The magazine publisher is the copyright holder of this article and it is reproduced with permission. Further reproduction of this article in violation of the copyright is prohibited.

ABSTRACT
Technology and instruction have recently entered an alliance of reciprocal influences. Technology serves instruction and at the same time opens up novel opportunities. Concerning the former, a major justification for the employment of computers is the acceptance of constructivist conceptions and a growing understanding of learning as a social process. Technology thus comes to facilitate the realization of the learning environments that emanate from constructivist conceptions. Concerning technology's influence on education, ever-newer technological affordances pull instruction in new and promising directions. However, many of these lack purpose or rationale. Why, for example, should students design their own wWeb sites? New questions arise that need to be answered, such as whether hypermedia programs offer frail and casual webs of information that lead to the cultivation of similarly flimsy mental networks (the "Butterfly Defect"), or whether computer-mediated communication (CMC) might create virtual, faceless learning environments. It also becomes evident that the new learning environments rely more heavily than their predecessors on students' proclivity for self-regulated and mindful learning. Can technology facilitate the cultivation of these? Educational psychology and technology are now engaged in an intensive duet that, if seriously studied, explored, and evaluated, may offer novel and improved instruction.
    Technologies and prevailing psychological conceptions of learning, thinking, and instruction have always served and inspired each other in reciprocal ways. On the one hand, technologies in education have served to facilitate and realize the kinds of pedagogies that emanated from the changing zeitgeists and from prevailing psychological conceptions. On the other hand, and possibly only recently, technologies have been imported into education, challenging it and requiring novel psychological explanations and pedagogical justifications. Concerning technology as a means for the realization of pedagogies, think of the Skinner box, educational television, and LOGO programming. Each of these was inspired by a particular psychological conception of learning--conditioning, knowledge-transmission, and learning-as-problem solving, respectively. Little wonder that in its early days, the educational applications of computer technology for drill and practice and programming raised fears of dehumanization and mechanization as realizations of some kind of Skinnerian nightmare (e.g., Cuffaro, 1984). Concerning technology as a challenge, we might think, for example, of students' surfing the Internet and our need to find proper usages and justifications for that kind of previously unknown learning activity quickly.
    By now it is commonly understood that although the employment of technology in education does not lead to dehumanization, it cannot be justified in and of itself except on the basis of a nontechnological educational rationale. Such a rationale--psychologically or philosophically grounded--provides the conceptual underpinnings from which pedagogical implications and designs are derived. However, although technology is helpful, sometimes even essential, for the realization of these (e.g., Salomon & Perkins, 1996), top-down rationales do not translate unequivocally into particular pedagogical implications and designs. Thus, much is left undetermined and open for the novel technological possibilities that may be suggested or afforded in a more bottom-up fashion.
    The pedagogy that develops around students' opportunity to design their own Internet Web sites is a case in point. It does not derive from any superordinate conceptual rationale. In fact, such a technological possibility needs educational psychology for rationales and persuasive justifications. As technologies and educational usages develop, and particularly in recent years when those developments have outpaced developments of our psychological conceptions, technology comes to challenge educational psychology. And it challenges educational psychology by both reawakening old and partly dormant issues (such as transfer of learning or the roles of intentionality and mindfulness) and by demanding new conceptions and novel understandings of human behavior, learning, and instruction.
    In this article we wish to describe this reciprocity of relationships between recent educationally relevant psychological conceptions and educationally oriented usages of technologies. On the one hand, we wish to show how technology serves to realize psychologically guided pedagogical conceptions. On the other hand, we wish to illustrate the ways in which technology challenges educational psychology and some of the questions and concerns it raises. No claim for exhaustion is made; we merely illustrate here the reciprocal relations between technology and educational psychology.

TECHNOLOGY REALIZES PEDAGOGICAL CONCEPTIONS
    The appearance, often dramatic, of a novel technology on the educational stage raises high hopes for rapid and profound effects. This was the case with film, educational television, computer-based instruction, programming, intelligent tutoring systems, and--most recently--with hypermedia programs, the Internet, and computer-mediated communication (CMC). All these were expected to make a difference by their very introduction into an otherwise unchanged pedagogy. A paradox gradually became evident: The more a technology, and its usages, fits the prevailing educational philosophy and its pedagogical application, the more it is welcome and embraced, but the less of an effect it has. When some technology can be smoothly assimilated into existing educational practices without challenging them, its chances of stimulating a worthwhile change are very small. Domesticated technologies, such as educational television, do not affect life in the classroom much: The basic philosophy of one-program-fits-all and the commonly practiced pedagogy of top-down knowledge transmission remain unchanged; thus, the addition of this or that means of delivery cannot really "make a differenc." Much research on media in the classroom supports this conclusion (e.g., Salomon, 1994). Only so-called subversive technologies have a chance of stimulating a process of pedagogical change as they affect the whole classroom culture, practice and atmosphere. As has been correctly observed by Papert (1987), if the addition of a technology to instruction is so harmless that it can be easily assimilated into existing instructional practices without much changing them, then it will be equally harmless in making an instructional difference. Indeed, if the same kind of practice sheet is given via computers instead of in a booklet, why should this affect the comprehension of mathematics?
    Triggering a change for the sake of change is not a very convincing justification for imposing a new technology, shiny as it might be on education. The argument "because it is there and available" (in- and out-of-school) is often made nowadays to justify the adoption of novel technological offerings, yet this in itself produces nothing but the above-mentioned much-ado-about-nothing paradox and disappointment. Why, for example, should children in school surf the Internet? Why should they engage in computer-mediated communication with other students overseas? Or why should they design new data bases? The fact that such activities are possible is certainly not much of a justification and not much can be expected to result from them unless they come to serve a purpose beyond them. The paradox mentioned above can be avoided only when there is a justification that transcends the technology itself and provides a rationale for particular kinds of technology employment that strongly and justifiably deviate from pedagogical routine.
