Here follows information on a book from Phil Agre's Red Rock Eater that bears on the recent thread ... Date: Tue, 10 Oct 2000 14:46:03 -0700 To: "Red Rock Eater News Service" <[log in to unmask]> Subject: [RRE]Social and Technological Resources for Learning [A lot of useful work is happening in the community of education people who descend intellectually from a Russian psychologist named Vygotsky. According to Vygotskian psychology, thinking has its origins in social interactions: you interact with other people, you internalize those patterns of interaction, and that's what thinking is. This theory has innumerable consequences, all of which are illuminating and productive. For example, Vygotsky's theory directs our attention to all of the informal apprenticeship that happens in everyday life, whether between adults and children or simply between people who have different degrees of expertise in a particular area. In situations like that, the people often spontaneously and dynamically adjust the division of labor so that the learner is operating right at the outer limits of his or her current skill level. That's an ideal environment for learning, and many studies have been conducted of its internal architecture. Vygotsky's theory also directs our attention to the social organization of the environments in which learning takes place: if you learn to think by internalizing your interactions with others, and if your interactions with others are shaped by institutions, it follows that your patterns of thinking will be related in some important way to the institutional structure of your society. The relationship need not be obvious, but it is worth looking into. The enclosed message contains lengthy excerpts from a new edited book that applies these ideas to the study of learning environments, and especially ones that involve computers. The whole way of thinking and writing will seem like a foreign language to a lot of people, but I am hoping that my brief summary here will provide some sense of the overall point. Although I have heavily reformatted the text, it does contain an unusual number of glitches and typos, most of which result from it having been converted from one too many format to another. The manuscript was also modified in small ways in proof. So if you want to quote a passage from this material, you should refer to the printed book for the authoritative version. Also, to fit this message into your mailbox, I have taken it upon myself to cut out several long passages; each of the cuts is marked with [...]. These cuts necessarily mangle the message to some degree. So, again, refer to the printed book for the complete story. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= This message was forwarded through the Red Rock Eater News Service (RRE). You are welcome to send the message along to others but please do not use the "redirect" option. For information about RRE, including instructions for (un)subscribing, see http://dlis.gseis.ucla.edu/people/pagre/rre.html =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Date: Fri, 06 Oct 2000 23:23:43 +0100 From: [log in to unmask] (Joan Bliss) Learning Sites: Social and Technological Resources for Learning Edited by Joan Bliss, Roger Saljo and Paul Light Pergamon Press, 1999 ISBN -- 0080433502 Table of Contents 1. The Human-Technological Dialectic Joan Bliss and Roger Saljo 1 PART 1: Learning and the Negotiation of Meaning Introduction by Roger Saljo 13 2. Learning Mathematics in and outside School: Two Views on Situated Learning Guida de Abreu 17 3. Situated Selves: Learning to be a Learner Chris Sinha 32 4. Negotiating Identities and Meanings in the Transmission of Knowledge: Analysis of Interactions in the Context of a Knowledge Exchange Network Nathalie Muller and Anne-Nelly Perret-Clermont 47 5. Real-world Knowledge and Mathematical Problem-solving in Upper Primary School Children Erik De Corte, Lieven Verschaffel, Sabien Lasure, Inge Borghart and Hajime Yoshida 61 6. Quantifying Time as a Discursive Practice: Arithmetics, Calendars, Fingers and Group Discussions as Structuring Resources Jan Wyndhamn and Roger Saljo 80 PART 2: Learning and Reasoning in Context Introduction by Paul Light 97 7. Situated Learning in Instructional Settings: From Euphoria to Feasibility Alexander Renkl, Hans Gruber and Heinz Mandl 101 8. Engaging with Organisational Memory Antonio Rizzo, Patrizia Marti, Vito Veneziano and Sebastiano Bagnara 110 9. The Relevance of Relevance in Children's Cognition Agnes Blaye, Edith Ackermann and Paul Light 120 10. Empirical Abstraction and Imaginative Denial of Rules Joan Bliss, Jon Ogborn, Orla Cronin, Will Reader and H.A. Tsatsarelis 132 11. Enactive Representations in Learning: Pretence, Models and Machines Edith K. Ackermann 144 12. Contextual Knowledge in the Development of Design Expertise Anneli Etelapelto and Paul Light 155 PART 3: Learning with and by Machines Introduction by Joan Bliss 165 13. Gender and IT: Contextualising Differences Karen Littleton and Maria Bannert 171 14. Information Technology and the Culture of Student Learning Charles Crook and Paul Light 183 15. Assisting Child-Computer Collaboration in the Zone of Proximal Development (The Vygotskian Inspired System (VIS)) Rosemary Luckin 194 16. Learning and Discovering with Computational Aids Vincent Corruble and Joan Bliss 210 17. Situated Cognition: A Challenge to Artificial Intelligence? Dolores Canamero and Vincent Corruble 223 18. Situated Learning in Autonomous Agents Bart de Boer and Dolores Canamero 236 19. Situated Learning at the Threshold of the New Millennium Yrjo Engestrom 249 The Human-Technological Dialectic Joan Bliss and Roger Saljo Human learning, reasoning and thinking, and the different sites in which these activities take place, are the focus of this book. While the different authors do not share one single perspective on how to approach and understand these issues, they have come to articulate a framework which they both use and criticise constructively in order to develop it. Importantly, human learning is seen not only as individuals actively making sense of their world in solitary activities, but it is seen also through the manner in which the social and cultural world, its practices and its artefacts, codetermine the way in which people approach learning in various settings, inside and outside formal institutions. The social and technological contexts in which individuals