Developments in the fields of psychology of learning, technology, and philosophy, have recently had a growing love affair with each other. Novel conceptions about learning have become coupled with constructivist philosophies, leading to the design of novel pedagogies. Technology became a useful partner when it was invited to serve as the means for the practical realization of those novel pedagogies. The results of this menage a trois are interesting novel learning environments, such as Scardamalia and Bereiter's CSILE (1996), Brown and Joe Campione's Communities of Learners (e.g., Brown, 1994), Anchored Instruction and other designs of the Cognition and Technology Group at Vanderbilt (1996), or our SELA project in Israel (Salomon & Almog, in press). These and others like them are learning environments which are based on constructivist psychological and philosophical principles, team-based, often interdisciplinary, oriented towards the solution of complex, real-life problems, and utilizing a variety of technological means.

Having reached relative maturity, it may be time to raise a few questions about such learning environments. The purpose of raising these questions is not to undermine them, but to better understand their potentials and limitation, and to serve as safeguards against the lure of pedagogical and technological fads and fashions. And since the design and practice of the novel learning environments greatly depends on the employment of novel technologies for their realization (e.g., Salomon & Almog, in press), the issues to be raised pertain to both the learning environments and the technology. This close alliance is further enhanced by technology's rapid developments, providing challengingly novel tools and ways of accessing and structuring rivers of information such as the hypermedia and the Internet. Technology thus comes to shape, not just enable the design of constructivist learning environments.

The issues to be addressed here are as follows: (a) The nature of the learning outcomes expected from novel constructivist learning environments; (b) the kind of mental organizations of knowledge students are likely to prefer, even to adopt as a consequence of constructing; and (c) the possible effects of navigating the information avalanche of the internet on the construction of knowledge.

THE ISSUE OF LEARNING OUTCOMES

Two versions of constructivism
Since the construct of constructivism has various meanings various communities, it is beyond the scope of a single paper to attempt to describe, let alone define it. Apparently, true to the spirit of constructivism, most everybody constructs his or her own particular meaning for it. Nevertheless, there appears to be a core of common understandings of constructivism in the area of learning and instruction (e.g., Phillips, 1995, p. 5). Knowledge is believed to be actively constructed, tightly connected to the individual's cognitive repertoire and to the context within which this activity takes place, hence it is situated. Accordingly, the learner is more a maker than a spectator. And, to quote Von Glasersfeld (1990), knowledge comes to serve one's organization of the experiential world, not the discovery of an objective ontological reality with its universal Truth value. Rorty (1991) has put it very succinctly: Knowledge is in the constructive process rather than a finding; it is not the content stored in memory but the activity of constructing it. In short, "children don't get ideas; they make ideas" (Kafai & Resnick, 1996, p.1).

But constructivism is in fact a two headed creature. While it may have initially been strongly influenced by a Piagetian cognitive developmental approach, emphasizing the individual and the way he or she constructs knowledge, a socio-cultural competitor has developed besides it. The socio-cultural approach, influenced by the Soviet school of thought of Vygotsky, Leont'ev, and Bakhtin (Cole, 1991; Rogoff, 1991; Wertsch, Pablo, & Alvarex, 1995), is concerned mainly with the social process of interaction and participation, the socially-based appropriation of meaning, that is—with the way the situative social system, not just the individual, interactively operates to construct shared knowledge (Greeno, 1997; Sfard, 1998).

This "situative" approach with its emphasis on the here and now of the operating system differs epistemologically in important ways from the cognitivist-individually oriented one. Most importantly, as Hewitt, Scardamalia, and Webb (1997) point out, the shift of attention from the cognitivist to the situative approach changes the meaning of "knowledge" and of "learning" from pertaining to individual attainments to being contextually tied, distributed activities of participation.

Relatedly, the more traditional constructivist view emphasizes learning activities designed to result in the cultivation of individuals' skill and knowledge. Students should be capable of walking away with new capabilities which they can then apply elsewhere. The socio-cultural approach, on the other hand, emphasizes the social process of participation and changes that take place while students are engaged in collaboration, problem solving, and team-based design activities. Thus, whereas the traditional approach strives to attain solo effects of (resulting from) constructivist learning activities, the socio-cultural approach aims at effects with them (Salomon, Perkins, & Globerson, 1991). The difference is akin to that between being able to play the piano more rhythmically with the aid of a metronome, vs. one's improved generalized sense of rhythm as a consequence of such experiences.

