Full Paper

Mark McMahon

m.mcmahon@cowan.edu.au

School of Multimedia & Learning Technologies

Edith Cowan University

The growth of the Internet and the Word Wide Web, in particular, are attracting the attention of tertiary educational institutions worldwide. This is manifest in the increasing number of distance education courses being offered in this medium (University of Texas, 1997; Pagram & McMahon, 1997). It is significantly less expensive to produce materials electronically than in printed form, and the material may easily be kept up to date (Eklund, Garrett, Ryan, & Harvey, 1996). These reasons, combined with the cost savings of a 'virtual campus' in real estate and contact time for the university, are leading to the Web being seen as an effective alternative to traditional face to face modes of education. It has been argued that students do not like to learn at a distance (Simonsen, 1995), but the convenience and flexibility of an external mode of delivery for those with busy life styles is making distance education an attractive proposition for students (Truman, 1995).

Caution is required to ensure that these financial imperatives do not dominate the push for Web based learning. The proliferation of research which finds "no significant effect" for technology still raises concerns (Russell, 1997). The Web and the Internet itself is, after all, another in a long procession of technologies which offer much but whose promise not always fulfilled; and the rabid enthusiasm of many Internet proponents is tempered by the jaded cynicism of others. For every Nicholas Negroponte espousing the Internet as "humankind's best chance to respect and nurture the most obscure languages and cultures of the world" (Negroponte, 1996) there is a Clifford Stoll, presenting the Net as a chimera of unfulfilled promise, which actually works against literacy and creativity rather than promoting them (Stoll, 1995).

There is little doubt that the Web is a significantly different medium to CD-ROM based Interactive Multimedia (IMM). While some argue that the Web is becoming a strong multimedia platform (Shotsberger, 1996), slow response times often make such environments impractical. In essence, the Web remains true to its initial objective of being a means of linking documents across a diverse network (Berners-Lee, 1989), and this raises concerns over the level of interactivity and engagement that can be supported. While there is no doubt that the potential of the Web as a global resource of information can have a strong potential for learning, it is worth being mindful of the fact that the Web does not ensure learning any more than a library on a university campus does (Reeves, 1996). Any approach to Web based learning must be guided by assumptions of what is to be learned and how learning itself comes about.

I have argued elsewhere for the need to find Web learning solutions that are explicitly grounded in theory, since learning strategies are informed by specific epistemological assumptions (Ring & McMahon, 1997). At the risk of oversimplifying a complex issue, much learning can be defined within the parameters of one or other of the three broad theoretical approaches of Behaviourism, Cognitive Theory, and Constructivism.

Behaviourism argues that learning takes place through a mechanism of stimulus and response - a convenient approach since both the stimulus and response are manifest and therefore measurable, and offer an empirical legitimacy to the 'soft' science of education. The operant conditioning of Skinner (1974), with its focus on unpleasant and pleasant consequences (reinforcement) as a means of shaping behaviour is perhaps the best known educational application of behaviourism, and has lead to the development of tangible guidelines for learning strategies such as a focus on incremental learning (Slavin, 1991) and the need for consequences to be intermittent and timely (Slavin, 1991; Langford, 1989).

These tenets have been rapidly adopted into models for instructional design and can easily be adapted to the Web. Skinner himself advocated the use of teaching machines to provide modularised learning based upon concrete behavioural objectives (Kratochwill & Bijou, 1987). Behaviourism does appear to be limited, however, in the types of learning it supports:

The origin of the decline in adherence to behaviourism was not that classical conditioning, operant conditioning and imitation do not exist but that behaviourists made the mistake of thinking that these three learning processes could explain all learning. (Langford, 1989, p. 4)

Since it relies almost exclusively on observable behaviour and does not account for individual thought processes, the roll of behaviourism in learning is necessarily limited to the types of learning which can be easily observed such as factual recall, rather than less clearly defined learning which involves internal conceptual change within the learner.