    Recent developments in educationally relevant psychological understandings of desirable learning, coupled with constructivist philosophies, appear to offer precisely the kind of rationale needed for intelligent and effective employment of technology in instruction (e.g., Brown, 1992). As we shall try to sketch out below, such developments suggest (although they do not dictate) particular designs for education. The realization of these designs is greatly helped, however, by the employment of technology. In fact, the coupling of new psychological and philosophical conceptions with technological possibilities has led to a shift in the kinds of scholarship in which many educational psychologists engage. Attention of many pioneering scholars in the field has shifted from the analytic study of single variables (e.g., anxiety, reading difficulties, intrinsic motivation) under relatively controlled conditions to the design and study of whole, composite instructional learning environments (e.g., Brown, 1992; Chinn & Anderson, 1998; Salomon, 1996). Such "design experiments," as they have come to be called, are what Herbert Simon once described as "the science of the artificial" (Simon, 1982). Specifically, this means that scholars attempt to weave psychological, instructional, curricular, interpersonal, and organizational considerations into new, workable, and effective learning environments manifesting a variety of constructivist approaches. Thus one finds the CSILE environment of Bereiter and Scardamalia (in press), Ann Brown and Joe Campione's Community of Learners (e.g., Brown & Campione, 1994), and the design of similar learning environments in which constructivist philosophy, new psychological understandings, and technology meet each other in what appears to be a rather promising new integration.
    Looking specifically at the psychological member of the threesome, there are at least three important developments that deserve special mention in this respect: Learning as a constructivist process, learning as (partly, at least) an interpersonal, often socially distributed, process, and human ability as (again, partly, at least) context-bound.

LEARNING AS A CONSTRUCTIVIST PROCESS
    Inspired to an important extent by Piaget's work and by philosophical perspectives, such as that of Von Glasersfeld (1990), there is a growing agreement among psychologists and educators that learning is essentially a process whereby learners construct their own knowledge by applying their existing knowledge and mental skills to novel incoming information, constructing their own meanings as they go along. The knowledge that students finally acquire is only the knowledge they have actively constructed themselves, not the information transmitted to them ready-made. Learning-as-construction thus contrasts with conceptions of learning as the relatively passive acquisition or internalization of ready-formed bodies of handed-down information (e.g., Phillips, 1995).
    One of the important underlying assumptions of this view of learning is that learning is not to be seen and assessed as the acquisition of knowledge, and, by implication, instruction is not to be conceived of as the well-structured, appealing presentation of information-to-be-acquired. Rather, learning is to be seen as the activities of constructing meanings and understanding within a particular context and situation (Duffy & Cunningham, 1996). What then would trigger these activities? Underlying the different constructivist approaches is the conception of discrepancies, conflicts, contradictions, unsolved problems, or in short Piaget's (1957) "disequilibria" and Bruner's (1960) "perturbations," which are sensed by learners (not necessarily the teachers!) and cause them intellectual uncertainty, annoyance, curiosity, or at least puzzlement.
    It follows from these conceptions, for example, that contrary to common lore, the acquisition of knowledge and the activities of construction (e.g., problem solving, designing) are inseparable (Perkins, 1992): One acquires knowledge while attempting to solve a problem or design something new, often as a result of experienced uncertainty or of a routine run aground. Instruction, then, is seen not as the effective transmission of knowledg,e but rather as setting the stage, providing some guidance, and offering the raw information for the activities of problem solving and design to take place.
    Related to the above, although not a necessary corollary, is the idea of comprehension or meaningfulness as the active construction of a network of connections between nodes of knowledge. No single bit of information can be meaningfully understood unless embedded in a rich network of relations. It is the causal, correlational, part-whole, rule-example, associational, or sequential links connecting a bit of information to others that give that bit its meaning. And, the denser, better organized, and less random the web of connections, the more meaning each part would have for the person. Following Geertz (1973), we might call such networks "Webs of Meaning" (Geertz called them "webs of significance"). Such webs are akin, to some extent, to the older notion of "cognitive maps," explicated by Neisser (1976) as "orienting schemata," active information-seeking and organizing structures. But whereas for many scholars, cognitive maps concern mainly spatial relations, cognitions, and orientations (e.g., Nadel, 1994), the webs of meaning alluded to here are far more general, pertaining to the way all kinds of information units relate to each other. Entwistle (1996) describes how students report the way they represent knowledge to themselves as precisely such "webs" of connections between nodes of knowledge. Given a constructivist view, the emphasis is on one's activities of constructing one's own weblike structures of knowledge rather than on the acquisition of a ready-made one. This, then, implies the necessity of engaging in activities intended to interrelate bits and pieces of knowledge such that a rich web can emerge. This, in turn, has strong implications for the social aspects of learning and for the interdisciplinary nature of the contents to be dealt with. To these we turn next.

LEARNING AS AN INTERPERSONAL PROCESS
    Traditionally, students have been perceived as isolated entities and their learning as a solo process. The interpersonal context in which learning takes place was usually ignored or, at best, seen as mere background, not really part of the actual learning process (e.g., Cole, 1991). Hence followed individualized instruction and--even more common--individual testing. But at least two sources challenged this view. One such source was Vygotsky's (1989) theory according to which development is, to a large extent, a matter of interpersonal interactions becoming internalized to serve as cognitive tools. A second, and clearly not unrelated source, was the growing conception of learning as situated rather than decontextualized (Brown, Collins, & Duguid, 1989; Greeno, 1989, 1997). According to this view, one's cognitions are so tightly connected to the situational context in which they are employed, to the specifics of the issue at hand, and to the activity one is engaged in that in-the-head cognitions and in-the-world activity should not be treated as separate entities (e.g., Lave & Wenger, 1991).
