The great promise of the world wide web and internet information sources in general is that learners will be able to access meaningful information on any subject imaginable. But how will learners access this information? This article examines educational technology issues related to accessing meaningful information in an elaborate information space. In particular, the problems associated with an over-simplified view of interactivity and access are discussed and suggestions given for introductory college courses in new media and human-computer interaction.

Currently, the primary access path to internet-based information is the world wide web. Documents on the web are essentially static, in the sense that they offer a fixed set of data. The pages may be composed of graphics, text, animation, sound, and video elements but they are used to disseminate a specific “brick” of information. They are fixed with respect to content. Interactivity on the world wide web is handled primarily by “hot” hypertext or hypermedia links and is nearly always synonymous with navigation. Navigation, in turn, is often thought to be related to “going somewhere” as in “When I click this underlined word, we will all go to Germany”. Words and phrases such as “teleported,” “whisked off to,” “brought us to,” and “took us to” are often used by people explaining how hypertext works. Allowing the misunderstanding that the data is actually coming to the user (and not the other way around), these statements still promote an overly simplistic view of interactivity.
Use of the world wide web not only promotes the idea that interactivity triggers a navigational event, but that clicking (instantiating the event) will “take them to” documents similar to the one they are currently viewing. Users settle into a generalized, fairly simplistic notion about the tasks they will be given, the representation of the information provided (i.e., similar pages), and the level of the interaction (clicking designated areas). Rather than using current web browsers to introduce the concepts of access and interaction, instructors can develop a more sophisticated conceptual context which focuses on accessing a variety of information types, the terms of the interactions involved, and the characteristics of the user.
The problem of learning an oversimplified view of a given phenomenon has been given a great deal of attention. Recent work in the field of cognitive flexibility theory (Spiro, Feltovich, Jacobson, and Coulson, 1992) suggest that oversimplification may come back to haunt future users of more complex systems. The theory suggests that how efficiently a person releases misconceptions is due in part to how the misconceptions were learned. When a person learns that a concept is simple, and later finds that it is more complex, the learner must not only learn the “new” aspects of the concept but “unlearn” the old conceptualization scheme. “Unlearning” a concept may be much more difficult than learning it.

In a typical world wide web session, “hot” (or linked) means a forward connection to a document. The term forward refers to the idea that the link will advance an idea. It is not quite like a footnote, for example, which might explain a mode of thought or a bibliographic reference, or discuss the origins of an idea. But “hot” does not necessarily mean any of those things; the link might provide bibliographic information, or background information, or simply more information. The link might be a see also reference, or a more like these reference. It is possible that the same word used twice in a paragraph might be linked to different documents. In reality, there is a certain ambiguity about links, even when they are textual links used in a text document.
Hot text, however, is a small subset of the universe that represents hypermedia links. A more useful description of “hot” equates the term with a hypermedia button. A button may be thought of as an artifact that is designed to trigger an event. In this sense, the button may be viewed as a part of a larger dialog, namely, the interaction. The event instantiated by the button may open a path to the interaction’s goal state, but it may serve another function. For example, the button may supply the system or contextual help needed to reach the user’s goal.
In fact, a hypermedia button may define a function that has no counterpart in the user’s experience, or for that matter, in reality. We typically use hypermedia tools in such a limited context (to navigate the web) that we forget their most important characteristic: they may be invented. For example, a hypermedia button encasing a word might provide a wide variety of experiences. Suppose that when a user holds a mouse button down over the targeted word, a menu of options pops up which includes: dictionary definition, synonyms, other language translations, navigational links, background information, and diagrams. The choices in the menu represent part of a negotiation between the user and the information provider, representing an interaction that is much more elaborate than a simple navigational link.
Although the goal of computer based interactivity is to provide a continuous two-way communication consistent with the user’s needs (Jonassen, 1988), there are several places where breakdowns may occur. Building interactive artifacts involves turning abstractions into clear, functional expressions of ideas which can augment the performance of the user. However, what the user sees depends on how the user looks, and even well-designed interface objects may end up being useless or not fully functional. Research abounds which analyzes interactivity (Schwier & Misanchuk, 1993; Bellotti, Blanford, Duke, MacLean, May, & Nigay, 1996), but a few further examples may help identify key issues. A metaphor being applied may be misunderstood and the user may be unable to continue with the interaction. A system response may not make sense to the user, creating what Don Norman (1990) calls a gulf of evaluation. Or, the user simply may not know how to proceed, creating a gulf of execution (Norman, 1990). These phenomena may occur because the user is experiencing cognitive overload, which Brenda Laurel (1991) defines as “too much to do or see,” or they may exist for some other reason. Interactivity, being a two-way communication is very much dependent on the system being used to access the information, the information itself, and the user’s characteristics.

