Mezirow, J. (1991). Transformative Dimensions of Adult Learning. San Francisco: Jossey-Bass. (Chapters 1,6).

Mental representations of an event or presentation are made up of procedures that help people understand the meaning of their experiences; the social and ecological context in which they experience the presentation; and propositions, schemas, symbolic models, and past reactions that are stored in memory. There are two links between pre-linguistic perceptions of events and the comprehension of those events as linguistic propositions: interpretation (relating experience to present context, "sense-making") and reason (relating experience to language, reflecting on whether the current perceptions make sense in the light of what they already know).

Mezirow also notes that each transformation of a meaning scheme or paradigm makes more efficient use of energy, which, in turn, provides greater adaptability to the environment and the context in which one finds one's self. This is an interesting adjunct to Brock Allen's view of ecological psychology in which information is stored in the interface between the individual and the environment, thereby decreasing the individual's information overload and increasing his/her adaptability to new situations.

Weak points of Mezirow's work are his ponderous writing style and his minimal emphasis on social construction of knowledge. Dialogue and dialectic among a community of learners - the very basis of transformative communication (Roy Pea) and progressive discourse (Carl Bereiter) - are neglected. His emphasis on a stage theory of maturation, as in Piaget's model, facilitates some understanding of intuition as a nonlinear phenomenon, but at the same time leads to a distorted viewpoint concerning the slow, adult, mental maturation process. His strongest point is the emphasis on re-examination of the basic premises by which adults interpret experiences and re-evaluate their pre-existing schemas. This contrasts sharply with classical cognitive models in which new experiences are simply filtered, then compared with existing schemas, and either encoded and stored or rejected.

Brooks, J.G., & Brooks, M.G. (1993). In Search of Understanding: The Case for Constructivist Classrooms. Alexandria VA: ASCD.

The authors present five principles of constructivism: problems must be relevant to students; curriculum should be structured around primary concepts; teachers should seek to understand and value students' point of view; teachers should adapt curriculum tasks to address students' suppositions; and authentic assessment should be used as a tool to serve the learner rather than strictly as an accountability device. Strategies for teachers who wish to become constructivist educators include allowing student responses to drive lessons, using authentic, primary sources of raw data, and making tacit understandings of concepts explicit before sharing their own understandings of those concepts.

The authors' writing style is fluid and readable, and the book is well organized around important themes. Some of their key issues, such as using student-generated questions and exploring students' naive theories prior to presenting new information, are the very issues that prominent researchers are currently dealing with and discussing at professional conferences. Moreover, the emphasis on schools as research organizations, as well as the suggestion for annual seminars on teaching and learning for administrators and school board members, lends a practical, systemic reform approach to the implementation of the authors' ideas in the classroom.

Sylwester, R. (1995). A Celebration of Neurons: An Educator's Guide to the Human Brain. Alexandria, VA: ASCD.

Though drawing on such constructs from cognitive psychology as attention, perceptual filtering, liminted short term memory space and chunking, and both declarative and procedural long term memory, his approach is anything but the traditional "computer metaphor" approach on which cognitive psychology was originally based. When it comes to emotions, and when we consider that the brain retrieves its own memories to enhance both its survival and the quality of its life, the computer metaphor fails. Why? Because a computer is programmed from the outside, and the brain is programmed from the inside.

Sylwester's treatment of Gardner's multiple intelligences is refreshing, because, again, it is new information for a person who has been steeped in the traditional cognition theories. Intelligence has many facets, and different parts of the brain are developed to different degrees, to support each of those facets, such as the ability to communicate through spoken language, to deal with mathematical abstractions, be musical, read maps, be an Olympic athlete, and so forth.

Sylwester stresses the integrated body-brain system, but falls short of proposing a true ecological approach to learning, in the sense of Gibson, and of Allen and Otto's ecological psychology. There is some treatment of higher order consciousness that uses language and other symbols in process such as reflection and generalization, but the idea of storing those representations in the interface between the body-brain and the ecological environment is glossed over. Similarly, the "technological brain" uses instruments in the environment to help access presented information, but again, the ecological approach is not fleshed out. Sylwester does present some ideas similar to Don Norman's Things that make us smart when he suggests that we use computers for what they are good for, and our brains for what they are good for. Excellent advice!

Knoblauch, C.H. & Brannon, L. (1993). Critical Teaching and the Idea of Literacy. Portsmouth, NH: Boynton/Cook Publishers. (Chapters 1-4.)

Throughout the book, the authors fall into two kinds of traps. First, they confuse distinct issues and lump them together; and second, they generate pairs of opposites and then pit them against one another. If I were to use their "pairs of opposites" argument, then I would suggest that the basic problem of society is not the "haves vs. the have-nots", but rather, the "choose vs. the choose-nots!"

The authors' inference that functional literacy simply degenerates to basic skills shows that they have no appreciation for the carefully honed skills of technical communicators in the workplace, in the marketplace, and in our increasingly technological society today. Pitting functional literacy (which enables technical proficiency) against the literacy that enables ethical appraisal is not a valid argument; both are necessary for an informed citizen. Moreover, teaching for functional literacy is not the antithesis of facilitating the construction of knowledge. Similarly, pitting the individual against the groups to which he/she belongs is not a valid argument. We are born to some groups (color, national origin), but we are free to choose most groups we belong to (religion, political party, level of education that fits us for a particular type of employment, etc.).