The two approaches in reciprocal relationship
The emphasis on effects "with", that is - on the systemic, participatory nature of the interactions in a learning environment, and the ways socially shared knowledge is constructed, is quite congruent with the situative branch of constructivism. Greeno (1997), a most eloquent spokesperson for the situative approach, raises the question of the desired goals of education, as follows: "Should we consider the major goals and outcomes of learning as primarily collections of sub-skills or as successful participation in socially organized activity and the development of students' identities as learners?" (p. 9). In addition, he recommends to focus on "the contribution of learning activities to the learners' development of greater efficacy in their participation in valued social practices and to the development of their identities as capable and responsible learners" (1997, p. 9).

Such a focus is an essential element for the understanding of a constructivist learning environment and its improvement, but it is not clear whether this is all one really wants from a constructivist learning environment to yield. Thus, we'd need to ask what else are the students supposed to learn? And this implies what is characteristically meant by the more cognitive-developmental branch of constructivism: Individual attainments that can be carried away and utilized in new situations as partly abstracted, decontextualized knowledge and skill.

It becomes an interesting challenge to interrelate the situative approach of Rogoff (1991), Pea (1993), and Greeno (1997) with the more cognitively oriented one as already hinted at by Cobb (1994) and Sfard (1998) and further elaborated upon by Salomon and Perkins (in press). In this light, it seems that the principle of reciprocal relations and spiral developments (e.g., Bandura 1989; Weick, 1979) appropriately accounts for the relationships between the cognitive and situative approaches to constructivism. Indeed, as Damon (1991) has forcefully argued -

[E]ven when learning is fostered through processes of social communication, individual activity and reflection still play a critical role. Sometimes... individual activity may build on collective questions and insights. Other times, however, individual activity actually may need to resist the collective illusions created by a group... Any paradigm that assumes a one-way, deterministic relation between the collective and individual knowledge construction is overly simplistic.(p. 392)

It is easy to see that the two views of constructivism are not only complementary, but they represent two sides of an ongoing dynamic process of reciprocal influences. Thus, in a constructivist learning environment we have a cycle of acquiring that which is socially constructed as a shared meaning or a shared artifact, affecting a child's cognitive repertoire, which is then used to affect that child's social participation. In other words, the idea of reciprocal relations suggests the acceptance of both approaches to constructivism, implying that we need to look at both the participatory effects "with" as well as solo effects "of" such learning environments—the way they work together and the way they affect each other reciprocally.

What learning outcomes?
But even if one is to grant reciprocal relations between the cognitive and the socio-cultural approaches to constructivist learning environments, one is still left with the question of what learning outcomes—with the learning environments and of them are we to expect and evaluate? Perhaps with the exception of a few projects, most often in mathematics (e.g., Cobb, 1997; Schwartz, et al. 1995), we tend to fall back on what Petraglia (1998) has described as the domesticated version of constructivism: The application of a constructivist pedagogy to attain traditional goals. In such cases, and they are the majority, the changed instructional practices are just a changed means to attain old ends, or, as Petraglia puts it, "a theoretical mascot to be trotted out to boost intellectual morale, but ensuring that it is rendered incapable of posing a real challenge to the basic educational framework with which we are most comfortable" (p. 5). One common way out of this dilemma is to temporarily ignore the relatively radical philosophical underpinnings of constructivism, both cognitivist and situative, and limit constructivism to its practical delivery and technological aspects.

The issue of learning outcomes can be easily answered where only domesticated constructivism is practiced. For better or worth, domesticated constructivism is a prime example of a wasted opportunity. All those changes of the social interaction in the classroom, of curricula and rules of conduct, of teacher training and teaching practices, of the disciplinary boundaries and most importantly—of the learning activities themselves, just to attain the same old learning outcomes? But not all novel learning environments are of this domesticated variant; what about learning in its less tame version?