Students need a paradigm for knowledge representation which accommodates conceptual modalities that have never been encountered. The methods which students will use to access information in the future are somewhat unpredictable. New representations for knowledge will be (and are being) invented, and it is likely that interacting with those representations will be highly contextual.
Access to computerized information is a component of a human-computer or a human-computer-human dialogue. Characteristics of such a dialogue are dependent on the representations of the information provided, user attributes (including perception), and an event horizon defined by the level of interactivity possible in the dialogue. Interactivity provides a user with tasks, and it is quite clear that the tasks themselves also influence the nature of performance.
When we attempt to access the massive amount of data now online, we engage in a dialog. As an illustration of this type of dialog, consider the interactivity inherent in a person’s approach to a carnival entrance. Think of the information channels in operation: signs, whirling ferris wheels, calliope sounds, a carny barking “step right up” into a megaphone, the colorful ticket booth, people standing in line, people inside the gate carrying stuffed animals and eating cotton candy, the smell of, well...the smell. Notice how much of the dialog is dependent on the characteristics of the person approaching the gate. An adult might see sleazy people everywhere, and aging ride equipment. It is unlikely that a child would focus on these particulars. Essentially, the information presented by the carnival is filtered by the person’s perspective and becomes part of the interaction.
The ticket booth represents a key component of the interaction. Buying a ticket is the trigger which instantiates the person’s goal state. Even so, characteristics of the person are once again involved, including the person’s financial state. Other events are possible, the person may wish to talk to the barker, or to others in line, etc. Notice that the dialog becomes less reflective in the event space (when buying a ticket). Clearly, claiming that “buying a ticket equals access” trivializes or at least understates what is really involved in access to the carnival. Similarly, access to a sphere of information is an oversimplified by those who claim that access equals a connection in every home or classroom.
Future knowledge domains will contain a wide variety of information types or channels, and content providers (i.e.students and teachers) will need to communicate using these channels. An introductory course in new media, instructional technology, multimedia, or use of the internet should therefore include a discussion on the use of appropriate information channels. This discussion sets the stage for the use of multimedia and mixed media and helps the student develop a wider understanding of the word literacy. In addition to identification of gross media types, examples illustrating the strengths of those types may be given. It is easy to think of examples in which a short video can take the place of several pages of text, or an audio clip can provide information that would be impossible to communicate using text. Gesture, voice inflection, facial expressions may communicate a sense of humor or other personality characteristic that it would be difficult to capture using text. Conversely, inappropriate use of an information channel may add unnecessary burden on the user. The issues of tradeoffs, bandwidth, and necessary hardware (as well as who can afford it) are related. Students carrying out such study should be able to envision an internet far more sophisticated than the one in use today.