    Cognitions situated in the social context of some activity can be said to be distributed in the sense that the social processes entail the shared co-construction of knowledge. Learning is thus a socially distributed process of meaning appropriation (Newman, Griffin, & Cole, 1989). Whether it is the individual's solo learning that is facilitated by interpersonal processes, or whether the learning process and the resulting knowledge are both distributed, emerging "in between" the participants (Pea, 1993), much of the learning is due to the distributed mutual scaffolding afforded by the interpersonal activity.
    In light of these conceptions, learning becomes understood as a process for which social interaction serves a variety of crucial functions. These range from the provision of feedback and mirroring to mutual intellectual stimulation, instruction, and correction, and from mutual scaffolding of comprehension to the socially shared construction of meanings. Research on collaborative and cooperative learning generally tends to support such conceptions, showing that under certain conditions and with particular learning tasks, team work, collaboration, reciprocal teaching, and the like are beneficial for the learners (see, for a recent review, Hertz-Lazarowitz & Miller, 1992; Slavin, 1996).
    It might be argued that constructivist and interpersonal views of learning are somewhat contradictory: Constructivism assumes the dominance of in-the-head and transferable cognitions with activity being subservient to thought, while the interpersonal views of learning assume cognitions to be situated in particular activities, being socially distributed (Hewitt & Scardamalia, 1996). But this contradiction may be more apparent than real. For one thing, solo cognitions and distributed ones are likely to be interdependent, developing each other in a reciprocal spiral-like manner (Salomon, 1993; Salomon & Perkins, in press). For another, the ideas of learning-as-active-construction-of-knowledge and as a social process do not rule out each other. As a matter of fact, the newly designed "constructivist" learning environments, to be described below, that realize the new psychological conceptions succeed in effectively integrating the two views.

HUMAN ABILITY AS CONTEXT-BOUND
    Old conceptions according to which there is a clear distinction between content knowledge ("knowledge that") and abilities and skills ("knowledge how") have come under growing criticism. Although views and the findings on which they are based leave much room for disagreement, it has become increasingly common to view skills and abilities as less decontextualized than traditionally assumed. In the competition between a well-mastered abstract skill (say, a general problem-solving ability) and rich knowledge of a particular field (e.g., knowledge of soccer rules), the latter comes out the winner: it is better to know something than to be ignorant but equipped with a general, decontextualized skill (Glaser, 1990; Weinert & Helmke, 1995).
    For some researchers, general, decontextualized skills are of no interest. Thus, for example, Lave (1988) disagrees with the conventional views of abstract skill acquisition and transfer, arguing that "knowledge-in-practice, constituted in the settings of practice, is the locus of the most powerful knowledgeability of people in the lived-in world" (p. 14). Such views are supported by the relatively poor yield of research into transfer of training. If abilities are decontextualized, why does their deliberate training fail to transfer to new situations, contents, and contexts? Other support comes from research on the highly intelligent performance on the job of otherwise poorly educated individuals (e.g., Lave & Wenger, 1991).
    However, for other researchers, giving more credit to the long tradition of psychometric research into the nature of human intelligence and other relatively general abilities, the question is not an all-or-none one. Rather, the questions they ask concern the interplay between general skills and specific knowledge and the opportunities when either one of the two assumes dominance. For example, could it be that general abilities come into play only when specific knowledge is lacking? Or could it be that abilities serve as general levers for, say, problem solving, analogous to one's arms, whereas specific knowledge serves as the particular adaptations of such levers to the specifics of a situation, analogous to one's fingers (Perkins & Salomon, 1989)?
    Despite such difference of opinion, specific knowledge and activity within particular contexts have recently gained a more central role in the understanding of human learning and intellectual activity. Skills and abilities are to be cultivated within a variety of particular contexts, and if transfer is desired, it itself needs to be cultivated within particular situations. The generality of skills is not a given, and their cultivation cannot be a decontextualized educational activity. One major implication that follows is that learning is to take place within rich and complex real-world contexts, rather than with decontextualized skill-building materials. It also follows that learning is to take place through learners' active and personally consequential interactions with peers and within particular, content-rich contexts, rather than through the training of abstract subskills (Greeno, 1997).

PEDAGOGICAL IMPLICATIONS
    There is an impressive congruence among the three psychological perspectives discussed above: Good learning is a process of socially based, active co-construction of contextualized knowledge and webs of relations among its nodes. The translation of these underlying psychological conceptions and their pedagogical implications into classroom practices concerns the design of whole learning environments that integrate these implications. Indeed, a variety of novel, constructivist learning environments, although differing from each other in detail, entail a number of common elements and practices that manifest these psychological perspectives. These are team-based collaborations (communities of learners according to Brown & Campione, 1994; knowledge-building communities according to Scardamalia & Bereiter, 1994), whereby students tackle real life-like rich and often interdisciplinary problems, while engaging for a significant amount of time in intensive search for pertinent information, exchange of data, and the design of hypermedia and other knowledge products that can be co-constructed and shared, with the teachers serving as guides and consultants (see, also, Almog & Hertz-Lazarowitz, in press; Cognition and Technology Group at Vanderbilt, 1992; Salomon & Perkins, 1996). In these environments, teachers often work in teams, and learning becomes an interpersonal process whereby knowledge is jointly constructed, thus integrating the psychological conceptions discussed above.
    The actual practice of the new learning environments requires a number of major shifts--a conceptual and cultural shift from teacher-led instruction to an interactive community of active learners; from a highly structured curriculum to an emerging, often improvised one; from knowledge as the accumulation of discrete units to the tackling of whole issues; and from the acquisition of handed-down knowledge to the handling of information to be sought and processed (e.g., Brown, 1992).