learn and work play a decisive role in the development of the individual. To understand how people learn is to simultaneously understand how they are able to adapt to (and sometimes to resist) the practices of various social institutions, and to appropriate and operate with the technological and intellectual tools that are salient in these environments. In Vygotskian terms (Cole, 1996; Vygotsky, 1986), artefacts mediate the world for us in accordance with the traditions and knowledge generated by previous generations. For example, the compass and the clock as artefacts are interesting and useful to us precisely because they mediate direction and time in a culturally meaningful and useful manner. It is the co-ordination of a physical device (which is not particularly interesting per se) with categories such as north and south, hour and minute that creates an interesting and highly significant artefact that is of considerable use to people. Although material in their appearance, such artefacts are designed to give information along the lines suggested by categories developed by humans. Thus, in some sense reasoning and artefacts co-evolve during human history. And the more sophisticated artefacts become, the more our learning and reasoning will involve the mastery of such resources. But in order to be able to consider the link between humans and technologies, it is obviously necessary to pay attention to the most powerful system for mediating the world that humans have created: language. Concepts and the manner in which these codify reality are the link between human reasoning and artefacts. As we have already hinted at, even technologies are impregnated with human signs, and when we encoun-ter them it is precisely their sign-giving function in which we are interested. Thus, discursive constructions are very much a part of our practical activities in everyday life whether we are talking or not. 1 Communication as Practical Action [...] In order to underline the similarity between physical artefacts and conceptual constructions codified in language, Vygotsky (1986) argued that both these elements of human practices should be thought of as tools. Language is a tool used for practical purposes, and it has furthermore been created by human beings to account for reality in a manner that is congenial with a certain culture and its basic assumptions. Concepts and linguistic expressions can, following this mode of thinking, be thought of as psychological tools that have emerged through the history of a society. They help us to do things in a manner that is analogous to how physical tools can be used for chopping trees, cultivating land or mending engines. The psychological tools of language are used both for thinking and for communicating. But, and as we have already pointed out, there is no strict line of demarcation between psychological tools (concepts) and physical tools either. Concepts are embodied in physical artefacts, and they are part of what makes such artefacts meaningful to people. When using the computer, we are not relating to an object that we conceive of in terms of electric circuits. Rather, we are writing texts using a word processing programme that represents language in a manner that is meaningful to us or we are making calculations by multiplying or adding digits that represent the numerical system that we know from other practices. Technological devices as such, and in their physical appearance, are largely uninteresting to us and we generally do not want to bother about such matters. As Ogborn (1996, p.3) points out: "commercialised technology puts a lot of effort into making it not necessary to understand the workings of its artefacts. Technologies often make the science and the technique behind them invisible." But the focus on what people do with words and technologies must not be seen in terms of individual activities only. A third component of our triadic unit of analysis of people using tools when acting in social life is the collectively shared cultural practices that make up society. Thus, we have to understand how modes of communication and technologies are inserted into and help sustain collective human action in enduring patterns. Put differently, when focusing on learning and reasoning, we have to consider how human activities are institutionalised. 2 Social Institutions as Contexts for Reasoning Schooling is one of the most apparent examples of how societies create stable social practices that are continuously produced and reproduced over long periods of time. But we find enduring arrangements of how to conduct social life in many sectors of society: in production, bureaucracies, science and research, and health care, to mention just a few examples. These institutions are interesting from the point of view of learning and reasoning, since knowledge is to a large extent developed and maintained within such enduring social structures. Science and research are obvious examples of this kind of institutionally produced knowledge and skills. The ability to act competently within these institutions requires familiarity with the traditions and knowledge systems that are dominant. And learning in modern society is largely a matter of mastering the knowledge and skills that have been generated through such social institutions. In schools, for instance, children's ability to appropriate different kinds of knowledge is decisive for the success that they will experience whilst being a student, and this, in turn, will have consequences later in life. At first sight it might seem as if the relationship between knowledge acquired in various settings is obvious. What we learn in school will be readily translated to other settings. A key feature of traditional learning theories has been the concept of transfer, and the assumption that people will understand new problems and events in terms of the relevant knowledge they already possess. But, as will be discussed by several authors in this volume, transfer is by no means automatic or simple. What we learn in one setting does not always translate easily into other settings where the conditions and institutional arrangements may be slightly different. Furthermore, there may often be conflicting assumptions regarding what is the relevant knowledge to transfer to a particular problem in a specific setting. Reasoning is situated and knowing what to do is more than knowing a rule, an algorithm or a principle. What is required is generally familiarity with a certain set of social practices and a sense of what is the expected mode of acting and reasoning (for empirical research illustrating the situ-atedness of human reasoning, see, for instance, Chaiklin & Lave, 1993; Resnick et al., 1997). 