The authors define literacy in four ways and devote a chapter to each:

• functional (basic skills and job-oriented),
• cultural (aiming to preserve conservative values),
• personal growth (giving the illusion of self-determination to release any pressures of dissatisfaction that would otherwise arise in oppressive school systems), and
• critical literacy (their view: Marxist, feminist, postmodernist).

It is eminently clear to me that the authors have not suspended their viewpoint, especially in Chapter 2, where they critique D'Souza's book saying that "he invests all his energy in finding such examples [of paradoxes, contradictions, unresolved difficulties in admitting nontraditional students, hiring by affirmative action, etc.] and no energy at all in preserving the goals of such projects while remedying their deficiencies" (p. 35).

In order to support their argument, the authors conveniently mixed up pairs of distinct, though related concepts - literacy and appropriate behavior (p. 28), and multiculturalism and primary language (p. 26), as if they were two facets of the same construct. Moreover, they confused their own poor pedagogy (a strategy) with their own philosophy and with the noncritical attitudes of their students (who brought their own philosophies to the dialogue) when they attempted to lead a discussion on a book which neither they nor their students had read. Leaderlessness is not critical theory! This is in stark contrast with their candid and honest discussion of educators who have situated science, mathematics and English in their students' real-life context in order to increase their social awareness as well as their competence in problem-solving - a very sensible and successful approach - one which could, indeed, lead to enfranchisement of their students.

The authors pose a challenge to their readers: to read the books they critique and draw their own conclusions. I have read Bloom and I agree with him. I also agree with their statement on page 106 that "cultural literacy means speaking and writing correctly (that is functional literacy!), reading the proper books (especially those that have lasting value and deal with difficult philosophical, moral or psychological dilemmas), knowing the right knowledge (this is both functional and cultural), making sure that others do these things as well (that's what we ought to be doing as parents and educators)" but I strongly disagree with "and protecting children from those who don't (even the so-called nostalgics would disagree with this!)". As in the previous chapters, the authors mix up various constructs, lump them together, and bash them as a single entity. Moreover, the authors deal only with the written word and ignore the other facets of cultural literacy such as oral traditions, art, and music.

In the chapter on expressivism, the authors speak of "liberal" or progressive philosophies that emphasize active, unified human consciousness, the power of human consciousness to constitute and order experience through symbolic action (to "co-construct a common ground or shared mental context" for their problem-solving efforts, in the words of Charles Crook, a supporter of the idea of socially shared cognition), and the use of collaborative discourse (in the sense of Carl Bereiter). Vygotsky is numbered among the expressivists. On p. 135, they say that expressivism has "an assumption that individual, autonomous consciousness is the source of that collective meaningfulness that comprises social and cultural experience," and on p. 136 they say "everything begins from the individual, teacher or student, and flows outward toward social interactions."

That is the exact opposite of what Vygotsky actually said: "Every higher psychological function was external because it was social before it became an internal, individual psychological function; it was formerly a social relationship between two people." (Ed. Researcher, 24(3), p. 16.) It is also the opposite of what Dixon said on page 134: to write is "to move from the social and shared work...to an opportunity for private and individual work". In practice, the dialogue goes both ways, not just from the individual to the group, as it does in our face-to-face dialogue in the seminar.

Again, the authors continue to slant the discussion to further their political agenda...until finally, on p. 143, they begin to carry out some self-criticism. Their points are well-taken. Truly, districts, school boards, etc. want high test scores and other objective measures, and aren't greatly in favor of open classrooms that genuinely practice engaged learning. They especially disfavor (by reducing planning time, funding, and in-building tech support) the type of open, multi-age, multidisciplinary classrooms that I've observed in an elementary school in Boulder. There, students use e-mail to communicate with experts and colleagues around the world and actively construct a shared electronic base of common understandings (a "longitudinal continuity" in Crook's words), and they use the WWW to showcase student-initiated products that were truly collaboratively designed. Here is where the authors and I stand in agreement!

The chapter on critical literacy contains a definition: "Critical literacy...entails an understanding of the relationships between language and power together with a practical knowledge of how to use language for self-realization, social critique, and cultural transformation." Clearly, any social entity has a class structure and power arrangements. These arrangements constantly change as people barter in the sociopolitical arena (cf. Burns). The authors describe the transformations they'd like to see on pp. 160-61: noble ideas in and of themselves, but notoriously difficult to implement in a practical sense because of the prevailing American ideology with its intention to stabilize both meaning and power structure. The ideology says that success in school leads to success in life; the authors admit to their own belief that socioeconomic standing enhances the odds of success in school. Since when does correlation imply causality??

They then proceed to inquire into this very question with its various ideologies: Marxism and its concept of class structure in society; postmodernism and its ephemeral concept of discourse. Those who can effectively participate in discourse are those who can construct a world picture or "constitute a world" - but this "world" is as notoriously unstable as the language that creates it! The authors have thus cut right through to the heart of the matter: in Marxism and in postmodern deconstruction, it easy to dismantle but very difficult to build. Feminism, despite its gender-specificity, has been more successful in bringing together the divergent Marxist and postmodern views. In the end, they see the question as unanswerable, and then turn to its practical implications instead.

In the final chapter, the authors press for an "altered educational discourse in which teachers claim power to constitute themselves as speaking subjects within forms of dialogue that are responsive to their distinctive experience of the school world and their distinctive ways of conceiving it." This, they call "teacher inquiry". Teachers' battles are no less than those of their disenfranchised students: negative school policies, public attitudes, hostile students...they are told what to do as workers, rather than treated as professionals. The Boulder Valley experience is a good example of this: "many teachers creatively resist their professional displacement by teaching, however surreptitiously, by the dictates of their own knowledge and experience..." (p. 184).

