire the ability for systematic structuring and develop procedural thinking - this is the impression given. The implication is that the learning process is osmotic, that i s learning mathematics and programming without knowing that they learn, "like learning French by living in France ..... without being taught". (Papert [1], p. 4). According to Papert [1] the process in the development of cognition moves from concepts to an understanding of the relation between concepts granted that the concepts are founded in body knowledge. Papert argues this by referring to his own childhood fascination with gears, where he knew the names of all the parts in the transmission system, the gearbox etc., before he understood the relation between them. An understanding came about by "being the gear", by relating the process to his body knowledge. With this as point of departure, the suggestion is to get children to "do the turtle" e.g. walk in a circle, when they have problems in programming a circle: or writing a program which will simulate the juggling with balls before actually juggling. The power of the computer is its ability to simulate these processes, and that it works with uniquely defined concepts in a systematic sequential structure. Through this it becomes a transitional object, an object to think with, to understand other abstract ideas with. These two aspects account for a general cognitive gain, a procedural thinking. As most processes can be simulated by the computer, starting out with the concepts, and the 56 J. Nielsen / Human Interaction: the Basis of Development abstract and formalized analyses of the processes, through which a program may be written, the main point is to focus on the form, organizing it into a systematic procedure, ignoring the contents (the peculiarity of the specific domain). Through this a certain style of thinking can be learned [1,3,8], and the ability of procedural thinking developed. R. Lawler [9,10] also emphasizes the systematic procedure built into the programming. With a simple word-processing system he separated content and structure. The structure he presented as a preprogrammed composition for writing stories: the beginning (Once upon a time); the middle part; the end. Lawler's research builds on the intensive work he did together with his daughter (then 6 years old) for half a year. And Lawler suggests that by working within this systematic procedure the child " m a y gradually perceive the structure of the text" ([9] p. 16). Again one sees the osmotic process through which the structure was acquired by the child and became the structure within the framework of which she several years later generated her writing [9]. R. Noss [2,6,7] has worked with LOGO and younger children in primary schools. Noss's research project took place in a normal school setting, and as the necessity of helping children " t o fully explore their new learning environment" [6] was realized, the classes alternated between unstructured free playing, and structured with a curriculum. Noss also values the systematic procedure and the computer as a transitional object. Noss's point of departure is that children already structure their own knowledge, and that their thinking is logical. Children generalize, specialize, look for patterns, verify etc. in their everyday life. It is, according to Noss, the same logical process that working with mathematics demands. However, as mathematics is abstract and decontextualized, not concretely founded like everyday knowledge and thinking, this has hitherto been difficult for children (and for adults, one may add). But with the computer and LOGO, it has become possible to present the formalization of mathematical thinking in a concretized way. Learning to program means learning to deal with the rigour and formalism which is essential in programming, learning to deal with decontextualized symbols and knowledge - developing higher cognitive skills. The research is very promising, yet many of the teenagers in my research project did not get very far, despite the fact that they are much older than the children in the research mentioned and presumably have reached a higher level in cognitive development. They had problems in understanding the programming language, the necessary temporal structure, the subroutines, and often the text picture appeared incomprehensible to them. To a large extent they did not seem to have any mental images of what a program might execute, just as they had problems in fully grasping the relationship between the drawing and the program. Very often the Systematic procedure necessary created problems or was not understood, as in the case of Susan and Jane, though they had perceived something about being systematic and working sequentially. Very carefully, CATALOGUE, FORGET, END etc. were entered. In order to explain their programming activity, one has to move beyond the theoretical assumptions and the results presented in the quoted literature. To stay within these frames would almost inevitably lead to the conclusion that these girls are plain stupid. They just happen to belong to a group of pupils - but unfortunately there are many of them - who neither succeed with programming nor in school. 