3 A Sociocultural Perspective on Learning and Reasoning What has been said so far indicates our ambition to build a bridge between people, their intellectual and practical actions, and sociocultural reality. From such a perspective artefacts in the shape of physical objects, discursive constructions, and social institutions play an essential role. Our culture in its material and immaterial appearance is to a large extent artificial in the sense that it has been produced by human beings. White (1996, p. xiii) expresses this by saying that "[a]rtefacts are the fundamental constituents of culture". Consequently, the "... growth of the human mind, in ontogeny and in human history, must be properly understood as a co-evolution of human activities and artefacts. The words we speak, the social institutions in which we participate, the man-made physical objects we use, all serve as both tools and symbols". And they are all essential to our daily actions. This conception of thinking and reasoning as tool-dependent is fundamental to a sociocultural understanding of the human mind and human practices. Our intellect is social in the sense that it relies on conceptual and material resources and tools that emanate from our culture. As Cole (1996, p. 137) puts it, "... what we call mind works through artefacts". This implies that "... it cannot be unconditionally bounded by the head or the body but must be seen as distributed in the artefacts which are woven together and which weave together individual actions in concert with and as part of the permeable, changing events of life". Thinking and reasoning are better conceived of as distributed over individuals operating with tools in social activities rather than as confined to the inside of our head only (Lave, 1988). This notion of the distributed nature of human thinking between people and across artefacts and social institutions is essential to a socio-cultural perspective on the human mind. Artefacts and technologies are some of the places where humans store their knowledge and experiences, as we have already pointed out. 4 Past Technology and Social Practices [...] 5 Technology and Social Practices: the Human Urge to Communicate and its Technological Manifestations [...] 6 Technologies as Tools for Reasoning and Social Action The emphasis in this book on technologies should not be read in a narrow manner as signalling a clear boundary between human beings thinking and reasoning on the one hand, and a set of sophisticated, but dead objects, on the other. In fact, our intention is exactly the opposite. It is not the 'human thinking plus technology' conception of learning and reasoning that is being presented. Nor do we claim that learning becomes very much easier or more pleasant through the use of technological tools, which seems to be the favourite metaphor behind the products being offered on the market through the edutainment industry. Rather, what is being considered is the inclusive nature of the link between people and artefacts in various social practices. When technologies change, so does the nature of human thinking and learning and so do our practices. It is, to take a rather trivial example, one thing to solve a multiplication problem involving numbers with several decimals using nothing but what is inside our heads. It is another thing to solve the problem when paper and pencil are available. And it is yet a different matter to do it with a mini-calculator as a tool to think with. What is impossible to manage in the head, and without external resources, becomes manageable with paper and pencil, and may very well be trivial to handle if there is a calculator around. Following this line of reasoning, physical artefacts at various levels of complexity (watches, rulers, computers, maps, television sets and so on) will be conceived of as "the manufactured objects that silently impregnate the furniture of the world with human intelligence" (White, 1996, p. xiii). In this sense, human beings convert their ideas, concepts and methods into artefacts, and they thus have "the power to endow the material world with a new class of properties that, though they owe their origin to us, acquire an enduring presence in objective reality, coming to exist independently of human individuals" (Bakhurst, 1991, p.l79-180). This volume has three parts, and in the first two parts artefacts are part and parcel of the contexts and social practices examined and discussed. These chapters show that artefacts both ideal and material (Cole, 1996, p.117) have always been part of our lives, have shaped our lives, and have been shaped by lives. Some chapters will focus on specific ones, like the use of the braca, the measuring and counting system of Brazilian sugar-cane farmers in Chapter 3, the use of the calendar, or the fingers in Swedish pupils' mathematics lessons in Chapter 6. In other chapters they are just a normal part of the context, like the teapot in the tea-making ritual in the 'topsy-turvy' children's task in Chapter 10. In the last section, one of the artefacts of our new technologies -- the computer -- is the major focus. While artefacts are an integral part of the whole volume, each section has a different and more specific theme, which is developed and illustrated through a number of pieces of research. The editor of each section introduces and situates each of the chapters in relation to the underlying theme. In Part I the theme running through all the chapters relates to how meaning is created and negotiated by people operating with cultural tools, carrying out concrete activities. In other words, thinking, learning and reasoning which are considered as an essential part of any human meaning-making activity are also determined by the social and cultural world, its practices and its artefacts. More specifically some of these studies examine the extent to which the context determines the nature of the knowledge to be learned and the degree to which it is respected and valued. They show how assumptions about learning, and thus the actions expected from learners as participants in social and learning