    What roles does technology play in all this? The shifts just mentioned could not practically be carried out in real classrooms without technology that serves a number of functions. One would need tools that enable the gathering, processing, and construction of information, some of it imported from afar; one would need design tools; and one would need tools that facilitate these activities as social ones. Thus, the designs and usages that serve these functions are numerous and varied. In some cases, technology provides the forum in which the knowledge-oriented interaction and the explicit co-construction of that knowledge take place; in other cases, technology offers a series of tools with which information can be sought, sifted through, processed, and designed (Hewitt & Scardamalia, 1996). In the case of the Israeli project SELA, much of the learning process consists of teams of students jointly designing hypermedia products (in this case--planning the city of the future, avoiding current urban plights). The hypermedia, serves as the stage on which carefully selected and formulated information becomes woven into a gradually unfolding plot. In yet other cases, technology is the means of accessing worldwide information, distance communication with peers and experts, lab experimentation, and the like.
    In all, the proper harnessing of technology makes possible the practical realization of those learning environments that are based on the relatively novel psychological understandings of learning. Technology is thus subservient to pedagogy, with a conceptually-based pedagogy providing the rationale, and technology, the means. This division of labor is not coincidental. The development of psychological and pedagogical thinking, often associated with the "cognitive revolution," was inspired by developments in computer technology, while the directions taken by the latter were clearly inspired by developments of our understanding of (and debates about) learning and thinking.

TECHNOLOGY CHALLENGES PSYCHOLOGY
    Whereas, it is easy to see how technology manifests and realizes prevailing views of learning and teaching, the reciprocal--namely, the way technology in education inspires and challenges educationally related psychology--is less obvious. Technology, throughout history and in many fields, tended to become developed quite independently of economic, managerial, psychological, or educational considerations or, for that matter, independently of any top-down planning. Education, perhaps unlike some other fields, is an extreme case in point; it adopts existing technological innovations and adapts them to its prevailing conceptions, philosophies, and practices.
    However, as the development of information technologies is becoming increasingly rapid, outpacing developments of pedagogical thinking, and as these technological innovations engulf the lives of children and adults, psychological and pedagogical thinking cannot but attempt to catch up. Indeed, what psychological wisdom do we have to deal with children's surfing of the Internet, wandering through virtual MOO "rooms," or entering night-long chats with strangers?
    Let us consider three prototypical examples in which recent technological innovations, gradually penetrating education, afford new kinds of learning experiences, thus possibly challenging prevailing psychological conceptionss and common understandings: intellectual partnerships between technology and learners, the exploration and design of hypermedia, and computer-mediated communication. These cases illustrate how a new educational reality gradually becomes possible and takes shape, leading to relatively novel pedagogical practices and roles. Existing psychological constructs, theories, and understandings turn out to fall short of what is scientifically desired and practically needed. New technological affordances open up novel pedagogical possibilities that need fresh psychological explanations and justifications.

INTELLECTUAL PARTNERSHIP: EFFECTS "WITH" AND "OF" NEW TECHNOLOGY
    Intellectual partnerships with technology are manifested, for example, in writing partners that provide metacognitive-like guidance in computer-based laboratories, in interactive visualizations of difficult-to-imagine physical processes, in collaborative construction of novel knowledge by means of a specially designed interactive tool (e.g., Lajoie & Derry, 1993), and so forth. Typical of such tools is the distribution of cognitive activities between users and tools. On the one hand, nothing takes place without the active participation and intentional guidance provided by the users. On the other hand, the technology offers semi-intelligent help, affords the offloading of irrelevant menial cognitive tasks, provides intelligent feedback, or affords rational organization and clever ways of interrelating participants' input. Research and evaluation of such tools for partnership clearly show that the learning students experience, the kinds of activities they engage in, and the effects of these activities are diverse and profound (e.g., Jonassen, 1996). Such effects may range from greater team interdependence to more intensive metacognitive activity, and from increased motivation to higher levels of thinking.
    But not all effects are created equal, and thus the daily meaning of the term effect has come to mean two things: the effects attained during partnership with a semi-intelligent tool as contrasted with those that develop as a consequence of that partnership and appear later on as a changed capability. A new distinction had to be introduced, a distinction between effects with a technology and the more lasting effects of it (Salomon, Perkins, & Globerson, 1991). Whereas the former pertains to the immediate changes that take place while students engage in an intellectual partnership with a helpful tool (e.g., better essay writing when a tool provides metacognitive-like help), the latter pertains to the more lasting changes that this partnership may lead to (e.g., improved tendency for self-regulation, independent of that help; better essay-writing ability). This distinction raises a number of new questions. One of these is whether effects with a tool are a necessary condition for the attainment of effects of it. A second question concerns the role of mindful or intentional engagement in the intellectual partnership: Could lasting cognitive effects be attained while one is less than mindfully engaged in an intellectual partnership with a tool? A third question revisits an old dispute in education and psychology--the issue of how general (hence transferable) versus context-specific one's cognitive skills are (e.g., Anderson, Reder, & Simon, 1996; Greeno, 1997; Perkins & Salomon, 1989). The very mention of effects of technology, in the form of some decontextual cognitive residue that transcends the tool and the situation in which it was first attained, assumes some generalized transfer of learning to new situations and tasks. How does such logic square with the growing acceptance of knowledge and skill as being situated and contextualized? As pointed out earlier, it might well be the case (and there is evidence to support it) that skill acquired during activities with a tool can become generalized under conditions of varied and repeated experience, and become detached from the original situation or tool that facilitated its cultivation (e.g., Salomon, Globerson, & Guterman, 1989).
