Hypermedia and Cognitive Mapping

Abstract:  The discussion of the impact of media on learning has shifted from an examination of the media attributes to the active and supportive role of hypermedia technologies as cognitive learning tools.  In particular, the literature on the mapping of knowledge structures to hypermedia is reviewed and found generally to support cognitive mapping, especially when learners become hypermedia authors.  Conversely, cognitive mapping from hypermedia to existing human knowledge structures is proposed by the author to afford a coherent cognitive and semantic explanation for hypermedia-based learning.  This is based on his experiences with student-authored hypermedia production in the 1990's and through research papers supporting the writer's thesis.  It is suggested that further intensive research in this field might reveal additional evidence of cognitive mapping that mirrors the semantic structure of hypermedia projects.

Introduction

There has been much debate about the impact of media on learning in the literature (Clark, 1983; Kozma, 1994).  This debate has been characterized by a discussion on the effect of different media upon the learning process.  Clark, using a grocery van as an effective metaphor, asked which has the most impact on nutrition (learning); the van (media) or the groceries (materials and pedagogy)?  Kozma, on the other hand, cited studies that reported noticeable increases in student learning attributed to hypermedia.  Researchers such as Mayes (1994) and Jonassen (1994) have consistently suggested that learning with computer technology, especially "putting authoring tools into the hands of students" is likely to lead to a deeper understanding of the material.  More recently, Liao (1998) suggested the "effects of using hypermedia in instruction are positive compared to the effects of traditional instruction", additionally providing some evidence to contradict Clark's thesis.

This debate has focussed primarily on the characteristics and attributes of the media.  Increasingly, Jonassen et al (1994) and others have shifted the focus to the role of media in "supporting, not controlling the learning process."    They argued that the focus should shift to how media can facilitate the construction of knowledge by the learner.  In their view, designers should “examine the process of learning first, then the role of context and the kinds of environments and cognitive tools needed to support that learning” (p. 38).   Cognitive tools may include semantic networks, concept maps and multimedia construction (Jonassen and Mara, 1999).  Additionally, a supportive effect hinted by researchers (Burbules and Callister, 1996); Kayama and Okamoto, 2001)  lies in the way hypermedia appears to enhance the learning process by mirroring explicit structured media to existing cognitive structures.   This paper seeks to establish a link between  hypermedia and cognitive tools such as multimedia construction and cognitive maps through a thorough review of published papers. 

The author's experiences with the use of hypermedia in a variety of educational settings have identified possible causal relationships between a number of variables which will be examined in the literature.  In particular, the structure of learner-created multimedia projects, actively conducted in the early 1990's, may assist with cognitive mapping whereby the semantic structure of the project is mirrored in the mind of the learner. 

Answers will be sought to the following questions:

Hypermedia is a portmanteau which combines the words hypertext and multimedia.  Typically, multimedia, hypertext and hypermedia are terms that have been used to denote similar, even parallel concepts.  In this paper, I shall use the word "hypermedia".

Hypermedia is defined in E-commerce's Webguide as the use of computers to present text, graphics, video, animation, and sound in an integrated way.  While many readers may assume that hypermedia is a modern phenomenon associated with the World Wide Web, it may surprise many that this notion was first described by Vannevar Bush in 1945 and extended by Ted Nelson who coined the term "hypertext" to describe a non-linear form of writing "text that branches and allows choices to the reader."

In many ways, different media have been employed as teaching aids for a long time - most of us went to schools where text, graphics, television, film and video were used to assist student learning.   Apart from some integrated social studies and language packs that contained workbooks and audio tapes, the media were not integrated in any purposeful way.  The collision of new powerful personal computers in the 1990's capable of displaying integrated multimedia and the Internet changed the educational information technology landscape forever.  This had an interesting and unforeseen side-effect as the technology appeared to drive the teaching and learning paradigms in classrooms around the globe.   During the last decade of the 20th century, researchers began to describe the phenomenon and to tentatively explore the causative link between hypermedia and learning - David Jonassen has probably travelled this road further than most others.