In A Study of Thinking, Bruner (1956) provided a strong argument for why behaviourist theories have fallen from grace in favour of those that acknowledge the role of the individual in mediating learning, claiming:

It has resulted from a recognition of the complex processes that mediate between the classical 'stimuli' and 'responses' out of which stimulus-response learning theories hoped to fashion a psychology that would by-pass anything smacking of the 'mental.' This impeccable peripheralism could not last for long. (p. vii)

More recent developments in understanding of how the brain processes information have become influential to the extent that techniques of encoding and retrieval from memory have become integral to most models of Instructional Design. Gagné's Events of Instruction (Gagné, Briggs & Wager, 1998), for example, has a strong focus oon cognitive aspects such as stimulating recall of prior learning and enhancing learner retention, as well as behaviourist sequences of presenting stimuli and eliciting performance.

Such approaches can be easily supported by the Web. The implementation of Common Gateway Interface Forms as well as Shockwave (Macromedia) and Java (Sun Microsystems) applications as well as commercial applications available such as Web Test (University of Waterloo, 1996) greatly assist with the development of tutorials and assessment within such a paradigm However, while the Web in its current form is a good delivery medium for finite knowledge and can incorporate strategies to aid encoding and retention, the goals of cognitive learning are often broader, and incorporate skills such as problem solving and critical thinking. It is when these educational outcomes are confronted that the Web becomes severely challenged.

Constructivism goes beyond the study of how the brain stores and retrieves information to examine the ways in which learners make meaning from experience. Rather than the transmission of knowledge, learning is an internal process of interpretation:

Learners do not transfer knowledge from the external world into their memories; rather, they create interpretations of the world based upon their past experiences and their interactions in the world. How someone construes the world, their existing metaphors, is at least as powerful a factor influencing what is learned as any characteristic of that world (Cunningham, 1992, p. 36)

While Constructivism does not necessarily deny the existence of an objective reality, it does deny the existence of an objective knowledge since "there are many ways to structure the world, and there are many meanings or perspectives for any event or concept." Thus, "there is not a correct meaning that we are striving for" (Duffy & Jonassen, 1992). It is this rejection of absolutism that characterises constructivist approaches to learning, and it is a radical ontological departure from the previous theories discussed.

Most cognitive theory, and the constructivist approaches that have grown out of it, argue that learning should be durable, transferable and self-regulated (Di Vesta and Rieber, 1987 cited by Hannafin & Rieber, 1989). Mechanisms need to be in place to promote the deeper internal processing required for such learning to occur. The high level of interaction required for such processing, however, clearly demands more of the Web than merely being a delivery vehicle for information. While some view the Web itself as a cognitive tool for investigating and representing knowledge (Reeves & Reeves, 1997) and as a semantic knowledge space which will mirror learners' own developing cognitive structures (Lambert & Walker, 1996), attempts to create specific Web sites which are constructivist in nature are rare. While high fidelity simulations and microworlds are now common place with CD-ROM based Multimedia, bandwidth issues conspire with the primitive nature of HTML to create pages that are usually flat and lacking in the interactivity required within a constructivist approach.

In many respects the Web is an ideal forum for constructivist learning, and despite its limitations, HTML does offer some interesting opportunities. Hypertext links work by association rather than indexing and it could be argued that this "free association" can be disorientating. Yet, the counter argument that it operates much like the way humans think (Gygi, 1990) suggests intriguing possibilities for the meaningful linking of data required for the information processing within a cognitive framework.

Cognitive Constructivism, as derived from the work of Piaget (1977) defines learning as a process of accommodation, assimilation, and equilibration (Piaget, 1977). This is a "dialectic process in which the subject resolves perturbations in the coherence of his or her structuring activities by coordinating and constructing new, more adequate cognitive structures" (Saxe, 1991). One complementary approach, Cognitive Flexibility Theory (Spiro, 1995), may be particularly informative. This theory argues for multiple representations of content where knowledge is highly interconnected and complex (unsimplified) (Archee & Duin, 1995). The potential of the Web to present a variety of information sources may help to stimulate the cognitive conflict required within a Piagetian approach. This theory, though, also calls for cased based authentic learning, and does not provide specific strategies for how engagement with the disparate complex information that the Web offers can be ensured.