    The distinction between effects with and of is clearly relevant and useful to other educational interventions, such as new curricula or collaborative learning. In studying most educational interventions and innovations, effects attained during an intervention should not be confused with the intervention's long-range cognitive residues, although such a confusion is often the case. Why, then, was this distinction made only in the context of studying technology's educational effects? It seems that what has stimulated the making of the distinction is the gap between the observable novelty of students' activities during intellectual partnership with computers, on the one hand, and the paucity of lasting cognitive effects that were expected to emerge, on the other, but did not (see, e.g., Oppenheimer, 1997). Could it be that when it comes to the study and measurement of common achievements resulting from computer-related activities, the results are less impressive than the learning activities students are observed to egage in?
    Herein lies the challenge: If students engage in genuinely novel activities during intellectual partnership with technology, capitalizing on the novel technological affordances, what kinds of learning outcomes ought to be expected and measured? Are the desired learning outcomes (effects "of") to be identical to the ones aimed at before the new affordances were available? Or, to put it differently, should all those technological novelties (with the new communication and information-access possibilities they afford) come to serve no more than traditional educational goals crystallized in a much earlier era? Such questions echo to an extent the debate between the "cognitive" approach, espousing more traditional skill training (Anderson, Reder, & Simon, 1996) and the "situative" approach, espousing more social-participatory goals (Greeno, 1997). Although that debate is not directly linked to technology, it nevertheless reflects the challenges that novel technologies pose. What educational achievements should we try to attain and measure in the age of constructivist, socially shared, situative, technology-intensive learning environments?

HYPERMEDIA: THE CONSTRUCTION OF FRAIL KNOWLEDGE NETWORKS
    There is a potentially interesting affinity between cognitive networking that underlies comprehension and the network-like structure of hypertext and multimedia. As pointed out earlier, comprehension, to an important extent, is likely to be a matter of cognitive networking; that is, the construction of a network of relations between nodes of knowledge (Entwistle, 1996; Salomon & Perkins, 1996). Hypermedia and related genres appear to afford exactly that kind of web in the sense that they are constructed as networks of interrelated information items allowing free movement from one item to another, not necessarily adjacent, one. Thus, it appears that hypermedia programs reflect a mode of knowledge organization that could be isomorphous with, or correspond to the cognitive webs of meaning described earlier. The hypermedia students work with, even if only exploring an existing program, possibly serves as mirrors of their own minds--the network they explore, or, better yet, the one they design may well resemble the one they simultaneously construct in their own minds. Technology and mind work together; one may be mirroring and scaffolding the other.
    The distinction between effects with and of technology, described earlier in the context of intellectual partnership, is relevant in the present case as well. Mindful engagement in the design of a hypermedia product may afford the opportunity of higher-order thinking about the logical ways in which the particular information components of a body of knowledge relate to each other. Clearly, the effect attained with a hypermedia program, while working with it, might well be the construction of a better-organized and more meaningful domain-specific cognitive knowledge network (e.g., Lehrer, Erickson, & Connell, 1994). A more lasting cognitive effect of the active construction of hypermedia knowledge maps may well be an improved, relatively general ability, or greater disposition, for the construction of logical cognitive webs of meaning. In other words, students may become better able to construct for themselves interrelated networks of knowledge and acquire the disposition to think of ways to logically link bits of knowledge to each other.
    The requirement to define relations among components of a hypermedia network explicitly would be expected to be internalized, thus making hypermedia "cognitive tools" (e.g., Lehrer, 1993). Is this a viable possibility? Can interactive technological devices designed to handle information become internalized, to serve as cognitive tools the way, say, media's symbol systems (Salomon, 1994) or statistical tools (Gigerenzer, 1991) are claimed to serve? Research is badly needed to address such questions.

THE BUTTERFLY DEFECT
    However, not all of the potential effects with and of learning by means of multimedia and hypermedia are likely to be positive. There might also be a downside to hypermedia. One of the outstanding attributes of typical hypermedia programs, as well, as the Internet, is their nonlinear, association-based structure. One item just leads to another, and one is invited to wander from one item to another, lured by the visual appeal of the presentation. In fact, surfing the Internet or hypermedia programs is a good example of a shallow exploratory behavior, as distinguished from deeper search. The distinction between exploration and search was formulated by Wright and Vliestra (1975) to describe a developmental sequence. Whereas exploration is greatly influenced by visual appeal, a bottom-up unsystematic hopping around, search is focused, goal-directed, and metacognitively guided. Thus, developmentally, exploration precedes search and is replaced by it. Surfing the Internet and exploring hypermedia are activities that seem to faithfully manifest the exploratory behavior as conceived by Wright and Vliestra: a butterfly-like hovering from item to item without really touching them.
    One may need to entertain the hypothesis that based on the affinity between the network-like structure of comprehension and that of hypermedia products, intensive interaction with the latter might facilitate the construction of rather shallow associationist cognitive networks. Such networks would consist of trivial, frail connections, having no intellectual merit. One piece of information leads to another by virtue of some fleeting association without much rational justification, reflecting the aimless, visually-lured wandering though the screens of a hypermedia program, hence the Butterfly Defect.
    How would the cognitive Butterfly Defect be manifested? Although one should be careful not to entertain doomsday predictions of profound changes in the hard-wiring of students' minds, there is still reason to hypothesize that students' conceptions of what knowledge consists of might well be affected. Such conceptions are likely to be amenable to external influences. Thus, students might well come to believe that knowledge is a hypermedia-like structure, the links of which are not the kind of logical connections science is designed to construct. Similarly, students may come to prefer to learn from sources that present fields of knowledge in a hypermedia structure, thus sidestepping the acquisition of the logical, hierarchically structured connections and links that constitute science as we know it. To the extent that intensive interaction with unstructured, association-based tapestries of information can affect cognitive preferences or conceptions of knowledge, one might hypothesize that hypermedia-like minds are being cultivated by surfing Internet-like webs.

COMPUTER-MEDIATED COMMUNICATION (CMC): INTERNETED, FACELESS LEARNING?