3. Cognitive Development As a possible model for understanding the cognitive development and the process of concept formation, I operate with a paradigm with three fundamental ways of cognition [11]. They are presented independently, but the development of and interaction between them is the basis for human cognition, for the person's comprehension of a coherent world throughout life. The sensori-motor way of cognition has body and senses as its basis, and it is through the individual's practical activity, by laying hands on reality, that this qualification is developed. We may call it the common sense of the body. We see it most clearly in children, while in adults it is expressed to a lesser degree. Emotions also function as a means of cognition. The emotive way of cognition grows out of human interaction, and is qualified through the necessary emotional relationships formed with other human beings. A new-born infant can survive J. Nielsen / Human Interaction: the Basis of Development only if an interaction is established between the mother (primary person) and the child, in which the physical as well as the emotional needs of the child are met. The emotions function as an orienting means for the child's way of relating to the world, and through this is created the possibility of dissolution of borders between self and others, hence the ability for caring identification. Neither of these two cognitive ways enjoys a high status in our culture, despite the fact that they maintain their essential role in the development of cognition throughout life. The third cognitive way is the symbofic. It may be transformed through language, images, colours, music etc. I shall deal only with language here. Cognition is - in our culture - often equated with the mastering of language, and of meaning: the ability to let the world be represented in words, to allot a meaning to the symbol and to understand on this basis. Language is introduced in the mother/child interaction, and moves from being merely sounds to the child, to become the naming of the interaction - the wholeness. But through the use of language in many different situations, the sounds are differentiated, and gradually they are allotted a meaning, they become concepts: not, however, uniquely defined concepts, because there are no sharp borders in everyday life, but blurred, and the meaning of the concept is derived from the context in which sensori-motor and emotive dimensions are embedded. This may be illustrated in the following way [12,13]: /~ tacit process of the concept formation "~ symbolic Sensori-motor and emotive cognition I have defined as the tacit dimensions [14]. Of course, part of this knowledge may be explicitly verbalized, but it is important to bear in mind that we know much more than we can say. But language has a double function. Language is a means of communication, where it is closely related to human interaction. However, language may become detached from the direct experiences. Through language it becomes possible to transcend the immediate, voice the past and the future, create theoretical constructions, etc. As such, the language becomes a very essential tool for the 57 development of cognition. Development of theories and scientific concepts may be an example of an abstract language in which the concepts are defined rather by their relation to other concepts, and to a lesser degree by their relation to reality. Yet they must refer to the reality, else they cannot become the means for grasping and explaining it. The results of mental processes must be confronted with reality, otherwise one may speak of text "sine con", merely figments of the brain. Formation of concepts thus takes place on the basis of human interaction and human action, that is by laying hands on reality, and derives its meaning from the context. This is the basis for cognitive development. Language may, however, become detached from its basis to the extent that it becomes a pure abstraction. No body, no senses, nor human relations hold it. When nothing embodies the language, it is without reference and must be defined, and rules and principles must be made for the use of it. It is no longer language but signs, like mathematical and computer languages. This does no imply, of course, that they cannot be used as tools. 4. Computers and Human Interaction It is the latter, the signs, that children have to work with when learning to program: not the fuzzy concepts of everyday life, but symbols without reference, and only a few (if any) may be referred to body knowledge. Papert suggests walking the circle like the turtle; but children do not walk circles like the turtle. To do this children have to break down the process of walking a circle into the smallest components possible. However, one does not walk a circle by taking one step, turning a little, one step etc. On the contrary, the process is perceived and executed as an organic whole. And one does not naturally think of the process explicitly verbalized in fragments. To walk a circle is a qualification one possesses through laying hands on reality, an experience acquired in the process of learning to walk; a process which takes place long before one possesses the complex language necessary for voicing the movement - in detail. This does not imply that one cannot learn to reduce reality to fragments, but it is a very special approach which does not come naturally. 