It is here that Social Constructivism may offer some hope. Pioneered by theorists such as Vygotsky (1978), this paradigm argues for the importance of culture and context in forming understanding. Learning is not a purely internal process, nor is it a passive shaping of behaviours. Vygotsky favoured a concept of learning as a social construct which is mediated by language via social discourse.

While Piaget did account for the social transmission of knowledge (Langford, 1989), "the interplay between social life and cognitive development processes was not a core concern", his focus instead being on "the formal properties of action without regard for the situatedness of actions in a sociohistorically articulated web of meanings" (Saxe, 1991, p. 6). Traditional behaviourist/instructivist approaches strive for context independence, whereas a Social Constructivist paradigm views the context in which the learning occurs as central to the learning itself.

Underlying the notion of the learner as an active processor is "the assumption that there is no one set of generalized learning laws with each law applying to all domains" (Di Vesta, 1987, p. 208). Decontextualised knowledge does not give us the skills to apply our understandings to authentic tasks because "we are not working with the concept in the complex environment, experiencing (exploring, evaluating) the complex interrelationships in that environment that determine how and when the concept is used" (Duffy & Jonassen, 1992). One Social Constructivist notion is that of authentic or "situated learning", where the student takes part in activities which are directly relevant to the application of learning and which take place within a culture similar to the applied setting (Brown, Collins, & Duguid, 1989). Cognitive Apprenticeship has been proposed as an effective constructivist model of learning which attempts to "enculturate students into authentic practices through activity and social interaction in a way similar to that evident -- and evidently successful -- in craft apprenticeship" (Ackerman, 1996, p. 25).

Reeves claims that "most existing examples of WBI [Web Based Instruction] employ academic tasks, but WBI can be designed to focus on authentic tasks relevant to learners" (Reeves & Reeves, 1997). Yet this assertion is yet to be demonstrated with research and existing examples on the Web. Some sites, such as Virginia University's Interactive Frog Dissection (Kinzie, 1994), use video to provide realistic representations of content, but interactivity is limited to "hotspots" which judge responses and dictate navigation through the tutorial. Also, video is slow to download and does not account for authentic activity rather than merely authentic representation.

In highly concrete knowledge domains this is a very real problem. A trade such as Plumbing, for example, requires skills which involve enactment on a physical environment. The difficulty the Web has in creating such authentic simulations appears to limit its capacity to accommodate this. Even more ill-structured academic areas such as History and Philosophy require a level of context, even if that context is the academic one requiring the implementation of a culturally specific methodology and use of language. It is here, perhaps, that a Social Constructivist approach may be particularly useful. It could be argued that the use of the Web is best suited to that of a communications medium for collaborative approaches to learning rather than as a "24 hour a day glorified whiteboard" (Archee & Duin, 1995). Such a use would involve a high level of social rather than physical interaction; an aspect well supported by the Web and integral to a Social Constructivist approach.

Going so far as to reject the 'botanical' and 'zoological' models of physiologists such as Piaget that see maturation as a passive biologically fixed process, Vygotsky favoured of a concept of learning as a social construct which is mediated by language via social discourse. In our complex use of language, humans are unique and it has become the primary enabling tool of learning:

The most significant moment in the course of intellectual development, which gives birth to the purely human forms of practical and abstract intelligence, occurs when speech and practical activity, two previously completely independent lines of development, converge. (Vygotsky, 1978, p. 24)

Language and communication become the principle focus, and if one is to believe the claim that "throughout most of their lives people learn and work collaboratively, not individually" (Resnick, 1988, cited by Brown et al, 1989), then a framework based upon co-operative learning and social negotiation may be useful. While such an approach does not deny a Piagetian perspective, it also contains a major strength in that it can help to form learning is appropriate to the culture in which it is to be implemented (Tudge, 1990).

Laurillard emphasises learning as an iterative process, involving discursive, adaptive, interactive, and reflexive qualities, the main focus being on teacher-student relationship since "academic knowledge consists in descriptions of the world, and therefore comes to be known through a discursive interaction between teacher and student" (Laurillard, 1993, p. 89). Other studies, too, argue for the importance of mentoring in the process of learning (Archee & Duin, 1995; Brown et al., 1989).