Due to the large number of variables involved, study of user behavior and user characteristics in computer mediated communication is very “soft.” It is difficult to quantify the user’s relationship to a human-computer interaction. Schoenmaker (1993) summarizes this perspective by claiming that “although the field of cognitive psychology includes a number of interesting viewpoints, there is no unified theory supporting formal design methods...” Nor is there likely to be such a theory. In the future, many human-computer encounters will involve abstract virtual environments that are task specific. It is becoming apparent that the human dimension of communication will be accentuated rather than diminished by computer mediated communication.
The current, overly simplistic assumption that interactivity means navigation limits the relationship between the information and the user, diminishing the user’s perceived options. It is more appropriate to view interactivity as contextual, and interactive artifacts (such as buttons and slide controls) as context sensitive devices which need to be mastered by the user in order to use the information effectively. For example, today’s web browsers use many of the same interactive artifacts common in other applications: scroll bars, resize boxes, and copy functions. They also include buttons sensitive to the context in which they appear (namely “online”), such as go back buttons, reload buttons, stop buttons, etc. New users of the world wide web learn to use the unfamiliar tools and apply them in the context of web access, with varying degrees of success.
Although today’s browsers are adequate for information that is presented in documents similar to paper documents, it is not difficult to see the need for more interactive “triggers” in a document involving multiple video and audio sources. In other words, as internet material becomes more complex, users will need more sophisticated interactive tools. It is quite likely, and appropriate, that certain internet environments such as simulations will use specific, unique interactive controls. The proliferation of more complex information types over the internet will widen the scope of marketing and distance education efforts, but it will also place an increased cognitive burden on the user and the developer.
Traditionally, computer research has accounted for variables such as previous computer experience of the user, socioeconomic background, locus of control index of the user, domain expertise of the user, and measures related to the level of engagement experienced by the user. The needs of the user are of prime concern, but often overlooked in controlled studies. Stephen Weyer (1988) thoughtfully described several goal possibilities as: tell me, inform me, amuse me, challenge me, guide me, and teach me. Obviously, the needs of the user interact with the knowledge domain, and affect the interaction in a human-computer dialog.

We are entering a new era of communication, unlike anything we have experienced before. The challenge facing us is to develop a new brand of “common sense” that is dynamic; one that integrates new forms of communication into existing streams of discourse. To rise to this challenge, we must build sophisticated frameworks capable of supporting a large number of variables simultaneously. It is up to institutions of higher learning to help students develop models useful enough to support a “gestalt leap” into the 21st Century.

Bellotti, V., Blanford, A., Duke, D., MacLean, A., May, J. and Nigay, L. (1996) Controlling accessibility in computer mediated communications: A systematic analysis of the design space. Human Computer Interaction, in press.

Jonassen, D. (Ed) (1988). Instructional designs for microcomputer courseware. Hillsdale, N.J.: Lawrence Erlbaum Associates

Jones, M.K. (1989) Human Computer Interaction: A design guide. Englewood Cliffs N.J.: Educational Technology Publications

Laurel, B. (1990). Interface agents. In B. Laurel (Ed.), The art of human&endash;computer interface design (pp. 355&endash;365). Reading, MA: Addison&endash;Wesley.

Norman, D. A. (1991). Cognitive artifacts. In J. M. Carroll (Ed.), Designing Interaction: Psychology at the human&endash;computer interface (pp. 17&endash;38). N.Y.: Cambridge University Press.

Schoenmaker, J. (1993). Linking new applications to new design paradigms. Computers and Education, 21 (1/2), 181-192

Schwier, R.A. & Misanchuk, E. (1993). Interactive Multimedia Instruction. Englewood Cliffs, N.J.: Educational Technology Publications.

Spiro, R. J., Feltovich, P. J., Jacobson, M. J., & Coulson, R. L. (1992). Knowledge representation, content specification, and the development of skill in situation-specific knowledge assembly: Some constructivist issues as they relate to cognitive flexibility theory and hypertext. In T. M. Duffy & D. H. Tonassen (Eds.), Constructivism and the technology of instruction: A conversation (pp. 122-128). Hillsdale, N.J.: Lawrence Erlbaum Associates

Weyer, S. (1988). As we may learn. In Ambron, S. & Hooper, K. (Eds.), Interactive Multimedia (pp. 88-101). Redmond, Washington: Microsoft Press.