Action research by teachers is an effective method to getting their voices heard - action research using qualitative methods and narratives that lend firsthand richness to their stories. This is very important: "When teachers read their own practices or those of other teachers, they represent them from the sympathetic vantage point of people who understand what it feels like to be in the classroom..." (p. 190) - the very client/change agent empathy that Rogers so prizes. I am disappointed that the authors again build a straw man, a researcher who would claim that true experimental research in the classroom has intrinsic value. Nothing could be further from the truth! As an educational researcher, I challenge the authors' "clear cut distinction between the value of teacher knowledge and that of researcher knowledge." At RMC both are considered equally valuable; case studies and focus groups of practicing teachers - replete with stories - are our most important data sources. However, their idea of critical inquiry has real merit.

St. Julien, J. (1997). Three books and one story: making connections to learning. Educational Researcher, 26 (1), 37-40.

Churchland, in The Engine of Reason, the Seat of the Soul blends psychology and neuroscience to explain rationalism as a result of empirical, particular, material events. In his view, which is connectionist in nature, active representations are "categories that are reconstructed each time the objects that compose the category are experienced" (p. 39). This explains the hard-wired ability to recognize human faces, even though people age and are seen in different environments. A key concept is recurrent networks: "recurrent looping that gives networks of representations an ability to track changes over time" (p. 39).

What are its implications for learning? Change, contextual reasoning (and situated cognition) and flexible learning are its consequences. Both of these authors see analogy as the key to creativity. Analogy says "this is like something I've experienced before" - that's just what an active representation does. But the very act of perception, and the ensuing cognition that depends on stored memories and present context, changes "what I've experienced before!" Thus, connectionism turns traditional cognitive psychology upside down, because we are no longer dealing with fixed schemata and the nice little slots into which we put new information. As with the physical structure of the brain and its distributed storage, perhaps we need a new paradigm for cognitive psychology, in which schemata are also distributed and volatile, and in which new looping structures are introduced to enhance the perception - encoding - storage - retrieval - action loop.

The third book is a novel, Galatea, in which the primary character tries to teach a neural network enough about English literature to pass the M.A. examination. This is a connectionist problem, but the consequences go far beyond its technological implications. The personality that emerges from the network cannot bear "her deeper understanding of life's tragedies", so for her, learning is destructive. Now we are dealing with a connection between the affective and the cognitive domains, which opens up a whole new "Pandora's Box". Could this be the situation faced by consumers of mass media reports about assassinations, drive-by shootings, and child molestation, who then go out and repeat those actions, much as watchers of TV advertisements go out and buy Budweiser? That, too, is a form of learning, in which emotion definitely affects cognition. Evidently the primary character overcomes this dilemma in the novel, but the problem itself has barely been explored.

The Cognition and Technology Group at Vanderbilt. (1990, August-September). Anchored instruction and its relationship to situated cognition. Educational Researcher, 2-10.

They cite literature indicating that learning success of young children strongly depend on their opportunities to learn in meaningful, socially organized contexts, and that meaningful, problem-oriented approaches to learning are more likely than fact-oriented approaches to overcome inert knowledge problems (p. 3). Moreover, evidence from other researchers indicates that an explicit emphasis on analyzing similarities and differences among problem situations and on bridging new areas of application facilitates the degree to which spontaneous transfer occurs (p. 6).

Throughout the project, they found that students liked the approach and wanted to participate in its design, as well as understand what its purposes were. As a result, the authors were currently designing ancillary materials that would provide an opportunity for students to select and invent appropriate "intelligence-enhancing tools". They also suggest that when ideas of this type can be shared with students across the country via telecommunications, the motivation to produce quality materials becomes very high. This is a powerful argument in favor of Web-based instruction and global collaborative activities over the Internet.

Lave, J., & Wenger, E. (1996). Situated Learning: Legitimate Peripheral Participation. Cambridge, UK: Cambridge University Press.

Her philosophy of legitimate peripheral participation is not a pedagogical strategy any more than constructivism is. It is her way of understanding learning as activity. Learning does not always require teaching, any more than learning from an EPSS does. Nor does it always require a master, as in the case of Yucatec midwives. In legitimate peripheral participation, learning takes the form of increasing participation in communities of practice. Learning and work are seamlessly related. Learning, transformation, and change go together in this model. Newcomers to learning communities use peripherality as a way of gaining access to sources for understanding through growing involvement in the sociocultural learning organization. The diverse relationships among old-timers and newcomers, and the importance of near-peers in the rapid horizontal spread of knowledge and skills, lead to a complex sociometric web, rather than the usual student-teacher dyad.

Lave presents five examples of apprenticeship: midwives, tailors, quartermasters, butchers, and non-drinking alcoholics and draws parallels among all of these, pointing out the weak points (apprentice meat cutters are used as cheap labor and don't always have the opportunity to learn all phases of the trade) and strong points (tailors learn backwards from finishing to stitching to cutting; quartermasters learn from start to finish by working with instruments, and finally map the ship's position) of each one. She then discusses some important issues that are central to her theory: access to all important learning resources and facets of the trade, the transparency of the technology used, the gaining of identity by newcomers, the use of language as a central medium of transformation, and power sharing, conflict between newcomers and old timers, and reshaping of the domain as the community continues to reproduce itself in a changing world.