Different means for different ends
Different means are best designed to serve different ends, and the more one changes the means, the less sense would it make to adhere to old ends better served by older means. Radical changes of pedagogy—turning a didactic classroom into a constructivist one, entailing teams, interdisciplinary problems to be solved cooperatively, technology, and all, should urge us to think of new criteria and goals which are more suitable for the new pedagogy. It is beyond the scope of this paper to suggest a list of such criteria, but a few of the new learning environments designers have already started to do that. Thus, for example, Scardamalia et al. (1992) have started to design and use nontraditional outcome measures such as graphical literacy, coherence of knowledge, and level of construction activity. Schwartz el. al, (1995) in their Harvard based project of Balanced Assessment in Mathematics Project have devised novel measures such as mathematical estimation and problem-solving, and we, in our SELA project, have taken measures of students' changed conceptions of what learning and knowledge are all about.

It might be argued that one of the most important and interesting outcomes of constructivist learning environments might be students' improved ability to work in a team to solve new, complex and ill-structured real-life like problems, showing their coordinated abilities to access information and turn it into viable knowledge. Knowledge, then, would not be something possessed for its own sake, but rather something accessed and constructed when needed to solve a problem or design something useful. While this, then, emphasizes the utilitarian and situative side of knowledge, it also acknowledges the possibility of some individual generalizable cognitive residue, such as one's improved ability to work in a team, to access needed information, to formulate an intelligent question, or to redefine the focus of a futile search.

What is suggested here is not just a list of new final exams, but rather an outline of a conception of knowledge: Knowledge as a social construction in the dual sense of being a verb denoting a process as well as being a noun denoting an object the learner comes to possess. The constructive-process aspect of this conception emanates from the socio-cultural partner of the constructivist duo, while the constructed object aspect emanates from the cognitive partner. One would have difficulties studying one without the other. For example, it would be difficult to design or to study a new constructivist learning environment without paying attention to outcomes such as the emergence of shared understandings, jointly developed modes of working together and standard operating procedures. It would be equally unsatisfactory to study social participatory processes without the examination of solo learning products to which that participation may have led. In this light, the idea of reciprocal relations between the two conceptions of constructivist and the effects with and of that they imply seems pretty natural, even trivial.

TECHNOLOGY AND MIND

Technology in instruction rarely functions as a silent, unfelt and indifferent servant of constructivist pedagogy; it affects learning and thinking on its own right, and might be doing it in more profound ways than we are willing to admit. Technology, as employed for the realization of a constructivist learning environment can affect minds in a number of principled ways. One of these is the opportunity that it affords of actively constructing knowledge in particular symbolic forms (word, graph, picture), and structured in particular organizational ways (databases, hypermedia), available for exploration and manipulation.

Not just being exposed to the symbol systems of technology and the media, but being actively involved in construction, either individually or collaboratively, makes this possibility a most powerful one. As it has already been shown (Lehrer, et. al., in press), it is one thing to cruise a hypermedia program; it is an entirely different thing to actually construct it. And constructing multimedia programs is of course precisely what a constructivist approach would recommend.

Activities such as constructing a hypermedia or multimedia product are of particular interest here because exposure to the web-like, non-linear organization of information of hypermedia may be isomorphous to the way information becomes organized in one's mind. Such isomorphism can be derived from Vygotsky's (1981) theory of social tools and symbols that become internalized to serve in a cognitive function. Related arguments have been made about the possible affinity between the symbol systems of the media and the arts and the mind's symbolic modes of representation (e.g., Gardner, 1983; Salomon, 1979/1994). On this basis, a similar affinity can be suggested between the non-linear, web-like nature of hypermedia and the web-like cognitive organization of information.

Webs of meaning and hypermedia
Thus, whether we speak of stable, static representations of knowledge or ones that become constructed anew when needed, the web-like structures of knowledge provides meaning. Meaning, among other things, is a function of the density, complexity and organization of such a cognitive web. The denser and better organized the web, the more does it function as a web of meaning. Indeed, Entwistle (1995) has studied and found that students report organizing their knowledge in a field in a web-like manner he called "knowledge objects", webs that become denser as a course of study progresses (Jones and Vesilind, 1996). It is worth noticing how similar such webs are to the way information in hypermedia is organized, hence accessed.

