"Where is the wisdom we have lost in knowledge?
Where
is the knowledge we have lost in information?"
T.S. Eliot.
Most of us misunderstand the educational potential of the Information Superhighway (IS). We believe that it has the intrinsic ability to teach. If we did not believe that, why are so many of us creating Web pages for our students to study? Why are distance educators turning to the Web to teach our students located at distant sites? Yet the IS does no more than deliver information, like a carrier pigeon. What happens to that information when we receive it is not up to the IS. It is up to us.
The reason we are confused about the IS is that we have forgotten that information is not the same as knowledge and that knowledge is not the same as wisdom. Let me give a simple example. Every morning, before I leave for work, I read a small digital thermometer in my kitchen that tells me the temperature outside. My thermometer is a device that measures data - the temperature - and converts the data into information I can read - a number. When I read the number, I interpret its meaning. I can do this, first, because I am literate. I know that "40" is a value on a scale that represents all possible values of temperature. I obtain meaning from the number by associating it with what I already know about temperature - that 40 is fairly cool, that my tomatoes won't be frozen and that there may well be dew on the lawn. Only when I have decoded the information and drawn inferences about the tomatoes and the dew can I truly say that I "know" what the thermometer has told me, that I have learned something about the weather that morning. Then, if I am wise, I will wear a light sweater and put on shoes that don't leak when I leave for work.
This simple example illustrates a number of important points. First, information is no more than a structured representation of data. When course developers create Web pages, they are giving form to their ideas by selecting what information to give, by deciding how it will be divided among "pages" and arranged on them, by determining type sizes and fonts, by selecting graphics and so on. ("Information" is, after all, something that has been given a "form".) Second, information is learned only when students work at transforming it into knowledge. This can only happen if students can decode the information and can construct meaning from it in terms of what they already know. Third, wisdom is the use of knowledge in ways that are compatible with a person's needs and with the accepted norms and behaviors of the society in which the person lives. Knowledge that is constructed from information but not used wisely (or not used at all) is said to be "inert" (Grabinger 1996, p. 666).
It follows that learning occurs when students generate knowledge from within, not receive information from outside. I will go into this in more detail later. Suffice it to say for now that learning is an active process not a passive one and that it can take place a long time after relevant information has been received. Students need time to think, for information to age in the cask. It also follows that information may be received but not learned, even though it may by retained, through brute memorization, just long enough for the test.
From this, I would like to suggest some principles about learning that underlie the rest of this paper.
Access to the IS is not a sufficient condition for learning, though, for students in distance learning programs, it may be a necessary one.
The information that we prepare for students comprises data to which we give a structure that is determined in no small part by the medium in which the information is presented. This structure will be influential in how students understand what we tell them.
The acquisition of knowledge from information requires effort and involves perceptual and cognitive processes that decode symbols, deploy literacy skills to interpret them, and apply inferencing abilities to connect them to existing knowledge.
The acquisition of wisdom from knowledge requires practice in the judicious application of that knowledge in the personal and social context in which the student acts.
By implication, testing just to see whether a student has received information, which is not atypical even in higher education, sheds absolutely no light on what students really know nor on whether they can use any knowledge they have acquired productively.
These principles are clearly concerned with learning rather than offering prescriptions for teaching university students via the Web. However, it is extremely important to understand what must happen for learning to occur before we start suggesting ways to support it. The following section therefore sketches out components of a theory of learning that can subsequently guide instructional design for Web-based courses and how to teach them so that the intrinsic inability of the Web to teach ceases to be as large a problem as it seems at first to be. How this theory might translate into instructional strategies is tackled in the section after that and forms the focus of discussion and activity for the rest of the workshop.
In this section, I look at three related topics: Theories of knowledge construction ("Constructivism"), the importance of context in learning (theories of "Situated Cognition" and "Situated Learning"), and the social context of learning. To set the stage a little more completely, though, I begin by looking more at the outcomes of learning which I have so far neglected in favor of its basic mechanisms.
As we have seen, learning is the process through which information becomes knowledge. But knowledge is not the same as memory which means that tests of memory for information do not tell us what students have learned. So what actually results from knowledge construction? Current thinking is that as students learn they construct internal representations of knowledge, commonly referred to as "mental models" (Johnson-Laird, 1983; Rouse & Morris, 1986). Most simply conceived, a mental model is our way of knowing the world. It is our own personal explanation of a particular phenomenon or set of concepts. To obtain a better of idea of what mental models are, it is useful to look at particular attributes good models have and the functions they can perform.