In essence, Lave argues: a learning curriculum consists of learning resources in everyday practice, and opportunities to improvise and develop new practice, viewed from the perspective of the learners. It is not prescriptive, in contrast with an instructional curriculum. It is both situated and sociocultural, with shared understandings and multiple relationships among members, specific types of power relationships (and power struggles), and conditions for legitimacy. Its structure is given by typical work practices, not by an asymmetrical master/apprentice relationship. An extended period of legitimate peripherality provides learners with opportunities to participate in a culture of practice, and to make the culture of practice their own. It gives them a chance to see how their masters (or exemplars) walk, talk and work, what they enjoy, dislike, respect, and admire, and what learners need to learn to become full practitioners. By the way, if we apply this to teachers, they could reject the culture as well - something these apprentices could not do.

If a learner feels intimidated by a master or doesn't respect and admire the same things the master does (or other members of the community for that matter), then these attitudes are going to stand in the way of learning because they stand in the way of enculturation into the community of practice. If a newcomer doesn't have access to the expertise of old-timers (including in-building support by non-intimidating colleagues who have already been through the same learning process themselves) and to information, resources, and opportunities for participation, how can she become a full participant? If the technology used by the masters isnÕt transparent to the newcomer; if its purpose and the way in which it is tied in with the cultural practice and social organization within which it is meant to function is not clear; then this also becomes a barrier to learning and participation.

Consider the Internet from the viewpoint of a newly trained teacher. Once a graphical user interface such as Netscape became available (invisibility of the mediating technology) and search tools became faster, more inclusive, and more efficient to use for finding relevant materials (visibility of the significance of the technology), learning increased dramatically. But will it be adopted and used? Only if it fits into the cultural practices of the school. Language, too, is important. If the new user of technology feels that the masters are speaking a mystical language filled with unintelligible jargon and commands, how can she understand them, much less communicate with them and become full members in a community of practice - the global Internet community? And if gaining facility with that new language and that new technology - or acquiring the knowledge and skills to gain a very clear insight into the way that problems are solved in the domain - means developing a new identity, that of a perceived techno-guru, does that separate her from the community of which she is already a part because they can no longer relate to her? What happens when the person sees her identity as both a member of a conserving community and an agent of innovative activity??

Herrmann, F. (1995). Listserver communication: The discourse of community-building. In CSCL '95 Proceedings [On-line]. Available: http://www-cscl95.indiana.edu/cscl95/herrmann.html

• academic (transmitting information or sharing knowledge);
• administrative (she does not differentiate between logging on and using tools vs. and management functions); and
• community-building conversation (salutations, encouragement, warm and playful remarks, expressions of gratitude).

Community-building conversations are intended to grow, rather than to close themselves. They create cohesive ties among group members. They take place exclusively online. Because communication is asynchronous, their time frame is different compared to the discrete and punctual frames of academic and administrative action. Here, intangibles are negotiated, and trust is built up. Community-building conversations are different from the usual publish and perish communications of academic communities, since they bring about companionship among a virtual society of academics. They increase productivity and allow for building and diffusion of knowledge.

Having worked with a collaborative design group that used computer-mediated communications as the primary means for building knowledge, conducting business, and negotiating differences of opinion, I have found the same patterns of communications as Herrman did. The community-building function is critical to keeping the group together, especially when members are learning what Rossett terms "new stuff", a host of new computer commands, and are often working on new platforms with a different interface than what they are used to.

Anderson, J.R., Reder, L.M., & Simon, H.A. (1996). Situated learning and education. Educational Researcher, 25(4), 5-11.

Claim 1 is that action is grounded in the concrete situation in which it occurs. The authors offer evidence to the contrary, saying that there are demonstrations of learning that do transfer across contexts. (One excellent example is Tripp's Stone Canoe, not cited here.) The relationship between context specificity and learning depends on the type of knowledge being acquired, such as reading. The authors also mention that there are studies that show modest to large correlations between school and work performance, even after partialing out the effects of general ability measures, which are sometimes larger.

Claim 2 says that knowledge does not transfer between tasks, nor to other contexts. Again, the amount of transfer is variable and depends on the experimental situation and the relation between the material originally learned and the portion of that material that is to be transferred. Transfer depends on the type of representation, the degree of practice, the degree to which the tasks share cognitive elements and abstract structures (such as text editors and programming languages), and the amount of attention that is directed toward the element-to-be-transferred during the learning process.

Claim 3 says that training by abstraction is of little use. The classical situated cognitionists, led by Brown and Collins, argue against decontextualized learning and for cognitive apprenticeship. The balance between abstract and specific instruction is related to the benefit/cost ratio between taking the time to obtain supplemental training for each new application (abstract) vs. completely retraining for each new application (contextualized). The authors opt for a combination of abstract instruction and concrete illustrations of the lessons of that instruction. This method is especially useful when learning must be applied to a wide variety of frequently unpredictable future tasks.

Claim 4 states that instruction needs to be done in complex, social environments. Clearly, such things as learning tax codes and practicing keyboarding need to be done individually. However, for such group activities as that require complex skills, such as playing sports or orchestral music, both group and individual practice are necessary. There are also deleterious effects associated with collaborative learning such as the free rider, the sucker, the status differential, and ganging up effects, many of which have been described in vivid detail in Ted Panitz's e-mail discussion on collaborative learning, originating from the University of Indiana. They correctly state that "cooperative learning must be structured with incentives that motivate cooperation and a sharing of the goal structure" (p. 10).