The cognitive webs one would like students to construct in science classes are ones that entail logical connections: Cause and effect, temporal sequences, correlations, rule-and-instance, part-whole, and the like. The more logical the constructed web of meaning, the better it can be reconstructed when the occasion arises. Thus, dates in history, names or single formulae, are in fact pretty meaningless (inert) and therefore cannot be well remembered, precisely because they freely float unconnected to anything in one's cognitive space, much like satellites in outer space.

Given the possible affinity between hypermedia webs and cognitive ones, is it possible that the hypermedia programs that students cruise, particularly the ones they construct in typical constructivist learning environments, serve as external intellectual tools that facilitate the construction of their cognitive parallels? Would the cognitive web students construct be reflected in their computerized product, or vise versa—would the cognitive web reflect the external product? There is not enough research to date to offer a clear answer, but the little evidence there is tends to support this possibility (e.g., Lehrer, et. al., in press; Winnie, 1998). If this is a possibility when students cruise a hypermedia program the effects ought to be far stronger when they actively construct it. Through this activity of construction students affect the way their technological products are likely to affect them. One might go one step further. If hypermedia cruising or constructing may facilitate the weaving of students' cognitive webs, as already shown by Barnes (1994), might it not also influence students' subsequent way of organizing information in their cognitive webs of meaning?

The Butterfly Defect
This raises yet other questions. Hypermedia programs of the kind widely touted and widely used in education are non-linear, perhaps the way cognitive webs of meaning are. However, the connections they display, and particularly the ones students build into them, are anything but logical. In fact, such programs are deliberately based on casual associations and on visual fascination, luring the user to wander from one item to another which happens to be associated with it. In fact, this is not just a private case of hypermedia and multimedia; it is the defining attribute of the hottest thing in town: The Internet.

There is nothing wrong with bouncing around, as hypermedia and the Internet invite one to do, except that this is typical of bottom-up, unguided exploratory behavior, as contrasted with the developmentally more advanced search behavior which is top-down, metacognitively guided and goal directed (Wright & Vliestra, 1975). Search, unlike exploration, is neither guided by the lure of shiny buttons, nor does the finding of simple associations satisfy it. If students can emulate the organization of information in hypermedia for the organization of knowledge in their minds, matching their maps of meaning to those they construct on the computer, would they not organize it in the same associationistic way hypermedia are organized?

The questions thus concern two interrelated developments. One development concerns the structure of students' webs of meaning. Could students' cognitive webs of meaning come to reflect hypermedia characteristics, consisting of flimsy associationistic connections? The second development concerns the mental activity associated with those webs: Would students come to acquire a tendency to mentally hop around their own cognitive webs in a hypermedia-like manner?

These possibilities can be called The Butterfly Defect: Coming to think or to prefer the style of thinking in a hypermedia mode—"touch, but don't touch, and just move on to make something out of it". A teacher recently interviewed by Oppenheimer for a biting article about educational computing in The Atlantic Monthly (1997), proudly announced that his students have come to think in a multimedia manner. If he is right, then the danger of a Butterfly Defect may be more imminent than we think.

TECHNOLOGY: FROM REALIZATION TO PROVOCATION

As already mentioned in the introduction, technology's rapid development affords entirely new usages the pedagogical rationale for which is quite unclear. Suddenly, and perhaps for the first time in human history, education has at its disposal novel and very tempting tools and engines without, in many cases, having instructional rationales and psychological underpinnings to justify their employment. For the first time, education has to chase technology down the classroom aisles and the Internet channels to tailor old pedagogical rationales to the new possibilities or to make up new ones. Why indeed would anybody want students to learn by surfing the Internet or by exploring MOO rooms? And what purpose does the design of students' home page come to serve? And so technology, serving as a force that helps realize psychological conceptions about learning and instruction, has come also to serve as a provocateur—offering new possibilities that require new conceptions, new rationales, and new modes of usages.

The glut of information
The above pertains particularly to the much lauded opportunities afforded by the Internet. At first glance it appears to afford everything a constructivist approach would wish to have: Multiple sources of information to draw from, a whole information world to explore, and an invitation to become an active, fully participating member of the larger, virtual, and diverse information society. The problem, though, is that it is too much of a good thing. Too many sources, too much information, too many communication links, and too much excitement relative to the meager pedagogical rationale to justify the process of surfing this abundance for the purposes of learning.