Basic to cognitive theories of learning is the idea that there is a reciprocal and iterative relationship between new and existing knowledge. New knowledge is assimilated to old and old knowledge changes to accommodate new (Neisser, 1976; Piaget, 1968; Rumelhart & Norman, 1981). Imagine that you are looking at a Duckbilled Platypus for the first time and that you have never heard of one before. You have a mental model of "mammal" that describes warm-blooded air-breathing animals, most of which walk about on dry land, but some fish-like instances of which live in the ocean. The Platypus is confusing because it has some attributes of "mammal" but not others and a set of attributes uniquely its own. At first, you might try to assimilate the creature to your "reptile" model. When you test the validity of this hypothesis, however, you find that the Platypus is warm-blooded and has fur. So you revise your hypothesis and try again. This continues until, through knowledge-building that is almost certainly guided by some form of teaching, you alter your mental model "mammal" to accommodate the new creature, other monotremes and all non-placental mammals. In short, your mental model helps you acquire new knowledge and is changed in the process.
To be useful, some of your mental models must be sufficiently abstract to apply to a large number of cases. "Mammal" certainly fits this criterion. This means that accommodation to new knowledge can be achieved by altering the model, not replacing it. Rumelhart & Norman (1981) also suggest that new models may created ex nihilo through analogy to existing models.
The more abstract a mental model, the less information it contains about specific cases. Yet a good mental model will always contain enough information to tell the student the kind of information to expect. While the "mammal" model will specify that mammals have body hair, it will not specify how much or what color it will be. On the other hand, it will allow the student to infer the presence of body hair even though none may be apparent, as in the case of elephants or seals.
Extending this idea further, a mental model can serve as a tool for learning about the world. Neisser (1976) speaks of "anticipatory schemata" which are models that direct people's inspection of the world as they learn about it. In short, what we already know about the world will guide our search for new information and our acquisition of new knowledge. This activity, under guidance from a teacher, forms the basis of inquiry teaching (Collins & Stevens, 1983) that has been shown to be extremely effective for learning, particularly in the sciences. Indeed, the method of learning by inquiry is largely indistinguishable from the Scientific Method.
A well-formed mental model is a prerequisite for effective communication about a topic. It is extremely difficult for anyone to describe what they know about something if their knowledge is fragmentary and poorly organized. This has two implications. The first is that the act of describing what you know to someone else may have the effect of improving the mental model, particularly if your interlocutor keeps prompting for clarification. The second is that students' descriptions of what they know, whether spoke, written, drawn or presented in some other medium, can be very useful for someone who must determine the extent of their knowledge.
Mental models can be sufficiently abstract, or sufficiently complex, to map entire bodies of knowledge. For example, a student's model of "literature" might contain the concepts "author", "work", "theme", "character", "plot", as well as the major works, authors and movements in literature. This model can then help the student search through information about literature in indexes, databases, or on the Web. (As we shall see below, the development of such strategic skills is more important when students are less-well connected to on-campus teachers and other learning resources than when students have these facilities close at hand.)
With these ideas as background, let us now turn to the three major pieces of learning theory that underpin learning from the Web by university students.
As we have seen, knowledge is constructed by students. Theories of knowledge construction, commonly referred to as "Constructivist" theories, describe psychological processes by means of which students build their own understanding of content. Understanding arises as students work to reconcile what they already know and believe with information they are encountering for the first time, or old information on which they are gaining a fresh perspective. Each student brings to this process a unique aggregate of prior experiences and understanding as well as a set of aptitudes and beliefs about the learning process itself. Constructed knowledge is, within limits, personal and idiosyncratic.
At first blush, it appears that constructivist theories of learning are no different from the general cognitive theory I described and illustrated above. To some extent this is true. However, there are important differences that have a particular impact on the way knowledge construction is supported by instruction. The first difference is one of degree. To what extent does the personal nature of knowledge matter when it comes to teaching and learning? The second difference, which I discuss in subsequent sections, is the context in which knowledge is constructed and applied.