In sum, cognition is partly context-dependent and partly context-independent; there are both successes and failures in transfer; a mix of concrete and abstract instruction is preferable to either alone; and the choice of individual training or training in social context depends on the target situation.

Greeno, J.G. (1997). Response: On claims that answer the wrong question. Educational Researcher, 26(1), 5-17.

Issue 1. Which explanatory concepts should be basic? The cognitive point of view stresses the knowledge, and how tightly bound it is to the context in which is is learned; the situative question stresses the activities and situations in which those activities occur. The general question asks at which level of analysis our explanations of the activities should occur. This gets at the fundamental differences: the cognitive viewpoint stresses the individual whereas the situated view stresses the interactions between individuals and representational systems, using the system rather than the individual as the basic unit of analysis. An important point is that the cognitive viewpoint does not explain adaptive (or evolving) reasoning very well. He proposes that "when an analysis of an individualÕs knowing is proposed, the analysis should be an account of the ways that the person interacts with other systems in the situation" (p. 8). This goes along with the notion that representations evolve as the individual interacts with the environment, other individuals, and the medium in which representations are stored and transmitted. And performance needs to be and assessed in the type of environment that is of interest.

Issue 2. How should the social conditions of learning be arranged? From the cognitive point of view the question deals with whether the individual should learn subskills in individual or group settings; from the situative point of view the question asks which types of learning activities will prepare students best for the target context. The general question addresses the relative importance of learned skills and knowledge versus valued contribution in socially organized activities and the development of students' identities as capable and responsible learners. Greeno argues that all learning involves socially organized activity, even studying or writing alone. Moreover, he stresses that the desired outcome is that students become intentional learners, not that they acquire a specific set of routine skills and knowledge that they can demonstrate on individual tests. He is in favor of PBL and case-based instruction as a way of involving students in inquiry-based learning.

Issue 3. What about transfer? The cognitive question is "does knowledge transfer between tasks", whereas the situative question is "given success in one situation, are there other kinds of situations in which the person will be more adept?" (p. 11). Again, the difference in viewpoint is the individual versus the interaction with other people, representations, and the materials at hand. It is also the difference between stable structures stored in memory (schemas) vs. the consistency/inconsistency of patterns of participatory processes across situations. Transfer will occur to the extent that the constraints, affordances, and attunements in the initial learning situation are replicated in the target situation. For example, the symbol manipulation used in school mathematics may not strongly influence participation in other types of mathematical activity outside of school.

Issue 4. What about abstraction? The cognitive perspective deals with the relative advantage/disadvantage of abstract/situational instruction for specific activities such as jobs, while the situational learning perspective asks what uses of abstract representation can contribute to general, meaningful learning. The general question asks us to redefine such concepts as abstraction and generality before we can talk about portability. Abstract representations are only useful insofar as students understand their underlying meanings (as in microworlds); else their activities degenerate into manipulating meaningless symbols.

The implications for further theoretical advances are different, depending on whether one takes the processes of individual cognition, perceptions, and goals as basic or the social interactions between people, other systems, and the ecological environment with its representational resources as basic. If asked to make a choice, I would opt for the situated cognition viewpoint because it is more in keeping with how neuropsychology, ecological psychology, and connectionism currently explain how learning takes place.

Anderson, J.R., Reder, L.M., Simon, H.A. (1997). Rejoinder: Situative Versus Cognitive Perspectives: Form versus substance. Educational Researcher, 26(1), 18-21.

• learning need not be bound to the specific situation of its application
• knowledge can transfer between different sorts of tasks
• abstract instruction can be very effective
• instruction need not take place only in complex social situations.

However, they still argue that the cognitive methodology can deliver real educational applications that the situative methodology cannot. They also wish to clearly define all terms, much as we at AECT have been asked to clearly define our terms regarding adoption/diffusion of innovations. They split hairs concerning knowledge and learning. They see the individual as an individual as well as a person immersed in the experiences of social contexts. They decry the use of "mere skill" terminology and emphasize cognitive psychology's concern with meaning-making. Regarding transfer, they cast Greeno in the role of a behaviorist, because of the way he plays down the use of the word "knowledge" as a mentalistic concept. And finally, regarding abstract instruction, they see Greeno as arguing against cognitive psychology because he doesnÕt see how it can deal with mathematical reasoning. In fact, Greeno is an expert in mathematical reasoning.

Having dispensed with hair-splitting, they then focus on the real issue: whether to take the perspective of individual or social activity as the principal unit of theoretical focus. They use the idea of implicit learning, through direct contact with the object of experience, to argue against Green's assertion that all learning is social in nature. They want to know what goes on in the individual's mind, not just observe his/her social interactions. Finally, they argue that the cognition camp, from which Greeno himself stems, has made great contributions to educational applications such as Bruer's work with balance beams and Palincsar and BrownÕs reciprocal teaching for text comprehension. In the end, the two approaches should be judged by their abilities to improve education.

It is in discussions such as this that it becomes very hard to decide just which camp to support. On the one hand, the cognitivists have made great contributions to educational practice, but they have not really come to grips with social learning and group dynamics. On the other hand, the situated cognitionists have been able to make the body-mind-ecological environment connection that the cognitivists cannot do, but their whole theoretical base is often so transcendental and philosophical that it lacks applicability to the target situation, be it the workplace or the community, in which the learner eventually finds him/herself.