activities, vary across and within cultures and settings. The studies cover a range of cultures from the Zapotec Indians in Mexico and sugar-cane farmers in Brazil to school children and students in Belgium and Sweden, and informal learning organisations in France. Part 2 can, to some extent, be seen as a natural extension of Part 1, but it is more specific in that various of its studies examine the forms of reasoning that take place in social and physical contexts, including but not limited to logical reasoning, and ask how these particular contexts constrain or activate different forms of reasoning. Some of the studies also examine workplace contexts, and reasoning and expertise within these. An underlying theme common to all this section's chapters is the scrutiny of traditional psychological accounts of learning and reasoning which give little consideration to context and situation, and the search for a new framework which takes into account such perspectives. However, reservations are articulated by some authors about the lack of detail within the situated learning approach: for example, how learning is scaffolded in some of the more complex learning environments. Again in this Part the studies cover research in a range of countries: France, Finland, Germany, Italy, UK and US. Part 3 focuses on new technologies. The word 'new' is used in an attempt to distinguish between existing technologies which have been critical in the development of humanity, and recent forms of technology. The so-called new technologies do not merely 'silently impregnate' the context, they interact with us. Saljo (1997) commenting on these pointed out: "what we see in sociocultural terms is a new division of labour between the human mind, body and the artefact". With the new technologies, valued and powerful information and knowledge become accessible outside the standard recognised institutions that previously controlled and disseminated such knowledge and information. In other words, information and communication technologies challenge traditional boundaries between knowledge communities and the demarcation of knowledge domains. Consequently they call into question existing definitions of professional and institutional roles and functions. Part 3's theme is more diversified than the other two parts. First we examine learning with computers and, through two reviews that survey the integration c computers into education at all levels -- primary to tertiary, we consider not only the implications of these new technologies for how we learn, but also how well established learning practices are robust and do not submit to change easily. I two further studies, attempts are made to use ideas from situated cognition both t( better design software and also to better understand users' attempts to work with software programmes. In the final two chapters we go one step further and describe how the situated cognition ideas are developed through artificial intelligence's 'situated agents' approach. Such a step shows us the difficulty of designing autonomous agents, be they synthetic or robotic. The difficulty does not so much reside in either the development of hardware or the software, although clearly this is critical, but the complexity of human thinking, learning and reasoning and how much of this we take for granted. Chapters in this section cover research in Belgium, France, Germany, UK and the US. 7 Concluding Summary Saljo (1997) points out: "... learning is the simultaneous transformation of social practices and of individuals. Just as new technologies and new intellectual tools transform social practices, so do they transform individuals who walk away with a new set of instruments by means of which they can relate to the world" Throughout this volume we will attempt to show the symbiotic interaction of individuals, tools, artefacts and social practices, and the importance of this for the development of human thinking, learning and reasoning. We shall be striving to develop a new theoretical framework which can account for, and thus attempt to explain, the nature of these interactions in order to better understand the role of context and situation in thinking, learning and reasoning. The thrust of our search for a new framework stems from the frustrations of working with the more traditional psychological approaches which have only paid lip service to context and situation. In summary, all three parts highlight a range of issues of which only a few can be mentioned here, but which represent the research programme as a whole: (i) the inseparability of knowledge and values and the need for an understanding of which forms of knowledge are privileged, with reference to everyday and informal knowledge; (ii) the positioning of learners and an understanding of what it takes, on the part of the learner, to be able to act competently within one particular framing of what it means to learn; (iii) the importance of contextual factors: the shift from their being considered as moderators of thinking and reasoning to their being seen as a constituent of thought leading to the recognition that dealing with problems is very much a contextual affair. With new types of tools and technologies, we will show that although new information and communication technologies make possible novel kinds of learning interactions within and across learning sites, it is not so clear how some of the more traditional sites of learning accommodate these changes. In fact, well-established and well-respected learning customs and practices appear to resist many of these changes. Indeed, even though information and communication technology can be seen as providing direct access to knowledge, bypassing traditional forms of teaching and learning, in order for information to become knowledge it needs to be constructed and validated through communicative and interpretative processes in concrete human practices. Many of these issues will reveal the importance of the social and cultural world, its practices and its artefacts, and how they influence and often determine the outcomes. The world in which intelligence is exercised is a world which has been shaped, and continues to be shaped, by the intelligent activities of others. PART 1 Learning and the Negotiation of Meaning Introduction by Roger Saljo [...] The chapters of this first section of the volume all analyse how meaning is created and negotiated by people operating