The most powerful characteristic of hypermedia is its ability to link different media types in the same or other documents.  This ability to present and link different types of information in a variety of modes highlights three features of hypermedia:

These fundamental changes to the way users access and learn information may be summed up by the word, connectivity. 

A distinction between the two main types of hypermedia production is usually made.  Much of what is described as hypermedia comes to us as consumers in the form of professionally manufactured CD-ROM's or Web-based hypermedia, either containing reference material (for example, Microsoft's Encarta), or a commercial game such as Tomb Raider.  In educational settings, hypermedia production by students in schools and universities, using authoring software such as HyperCard, HyperStudio and Toolbook, form the second type of authored multimedia.   In this context, authoring multimedia projects give students an opportunity not only to integrate different media, but also, to afford the establishment of hyperlinks between the media.  Authoring in a hypermedia environment is used increasingly in educational settings (Gant, 1998; Bates, 2000). 

Liaw (2001) offers a detailed description of the types of hypermedia-based learning environments.   Not surprisingly, these  environments are developed on constructivist learning theory in that they support non-linear access to information, interactive communication, and integration.  Other learning theories that have been shown to support these learning environments are dual coding (Butler and Mautz, 1996, in Liaw, 2001), multiple representations and cognitive flexibility (Spiro, Feltovich, Jacobson and Coulson, 1991).   The advantages of hypermedia-based learning over traditional learning methodologies are summarized by Liaw (2001) as affording multiple perspectives, collaborative learning, learner-orientation and interdisciplinary learning.

Learners typically create or author these digital or Web-based hypermedia projects as part of or central to a theme currently taught in the classroom.  They incorporate collaborative learning which is manifested in small groups of two or three students working together.  This collaboration can be extended to the classroom level where small groups author parts of a composite class project.  Projects can be open- or close-ended in terms of content and deadlines, but they usually share the common feature of "the grand presentation" in front of the whole class, a time where the group can showcase their project and answer questions.

The authoring tools as mentioned above provide the interface in which media are "pasted" on to digital pages and then linked.  The project structure, in particular, the linkages between the different media and nodes, offers learners a unique opportunity to construct their world as "they see it".  Hypermedia environments tend to encourage the "creation of structure" rather than a certain number of pages of words (Pohl et al., 1995).  Learners can explicitly explore, experiment and develop the semantic and cognitive links of their pasted media and through this act of creation, it has been suggested they can mirror the semantic structure of their hypermedia project on to existing cognitive structures in their minds (Mayes, 1994).   Because of their explicit semantic structures, hypermedia projects can be regarded as cognitive tools (Jonassen and Mara, 1999).  How this "mirroring" works is examined in the following section.

The theories of learning are as abundant as they appear contradictory (Kearsley, 1997).  However, through the study of human cognition, it is now understood that "learning flows from understanding" (Mayes, 1994).  Mayes (1994) states that hypermedia has the potential to be useful if it promotes better understanding.  According to Mayes, "learning is an inescapable by-product of comprehension" ( p.5). 

Cognitive psychologists sort knowledge into two categories: declarative and procedural (Nichols, 2000).   Declarative knowledge consists of descriptions of facts and things or of methods and procedures while procedural knowledge is knowledge that manifests itself in the doing of something or about how to do something.  Jonassen and Mara (1999) suggest that declarative knowledge is "not a sufficient prerequisite" for procedural knowledge and insist that in order to acquire procedural knowledge, the "propositional relationship between the entities" in declarative and procedural knowledge structures must be understood.  They believe structural knowledge, the "understanding of these relationships", provides this link and assists learners to form connections between description of facts and actually "doing it".   Mayes (1992) recognized that deep mental processing can only be reached by doing things.

Where does this structural knowledge reside?  Taylor (1994) elaborates a model of the human information processing system that is based on the analysis of cognitive structures.  Most cognitive psychologists agree to at least three kinds of memory: executive, short-term and long-term.  The long-term memory is solely concerned with the content and organization of memory structures and processes.   Popper (1972, in Taylor, 1994) described knowledge associated with these memory structures as objective knowledge and equated it with theoretical systems, problem situations, critical arguments and the contents of journals and books.  From a learning perspective, most instructional systems "aim to embed such....objective knowledge in the minds of students" (Taylor, 1994).