Everyone has a different understanding of the world around them, knowledge that is embodied in sets of mental models that differ considerably from person to person. Since this is also true for scientists who describe the world for us, and for teachers and instructional designers who build courses to teach students, it can be argued that all knowledge is subjective and that there is therefore no objective reality to teach about. The conclusion that there is no objective reality, proposed by some constructivists (Bednar et al., 1995; Cunningham, 1992a, 1992b), does not, in my opinion, follow logically from the observation that knowledge is subjective. Nor does it help much in our quest for guidance about using the Web to teach in higher education. It does draw attention, however, to the undeniable fact that our students start our courses with a great range of knowledge and assumptions about what they are going to learn. If these are at variance with our own, we must expect that changing them will be difficult, if not actively resisted, and that the outcomes of our efforts will not necessarily lead to conformity to our own understanding and opinions.
In some cases, this may not matter. Some subject areas do not have "right answers" and are taught so as to encourage students to develop their own ideas and opinions. For example, Spiro et al. (1992) uses interactive hypermedia to help students learn about what he calls "ill-structured domain". Literature is an example. Spiro uses a videodisk of the movie "Citizen Kane" to help students construct knowledge of plot, them, character, and so on, by letting them browse through scenes in the film in any order they choose. He describes it as "crisscrossing the landscape". Each time students revisit a scene, their perspective has been changed by visiting other scenes in the movie. Knowledge construction in this scheme of things is therefore cumulative, iterative and idiosyncratic. And because students arrive at their own, personal understanding of literature, what they know is far more compatible with their own needs and values than if it had been taught to them didactically as objective truth.
The same is not true, of course, of other disciplines. A lot of the time there are right answers and correct procedures we want our students to learn. As Merrill has put it (1992), he would not want to undergo brain surgery by a surgeon who has constructed his own, idiosyncratic knowledge of neuro-anatomy. In these cases, our goal is a certain uniformity in the learning outcomes our students attain. However, this does not mean that we simply get our students to learn by rote. Work in our laboratory, albeit with younger students (Osberg, 1997; Rose, 1996; Winn, 1996, 1997), has shown that, with care and guidance, students can construct accurate knowledge about chemistry, biology and grammar using techniques similar to Spiro's. And that knowledge, too, is personal and compatible with students' interests and needs.
Another factor that affects when it is permissible to allow students to construct their own knowledge is their degree of expertise. Spiro limits his claim for success of his approach to knowledge construction to the acquisition of advanced knowledge. If learning occurs when new knowledge is assimilated to mental models, which change to accommodate it, we assume that there is already a reasonably robust set of mental models in place. One would not simply turn neophytes loose with a videodisk of "Citizen Kane" and tell them, "Learn about literature."
Research on the differences between experts and novices and on the development of expertise (Dreyfus & Dreyfus, 1986 and chapters in Chi, Glaser & Farr, 1988, and Ericsson & Smith, 1991) has led to a number of conclusions about expertise. Important here is that as expertise develops, declarative knowledge is transformed into procedural knowledge. In other words, novices construct knowledge about something. Experts internalize and automate that knowledge so that they can use it without much reflection. In our terms, they have acquired wisdom. The internalization and automation of knowledge makes it subjective as the constructivists claim all knowledge to be. This means that, even in cases where our students need to acquire knowledge of an objective world, if they are sufficiently expert already and have good mental models of the content domain, they may certainly profit from constructing knowledge for themselves using strategies, perhaps like Spiro's, that are not particularly constraining with regard to how knowledge is acquired. At the extreme, of course, this is exactly how good research is done even in the "hardest" of sciences.
From these comments, we may draw the following conclusions about knowledge construction:
Constructivist theories of learning are generally compatible with and build upon cognitive theories. Some radicals deny the existence of an objective reality, but this position is not particularly useful, and is probably wrong anyway.
Knowledge is nonetheless subjective and often idiosyncratic. This makes it difficult to teach didactically.
The subjective nature of knowledge does not matter when the content is ill-structured and where there are no right answers.
The subjective nature of knowledge does not matter when the students are well along the road towards expertise. They have already developed robust mental models and, at the university level, need to think creatively about their discipline anyway.
It may be more efficient to teach content directly to less advanced students.
Students need guidance as they construct knowledge. The degree of prescription in guidance will vary from discipline to discipline and from degree of expertise to degree of expertise. I will return to this later.
Knowledge is constructed in a context. Knowledge construction never takes place in a vacuum and when students work with information devoid of a context, the knowledge that arises is inert and unconnected to what is meaningful to the student. Theories of "Situated Cognition" describe how the context in which knowledge is needed determine how it is used. Theories of "Situated Learning" stress the need for knowledge construction to take place in a context that both connects new knowledge to what is already meaningful for the student and does so in a way that makes learning worthwhile.