58 J. Nielsen //Human Interaction: the Basis of Development Some kind of help or teaching is essential and, given this, working with body knowledge may support the learning process. Yet many concepts cannot be referred to body knowledge. The actions they cause are abstract and take place out of sight, as in the case of Jane and Susan, who did not seem to understand the system orders nor the function of these because they take place within the computer and cannot be seen, felt or executed by the girls. Thus there is nothing to hold on to, and the purpose of the activity is lost. Or rather, one does not work in order to acquire knowledge, but more in order to meet the requirements of the teacher. And as the words carry no meaning, the girls cannot operate with the systematic procedure either. But having heard something about the necessity of being systematic, they do their best: CATALOGUE, FORGET, etc. - figments of the brain. This may also be why the girls have problems in comprehending the relationship between figure and text. If words carry no meaning, one cannot imagine what will happen; therefore the girls did not have any mental pictures. However, as pointed out, some concepts are more readily operated with, e.g. forward, turn. But this is when working directly with the graphics. In the case of Jane and Susan, when making drawings, they did not seem to have problems hypothesizing on the basis of a few moves, systematizing, drawing conclusions and finally reducing the number of moves necessary. This, however, has very little to do with body knowledge, but much more to do with the immediate visibility of what a keyed order causes. The concepts were connected directly with the visual information - one may say that the sensori-motor and symbolic ways of cognition could interact. This brings me on the question of human interaction when learning programming. Neither Papert [1,3] nor LaMer [9,10] reflects on the human interaction. Yet Papert states that "gears were part of the world of adults around me, and through them I could relate to these people". ([1] p. 11). Gears were a specific knowledge domain in his childhood, and it was the adults around him who supplied him with names, and pointed out the relationship between the concrete elements. The "feeling" and "love" for gears came about because of, and due to, the human interaction, on the basis of which a positive feeling was developed. Human interaction was also an essential part of the Brookline project [3], where the ratio was one teacher, numerous researchers, and four children per session, which took place in a special room to which these children came from their normal classes. Without violating the research one can state that the adults were very engaged in the children's programming activities as well as the children themselves. No child is unaffected by all this attention from adults, which also very clearly had to do with the new tool. Thus it is not surprising that these children developed positive feelings towards the computer (Hawthorne effect). I would suggest that one may assume that a very high degree of human interaction and of teaching took place. Anyone who has tried to make observations in a classroom will know that it is very difficult not to become partly engaged in the activities. One very quickly becomes a par: ticipant observer - the children simply draw one into the activities. And neither does Lawler reflect on the human interaction, on the significance of "working with DAD": a father who worked together with his daughter in computer sessions which took place at his work-site. The interaction consisted not only of suggesting contents for her writing, but also keying the stories (she had difficulties in spelling) as well as explicitly pointing out to her when the preprogrammed structure was not kept in the generated text. The father expressed extreme interest in the child's activities, to the extent of keeping close track of these, which included keeping a log o f her activities every half-hour, whenever it was possible - for five months [10]. Thus I suggest that an essential part of the explanation for the children's engagement with computers may be derived from focusing on the emotive way of cognition: a cognitive way which functions as an orienting means for the child's way of relating to the world, the basis of it being human interaction. One also has to look at the question of domain-specific knowledge and the process of concept formation. The human interaction taking place while learning to program includes asking questions, pointing out problems, discussing etc. Learning programming comes about both by laying hands on, and also by programming concepts and processes being explicitly verbalized during the interaction. Working thus over extended time, J. Nielsen / Human Interaction: the Basis of Development children may gradually acquire some domainspecific knowledge. Programming abilities do not come about as a "fingertip effect" ([15], but through extended human interaction. In children's conceptualization of the world, that is in the concepts they develop, the tacit and the symbolic dimensions are not separated, just as the relationship between concepts grows organically on the basis of the practical interaction. But working with programming is working with decontextualized concepts, and the relationship between these signs is defined a priori. Programming demands unambiguity, formalization, hierarchical organization and temporal structuring. Hence programming demands skills such as reading and comprehending the problem, the ability to deal with uniquely defined concepts, decomposing the problem to elements, hypothetically planning steps ahead, following the temporal logic, and writing the program as a sequential instruction, and as prerequisite is the mastering of the programming language. It is therefore essential that we keep in mind that programming is a very complex activity which demands qualifications at what one may term a "metacognitive" level, and this is no small task to ask of children [16,17]. Writing programs means following the rules and principles of the formal logic. It is closed normative system which tells how conclusions ought to be drawn, and it does not include confronting the premises or the solution with reality [18]. To engage in an abstract work process is what children have to learn, and this is also what creates problems. Programming is not concrete not even when gears and juggling are simulated on the screen (as opposed to the concrete experience) it is an abstract activity, and aquiring the ability does n o t come