The Internet offers a seductive flood of information. This is where the rule of the tool plays out: For the educator with the Internet, the whole world looks like a well of infinite knowledge. But does a flood of randomly accessed information aid in the construction of knowledge? Not very likely. Fascinated by the information highway, and quite unknowingly, we are drawn back to greatly esteem factual information, assuming that it is the major source of knowledge.

Herbert Simon is alleged to have argued a few years ago that the concept of knowledge will change its meaning from being a noun denoting possession to being a verb denoting access: It won't be that important what knowledge a student possesses relative to the knowledge he or she is capable of accessing. But access to avalanches of information, loosely inter-connected by threads of casual associations, does not facilitate turning the information into knowledge. And it is school's mission, particularly when colored by constructivist hues, to enable students to think and to acquire skills of intelligently handling information, not to drown them in it.

We are now facing the danger that technology will come to pull instruction by its nose, redefining the nature of learning environments and turning the ideas of constructivism—the active and thoughtful construction of knowledge—into the active but thoughtless compilation of raw information. It is as if technology might take charge, demanding of constructivist philosophy and of the psychology of learning and instruction to follow suit and to adjust themselves to the novel technological affordances. However, as Sarason (1984) has already warned us, not everything technologically possible, wondrous as it might be, needs to automatically also become instructionally desirable.

On a more positive note, the new possibilities should indeed provoke us to rethink instruction and redefine the nature of school. The point to be made is not that we ought to stick with the familiar, but rather that we, the researchers and designers of constructivist learning environments, should beware of attempts to lead us into a technological promised land without checking the promises and the stability of that land.

REFERENCES

Bandura, A. (1989). Human agency in social cognitive theory. American Psychologist, 44, 1175-1184.

Barnes, W.G. (1994, April). Constructing knowledge from an ill-structured domain: Testing a multimedia Hamlet. Paper presented at the Meeting of the AERA, New Orleans.

Brown, A.L. (1994). The advancement of learning. Educational Researcher, 23(8), 4-12.

Cobb, P. (1994). Where is the mind? Constructivist and sociocultural perspectives on mathematical development. Educational Researcher, 23(7), 13-20.

Cobb, P., (1997). Learning from distributed theories of intelligence. In E. Pehkonen, (Ed.), Proceedings of the 21st Conference of the International Group for the Psychology of Mathematics Education, 2, (pp. 169-176). Lahti, Finland: University of Helsinki.

Cognition and Technology Group at Vanderbilt (1996).Multimedia environments for enhancing learning in mathematics. In S. Vosniadou, E. de Corte, R. Glaser & H.G. Mandl (Eds.), International perspectives on the design of technoloyg - Supported learning environments. (pp. 285-305). Mahwah, NJ: Lawrence Erlbaum Associates.

Cole, M. (1991). Conclusion. In L.B. Resnick, J.M. Levine, & S.D. Teasley (Eds.), Perspectives on socially shared cognition (pp. 398-417). Washington, DC: American Psychological Association.

Damon, W. (1991). Problems of direction in socially shared cognition. In L.B. Resnick, J.M. Levine, & S.D. Teasley (Eds.), Perspectives on socially shared cognition (pp. 384-397). Washington, DC: American Psychological Association.

Entwistle, N. (1995). Frameworks for understanding as experienced in essay writing and in preparing for examinations. Educational Psychologist, 30, 47-54.

Gardner, H. (1983). Frames of mind. New York: Basic Books, Inc.

Greeno, J.G. (1997). Response: On claims that answer the wrong question. Educational Researcher, 20(1), 5-17.

Hewitt, J., Scardamalia, M., & Webb, J. (1997, April). Situative design issues for interactive learning environments: The problem of group coherence. Paper presented at the Annual Meeting of the AERA, Chicago.

Jones, M.G., & Vesilind, E.M. (1996). Putting practice into theory: Changes in the organization of pre-service teachers' pedagogical knowledge. American Educational Research Journal, 33(1), 91-117.