Much of the criticism of higher education in recent years (and of basic education, for that matter) has arisen from concerns about the relevance of what students learn. Contrary to common belief, this issue is far broader than whether or not students can use the knowledge they acquire at university to get a job. It lies at the heart of the knowledge construction process.
Recent research on how people solve problems in the real world suggests that thinking is guided as much by the immediate context in which the problem-solver finds himself as by the application of general rules, procedures or plans (Lave, 1988; Suchman, 1987). Suchman reports that people in a weight-watchers program developed contextualized ways of measuring ingredients for recipes. For example, if a recipe calls for two thirds of three quarters of a cup of cottage cheese, you could multiply the two fractions and then measure out the amount found in the answer. Or, you could measure two thirds of a cup of cottage cheese, form it into a patty on the kitchen counter, cut it into quarters and take three of them. The first approach relies on the application of a general rule to solve the problem. The latter uses the materials to hand and is context-specific - it would not work with orange juice, while the mathematical approach would. This does not mean that we should not teach general procedures for solving problems, rather that people tend not to use them in the real world. The reason for this is often that people fail to see the connection between what they have learned in school, in Math and Science for example, and everyday cognition. Their knowledge is inert. It is here that educators have focused attention, developing theories of situated learning.
There are two aspects of situated learning that have a bearing upon knowledge construction in higher education. The first concerns ways in which information be constructed that is not inert. The basic strategy is to anchor knowledge in a context that is meaningful to the student and that requires the application of knowledge to real problems (Cognition and Technology Group at Vanderbilt, 1990). This approach has always been present in university education. Cooperative programs in engineering have been around since the 'sixties. Internships are common in many disciplines. And computer simulations are commonly used that foster knowledge construction in context and require the application of new knowledge to "real" phenomena. One particular advantage of these strategies is that they largely avoid the problem of "reductive bias"(Spiro, 1992) where content is artificially simplified to provide easier access for novices but never restored to its full, real complexity.
The second way in which situated learning comes into knowledge construction by university students is via the belief that knowledge of a discipline requires you to look at the world in the way that a fully-fledged practitioner working within that discipline does. Thus, an objective of the situated learning of Math is to know what it is like to be a mathematician (Brown, Collins & Duguid, 1989). In such cases, our students serve cognitive apprenticeships, or even professional apprenticeships, in which they acquire knowledge and, by applying it, develop wisdom and a knowledge of the culture of a discipline or a profession (Lave & Wenger, 1991). This is certainly a worthy objective to pursue in higher-level courses, where the students are more likely to become mathematicians or literature professors. However, I am not convinced that this is a goal for all students to pursue in all disciplines. We do not need to be experts or even advanced in all disciplines (Winn, 1994).
Finally, we know that as a person develops expertise, their knowledge is increasingly contextualized (Dreyfus & Dreyfus, 1986). A novice medical student may well be able to manage in the clinical setting simply by following procedures learned declaratively. With advancing skill and as knowledge becomes increasingly procedural, the student will discover that in some situations those procedures simply do not work and that it is necessary to improvise. At this point, the student experiences uncertainty about their ability to deal with problems and their performance may actually decline - they dither, as Dreyfus & Dreyfus put it, "like a mule between two bales of hay". With advancing experience comes confidence and the ability to adapt to different contexts. The declarative knowledge of the novice may eventually disappear entirely. Indeed, it is common experience among instructional designers that experts make very poor subject-matter specialists: They have a hard time explaining what they do because they just do it without thinking.
Theories of situated cognition and learning allow us to draw additional conclusions about knowledge construction:
People often do not use their formally-acquired knowledge of a to solve problems in the real world, using rather the affordances of the situation in which the problem occurs.
It is therefore necessary to make an effort to help students construct knowledge in such a way that its usefulness is readily apparent. This can be achieved by situating learning in a meaningful context in which knowledge is acquired and used.
The mechanisms for anchoring instruction in this way range from cooperative programs, internships, cognitive apprenticeships and simulations.
However, the degree of expertise students need to attain will determine the extent to which they acquire the knowledge and skills of a practitioner in a discipline.
The development of expertise is accompanied by the ability to apply knowledge in context.
There are, of course, many generalities and abstractions that students need to know that are hard to learn in specific contexts.