As early as 1982, Glaser suggested that the psychology of instruction should attempt to understand the development of memory or cognitive structures and processes (Glaser, 1982, in Taylor, 1994).  One of the dimensions of objective knowledge is relational knowledge which is characterized by the cognitive processing of common features shared by elements or events at input (Hunt and Einstein, 1981, in Taylor, 1994).  Relational or structural knowledge is the knowledge of how concepts within a domain are interrelated (Diekhoff, 1983 in Jonassen & Mara, 1999).  The underlying assumption of structural knowledge is that meaning for any concept is implicit in the pattern of relationships to other concepts (Jonassen & Mara, 1999).  Further, Elio and Anderson (1981, in Taylor, 1994) provide evidence that supports a "schema abstraction model" in which information transfer is a function of similarity.

Much of the research concerned with uni-directional cognitive mapping focusses on mapping the semantic network of an expert or knowledgeable person onto the structure of a hypertext (Burbules and Callister, 1996); Kayama and Okamoto, 2001; Jonassen, 1992).   Meaningful learning results when learners relate new information to their existing knowledge in order to accommodate it with what they know already.  In this paper, I offer the possibility of the reverse scenario;  that is, mapping the structure of the hypermedia on to the cognitive structure of the learner.  Through studies such as Jonassen and Wang's (1992), structural knowledge acquisition has been shown to improve significantly when learners are assigned the task of generating a semantic network.

Hypermedia systems may be particularly appropriate as learning environments because the associative knowledge structures of a topic or subject matter are made explicit (Jones and Baxter, 1999).    Learning and understanding operate by making connections.  We come to understand something when we can bring it into association with other things we already know.  The process of actively selecting and assimilating new information meshes the potential of hypermedia with constructivist learning theories, especially schema theory.

A schema consist of a network of interrelations among its constituent parts, which themselves are other schemata (Wang, 1999).  The schema theory of Rumelhart and Ortony (1977, in Jonassen and Wang, 1992) claims that knowledge is stored in information packets or schemas.  Each schema we construct represents a mini-framework in which to interrelate attributes of information about a topic into a single conceptual unit.   These mini-frameworks are organized by the individual into a larger network of interrelated constructs known as a semantic network. These networks are composed of nodes: representations of schemas.

It is the semantic structure of the media to which some writers (Kearsley, 1988; Jonassen, 1988) point to the potential of hypermedia to reveal and mirror the cognitive processing and mapping occurring in the mind of the learner.  Kearsley (1988) and Jonassen (1988) are quoted in Alexander (1996) as supporting the notion that hypermedia  mirrors human cognition.  Kearsley states that "writers have emphasized that hypertext matches human cognition; in particular, the organisation of memory as a semantic network in which concepts are linked together by associations."  Jonassen goes to say "because hypertext is a node-link system based on semantic structures, hypermedia can map fairly directly the structure of knowledge representing it." 

Semantic networks are transferable: as a result of instruction, learners' knowledge structures more closely resemble the instructor's knowledge structure.  So, learners are acquiring two things during instruction:  knowledge and the knowledge structures that mimic the teacher's knowledge structure (Jonassen and Wang, 1993) - it is therefore quite feasible that this flow of knowledge is bi-directional especially when learners have created their own semantic map, the hypermedia project (Jonassen and Mara, 1999, p.4).

Schemata can integrate new data into its organizational structure because of their ability to respond to the existence of relevant data.  The schema theory assumes there is a parallel between human memory and hypermedia - they are both networks in which concepts are all arranged in a network of interrelated concepts known as a semantic network.  The associative nature (linking) of hypermedia is similar to the functions of the human memory so presenting or constructing information in a way that resembles human memory may enable people to learn better.

Schema have received significant empirical support from studies in psycholinguistics. For example, the experiments of Bransford and Franks (1971) involved showing people pictures and asking them questions about what the story depicted; people would remember different details depending upon the nature of the picture.   Research on novice versus expert performance (for example, Chi et al., 1988, in Hassebrock and Winters,1995) suggests that the nature of expertise is largely due to the possession of schemas that guide perception and problem-solving.