with cultural tools in concrete activities (cf. Wertsch, 1998; Mercer, 1995; Saljo, 1996). In Chapter 2, Guida de Abreu reports a study of mathematics learning inside and outside the formal school system. This work has been carried out among sugar-cane farmers and their families in Brazil. De Abreu describes the nature of the traditional measurement and counting systems that are unique to this trade, and the skill with which farmers are able to use them when reasoning about their crops, the payment they receive, the size of their land and so on. In addition, the author illustrates how this system for measuring and counting is not valid in all contexts of the Brazilian society. For instance, new technologies or new systems of supporting and/or financing the farming industry may call for the adoption of more traditional measurement systems. Banks and state agencies may not recognise the indigeneous systems for measuring as valid, and they may demand conversion to a metric system with which the farmers are unfamiliar. The school is also a problem in a similar manner. The children's mastery of the traditional mathematical reasoning will take them nowhere in this particular setting. However, perhaps the most interesting aspect of the study is the manner in which the author documents how children of sugar-cane families learn to see their own knowledge as an inferior kind of knowl-edge, and that of the more formal mathematics taught in school as superior. This implies that issues of what de Abreu refers to as valorisation of knowledge must be added to the agenda when studying how knowledge is reproduced in society. It is impossible to separate knowledge and values, and understanding learning is very much about understanding what forms of knowledge are privileged, for instance, by formal schooling. In Chapter 3, Chris Sinha discusses what it implies to be a learner, and what one has to know in order to fulfil the communicative obligations of acting as a competent learner in a particular activity. Assumptions about learning, and thus the actions expected by learners as participants of social activities, vary across cultures and settings. The Western tradition of separating teaching and learning from production implies that children qua learners have a different relation to what they produce in comparison to what would be the case in societies in which teaching and learning are part of the primary production of goods and services necessary for daily living. Thus, when artefacts are produced in a school-like situation, the expectancy is generally not that they are going to work as a practical device in some future situation. Rather, the making and the creation of an artefact per se is the main point of the task. Sinha contrasts this attitude, and this positioning of the child as a learner, with observations made among pre-schoolers in a Zapotec village in southern Mexico. Here children, inducted into the skill of making a pot, are also learning to be producers in the sense of learning how to be actors in an economically important activity. These differences between cultures and social systems should alert us to the ways in which children (and adults) are positioned as learners in different activities, and what it takes on the part of the learner to be able to act competently within one particular framing of what it means to learn. The setting in which Nathalie Muller and Anne-Nelly Perret-Clermont analyse learning (Chapter 4) reflects cultural changes, albeit in a slightly different manner. The context here is the activities within a social movement originating in France and working towards creating new modes of communicating knowledge and skills. The idea behind so-called knowledge exchange networks (or KENs) is to provide a context in which potential teachers and learners can meet and exchange knowledge outside formal schooling. People who want to learn something simply put up a note about the theme or skill which they are interested in acquiring, and someone who feels they have knowledge to offer responds. And a certain reciprocity is expected -- those who want to learn should also be prepared to act as teachers in other fields. These spontaneously created groups with shared interests then continue their interaction as long as they find it productive. The initiator of this movement originally saw KENs as a way of making scientific knowledge available to a broader audience, but any topic of teaching and learning may be offered/sought: cooking, foreign languages, comput-ing, and so on. The authors have carried out fieldwork on KENs in Strasbourg, and they report on the manner in which people interact and construe the object of knowledge when learning under these conditions. They also illustrate how people acting as teachers and students, respectively, negotiate their positions during sessions, and how conflicts in expectations are handled by participants. The two remaining chapters of this part take us into the context of formal schooling and illustrate the nature of reasoning meaning-making that students engage in within such settings. But again, the core issue concerns how children operating in the context of school construe meaning, and how they are able to use linguistic and physical tools when reasoning. Erik De Corte, Lieven Verschaffel, Sabien Lasure, Inge Borghart and Hajime Yoshida report on a series of provocative studies on children's mathematical reasoning in the context of solving word problems in Chapter 5. Such problems are supposed to serve as contexts in which children learn to handle real-life situations by means of the mathematical reasoning acquired through schooling. Basically, the problem addressed originates in observations made in a number of studies in many countries on children's apparent insensitivity to what word problems mean. Even problems that do not make sense in terms of the situations they depict, or that do not provide relevant information, are solved as if they were. For instance, a problem such as "John's best time to run 100 metres is 17 seconds. How long will it take John to travel I kilometre?" is solved by an overwhelming majority of children as if it were possible to make a linear prediction from the former to the latter. It would be easy to interpret children's failures to handle these situations as a lack of knowledge of what constitutes appropriate mathematical reasoning, but the authors illustrate that