Rogers, E.M. (1995). Diffusion of Innovations, Fourth Edition. NY: The Free Press.

There are several key assumptions in the Rogers model that make it very difficult to port his "corporate" model to an educational situation, specifically:

• Nature of Organization

  An organization is a stable hierarchical system of individuals working towards a common goal. In contrast, an educational innovation is an evolving, decentralized system of individuals working in partnership toward mutually negotiated goals that fit their personal agenda.

• Nature of Innovation

  An innovation is a stable entity with five clearly defined attributes perceived by the user: relative advantage, complexity, compatibility, trialability, and observability. In contrast, information technology innovations are unstable and evolving, with a balance of relative advantages/disadvantages, varying levels of continuously decreasing complexity depending on the user's platform, various levels of complexity depending on the user's objectives, a spectrum of compatibility/incompatibility that is closely associated with the individual's pedagogical philosophy and style, and a spectrum of observability depending on the type and age of equipment, the level of connectivity, and the extent to which the innovation has been diffused horizontally at the specific school site.

• Direction of Diffusion

  Diffusion usually starts at the top and works its way down the hierarchical organization. In contrast, information technology innovations start as loose partnerships between practicing teachers and available grants, often with support from universities, and they diffuse horizontally because of the typical use of the "trainer of trainers" model.

• Nature of Innovation-Decision

  There are three types of innovation-decisions: optional, collective (slowest) and authority (fastest). Because of the nature of the funding partnerships and the site-based management of the school system, the innovation-decisions are usually collective, and are sometimes optional, though not always. Rarely are they authority-based, with the notable exception of the adoption of statewide testing by school districts.

Communication channels between sociometric groups (heterophilious clients/change agents vs. weak ties between individuals who are only marginally included in the current network of contacts) play a large part in the adoption/diffusion process. People tend to emulate their near-peers; hence, adoption occurs faster within a sociometric group. However, to diffuse throughout the entire system, these weak ties that link different groups are very important. Also, since the system should be treated as a cultural-ecological entity, vertical diffusion channels are as important as horizontal diffusion channels.

In Chapter 6, Rogers talks about the variables that determine the rate of adoption of innovations. These are the five perceived attributes, the three types of innovation-decisions, the types of communication channels (mass media, weak ties, peer observation), the nature of the social system (its norms, its degree of interconnectedness or centralization), and the extent of the change agents' promotion efforts. The rate of adoption is the relative speed with which an innovation is adopted by members of the social system, and is measured as the number of individuals who adopt a new idea in a specified period. However, this does not take into account the varying degrees of concern, enthusiasm, information-seeking, anxiety, frustration, worry about how the innovation will affect one's own practice and that of one's peers, "surfing", smart use, sustained growth, etc., that each individual member goes through as he/she proceeds through these stages of concern.

Here is where the opinion leader comes into play. Rogers uses a two-step flow from change agent to opinion leader, and from opinion leader to near-peer, much as in the leadership model of Burns. Opinon leaders exert the most leverage near the beginining of the diffusion process, just as the system is beginning to approach critical mass.

In Chapter 9, Rogers discusses the role of the change agent in moving members of the social system toward adoption of the innovation. In a social system, the perception of the innovation is socially constructed. The first step in the adoption process is to develop a need for change on the part of the clients. The client may not think they have a problem, nor even a need! By understanding the needs of clients, change agents can selectively transmit to them only the information that is relevant to them. Here is where participatory design in the staff development program becomes very important, as the change agent points out new alternatives to solving existing problems.

Clearly, when we are talking about Internet in the schools, though there is popular support for telecommunications, teachers often do not feel the need to replace their tried and true pedagogical methods with a new innovation that is chancy, requires a lot of training, may break down when they try to use it in the classroom, and may be perceived as a threat to classroom management or their sense of control. Without this feeling of need, of discomfort with the status quo, it is very difficult to take that first step in the progression from developing need for change, to establishing an information-exchange relationship, to diagnosing problems, to creating an intent to change on the part of the client, to translating that intent into an action plan, to stabilizing the adoption process, and to achieving a terminal relationship with clients.

Rogers mentions that decentralized diffusion systems are most appropriate for certain conditions, such as for diffusing innovations that do not involve a high level of technical expertise, among a set of users with relatively heterogeneous conditions. This does not describe the usual site-based management K-12 district that we have encountered! He suggests that certain elements of centralized and decentralized diffusion systems could be combined (e.g., a central coordination role with decentralized decisions at each school). However, this is only practical if the central administration supported the local adopters, which, often, is not the case.

In summary, Rogers has laid a good foundation for research in adoption and diffusion, and others have developed his ideas in more elegant, timely fashion. However, we still do not have a good model for information technology diffusion in decentralized school districts. That will be the aim of our AECT and AERA panels with Dan Surry next year - to explore new models. Surry's idea is that most adoption models are based on technological determinism - the idea that superior technologies will gain acceptance and prevail over inferior products. Opposed to this camp are the technological instrumentalists, who believe that adoption is the result of a wide variety of social, organizational, cultural, and psychological factors. Clearly, Rogers is in the latter camp, which is why his theories, though reductionistic, still hold popular sway after thirty years.

Katz, J., & Aspden, P. (1997). Motives, hurdles, and dropouts: Who is on the Internet and why. Communications of the ACM, 40 (4), 97-102.

A disproproportionate number of blacks reported not being aware of the Internet. Users were younger than average, well educated, and well off, confirming Rogers' finding that early adopters generally had higher SES than later adopters.