An important part of the context in which students construct knowledge is other people. Knowledge is therefore constructed socially. Three factors comprise the social dimension of learning that are relevant to our topic. The first of these is the social construction of knowledge itself. The second is the social nature of situated learning. And the third is concerns students' beliefs about other people and themselves and the reasons they give for their own and others' actions. I will now briefly examine each of these.
Knowledge is constructed socially. Although the knowledge our students construct is personal, there must be sufficient commonality of understanding so that they can communicate what they understand to other people. Indeed, this outcome is largely inevitable. Meaning is shared and therefore negotiated among members of a community that uses common knowledge (Vygotsky, 1978). Within a social group, this takes place through dialogue about information as new knowledge is acquired from it so that insights and perspectives are shared from the outset. The need to reach consensus in a group has the advantage of requiring students at least to consider others' points of view even if they do not agree with them. A side-effect of this process is the ability to communicate within the group about the new knowledge, to jointly situate it in a context and to use it for the benefit of the community. On the other hand, it does not preclude individuals from maintaining their own idiosyncratic ideas, even if they keep them to themselves. Thus, knowledge and opinions develop dynamically in seminars, formal discussions and informal interactions among our students.
A second aspect of the social construction of knowledge concerns the relationships that exists between teachers and students. Here, too, is a social relationship, though, traditionally, it is an asymmetrical one; the teacher "knows" and the student does not. Ideally, the interaction between the teacher and the student is such that the teacher acts beyond what the student already knows but within a "zone" demarcating the knowledge the student is capable of constructing (Vygotsky, 1978). Put another way, the teacher provides information and guidance about how to build knowledge from it that will certainly alter mental models but not break them. The skilled teacher (who could be a professor or another student) keeps just enough ahead of the student to maintain a manageable challenge. This is hard to do. The student is a moving target!
Situated learning is likewise a social activity. Nowhere is this idea better expressed than in Lave & Wenger's (1991) notion of "Legitimate Peripheral Participation". Admittedly, these authors describe various kinds of apprenticeships, few of which have direct connections to university education. However, their case studies make important points for our discussion.
We have already discovered that situated learning is "legitimate". The tasks that the student learns to perform and the knowledge the student constructs is useful and directly applicable, which means that the student is engaged in activities that are authentic and necessary to the success of the enterprise. Medical interns do real and useful work, for example, and by virtue of that feel that they belong to the community in which they are learning. Yet, initially at least, the knowledge and activities are still somewhat "peripheral" to the mission of the enterprise. While they are essential, failure to perform well is not fatal. A medical intern does not normally take prime responsibility with life-and-death situations in the early stages of training. Finally, knowledge construction and application require the student to participate in the group as a whole. This leads not just to seeing the relevance of new knowledge and skill but also to acquisition of knowledge about the culture of the enterprise. After work, apprentices learn a lot about the community they are learning to join by listening to stories told by the old-timers, for example.
Finally, people's knowledge of the world and of the society in which they live is mediated by their beliefs about the information from which they are constructing knowledge, about the people who provided it to them and about their own needs and abilities to deal with it (Bandura, 1977, 1978; Salomon, 1981). Consistent with constructivist theory, when we construct meaning from information, we do not extract knowledge from it that is invariable and absolute. We build knowledge from what we think the information means. Often, this involves attributions we make about the source of the information. In extreme cases, we may believe that the information is a lie and either construct knowledge from the opposite of what it says or disregard it altogether. Learning about political ideas through campaign literature is an example! But in almost all cases the knowledge we construct is influenced by our perceptions of the source of information, including the technology that is used to deliver it. For instance, people typically think it is easy to learn from television (Salomon, 1984).
A particularly important kind of attribution we make concerns our perception of our own ability. "Self efficacy", as this is called, is a powerful determinant of how well our students learn new knowledge and whether they enjoy doing so. For example, a large number of education students (including myself) have low perceived self-efficacy for statistics. In my case, I started my intermediate statistics course thinking, "I'm no good at Math." This turned out to be a self-fulfilling prophecy, as low self-efficacy usually is. The opposite is not true, however. Self efficacy that is undeservedly high, or overconfidence, also leads to less than optimal success; students think a course will be "a breeze" and fail to put enough effort into their work to succeed.
Again, we can add to our conclusions about knowledge construction from our consideration of the social dimensions of learning:
Learning takes place in a social context. The meaning we ascribe to information and the knowledge we construct from it emerge through dialogue with other people. Our knowledge is therefore, in part, held in common with others.