this is hardly the most likely explanation. Rather, by modifying the contextual conditions under which children work when solving problems of this kind, the results become very different and much more encouraging. Jan Wyndhamn and Roger Saljo (Chapter 6) analyse how children reason when quantifying time. The point of departure for the empirical work was the finding in national evaluations in Sweden that children have considerable difficulties in solving problems of the following kind: How many days are there from March 24 until June 18? The performance level for this kind of task is surprisingly low, and, furthermore, students do not seem to get very much better at handling them as a consequence of further exposure to mathematics teaching. In the study reported, it is shown that the issue of the extent to which children are able to deal with this problem is very much a contextual affair. For instance, when given an artefact such as a calendar as a resource, the problem becomes very easy to handle successfully because the calendar invokes a counting rather than a subtraction strategy. In a similar fashion, using your fingers for counting the number of days, which is generally considered a primitive method in formal mathematics teaching, is also efficient. PART 2 Learning and Reasoning in Context Introduction by Paul Light In introducing the first part of this book, Roger Saljo highlighted the fact that much of psychology is unduly preoccupied with what takes place inside people's heads. In a variety of ways, the chapters making up that first Part illustrated the fact that understanding psychological activity often depends at least as much upon understanding social and cultural processes of negotiating meaning as it does on individual psychological processes. This theme is continued in this, the second part of the book. Once again we shall see a recurrent tension between traditional psychological approaches which take little account of context, and other recent approaches which, by contrast, give context and situation pride of place in accounting for learning and reasoning. These tensions are not addressed as a one-sided argument in favour of 'situated' learning and reasoning. The chapters making up this Part find weaknesses as well as strengths in situated approaches to both learning and reasoning. Nonetheless, the extent to which such approaches pervade contemporary analyses of both education and psychology is well illustrated here. The chapters themselves range in subject matter from early conceptual development to adult reasoning, from the classroom to the company archive, and from children's play to the work of a systems analyst. Yet, despite this diversity the commonality of the underlying themes remains clear. Perhaps the central issue is how to reconcile the evidence for situational dependency in so much of what we do or say with the evidence for continuity and coherence in our experience and behaviour over time. [...] Renkl, Gruber and Mandl (Chapter 7) explore the tension, which has been apparent for some years now, between established approaches to the psychology of instruc-tion and contemporary situated approaches. The proponents of the latter criticise traditional methods of classroom instruction for delivering only 'inert' knowledge -- knowledge which may be reproducible to order in school tests but which does not effectively engage with or support effective practice in the domain in question. There are clear echoes here of those chapters in the previous section which addressed the nature of 'school knowledge'. A number of psychologists and educationalists, especially in the United States, have come up with alternative approaches to learning in the classroom which seek to anchor instruction in authentic practice. For Renkl and colleagues, though, these approaches are seen to suffer a number of limitations. Situated approaches to instruction, they argue, tend to be insufficiently specified, particularly in respect of how learners are to be supported in their exploration of learning resources. This issue is perhaps particularly critical in the computer-supported learning environments which Renkl and colleagues are concerned with. [...] Whether these criticisms are properly directed at situated approaches to learning per se, as against certain attempts to apply them in classroom education, is a moot point. However, Renkl and colleagues render a useful service in reminding us of some of the enduring realities of educational experience and of the need to examine empiri-cal evidence before plunging too enthusiastically into wholesale curriculum change. The chapter by Rizzo, Marti, Veneziano and Bagnara (Chapter 8) explores the relationship between the psychology of reasoning, which typically focuses on the individual, and working settings in which most reasoning takes place in shared contexts. As Rizzo and colleagues point out, there is a sense in which not only individuals but also organisations can be said to learn. They may also be prone to forget. The collective aspects of knowledge production and maintenance, and the way in which collective knowledge resides in the practices of the workforce, have become much more widely recognised in recent years. Rizzo and colleagues argue that the compu-terisation of the workplace may prejudice some well established processes of collective learning, including the induction of new staff into existing patterns of practice. [...] The emphasis on logical inference in social context is continued in the following chapter, by Blaye, Ackermann and Light (Chapter 9). Drawing on both Piagetian cognitive developmental psychology and the psychology of adult reasoning, Blaye and colleagues critique the idea that logic can be seen as offering an abstract 'syntax' of mature thinking and reasoning. Their suggestion is that, on the contrary, pragmatic considerations of relevance-to-context actually seem to shape much of the thinking and reasoning not only of children but also of adults. [...] The same issue of whether psychological development can be seen in terms of abstraction of one kind or another exercises Bliss, Ogborn, Cronin, Reader and Tsatsarelis in the following chapter (Chapter 10). The essential tension addressed in this chapter is between the Piagetian 'epistemic subject', developed through a process of reflective abstraction, and the 'actual subject' engaging in physical abstraction on the basis