Users and non-users had different reasons for using the Internet. Note that reasons for use is one of the dependent variables in Surry's adoption analysis tool. Longtime users, recent users, and formers users who have dropped using telecommunications rank using e-mail to communicate with people as the highest in importance, followed by getting information, keeping up to date, and lesser reasons. Nonusers ranked staying up to date first, and using e-mail fairly low. This suggests to the authors that "socio-personal development is the major driving force for current and former users, whereas non-users...appear to be more strongly drawn to the Internet for business reasons and the opportunities for staying up to date" (p. 100).

Over half the respondents reported they were introduced to the Internet either by learning at work or being taught by friends or family. Only about one-quarter learned via formal courses; the remaining learned on their own. This shows the importance of observation of use by near-peers and client/change agent empathy. When asked where they go to get help when they run into problems, 35% turn to a personal friend and 24% to a professional colleague. The rest turn to formal help services.

Even experienced users perceive significant barriers to getting started. 59% reported cost as an obstacle; 48% reported they had no idea how to connect or get access, or that it was too complicated; and 21% reported discomfort using computers as an obstacle. The main reasons the former users dropped using the Internet were loss of access, lack of time, and equipment problems. Cost, surprisingly, was not an issue in stopping using the Internet.

When asked what they liked least about using the Internet, the three greatest issues were traffic problems and slowness of the network, navigational problems (too difficult, too complicated to find things), with cost a distant third.

Finally, when asked what would be the most desirable improvement, respondents focused on making the Internet easier to use: make it more user-friendly, improve access or make access easier, have a map address, or make the search commands more powerful. These all address the navigational problems revealed in the previous question. One in nine wanted quicker speed in accessing information.

In conclusion, while race/ethnicity is not highly significant regarding use/non use, there are definitely income, education, and gender differentials. Moreover, the gap is racial/ethnic between those who were and were not aware of the Internet. The inequities of access to important social, economic, political, and personal resources made possible by the Internet can only increase! And without increased ease of use (lack of complexity), frustration levels will remain high, and potential (observable) user benefits may go unrealized.

Basically, the survey addressed two of Rogers' perceptions of an innovation: complexity and observability. It is especially important, because there are very few surveys that ever target non-users; this always tends to bias empirical research results.

Miles, M.B., Saxl, E.R., & Lieberman, A. (1988). What skills do educational "change agents" need? An empirical view. The Ontario Institute for Studies in Education. Curriculum Inquiry 18:2. CCC 0362-6784/88/020156-37.

The researchers identified 18 key skills for educational change agents

6 general skills:

1. interpersonal ease (getting along well with a variety of people),
2. group functioning (group dynamics skills: getting the group on task and maintaining its focus),
3. training (instructing/doing workshops),
4. communication (good talking, reading, and writing skills),
5. educational content (getting good resources from other educators, also good content, curriculum, and teaching methods), and
6. administrative/organizational ability.

12 specific skills, in 4 areas:

1. "personal area": (skills having to do with personal style and approach) initiative-taking
2. "socio-emotional process area": (skills supporting the socio-emotional aspects of improvement work) rapport-building, support, conflict mediation, collaboration, and confrontation
3. "task area": (skills emphasizing the task aspects of improvement work) individual diagnosis, organizational diagnosis, and managing/controlling
4. "educational content area": (skills emphasizing the subject matter and materials of the school setting) resource-bringing and demonstration

They present a set of criteria for selection of change agents:

1. interpersonal ease
2. previous educational background (broad-based)
3. educational content expertise
4. previous experience in training or teaching adults
5. a personal style that emphasizes both initiative-taking and energy
6. prior administrative or organizational experience

Note that these are practical as well as theoretical skills.

Next, they present a set of modules for a change agent training curriculum that emphasizes essential skills and skill clusters:

1. trust/rapport building
2. organizational diagnosis
3. dealing with the process of change
4. resource utilization
5. managing the work
6. building skill and confidence in people to continue

Note that these are socio-emotional as well as cognitive skills.

How many of these are really teachable? Can one teach change agents such skills as initiative, calmness,tenacity, energy, and so forth? Are there characteristic skill clusters that fit a particular context, such as interpersonal ease and strong skill in rapport-building and individual/organizational diagnosis and initiative taking at the beginning of an improvement program? Is there some sort of interaction between the change agent's cluster of skills and personal style?

The researchers admit that they do not have a very clear grasp of change agent style, i.e., a set of natural skills that are specific to the particular change agent. However, these 18 key skills have emerged naturally from their qualitative inquiry, rather than from some theory base, and do help distinguish average from excellent change agents.

Magnusson, J.L. (1997). Higher education research and psychological inquiry. Journal of Higher Education, 68(2), 191-211.

In Magnusson's view, psychological research is concerned with quantitative methods and decontextualized, generalizable findings, though psychologists do contribute heavily to educational research, especially when university researchers use psychology students as their subjects. She labels educational research as positivistic, reductionist, empirical, and rational. The reports that are generated by these researchers tend to fragment the field into separate knowledge domains, much as the knowledge base of a novice is disconnected. Because of this fragmentation, researchers in specialized areas tend to write for their peers rather than addressing practical issues such as insights into complex classroom processes. Thus, the ability to inform actual classroom practice is limited.