naturally. On the contrary, by decontextualizing and separating structure and contents, a method for treating data has been developed, and this is what children have to learn. To acquire this method is not to be comprehended as possessing procedural thinking. Or rather, procedural thinking is not a psychological category, but belongs to the world of mathematics and computer science, and characterizes the way in which processing of data takes place in a computer. The logical thinking derived from everyday life grows out of the practical concrete relationship with the world. It is the logic of the activities - 59 where concepts refer to context and may embody both tacit and symbolic dimensions. It is this ability to deal logically with the world, based on experiences gained through everyday life, that we see expressed in Jane's reaction to Susan's suggestion that they should not bother, "if they ask to see it is going to be a bit difficult". Real logic we may call it, as opposed to the logical or procedural approach to be mastered when programming. The latter is formal logic, and is inherent in computers. We may come to master it as a method, but in order to use it in a qualified way - and this means to pay attention to the peculiarity of the specific domain - we have to draw on the tacit knowledge, the indeterminance in knowledge, in the relationships we perceived, and in the formation, on which our discoveries rely [14]. Barbara McClintock - a Nobel prize winner - has described her work process as one where she cannot always explain precisely and explicitly why or how she knows what she knows. Because what is essential in the process is . . . . . . " t o hear what the material has to say to you" . .... " t h e openness to let it come to you", and more than anything else " t o have a feeling for the organism" [19]. Human logical processes embody much more and are much richer than the processes of formal logic. One does sometimes wonder if the promises of a cognitive gain labelled procedural thinking, or formal logical thinking, are derived not so much from an understanding of the development of human cognition, but more from an abstract analysis of the "cognitive computer" [20]. Much human interaction is necessary when computers are introduced into education, but this is not a challenge which can be met by the computer in the home, because there is no one there to teach. The parents are out at work, or out of work; besides only a minority of them understand and can work in a qualified way with computers. Only a very few children - mainly boys - turn into hackers. Most children do no get hooked on the computer at all, just as many youngsters are not hooked on education. Thus the important question to pose is not really whether computers in the home will be a challenge to education, instead we shall have to look at education and the essential question to pose is: What kind of education do we want? To answer that, we must ask ourselves: What is the most essential knowledge we wish our children to develop, and how may 60 J. Nielsen / Human Interaction: the Basis of Development ir/stead we shall have to look at education and the essential question to pose is: What kind of education do we want? To answer that, we must ask ourselves: What is the most essential knowledge we wish our children to develop, and how may they acquire it? And this, I will suggest, cannot be answered by technological junk. References [1] Papert, Seymour (1980), Mindstorms, Children, Computers and Powerful Ideas, Basic Books, N.Y.. [2] Noss, Richard (1985), Creating a Mathematical Environment through Programming: A Study of Young Children Learning LOGO. University of London, Inst. of Education, Doctoral Dissertation. [3] Papert Seymour, Watt Daniel, diSessa Andrea, Weir Sylvia (1979), Final Report of The Brookline Logo Project. Logo Memo no. 53, Massachusetts Institute of Technology. [4] Papert, Seymour (1981), Computers and Computer Cultures, in: Creative Computing, 7,2, 82-88. [5] Papert, Seymour (1984), Computer as Mudpie, in: Classroom Computer Learning. January, vol 4, (6), 37-40. [6] Noss, Richard (1983), Beginning Longo in the Primary School: Exploring with a Floor Turtle. A UCBE paper no. 1, Chiltern Project, England. [7] Noss, Richard (1983), Learning with Logo - Is the Teacher Necessary? A UCBE paper no. 2, Chiltern Project, England. [8] Watt, Daniel (no year), A comparison of the Problem Solving Styles of Two Students Leanaing LOGO: a Computer Language for Children, Logo group, Massachusetts Institute of Technology. [9] Lawler, Robert (1980), One Childs Learning: Introducing Writing with a Computer, LOGO Memo no. 55, Mass. Inst. of Technology. [10] Lawler, Robert (1985), Computer Experience and Cognitive Development, A Child's Learning in a Computer Culture, Ellis Horwood, N.Y. [11] Nielsen, Janni and Roepstorff, Lisbet (1985), Girls and Computers - a world of difference? Contributions to the Third GASAT Conference, Kingston, England. [12] Nielsen, Janni (1983), Teknologi, Videnskab, Kon. - rapport fra et studieophold i USA, PPP no. 20, The Royal Danish School of Educational Studies, Copenhagen. [13] Nielsen, Janni (1983), Hverdagsliv, snusfornuft og sandheden, in Gunnar Green, Chresten Kruchov, Lejf Moos and Jens Rasmussen, eds., Mikroteknologi, skole og opdragelse, Unge Paedagoger, Copenhagen, 123-139. [14] Polanyi, Michael (1968), Logic and Psychology, American Psychologist, 23, 27-43. [15] Perkins, D.N. (1985, The Fingertip Effect: How Information-Processing Technology shapes Thinking, in: Educational Researcher, vol. 14, 7, August/September, 11-17. [16] Pea, Roy and Kurland, Midian (1983), On the Cognitive Prerequisites of Learning Computer Programming, Bank Street College, N.Y. [17] Pea, Roy and