of particular lived experiences in a particular physical world. Although Piaget is often said to have pictured the child as a 'scientist', in his own work he made surprisingly little reference to the development of children's understanding of physical entities and their properties. Whereas Piaget's 'logical schemes' may be context free, 'physical schemes' cannot be. Similarly, while the process of reflective abstraction which generates logical schemes can be seen as an activity of a notional or 'epistemic' subject, the process of empirical abstraction which underpins physical schemes must be domain specific, and must rest on particular experiences. [...] In the final chapter of the second part (Chapter 12), Etelapelto and Light return to the issue of instruction with which this Part began. They explore the limitations of context-independent models of design expertise, and how such expertise is acquired. Etelapelto and Light argue that design expertise is typically opportunistic in character, and responsive to context and situation. But 'context' has diachronic as well as synchronic aspects; it develops its significance over time through the life and experience of the individual. Learning is part and parcel of identity formation, and the development of a particular professional identity has considerable implications for the uses to which acquired expertise is put. Etelapelto and Light seek to illustrate this thesis through a study of software developers and systems analysts as they move through the transition from instruction to early work experience. At a time when there is so much public and political debate about the functions of higher education in relation to subsequent employment, it is perhaps surprising that there is not more research on the continuities and discontinuities marking the transition from study to work. Not only could such work be highly policy-relevant, it could also provide a valuable test-bed for ideas about situated learning and the transfer-ability of understanding. The issue of transfer is not extensively or frequently addressed in the chapters which make up this part. Underlying all of the tensions explored in these chapters -- between cognition as abstract ('in the head') and cognition as inhering in the relationship between an individual and a particular social or physical context -- there is an issue of 'what transfers' as individuals move through their lives, encoun-tering and participating in one situation after another. Arguably, none of the chapters offers any wholly satisfactory solution to this problem (if problem it really is). However, each of them in its different way contributes valuable illumination of the broader issue of how learners' engagement with their physical and social worlds conditions their learning, their reasoning and their behaviour. PART 3 Learning with and by Machines Introduction by Joan Bliss The focus of this book is on human learning, reasoning and thinking in different social and technological contexts, and the manner in which these practices and artefacts together help to determine the way in which people approach situations. In the first Part there was a specific focus on the manner in which meaning is created and negotiated by people operating with cultural tools in concrete activities. While Part Two continues this theme, there is a more specific focus on how contexts constrain or facilitate human thinking and actions. More particularly, it looks at the tension between traditional psychological approaches, which take little account of context and situation and the situated learning perspective, which gives context and situation pride of place in accounting for learning and reasoning. Part Three chooses a more specific focus in that it examines how new technologies, or new forms of artefacts, are becoming part of the contexts in which we learn, reason and think. It sets out to understand the extent to which people can appropriate these new technologies as tools for their thinking, and whether or not these new approaches can enhance their learning, reasoning and thinking. [...] New information and communication technologies make it possible to retain huge quantities of valued and powerful information and knowledge without a problem; space requirements are now of an altogether different kind. Also ICT transforms the way in which knowledge can be distributed in and between learning sites. Such knowl-edge and information, stored in personal computers or tracked through electronic mail or internet connections, is now accessible outside the standard recognised insti-tutions. Thus ICT challenges traditional boundaries between knowledge communities and the demarcation of knowledge domains. So we can ask: "What will count as epistemic authority in a society of distributed knowledge and how will this authority be exercised?" [...] The first four chapters in this Part involve learning with computers. Thus we examine the importance and the nature of educational settings or contexts in which computers are used: first in schools, and then in universities. Then we look at software tools: in a first instance at how ideas from situated learning have influenced their design, and in the second case, how the user's situated knowledge both enhances and hinders the use of IT tools. In the last two chapters we turn to learning by machines, and the situated agents movement in artificial intelligence which will be familiar to those in the field. While the situated agents approach may have had its roots in cognitive psychology, quite often its later developments are so different from its origins that it is not always possible to see the links. These two chapters about situated agents attempt to build these links. Recent research would indicate that learning with computers -- whether at a fundamental school level or in the discipline of, say, computer science at the tertiary level -- is becoming a genderised issue in much the same way as the learning of physics or chemistry, where boys and later young men would seem to dominate the scene, at least in many European countries. Littleton and Bannert's survey of the area shows, however, that the issue is subtle at two different levels. First, Littleton and Bannert refer to the work of Huff and Cooper (1987) which points out, "... that it is not the computer or even the software that is at the root of sex bias in software, but the expectations and stereotypes