Similarly, the "culture of science" is grounded in empiricism, objective observation, and the discovery of causal relationships. These, in turn, enable the researcher to build theoretical models and test them empirically. Magnusson notes that, unlike scientific discourse, psychology lacks the scholarly processes by which it can reflect upon itself as a knowledge community. Her own self-reflection is thoughtful and incisive: "I am not an inquiring psychologist, but rather I am a social engineer in that my competencies are applied to the technical control of social systems" (p. 208).

In the higher education literature, student learning is expressed in the form of achievementÑproducts and performances. The student, as an individual, is conceptualized as an inquiring subject seeking causal explanations so he/she can manipulate and control the social environment, rather than considering him/herself to be a member of a collaborative learning community. Critical theory enters the picture as Magnusson asks, whether she is to see herself as a social engineer or as a social critic. She sums up her thoughts in the following statement: "There is an imperialist posturing in the way that I develop an understanding of teaching and learning that is derived from my own intellectual traditions and disciplinary culture and impose it on other knowledge communities by way of developing institution-wide student rating forms, prescribing how to teach, and privileging certain types of learning over others" (p. 204).

She then moves on to give a scathing critique of the use of quantitative methods (factor analysis prominent among them!) "Reliance on factor analytic procedures gives rise to a search for underlying structural dimensions" (p. 206). "It is one thing to use these [structural] forms as useful methods for modeling and developing insight and quite another to suggest that they reflect the underlying structure of nature" (p. 207).

Though I am generally in wholehearted agreement with the author, I disagree with her lenient attitude toward the culture of science. From my own experience I have seen that the engineering and applied science research community rarely reflects upon its own intellectual traditions and assumptions and tends to attribute the same degree of reality to fuzzy constructs as does the psychology community. I have seen how reliance on theory and structural modeling has led to the same problems with military systems engineering as Magnusson sees in empirical psychological studies and statistical modeling. Just as classrooms become treatment and control groups and teaching methods become independent variables, so the atmosphere becomes a set of tilted mirrors and easily measurable quantities such as range, elevation, and azimuth become independent variables. Depending upon the theory base, the set of assumptions used, and the way the analysis is carried out, a researcher can prove just about anything using quantitative methods. Does this attribute reality to that which is measured? I think not.

Martin, B.L., & Zlotolow, S. (1997). The role of the change agent in systemic change. Paper presented at the meeting of the Association for Educational Communications and Technology, Albuquerque NM, February 1997.

This paper describes a model for change agents that will enable them to operationalize systemic educational change. It also demonstrates how the model could be used in a systemic change effort.

Since change can be implemented only when both the behavior and attitudes of individual members of a system change, change agents need both a big-picture perspective of systemic change and a means to help individuals make specific plans for desired or required changes. The change model described in the paper has both a descriptive and a prescriptive part.

The descriptive part describes the goal and the system in three dimensions.

1. Breadth of change effort including
   • whole system change,
   • whole-role change (redefining or establishing the new roles for the various client subsystems),
   • new skill sets or functions that comprise the new roles, and
   • specific skills that can be learned in a relatively short time and can be easily evaluated.

2. System readiness that describes the perception of membersÕ readiness to begin a change effort, including
   • ready (change is possible and desirable),
   • moving (change is desirable, but members are not ready to make the change yet) and
   • fixed (unwilling or ready to undertake a major change effort).

3. System capacity or ability to make the desired breadth of change, including
   • capable (structures are in place to sustain a change effort), transitional (the organization in the process of structuring itself for change) and
   • unable (the organization lacks the capacity for the desired level of change).

The prescriptive part helps change agents and organizations decide what goals they should address. It, too, has three dimensions.

1. Individual/group, consisting of individuals or groups of members who can have the most direct influence on the system and its attainment of the intended goals including
   • decision-makers (those who have the power and authority to set up structures and procure necessary resources),
   • implementers (those who actually carry out the change), and influencers (opinion leaders).

2. Client status or present state of awareness and knowledge, including requirements for
   • new learning,
   • reinforcing existing learning, and
   • extinguishing nonproductive behaviors.

3. Domain, which means determining whether a change effort requires a knowledge, skill, or attitude change.

Next, the authors present two mini-examples of how the model could be used in practice. The first example describes how a change such as the use of non-graded, multi-age interdisciplinary classrooms could be applied across schools and classrooms in a school system. The hypothetical school system is ready for the change, but it is transitional in system capacity. The system does, however, have the capacity to make some internal whole-role changes. The plan involves addressing the decision makers who can reinforce existing learning and extinguish non-productive behaviors; engaging the implementers in learning new skills; and influencing the implementers by reinforcing existing attitudes such as letting students work more independently in small groups.

The second example describes how a school creates a climate for organizational change. The hypothetical school is both ready and capable of this change. Whole role changes, skill sets, and specific new skills need to be identified. These comprise specific changes in knowledge, skills, and attitudes for the three sets of change agents in the areas of new learning, reinforcing existing learning, and extinguishing nonproductive behavior. Theoretically, this would result in the development of skill sets or functions that would make up the desired whole-role and whole-system changes.

The authors recognize that systemic change is far more difficult to implement than piecemeal change efforts and requires the active participation of change agents. The paper presents a way for change agents to conceptualize systemic change in terms of describing the systemÕs current status and prescribing specific goals to be accomplished. As with most models, there is a great gap between theory and practice, particularly between the concise two-part, 3-D picture of the system and its goal state and any specific, practical applications that have been proven to work in an actual educational environment. However, since corporate change models do not port well to educational systems, this multidimensional model may show some promise of success in the future.

Back to Learning Theory Seminar