Kurland, Midian (1983), On the Cognitive effects of Learning Computer Programming, Bank Street College, N.Y. [18] Nielsen, Janni (1986), Videnskab, teknologi og konsspecifikke universer, in Naturkampen, no. 1, 11, Copenhagen in press). [19] Keller, Evelyn Fox (1983), A feeling for the organism the life and work of Barbara McClintock, San Francisco. [20] Schank, Roger with Peter Childers (1984), The Cognitive Computer, on Language, Learning and Artificial Intelligence, Addison-Wesley, Massachusetts. [21] Pea, Roy (1983), Logo Programming and Problem Solving, Technical Report no. 12, Bank Street College, N.Y. Discussion Larson and Janni Nielsen, Denmark. Zimmer: Could you clarify what you mean by "authentic" and "derived" meanings for children? Larson: There are two ways in which concept formation takes place, the first is direct. For example, the child forms a concept of what a dog is after seeing many dogs over several days. After he sees many dogs, he may make an abstract concept of what a dog is. The second type of concept formation I call derived. For example, it might be impossible for a child to form a direct concept of what a giraffe is, without seeing one. You can ask him to think of a dog, and then modify this concept by telling him it is something like a dog but with a long neck and brown spots. We need to understand the difference between authentic and derived concepts, and that there must be a balance between these two types. Computers in the home may lead to a disturbance of such a balance by establishing too many derived concepts. D. Tagg: Would interactive video alter the analysis of learning with computers? Larson: Interactive video can give a wider base for learning. However, we must distinguish between bringing the world to the child and the opposite: to bring the child to the world. Nielsen: Denmark: 20th century education is more derived than authentic. How can we solve this authentic/derived knowledge problem? Larson: I don't know. We have to face this issue. We should not focus so much on information technology, but should ask what should go on outside of information technology. Lauterbach: We learn intuitively what is socially necessary for daily living, e.g. language, tool use, how to handle food. School is usually not concerned with these things; it teaches for a universe of possible individual futures at work and in public. What is missing are ways in which children can reflect what and how they intuitively learn, i.e. how they are influenced and affected. At home or in school: What are chances and limitations for making these experiences? How do they relate to New Information Technologies (NIT). Larson; We must give children the ability to look at themselves and relate to themselves. The are presently bombarded by the mass-media. Adults must try to cooperate more with their children. It is bad to have these groups separated like living in different corridors. Presently children are separated from adults rather than the opposite. W. Tagg: It seems to me that 15 years is the worst age to J. Nielsen / Human Interaction: the Basis of Development introduce Logo. At that age learning styles - good or bad have already been developed and are not going to be overturned or even re-examined as a result of a short introduction to Logo. I would suggest that the two girls in the project could see little point in the exercise that they were asked to do. Richard Noss's work, which was referred to, is now institutionalized in m a n y primary schools within the 'Chiltern' Region and the main issue is the level of teacher intervention which should be introduced. I think that as children become older, there is more and more pressure on teachers to become accountable. This tends to destroy the concept of encouraging children to set their own goals and I think that this is the main reason for education " t u r n off" amongst other children. Nielsen: Yes, the teenagers in m y research m a y be too old for Logo, as they are for m a n y other elements in education. I think it is Margaret Donaldson who has asked: W h y is it that so m a n y of our children begin school so eager for learning and why is it so m a n y of our older children leave school so bitter and disilhisioned? Mabe it is because the content of education has very little to do with their life. The teenagers I speak of have reached puberty; the body is changing radically and m a n y other things are on their minds. 61 So it is no wonder that they do not get hooked on the computer. This is true of both boys and girlo~. But with the young children, I wonder if it is the children who are the point of departure for the results forwarded in so m u c h research or if it is the cognitive computer. Children do not work, think or develop in terms of linear structures or formal logic. On the contrary, younger children m a y work with the computer - but what are they really doing and what are they really learning m a y be both higher cognitive skills; what has been seen m a y merely be the parrot effect. Schuyten: There is a basic need in the child to control his environment. This results in natural learning. Personal involvem e n t (emotion) is essential here. The grown-up child has to go to school where more systematic learning takes place. This shift in learning is not successful for every child. W h a t can the computer at home do to relate the organized education at school (more systematic learning) and education at home (more natural learning)? Shouldn't we learn from research concerning television at home and its challange to education? Lauterbach: Did you imply that for 15 year olds, Logo-like tasks are not fitting the complexity of problems they are used to? Nielsen: Yes, for 15 year olds, the tasks were too trivial.