    CMC is yet another technological innovation challenging educational psychology and pedagogy. Students can now communicate with each other over immense distances and in real time, consult vast data bases and libraries, surf the Internet, and gain easy access to an enormous variety of discussion groups, clubs, and institutions of learning. CMC and the Internet add interesting new dimensions to regular school learning: Students work in virtual teams, comparing freshly collected data and exchanging views on controversial topics; jointly write papers; and, perhaps most interestingly, conduct continuous conversations and construct whole projects (e.g., Hiltz, 1994).
    Such technological applications may appear to revolutionize not only classroom teaching, but also institutions of learning, by challenging the traditional foundations of schooling as we know it, particularly the foundations of institutionalized higher education. Suddenly the idea of a whole society becoming "A Learning Society," in which everybody has immediate access to whatever source of information one needs or desires, is not as fantastic as it was only a short while ago (Harasim, Hiltz, Teles, & Turoff, 1995). Remote participation in virtual and inexpensive university and high school courses and remote access to libraries, tutors, consultants, peers, experts, and numerous other sources becomes the hallmark of the new possibilities. As information can be accessed all the time and from everywhere, learning becomes removed from its traditional citadel-like locations of knowledge, freed from the constraints of place and time (Noam, 1996).
    All this calls for the formulation of new rationales to justify the novel possibilities: Why should virtual classrooms replace face-to-face ones? Why should students reach into virtual rather than real libraries or communicate with distant peers when they hardly ever communicate with those next door? On the other hand, what psychological constructs can account for learning via the new forms of CMC, Internet surfing, electronic communities of learners, and the like?
    As for rationales, recent designs of constructivist learning environments that emphasized the development of self-guided heavily interactive communities of learners make increasing use of the newly afforded communication possibilities. The need for interaction and communication, as well as the access to sources of information, provide the desired pedagogical and psychological justifications. As for the psychological processes that are of particular importance for CMC-intensive learning, it becomes evident, for example, that in the absence of a teacher, classroom regulations, and face-to-face contact, self-regulation may become truly crucial. So do student's disposition to be mindful learners. Individual differences that are only mildly implicated in learning in traditional classrooms are likely to become of central importance when CMC is involved. Indeed, in the absence of students' ability to monitor their own learning, or in the absence of sufficient motivation to become engaged in learning mindfully and intentionally, CMC-based pedagogy may be fun, but not very effective.

A NEW ALLIANCE: COGNITIONS NEED CONATION
    As pointed out earlier, although the introduction of novel information technology into classrooms was accompanied by high hopes for immediate breathtaking effects, computer technology was often employed in unimaginative ways, based on conceptions of learning as a matter of "spoon-feeding." However, with the development of increasingly interesting usages of technology for education, the gap between high hopes and the actual (disappointing) outcomes grew wider, echoing the realization that opportunities given are not necessarily opportunities taken (Perkins, 1985). Thus, for example, the opportunity to internalize a writing partner's metacognitive-like guidance is not always taken (Zellermayer, Salomon, Globerson, & Givon, 1991), nor are the opportunities to explore a rich multimedia program taken by all learners (e.g., Ross, 1996).
    Among the conditions that need to be met for such opportunities to be taken is mindful engagement in the activity afforded or demanded by the technology. This, in itself, is not new. The questions of intentional learning (Bereiter & Scardamalia, 1989) and of mindful engagement in learning (Langer, 1997) have been raised independently of any technological considerations, concurrently with a growing interest in the overlap between cognition and conation (e.g., Snow, Corno, & Jackson, 1997). Nevertheless, the design of novel, constructivist, and technology-intensive learning environments of the kind mentioned earlier highlights the role of self-regulated mindfulness in learning. In such environments, for their very constructivist nature, much of the responsibility for learning is shifted over to the learners, either individually or in teams, while teachers' control is relatively weaker than in more traditional learning environments. Indeed, one of our findings in a study comparing traditional classrooms with "constructivist" ones was that although measures of ability were the best predictors of learning in the former learning environment, measures of students' disposition to engage mindfully in learning are the best predictors in the latter. Volitional, motivated expenditure of mental effort, mindful engagement, and metacognitive self-monitoring thus become crucial in the kinds of learning environments that the novel information technologies require, not because of the technology in and of itself, but because of its conjunction with a more general constructivist pedagogy.
    The new constructivist learning environments rely more heavily on students' independent learning, increasingly carried out in classrooms without walls or schedules. However, it appears that at least two psychological forces may constrain the unlimited spread of distance learning in virtual classrooms. One such force is students' need for face-to-face, real (rather than virtual) contact. There is only so much distance learning and impersonal access to information that students are willing to tolerate. Thus, for example, Rheingold (1993), while describing the experience of learners in the San Francisco Bay Area electronic learning network (SFNET), observes that many of them often organize face-to-face parties to overcome the electronic lack of intimacy. Similarly, all courses of the Israeli Open University, which allege celebrating distance learning, had to add periodic, face-to-face class meetings to alleviate the loneliness of the long-distance learner. The second limiting factor is the difficulty many learners have with self-discipline and self-monitoring in the face of routine learning tasks. Students seem to need the boundary-setting, guidance, and motivation-sustaining functions that a regular classroom with its peer group and teacher usually provide. Despite technology's challenge, its possibilities are likely to be tamed by human frailty.

CONCLUSION
    Educational psychology and technology have reached a fruitful alliance under the umbrella of wider philosophical conceptions espousing constructivism. This alliance has a number of manifestations. The most pronounced manifestation is the design of novel learning environments that follow the new psychological understandings of what good learning (and hence instruction) is supposed to be, the realization of which greatly depends on technological affordances. Indeed, it would be most difficult to create the kind of team-based, interdisciplinary problem-solving and information-rich learning environment of the kind mentioned earlier in the absence of technology-enhanced search for relevant information, computerized lab simulations, data collection and analyses, semi-intelligent tools for design and presentation, communication, and the like.