Jonassen and Marra (1999) offer further support in their paper on concept maps that "address the critical issue of how we organize our knowledge structures..."  Concept maps are a part of a larger group of cognitive tools which they describe as "mindtools" in that "students cannot use these tools without thinking deeply about the content that they are learning."  The process of mapping the explicit structures (hypermedia) to the cognitive and semantic map within the learner is described in constructivist nomenclature as internal negotiation, a process involving the learner linking nodes (concepts), establishing the relationship between nodes, and resolving any conflicts with existing knowledge structures.  As Alexander (1996) concludes, "by apprehending structure, integrating parts, acting on the world and using reflection, learners develop an individual interpretation of reality."  In other words, what writers and researchers seem to be saying is that what matters most in learning is the construction of personally relevant knowledge structures.

Beginning in the late 1980's, the author began to experiment with hypermedia tools in a small country primary school in New Zealand (Hatherley-Greene, 1993).   After several years' of experimentation, the author developed a pedagogy centred on small collaborative groups working together on largely open-ended hypermedia projects focussed around a humanities, science or English language theme or topic.  It quickly became apparent that students, working with technology, changed into highly motivated, creative and innovative learners.  Their entire learning interface consisted of technology - computers, electronic keyboards, scanners, video digitisers, video cameras, recorders and monitors.  The new setup, whereby the teacher retreated from the "stage", afforded the author with the opportunity to observe individuals and groups.

Individuals, using a workshopping technique, were assigned "contracts" or tasks to complete for the group.   Workshopping involved group leaders, in discussion with the author, assigning task responsibilities to individual students - tasks were regularly reviewed. This assignment prevented much of the aimless time-wasting previously observed in group project work. The author often noted individuals spending much longer times concentrating on the task, with much of their free time additionally being spent on task work - they seemed very motivated and energized.

The group interaction provided further evidence of a new "focus on learning". The group often appeared to "grow", feeding upon the synergy created by highly motivated individuals. Questioning became more reflective, "why" and "how" began to dominate the more common "who, what, where and when" questions.  Discussions often grew heated as individuals sought to justify their placement and linking of media to others who strongly questioned the structural components of the project - the main focus was usually on the links and the overall structure of the project.  Interestingly, structure was not perceived as hierarchial, but more like an open three-dimensional learning landscape full of infinite linking possibilities (Hatherley-Greene, 1999).

The results of this experiment, lasting over three years, were not quantitatively measured. However, the quality of the digital projects, the obvious rise in personal self-esteem and the continual feedback, even after ten years, bear testament to a learning experience that continues to have a profound effect on the students. That learning took place is beyond doubt - that technology and the author's teaching technique assisted in that learning are unproven, but highly suggested. As most of their work consisted of hypermedia projects, it is reasonable to assume that cognitive mapping may have played an important role in their learning. Much of the group discussion around a computer centred on the structure and the links between individual media items - "why is this a better place to put this picture?" and "why did you link that picture to that word?" were the kind of questions often heard.

The author was able to implement similar learning environments at both secondary and tertiary levels.  The total learning experience was, however, lessened due to inflexible timetabling, traditional teaching cultures and external examinations.  At the secondary school, synergistic and spontaneous activities resulted from a term's work with hypermedia - the students organized a class stall at the school fair where digital photos were captured, edited and printed using computers.  In my present setting, a tertiary level college, hypermedia was introduced by the author six years ago - in that time, faculty have received training and many have gone to develop web-based hypermedia.  Students, working in their second language, have produced innovative work, but their school background (based on rote learning and memorization)and aforementioned factors present formidable obstacles in implementing a flexible learning model.