When a teacher is part of this social interaction, the opportunity exists for learning to arise when a tension is maintained between what the teacher knows and what the student knows that is sufficient to change mental models but not to break them.
In the special case of apprenticeships, students learn as much about the culture of the community they are apprentice to as they do about the knowledge that their work in it requires.
The knowledge we construct is influenced by our beliefs about the source and value of the information we receive and about the medium in which it was delivered.
Our success in knowledge construction is significantly influenced by our beliefs in our own ability to learn it.
We have seen that learning requires the transformation of information into knowledge and knowledge into wisdom. These activities are enabled by and mediated by three things: Cognitive processes that add to and alter mental models; the influence on these processes of the context in which knowledge is constructed and used; and social influences that enable and mediate the knowledge construction process. In one way or another, universities have always provided support for these three activities. Classroom instruction, seminars and work in laboratories can directly guide knowledge construction. All manner of activities, from formal exercises to internships to simulations, provide contexts for knowledge acquisition and application. The formal and informal interactions among students, faculty and the community provide a social context for directing knowledge construction along acceptable and meaningful paths.
The advent of the IS has changed all this. It has made it possible for students to receive information, in a variety of formats, at home and in the workplace. But because the Web provides just information, not instruction or any other support for learning, it is completely incapable on its own of supporting knowledge construction, of providing a context for learning and of providing the kind of learning community that universities have always nurtured. This means that whenever the Web is used in our courses, for whatever reason, we must deliberately add back to the learning experience these three kinds of support that we have traditionally provided as a matter of course. Without this, our use of the Web is bound to fail.
This section sketches out some general principles about how this might be done. Putting meat onto those bones will be a significant part of the activity in the workshop.
The strategies we can use to help students construct knowledge focus primarily on providing them with guidance that limits the scope of what they have to do. If one were, for example, simply to turn literature students loose with Spiro's "Citizen Kane" disk without any guidance, learning would be inefficient at best and probably quite chaotic. Perhaps, like the proverbial ten thousand chimpanzees, a student might eventually hit on an acceptable notion of literature. But it is very unlikely.
Also, we must use strategies that get students to work with the information they receive. As we have seen, knowledge is constructed through iterative interactions with material that force students to work with the information, to view it from different points of view and to associate it with what they already know. We must provide means for helping students to become active in the knowledge construction process.
The support we provide for students' knowledge construction falls into three general categories: Online as part of the information we provide; online "help"; and off line.
Online strategies designed to help students construct knowledge from information may take a number of forms:
Guidance may be embedded in the information itself. The way in which information is organized on the screen and the symbol systems in which it is encoded can, implicitly, direct the knowledge construction process. On the one hand, screen composition can guide eyeflow, directing students' attention to important information in a particular sequence. On the other hand, visual and audio cueing can direct attention, highlight information, suggest structure and sequence. As the multimedia capabilities of the Web develop, this aspect of guiding knowledge construction will become increasingly important.
Guidance may be provided on separate pages or in separate frames on the same page as the one that provides information. This might consist of explicit directions about how to process the information, using such comments as, "Be sure to notice that ...", or, "What does this make you think of?" to build connections to existing knowledge.
Guidance and interaction with information can be effectively implemented in tutorials that are either Java-based interactive Web pages or discrete downloadable programs written in any common authoring language. Right now, Web-based tutorials are extremely limited in the amount and type of tutorial interaction they allow. This will change. But the course developer should consider using computer tutorials to substitute for the kind of interaction that goes on in a face-to-face discussion, whether in a class or elsewhere. It goes without saying that the tutorials developed for Web-based learning should follow the precepts of good design and teaching that underlie the development of all tutorials.
Online help systems fall into two categories:
Additional online resources to which the student may turn for guidance. These may be links to additional Web pages that provide additional information, commentary on the knowledge the student is engaged in constructing, or explicit information about how to study.
A means to contact the instructor, teaching assistant or other students either in real time, through a chat facility, or by leaving email.
Offline guidance and interaction can be provided in a number of ways.
Through face-to-face "class meetings" that occur at one or more points throughout the course. Of course, it may not be possible for everyone to meet, physically, in one place and at one time. However, when meetings like this can be arranged, they can be very successful.
For distant students, through meetings with local tutors or visits to local study centers. The British Open University has organized this kind of support structure for its students very successfully.