    However, technology does not serve only as the lever for the realization of psychologically-based novel learning environments. Psychology and technology come to play a game of reciprocal influences. This reciprocity takes place when new technological affordances, resulting from the amazingly fast development of electronic technology, challenge education by offering it new tools, new usages, and new questions. Education must develop rationales and psychology needs to provide conceptual handles for these new affordances. Some aspects of these affordances look most promising--think of students communicating with their peers overseas to compare scientific data. But other aspects should perhaps worry us--think of the possible cognitive Butterfly Defect resulting from the ill-structured typical hypertext or of the poor value attributed to information when it comes without effort, without selection, and as a flood on the Internet.
    It becomes clear that unlike previous times, educational psychology and technology are now engaged in an ongoing duet.
ADDED MATERIAL
    GAVRIEL SALOMON AND TAMAR ALMOG
    The University of Haifa, Israel
    GAVRIEL SALOMON is dean of the faculty of education and professor of educational psychology, University of Haifa, Israel. He is the author of Interaction of Media, Cognition and Learning (LEA, 1994).
    TAMAR ALMOG teaches at the faculty of education, University of Haifa, Israel. She is the co-author, with R. Hertz-Lazarowitz, of "Teachers as Peer Learners: Professional Development in an Advanced Computer Learning Environment," in A. M. O'Donnell and A. King, Eds., Cognitive Perspectives on Peer Learning (Lawrence Erlbaum, in press).

REFERENCES
    Almog, T., & Hertz-Lazarowitz, R. (in press). Teachers as peer learners: Professional development in an advanced computer learning environment. In A. M. O'Donnell & A. King. (Eds.,) Cognitive perspectives on peer learning. Mahwah, NJ: Lawrence Erlbaum.
    Anderson, J. R., Reder, L. M., & Simon, H. A. (1996). Situated learning and education. Educational Researcher, 25(4), 5-11.
    Bereiter, C., & Scardamalia, M. (1989). Intentional learning as a goal of instruction. In L. B. Resnick (Ed.), Knowing, learning, and instruction: Essays in honor of Robert Glaser (pp. 361-392). Hillsdale, NJ: Lawrence Erlbaum.
    Bereiter, C., & Scardamalia, M. (in press). Rethinking learning. In D. R. Olson & N. Torrance (Eds.), Handbook of education and human development: New models of learning, teaching and schooling. Cambridge, MA: Basil Blackwell.
    Brown, A. L. (1992). Design experiments: Theoretical and methodological challenges in creating complex interventions in classroom settings. The Journal of the Learning Sciences, 2, 141-178.
    Brown, A. L., & Campione, J. C. (1994). Guided discovery in a community of learners. In K. McGilly (Ed.), Classroom lessons: Integrating cognitive theory and classroom practice (pp. 229-270). Cambridge: MIT Press/Bradford Books.
    Brown, J. S., Collins, A., & Duguid, P. (1989). Situated cognition and the culture of learning. Educational Researcher, 18(1), 32-42.
    Bruner, J. S. (1960). The process of education. Cambridge: Harvard University Press.
    Chinn, C. A., & Anderson, R. C. (1998). The structure of discussions intended to promote reasoning. Teachers College Record, 100(2): 315-368.
    Cognition and Technology Group at Vanderbilt. (1992). The Jasper series as an example of anchored instruction: Theory, program description, and assessment data. Educational Psychologist, 27, 291-235.
    Cole, M. (1991). Conclusion. In L. B. Resnick, J. M. Levine, & S. T. Teasley (Eds.), Perspectives on socially shared cognition (pp. 398-417). Washington, DC: American Psychological Association.
    Cuffaro, J. K. (1984). Microcomputers in education: Why is earlier better? Teachers College Record, 85, 559-568.
    Duffy, T. M., & Cunningham, D. J. (1996). Constructivism: Implications for the design and delivery of instruction. In D. H. Jonassen (Ed.), Handbook of research on educational communications and technology (pp. 14-39). New York: Scholastic.
    Entwistle, N. (1996, April). Conceptions of learning and understanding. Paper presented at the annual meeeting of the American Educational Research Association, New York.
    Geertz, C. (1973). The interpretation of cultures. New York: Basic Books.
    Gigerenzer, G. (1991). From tools to theories: A heuristic discovery in cognitive psychology. Psychological Review, 98, 254-267.
    Glaser, R. (1990). The reemergence of learning theory within instructional research. American Psychologist, 45, 29-39.
    Greeno, J. G. (1989). Perspectives on thinking. American Psychologist, 44, 134-141.
    Greeno, J. G. (1997). On claims that answer the wrong questions. Educational Researcher, 26(1), 5-17.
    Harasim, L. M., Hiltz, S. R., Teles, L., & Turoff, M. (1995). Learning networks: A field guide to teaching and learning online. Cambridge: MIT Press.
    Hertz-Lazarowitz, R., & Miller, N. (Eds.). (1992). Interaction in cooperative groups: The anatomy of group learning. New York: Cambridge University Press.
    Hewitt, J., & Scardamalia, M. (1996, April). Design principles for the support of distributed processes. Paper presented at the annual meeting of the American Educational Research Association, New Orleans.
    Hiltz, S. R. (1994). The virtual classroom. Norwood, NJ: Ablex Publishers.
    Jonnasen, D. H. (1996). Computers in the classroom: Mind tools for critical thinking. Englewood Cliffs, NJ: Merrill.
    Lajoie, S. P., & Derry, S. J. (Eds.). (1993). Computers as cognitive tools. Hillsdale, NJ, and Hove and London, England: Lawrence Erlbaum.