Summarizing the author's experience, it is clear that certain variables contributed much to the success of the hypermedia learning environment. The learner-centred approach, with the teacher playing a supportive and facilitative role, appears to be one of the most important variables. The choice of a learner-friendly authoring tool allowed students to quickly learn the fundamentals without interfering with the learning process. The establishment of a base level of technology skills additionally afforded the students key skills in a relatively short time-span. Collaborative groups offered interactive learning environments which gave learners many opportunities to question, negotiate and reflect upon their learning. Finally, the integrated and interdisciplinary learning approach encouraged learners to explore related knowledge without artificial "subject" barriers hindering cross-curriculum exchange (Liaw, 2001). These variables, acting together, appeared to have established the pre-conditions for cognitive mapping.

Discussion

Deep levels of mental processing and elaboration, both indirectly observed by the author and others, indicate some support from cognitive psychology for potential learning effects from a hypermedia learning environment (Pohl et al, 1995).  Jonassen and Mara (1999) further establish the credibility of concept maps as cognitive tools, especially when created by learners themselves. Cognitive tools for learning are not about learning per se (Mayes, 1994) - the tool makes it possible for the learner to acquire a "deeper understanding" of the material: the learning takes care of itself.   Finally, both writers (Jonassen, 1994; Mayes, 1994) make the telling point that the people who learn most from using technology as cognitive tools are the designers of instructional materials.  Mayes goes further and suggests that teachers "start putting authoring tools into the hands of students (p.5).

This paper has been able to demonstrate the possibility of bi-directional cognitive mapping by offering glimpses from supportive research.  However, the human cognitive structures, these "propositional networks in the mind" (Jonassen and Mara, 1999), are far more complex than what has been discussed here.  The link between explicit hypermedia structures and these cognitive networks remains tenuous.  There are strong hints, but much work is needed to establish a robust correlation.

Liaw (2001), concludes that "the (hypermedia) technologies may well be ahead of the educational applications".  It has further been suggested that, at times, the "tail wags the dog" in terms of technology driving learning paradigms (Laurillard, 1994).  I think that research in the last decade has played catch-up with the cognitive and pedagogical underpinnings of observed practice in classrooms and computer labs around the world.  Hypermedia users need further collaborative studies done to strengthen the relationship.

Most of the research is centred on uni-directional cognitive mapping, that is, mapping an existing "expert" knowledge structure to a hypermedia environment.  To a large extent, the findings have been supportive (Jonassen and Wang, 1992).   Multivariate analysis of data collected over an extended time-frame in a variety of educational settings is required to discover the significant variables and their effects when observing cognitive mapping from hypermedia to existing learner knowledge structures.   This would provide either confirmation or negation of bi-directional cognitive mapping - my opinion is that through David Jonassen and other's work, cognitive mapping has been accepted for providing some practical methodologies to facilitate learning in an increasingly digital and hypermedia world.

References

Alexander, S. (1995). Teaching and Learning on the World Wide Web. AusWeb95 (USQ online resource) http://www.scu.edu.au/sponsored/ausweb/ausweb95/papers/education2/alexander/

Bates, T. (2000). Teaching, Learning and the Impact of Multimedia Technologies. Educause Review, Sep/Oct 2000, Vol. 35 Issue 5, p.38 (USQ Online Resource)

Bransford, J. D., and Franks, J. J. (1971). The abstraction of linguistic ideas. The European Journal of
Cognitive Psychology http://www.sil.org/lingualinks/literacy/otherresources/LiteracyBibliography/BransfordAndFranks1971.htm

Burbules, N.C. and Callister, T.A. (1996). Knowledge at the Crossroads: Some Alternative Futures of Hypertext Learning Environments, Educational Theory, winter edition http://faculty.ed.uiuc.edu/burbules/ncb/papers/crossroads.1.html

Clark, R. E. (1994). Media will never influence learning. Educational Technology Research and Development, 42(2), 21-29. (81537 USQ online resource)

E-commerce Guide. [Web site]. http://e-comm.webopedia.com [15 August, 2001]

Gant, P.A. (1998). Maximizing Multimedia for Training Purposes. Vision, November, 1998. [Web site]. http://horizon.unc.edu/TS/default.asp?show=article&id=65

Hassebrock, F. and Winters, R. (1995). Assessment of Problem Representations and Mental Models of Physics Undergraduates http://www.denison.edu/fipse/psych/hasse1_present.html