Most Social Constructivist models, such as that proposed by Jonassen (1994), however, also stress the need for collaboration among learners, in direct contradiction to traditional competitive approaches. One Vygotskian notion, that has significant implications for peer collaboration, is that of the 'Zone of Proximal Development.' Defined as "the distance between the actual developmental level as determined by independent problem solving and the level of potential development as determined through problem solving under adult guidance or in collaboration with more capable peers" (Vygotsky, 1978), it differs from the fixed biological nature of Piaget's stages of development. Through a process of 'scaffolding' a learner can be extended beyond the limitations of physical maturation to the extent that the "the development process lags behind the learning process" (Vygotsky, 1978).

This has significant implications for the Web as a communications medium. While it may not be highly interactive in a physical sense, the Web has strong potential for social interactivity. The goal of this type of approach is the achievement of 'virtual communities' of learners on the Internet working in small collaborative groups to achieve a common goal (Dillenbourg & Schneider, 1995). While it would appear that care is necessary in the formation of such groups, it has been proposed that heterogenous grouping can assist in the creation of zones of proximal development (Walker & Lambert, 1996).

Traditional Internet communication tools such as E-mail, Newsgroups, Internet Relay Chat, and MOOs offer both the rapid synchronous communication of normal speech as well as asynchronous interaction which may help to promote a more reflective metacongitive approach. With the use of Web browser plug-ins and server software such as Ichat, such facilities are now becoming available in a more cohesive form on the Web. Examples of learning through communication can be seen in commercial environments such as TopClass (WBT Systems, 1997) which have no actual content but provide the functionality required for real-time communication and collaborative learning.

It would be convenient to see Social Constructivism as a single solution to the limitations of the Web, but no one theoretical approach is likely to achieve the broad range of educational outcomes required from tertiary study. Both Cognitive Theory and Constructivism are not without their critics, who are often damning in their observations of what are, after all, philosophies much less open to the rigorous scientific testing inherent in a behaviourist paradigm. Epistemologically, the relativism of Constructivism is particularly contentious. Constructivism focuses on the individual interpretation of a perceived external reality, and it has been claimed that "individual understanding and conceptualisation is parasitic upon this extra-individual scientific domain", leading to the criticism that Constructivism is "old unpalatable, empiricist wine in a new bottle" (Matthews, 1992)

One of the most tangible criticisms of Social Constructivism is the type of learning it supports. While it may be true that social negotiation is a useful approach to achieving consensual understanding of ill-structured subject matter, even in the 'softest' subjects there is often a body of undisputed knowledge. Constructivist strategies are often not efficient, resulting in "a trial-and-error approach to the performance in the real world" (Merrill, 1997).

Explicit guidelines for the use of real-time chat and threaded discussion lists have yet to be developed and there is little doubt that the unique attributes of on-line social interaction will play a large part in the effectiveness of the Web as a collaborative medium. Recent experiences with the use of Listservs at Edith Cowan University have produced mixed results. When a clear structure through prescribed activities is given to the interaction which takes place, students have anecdotally expressed positive outcomes from the experience. The discomfort which some students initially experience in posting their ideas to the 'ether' of cyberspace, as well as low participation rates where social interaction is seen as an adjunct rather than an integral component of study, however, suggest more research in this area is needed.

Finally, it must be noted that the Web, and Internet in general are likely to change significantly in coming years. Increasing bandwidth and processing power will make activities such as video conferencing and real time visual manipulation of data across large distances a reality. This will undoubtedly impact greatly on on-line educational practice. In the mean time, however, where the goals of instruction are broader than can be easily accounted within traditional approaches, it may be that instructional strategies informed by a Social Constructivist paradigm can help to lessen the inadequacies for highly interactive authentic learning which are manifest in the Web in its current form.

(c) Mark McMahon

The author(s) assign to ASCILITE and educational and non-profit institutions a non-exclusive licence to use this document for personal use and in courses of instruction provided that the article is used in full and this copyright statement is reproduced. The author(s) also grant a non-exclusive licence to ASCILITE to publish this document in full on the World Wide Web and on CD-ROM and in printed form with the ASCILITE 97 conference papers, and for the documents to be published on mirrors on the World Wide Web. Any other usage is prohibited without the express permission of the authors.