    Langer, E. J. (1997). The power of mindful learning. New York and Harlow, England: Addison-Wesley.
    Lave, J. (1988). Cognition in practice. New York: Cambridge University Press.
    Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. New York: Cambridge University Press.
    Lehrer, R. (1993). Authors of knowledge: Patterns of hypermedia design. In S. P. LaJoie & S. J. Derry (Eds.), Computers as cognitive tools (pp. 197-227). Hillsdale, NJ: Lawrence Erlbaum.
    Lehrer, R., Erickson, J., & Connell, T. (1994). Learning by designing hypermedia documents. Computers in the Schools, 10(1/2), 227-254.
    Nadel, L. (1994). Multiple memory systems: What and why, an update. In D. Schacter & E. Tulving (Eds.), Memory systems (pp. 39-63). Cambridge: MIT Press.
    Neisser, U. (1976). Cognition and reality. San Francisco: Freeman.
    Newman, D., Griffin, P., & Cole, M. (1989). The construction zone: Working for cognitive change in school. Cambridge, England, New York, and Melbourne: Cambridge University Press.
    Noam, E. M. (1996). Electronics and the dim future of the university. Science, 270, 247-249.
    Oppenheimer, T. (1997). The computer delusion. Atlantic Monthly, 280, 46-65.
    Papert, S. (1987). Computer criticism vs. technocentric thinking. Educational Researcher, 16(1), 22-30.
    Pea, R. D. (1993). Practices of distributed intelligence and design for education. In G. Salomon (Ed.), Distributed cognitions (pp. 47-87). New York: Cambridge University Press.
    Perkins, D. N. (1985). The fingertip effect: How information-processing technology shapes thinking. Educational Researcher, 14, 11-17.
    Perkins, D. N. (1992). Smart schools: From training memories to educating minds. New York: The Free Press.
    Perkins, D. N. & Salomon, G. (1989). Are cognitive skills context bound? Educational Researcher, 18, 16-25.
    Phillips, D. C. (1995). The good, the bad, and the ugly: The many faces of constructivism. Educational Researcher, 24(7), 5-12.
    Piaget, J. (1957). Logic and psychology. New York: Basic Books.
    Rheingold, H. (1993). The virtual community. Reading, MA: Addison-Wesley.
    Ross, J. H. (1996). The influence of computer communication skills on participation in a computer conference course. Journal of Educational Computing Research, 15(1), 37-52.
    Salomon, G. (1993). No distribution without individuals' cognition: A dynamic interactional view. In G. Salomon (Ed.), Distributed cognitions (pp. 111-138). New York: Cambridge University Press.
    Salomon, G. (1994). Interaction of media, cognition, and learning. Hillsdale, NJ and Hove, England: Lawrence Erlbaum.
    Salomon, G. (1996). Unorthodox thoughts on the nature and mission of educational psychology. Educational Psychology Review, 8, 397-417.
    Salomon, G., Globerson, T., & Guteman, E. (1989). The computer as a zone of proximal development: Internalizing reading-related metacognitions from a reading partner. Journal of Educational Psychology, 81, 620-627.
    Salomon, G., & Perkins, D. N. (1996). Learning in wonderland: What computers really offer education. In S. Kerr (Ed.), Technology and the future of education (pp. 111-130). NSSE Yearbook. Chicago: University of Chicago Press.
    Salomon, G., & Perkins, D. N. (in press). Individual and social aspects of learning. Review of Research in Education.
    Salomon, G., Perkins, D. N., & Globerson, T. (1991). Partners in cognition: Extending human intelligence with intelligent technologies. Educational Researcher, 20, 2-9.
    Scardamalia, K., & Bereiter, C. (1994). Computer support for knowledge-building communities (abstract). Journal of the Learning Sciences, 3(3), 265-283.
    Simon, A. H. (1982). The sciences of the artificial. London: MIT press.
    Slavin, R. E. (1996). Cooperative learning: Theory, research and practice. Englewood Cliffs. NJ: Prentice-Hall.
    Snow, R. E., Corno, L., & Jackson, D., III (1997). Individual differences in affective and cognitive functions. In D. C. Berliner & R.. C. Calfee (Eds.), The handbook of educational psychology (pp. 243-310). New York: Macmillan.
    von Glaserfeld, E. (1995). Radical constructivism: A way of knowing and learning. London: Falmer Press.
    Vygotsky, L. (1989). Thought and language (A. Kaszulin, ed. and trans.). Cambridge, MA, and London, England: The MIT Press.
    Weinert, F. A., & Helmke, A. (1995). Interclassroom differences in instructional quality and interindividual differences in cognitive development. Educational Psychologist, 30(1), 15-20.
    Wright, J. C., & Vliestra, A. C. (1975). The development of selective attention: From perceptual exploration to logical search. In H. W. Reese (Ed.), Advances in child development and behavior, vol. 10. New York: Academic Press.
    Zellermayer, M., Salomon, G., Globerson, T., & Givon, H. (1991). Enhancing writing related metacognitions through a computerized Writing-Partner. American Educational Research Journal, 28, 373-391.
    DIRECTIONS, B, ORAL
    I.E.R. Intelligence Scale C A V D (Completions, Arithmetic, Vocabulary, Directions) Levels A, B, C, D and E for grades K through 2. The directions for the above questions are the following: "7. Make two crosses, like these two. Make one here and one here, pointing to the spots below the two crosses ...)" (see above). "8. Make the other arm on this man (pointing to the one-armed man ...)" (see above). "9. Make the other leg on this man (pointing to the one legged man ...)" (see above). "10. Make 2 lines, like these two (pointing to the two lines...)" (see above).
    The test to which these questions belong had been created by The Institute of Educational Research, Teachers College, Columbia University, 1926.
    The property of Milbank Library, Special Collections, Teachers College, Columbia University.