of the designers of the software" (p. 519). Second, they analyse a number of studies, including their own, which indicate that even when software has no obvious gender connotation, a sex bias is introduced by the "particular contexts and situations which constitute girls' and boys' experience of computing activities". In other words, through the analysis of the social contexts of young people's computing activities we can begin to understand the importance of either the social situation or even the activity itself. For example, we learn that the framing of an activity as a 'game' or as a particular type of task can deter girls from or attract girls to working with computers and, of course, similarly for boys. Similarly through Crook and Light's examination of issues related to the greater deployment of computers in order to allow teaching and learning to be more flexible and more affordable, we become aware of the importance of the analysis of the learning contexts. Crook and Light show that learning with computers in universities does not take place in a type of neutral zone; existing educational practices have a long historical development and, whether we like it or not, these circumscribe the ways in which today's students learn. Crook and Light's studies are diverse -- from self-report surveys of study practices during examination revision to students working together on academic problems. They also paid particular attention to the informal, out-of-class experience of study, for example, the possibilities for informal collaboration with peers and informal contact with tutors that arise from participating in a full time community of study. Their results give food for thought when they state, "... our results confirm us in the view that students are enculturated into particular communities of learning, and the resulting practices will offer resistance to the 'bolting on' of new educational technologies". Both these chapters serve as a warning since through their analyses the authors are able to show that: "Far from revolutionising society, the computer has conformed to society, becoming another element of the status quo" (Reisman, 1990, p. 45). Yet, perhaps again they reflect the nature of the transition from an industrial to a technological society. Perhaps at the moment those who work with new technologies are, as Crook and Light point out, highly self-selecting and may still be small in number. Perhaps such studies can provide information about what is both sufficient and necessary to make the new technologies more powerful resources through being more properly integrated into a culture we understand more fully. Huff and Cooper point out that, "... gender stereotypes in educational software often arise as a direct result of the designers' expectations about potential users" (Huff and Cooper, 1987). Not only are the designer's expectations critical but also the theoretical perspectives that inform his or her implementation of the software design. Luckin's chapter takes inspiration from the situated learning perspective to design a new learning environment, VIS, which is part of a larger system called TRIVAR (Luckin, 1996). VIS allows the user (children between 10-11 years) to create a mini ecosystem through the selection of animals and plants which share a common habitat; these elements are then, through a range of activities, built into a food web. One of the key situated learning concepts which underlies the design of VIS is what Vygotsky called the 'zone of proximal development' or the ZPD. Crucial to the ZPD is the distinction Vygotsky draws between a child's actual level of development and his or her potential level, that which can be reached with the assistance of the able peer, parent, teacher or in this case, the computer. Wood, Bruner and Ross (1976), referring to the ZPD, coined, at that time, a term to describe this process of outside help. ... (it) involves a kind of 'scaffolding' process that enables a child or novice to solve a problem, carry out a task or achieve a goal which would be beyond his unassisted efforts (p. 90). Thus, understanding and implementing the strategy of an able teacher is essential to the VIS environment. Luckin has used Wood's later work (Wood et al., 1993) to implement the scaffolding approach in her software. This involves five levels of hints or scaffolding as well as ways of adjusting or differentiating the activities provided to the child, in other words, "... 18 possible combinations of help and activity adjustment". This is only part of the story in the design of this software -- for example, the student model and the knowledge base are equally as complex. Teaching is an activity that is often taken for granted. Most of us believe that, given we know some area reasonably well, we could teach it. Reading Luckin's chapter brings home very sharply the complexity of the activity of teaching and makes us rethink again the talents needed to do such a job. Let us hope that Luckin's software achieves at least some of the success of an ordinary school teacher! Corruble and Bliss in their chapter present two research directions which use computational methods to aid learning about the world. In both cases, computational modelling and simulation are helpful in externalising some key assumptions embedded in individuals' internal representations of the world. The approaches are however different in the specific tasks they assist. A dynamic modelling tool IQON is used in the classroom to help students to model phenomena they experience in their daily life (traffic congestion, fitness and diet, shop management). CHARADE and PASTEUR are used to aid researchers to develop causal hypotheses from a number of observations (leprosy research, cancer research). Yet, both systems address the situatedness of the task they assist through a constructive approach to modelling, by taking advantage of the trial-and-error capability of computational simulations to resolve the discrepancy between model's behaviour and modeller's expectations. Our examples illustrate how computational tools offer the possibility of being used simultaneously as a medium and as a mirror. As medium, users can externalise their thinking through modelling, as a mirror through the simulation of the model which reflects back to the user the true face of what has been externalised. [...] end -- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%