Hatherley-Greene, P. (1993). Classroom Management Techniques and Curriculum Delivery. Computers in New Zealand
Schools, November, 1993. http://www.oocities.org/peterhgdxb/Article/article.html

Hatherley-Greene, P. (1999). Student Multimedia Authoring: The New Literacy.  Unpublished paper presented at TESOL Arabia Conference, Al Ain, UAE, March, 1999

Jonassen, D.H. (1988). Designing Structured Hypertext, and Structuring Access to Hypertext. Educational Technology, November, 28(11)

Jonassen, D.H. and Wang, S. (1992). Acquiring Structural Knowledge from Semantically Structured HyperText. ERIC Document No. ED348000

Jonassen, D. (1994). Technology as Cognitive Tools: Learners as Designers. http://itech1.coe.uga.edu/itforum/paper1/paper1.html

Jonassen, D. H., Campbell J. P., and  Davidson M. E. (1994). Learning with media: Restructuring the debate. Educational Technology Research & Development, 42(2), 31-39. (81537 USQ online resource)

Jonassen, D.H. and Marra, R.M. (1999). Concept Mapping and Other Formalisms as Mindtools for Representing Knowledge. http://www.icbl.hw.ac.uk/~granum/class/altdocs/dav_alt.html

Jones, T. and Baxter, G.P. (1999). Student Explanations and Patterns of Use In a Hypermedia Learning Environment. Interactive Multimedia Electronic Journal of Computer-Enhanced Learning. http://imej.wfu.edu/articles/1999/2/05/index.asp

Kayama, M. and Okamoto, T. (2001). A Knowledge based Navigation System with a Semantic Map Approach. Educational Technology & Society 4(2)

Kearsley, G. (1988). Authoring Considerations for Hypertext. Educational Technology, November, 28(11)

Kearsley, G. (1997). Explorations in Learning & Instruction: The Theory Into Practice Database http://tip.psychology.org/theories.html

Kozma, R. B. (1994). Will media influence learning? Reframing the debate. Educational Technology Research & Development, 42(2), 7-19. (81537 USQ online resource)

Laurillard, D. (1994). Multimedia and the Changing Experience of the Learner.  Proceedings Asia Pacific
Information Technology in Training and Education Conference and Exhibition, APITITE 94. Brisbane, Aust. Vol 1: 19-24. http://snazzy.anu.edu.au/CEDAM/dl-apitite.html

Liaw, S. (2001). Designing the Hypermedia-based Learning Environment. International Journal of Instructional Media, Vol.28 Issue 1

Mayes, T. (1992).   Hypermedia and Cognitive Tools. http://www.icbl.hw.ac.uk/ctl/mayes/paper9.html

Nichols, F. (2000). The Knowledge in Knowledge Management (KM). http://home.att.net/~nickols/Knowledge_in_KM.htm

Pohl, M.,  Prenner P., Purgathofer P. (1995). Hypermedia in Education - Monitoring the Development of Hypermedia Documents Psychology Teaching Review, No 4,2  http://igw.tuwien.ac.at/igw/Menschen/pohl/yorkzwo.html

Rumelhart, D.E., and Ortony, A. (1977). The representation of knowledge in memory. In R.C. Anderson, R.J. Spiro and
W.E. Montague (Eds.), Schooling and the Acquisition of Knowledge. (pp 99-136). Hillsdale, N.J.: Erlbaum.

Spiro, R.J., Feltovich,   P. J., Jacobson, M. I. and Coulson, R.L. (1991). Cognitive Flexibility, Constructivism, and Hypertext:
Random Access Instruction for Advanced Knowledge Acquisition in Ill-Structured Domains. http://www.ilt.columbia.edu/Publications/papers/Spiro.html

Taylor, J.C. (1994). Novex Analysis: A Cognitive Science Approach to Instructional Design. [USQ Online resource]. http://www.usqonline.com.au/courses/81531_content/readings/Taylornovex.html

Wang, H. (1999). Theories and Hypertext/Hypermedia. [Seminar paper]. http://www2.educ.ksu.edu/Course/EDETC886/Hypermedia/Wang/paper.html