Through formal or informal study circles that consist of other students learning the same material. This provides not only the opportunity to get help with the material but also to establish the kinds of social contacts that the Web typically denies to students learning at a distance.
Let me begin by restating that not all learning need be situated in a particular context. There is a strong case to be made for teaching abstractions and generalities that are context free and that the student can bring to bear in a number of diverse situations (Merrill, 1992). Also, a lot of learning that goes on in higher education legitimately requires the construction of conceptual knowledge for its own sake. However, in many cases, constructing knowledge and applying it in particular contexts is both desirable and effective. In these cases, we need to assure the provision of support for Web-based learning.
In theory, because the Web is computer-based, it should be able to simulate a variety of situations in which knowledge can be acquired and used. In practice, given current technology, real-time simulations of processes that require rapid decision-making, such as flight simulation, are precluded. Nonetheless, it is possible to log in to remote sites and use apparatus online that is not available locally to interact with information and construct knowledge.
More simple to implement would be less complex simulations that might simply require students to play particular roles and make decisions - national leaders dealing with an international crisis, engineers deciding how to prevent flooding in the Midwest, NASA scientists planning and directing the exploration of a planet in the Solar System. In these cases, as part of a team, each student participant would be required to apply knowledge in order to solve problems and make recommendations and to do so as part of a team, each member of which had to meet obligations or jeopardize the entire activity.
Finally, the IS may be used to put students in touch with practicing members of a discipline or profession. To an extent, this can help in the student-apprentice's socialization to the community in which acquired knowledge is applied.
By default, the Web is a lonely place in the normal course of events. Each student sits alone at a terminal. If dialogue with someone else occurs, identities may be declared, concealed or altered without anyone knowing the difference. The bandwidth of any interaction with someone else is typically narrow, limited to the exchange of text typed either in real time or sent as email. (The transmission of real-time audio and video are becoming much more common, but at the time of writing remain the exception rather than the rule.)
On the other hand, the IS has encouraged the creation of all manner of virtual communities that could never have come into existence without it. These communities range from formally-constituted networks of colleagues working within a company or across organizations on a particular project, to loosely formed and often ephemeral communities that drift in and out, focusing on this topic or that, without any planned activities or even focus. These range from online clubs interested in the same investments or wines to the bizarre, the occult and the offensive.
Web-based virtual communities can support our students in the following ways:
Providing a forum for discussion of topics relevant to the knowledge all students in a group are working to acquire. In one sense, the Web inhibits the kind of social interaction that we have seen contributes so much to knowledge construction. It lacks the spontaneity of a good argument and, again, the bandwidth is narrow. What is more, people are freer to join or leave a discussion than they are in a classroom where convention requires even non-participants to remain until the bitter end. In another sense, participants in a Web discussion are less inhibited and more likely to say what is on their minds. This can lead to more open and honest exchanges, but also, sadly, to "flames".
Providing a less formal forum for discussion of the myriad things on-campus students talk about. Sports, politics, the arts, the formation and dissolution of friendships are all aspects of student life that can go on over the Web and can contribute to a student's sense of belonging to a virtual community.
Finally, it is certain that some students take more readily to using the Web for the social aspect of knowledge construction than others. A prudent course developer or instructor will provide alternative ways to socialize students who do not wish to belong to the virtual community.
The main point of this presentation has been to stress that simply converting traditional university courses to web pages, or worse, just putting lecture notes on a Web page, is a serious mistake. All the Web (and the other pieces of the IS) can do is bring information to students. To learn anything at all, students have to construct knowledge from that information. To do so, three things need to happen. First, processes need to become active that connect information with existing mental models and alter the mental models to accommodate the information. Second, often the context in which knowledge is constructed and applied has to be identified and brought into the picture. Third, the social nature of knowledge construction must be considered and opportunities provided for students to interact with other students, teachers or other members of the community. With the Web as a significant piece of the instructional delivery system, these important contributors to learning disappear and particular care and effort has to be given to providing them through other means. I made a few general suggestions on how this might be done at the end of the paper. It remains to develop more elaborate strategies for bringing interaction and guidance to our online courses.
Bandura, A. (1977). Social learning theory. Englewood Cliffs NJ: Prentice Hall.
Bandura, A. (1978). The self system in reciprocal determinism. American Psychologist, 33, 344-358.