Kafai, Y.B., & Resnick, M. (1996). Introduction. In Y. Kafai & M. Resnick (Eds.), Constructionism in practice—Designing, thinking and learning in a digital world (pp. 1-8). Mahway, NJ: Lawrence Erlbaum Associates.\

Lehrer, R., Erickson, J., Love, M., & Connell, T. (in press). Learning by designing hypermedia documents. Computers in the schools, 9(2/3).

Oppenheimer, T. (1997). The computer delusion. Atlantic Monthly, 280, 45-65.

Pea, R. D. (1993). Practices of distributed intelligence and designs for education. In G. Salomon,. (Ed.), Distributed cognitions—Psychological and educational considerations (pp. 47-87). Cambridge, UK; New York; Melbourne: Cambridge University Press.

Petraglia, J. (1998). Reality by design: Rhetoric, technology, and the creation of authentic learning environments. New Jersey: Lawrence Erlbaum Associates Inc.

Philips, D. C. (1995). The good, the bad, and the ugly: The many faces of constructivism. Educational Researcher, 24(7), 5-12.

Rogoff, B. (1991). Social interaction as apprenticeship in thinking: Guided participation in spatial planning. In L.B. Resnick, J.M. Levine, & S.D. Teasley (Eds.), Perspectives on socially shared cognition, (pp. 349-364). Washington, DC: American Psychological Association.

Rorty, R. (1991). Objectivity, relativism, and truth. New York: Cambridge University Press.

Salomon, G. (1979). Interaction of media, cognition and learning. San Francisco, CA: Jossey-Bass. (Published again by Lawrence Erlbaum, 1994).

Salomon, G., & Almog, T. (in press). Educational psychology and technology: A matter of reciprocal relations. Teachers' College Record.

Salomon, G., & Perkins, D.N. (in press). Individual and social aspects of learning. Review of reciprocal relations in Education.

Salomon, G., Perkins, D. N., & Globerson, T. (1991). Partners in cognition: Extending human intelligence with intelligent technologies. Educational Researcher, 20, 2-9.

Sarason, S.B. (1984). If it can be studied or developed, should it be? American Psychologist, 39, 477-485.

Scardemalia, M., & Bereiter, C. (1996). Adaptation and understanding: A case for new cultures of schooling. In S. Vosniadou, E. de Corte, R. Glaser & H. Mandl (Eds.), International perspectives on the design of technology—Supported learning environments. (pp. 149-163). Mahwah, NJ: Lawrence Erlbaum Associates.

Scardamalia, M., Bereiter, C., Brett, C., Burtis, P.J., Calhoun, C., & Smith Lea, N. (1992). Educational applications of a networked communal database. Interactive Learning Environments, 2, 45-71.

Schwartz, J.L., et al. (1995). Assessing mathematical understanding and skills effectively. Interim Report of the Harvard University Graduate School of Education: Harvard, MA.

Sfard, A. (1998, March). On two metaphors for learning and the dangers of choosing just one. Educational Researcher.

von Glasersfeld, E. (1990). An exposition of constructivism: Why some like it radical. In R.B. Davis, C.A. Maher, & N. Noddings (Eds.), Constructivist views on the teaching and learning of mathematics, (Monograph 4), (pp. 19-29). Journal for Research in Mathematics Education.

Vygotsky, L.S. (1981). The instrumental method in psychology. In J.V. Wertsch (Ed.), The concept of activity in Soviet psychology. Armonk, NY: M.E. Sharpe.

Weick, K. (1979). The social psychology of organizing (2nd edition), Reading, MA; Menlo Park, CA; London; Amsterdam; Don Mills, Ontario; Sydney: Addison-Wesley Publishing Co.

Wertsch, J.V., Pablo, D.R, & Alvarez, A. (Eds.). (1995). Sociocultural studies of mind. Cambridge, UK; New York; Melbourne; Cambridge University Press.

Winnie, P. (1998 , August). The allure of new technologies in modeling learners' minds. Symposium paper to be presented at the Meeting of the IAAP, San Francisco.

Wright, J.C., & Vliestra, A.G., (1975). The development of selective attention: From perceptual exploration to logical search. In H.W. Reese (Ed.), Advances in child development and behavior, 10, New York: Academic Press.