Bednar, A., Cunningham, D., Duffy, T., & Perry, D. Theory into practice: How do we link? In G. Anglin (Ed.), Instructional technology past, present and future. Englewood, CO: Libraries Unlimited.
Brown, J.S., Collins, A., & Duguid, P. (1989). Situated cognition and the culture of learning. Educational Researcher, 18 (1), 32-43.
Chi, M., Glaser, R., & Farr, M.J. (Eds.), The nature of expertise. Hillsdale, NJ: Lawrence Erlbaum.
Cognition and Technology Group at Vanderbilt (1990). Anchored instruction and its relationship to situated learning. Educational Researcher, 19 (3), 2-10.
Collins, A., & Stevens, A., (1983). Inquiry teaching. In C.M. Reigeluth (ed.), Instructional design theoriesand models. Hillsdale, NJ: Lawrence Erlbaum.
Cunningham, D.J. (1992a). Assessing constructions and constructing assessments: A dialogue. In T. Duffy & D. Jonassen (Eds.), Constructivism and the technology of instruction: A conversation. Hillsdale, NJ: Lawrence Erlbaum Associates.
Cunningham, D.J. (1992b). In defense of extemism. In T. Duffy & D. Jonassen (Eds.), Constructivism and the technology of instruction: A conversation. Hillsdale, NJ: Lawrence Erlbaum Associates.
Dreyfus, H.L., & Dreyfus, S.E. (1986). Mind over machine. New York: The Free Press.
Ericsson, K.A., & Smith, J. (1991), Toward a general theory of expertise. Cambridge: Cambridge University Press.
Grabinger, S. (1996). Rich environments for active learning. In D. Jonassen (Ed.), Handbook of Research for Educational Communications and Technology. New York: Macmillan, 665-692.
Lave, J. (1988). Cognition in practice. Cambridge: Cambridge University Press.
Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge: Cambridge University Press.
Johnson-Laird, P.N. (1983). Mental models. Cambridge MA: Harvard University Press.
Merrill, M.D. (1992). Constructivism and instructional design. In T. Duffy & D. Jonassen (Eds.), Constructivism and the technology of instruction: A conversation. Hillsdale, NJ: Lawrence Erlbaum Associates.
Neisser, U. (1976). Cognition and reality. San Francisco: Freeman.
Osberg, K.M. (1997, March). The effect of having grade seven students construct virtual environments on their comprehension of science. Paper presented at the annual meeting of the American Educational Research Association, Chicago, Il.
Piaget, J. (1968). The role of the concept of equilibrium. In D. Elkind (Ed.), Six psychological studies by Jean Piaget, New York: Vintage Books.
Rose, H. (1996). Design and Construction of a Virtual Environment for Japanese Language Instruction. Seattle, WA: University of Washington, College of Education. Unpublished masters thesis.
Rouse, W.B., & Morris, N.M. (1986). On looking into the black box: Prospects and limits in the search for mental models. Psychological Bulletin, 100, 349-363.
Rumelhart, D.E., & Norman, D.A. (1981). Analogical processes in learning. In J.R. Anderson (Ed.), Cognitive Skills and their Acquisition. Hillsdale, NJ.: Lawrence Erlbaum.
Salomon, G. (1981) Communication and education. Beverly Hills CA: Sage.
Salomon, G. (1984). Television is "easy" and print is "hard": The differential investment of mental effort in learning as a function of perceptions and attributions. Journal of Educational Psychology, 76, 647-658.
Spiro, R.J., Feltovich,P.J., Jacobson, M.J., Coulson, R.L. (1992). Cognitive flexibility, constructivisim, and hypertext: Random access instruction for advanced knowledge acquisition in ill-structured domains. In T.M. Duffy & D.H. Jonassen (Eds.), Constructivism and the technology of instruction. Hillsdale, NJ: Lawrence Erlbaum.
Suchman, L. (1987). Plans and situated actions: The problem of human/machine communication. New York: Cambridge University Press.
Vygotsky, L. (1978). Mind and society. Cambridge MA: Harvard University Press.
Winn, W.D., (1996, November). Learning virtually. Presented at the Annual Meeting of Northcon, Seattle WA.
Winn, W.D. (1994). Why I don't want to be an expert sitar player. Educational Technology, 34, (8), 11-14.
Winn, W.D. (1997, March). The Effect of Student Construction of Virtual Environments on the Performance of High- and Low-Ability Students. Paper presented at the annual meeting of the American Educational Research Association, Chicago, Il.