How do we move educational technology from the page-turning applications that have given CAL a bad name to an environment that helps retain the inherent interest of learning? The ideal would be an activity that intrinsically "engages" the learner, and leads them through an interactive experience that enhances their ability to solve problems. Are engagement and effectiveness mutually exclusive goals? I will lay out the different conceptions that contribute to an understanding of engagement, and then demonstrate the considerable overlap that exists. To begin with, however, I need to lay out the ground rules.
My premise is that learning does not have to be aversive. That does not mean that learning does not involve activity, but rather that we will invest considerable energy in activities that are engaging and fun. Learning enhanced to be fun can be more effective, as Lepper and Cordova (1982) have shown. Using some simple educational tasks, they demonstrated that learning embedded in a motivating setting improved learning outcomes.
This is not to minimize the complexities that are involved. Games and the contexts of their use are rich enough that we cannot easily control all variables and reliably conduct laboratory studies. Further, many of the elements that make an activity a game may arise external to the inherent nature of the task. The claim here is not that the games themselves are sufficient for learning to occur, but rather that they can provide elements of a learning process (notably the activity).
The definition of game itself is also problematic. "Game" has long been the canonical example of an undefinable concept. For example, the boundary between simulation and game is not clearly defined; when does a flight simulator change from a training tool to an entertainment activity?
I am going to take a very pragmatic view and insist that the only measure of whether an interactive activity is fun is the player's perception. You cannot decree that anything is a game, you have to ask or look for convergent evidence.
Note that I am not talking about so-called "twitch" games, but rather games that require cognitive effort and are not generally dependent on rapid response. There are learning applications there, but that is not where I think the interesting challenges and opportunities lie. That does not mean that I am therefore abandoning fun.
What elements comprise an engaging interactive multimedia learning experience? To be effective educationally, there must be reference to learning theories and methods of design. To be truly interactive, it must draw upon what is known about how to make computer systems usable. We also want to know what it is that characterizes activities that people classify as entertaining and engrossing.
I contend that, at least for cognitive skills and at least at a coarse level, learning approaches are converging on a model (as I was proposing as a response to Charles Reigeluth's recent discussion). This model includes motivating the learning by demonstrating the practical applications and importance of the knowledge, providing a conceptual description of the skill, demonstrating the application of the knowledge to practical problems, providing practice opportunities with support in the form of scaffolding, and facilitating transfer through guided reflection on the activity to integrate the practical issues with the underlying conception. Justification for the elements of this model are spread across approaches such as problem-based learning (Barrows, 1986), cognitive apprenticeship (Collins, Brown, & Newman, 1989), Laurillard's (1993) pragmatic approach, and others. There may be disagreement about which order elements are taken in, but I believe few would argue against the inclusion of any of these elements.
Increasingly, there is an emphasis on exploration and discovery, where the learner takes responsibility for constructing their own knowledge. There is similarly an increasing emphasis on the role of social interaction in learning. While the extreme view of this is contentious, equally so is the view that you can completely engineer the learning process. As in most things, the answer lies in between (e.g., Anderson, Reder, & Simon, 1996).
Specific elements to focus on here include motivating the learning, demonstrating application, and scaffolded practice. One implication of this model is that students should be assigned activities that reflect the application of the content knowledge as it is practiced outside the classroom. The goal is to induct the learner into a "culture of practice" which makes the knowledge meaningful. Another implication is that the feedback ideally should be intrinsically embedded into the context in which the activity is performed. A further implication is to have the learning challenge carefully balanced to keep it within a "zone" that matches the learner's ability. What we see are that elements that contribute to effective learning environments include a thematically meaningful story ("situating" the application of the knowledge), relevant and rapid feedback, and a carefully managed level of challenge.
Approaches to making computer-based tasks "direct" also inform the process of design. Innovations in interface design (interestingly, many were first seen commercially in games) were providing a new experience of using a computer, and several researchers tried to summarize the elements that contributed to the feeling of directly manipulating the computer environment. Shneiderman (1983) and Hutchins, Hollan, & Norman (1986) suggested that a tight coupling between action and feedback was important, both in the form of the communication, and in the time between action and response. In addition, complex syntax is replaced by direct manipulation on representations that are familiar from other experience.
Inspiration and clarification for the design of computer experiences have extended into examinations of other media. Two different perspectives have informed the investigation: Laurel (1991) has explored the implications of theater for interaction design, while Tognazzini (1993) has similarly explored stage magic. While these are passive activities from the observer's perspective, there are underlying principles that have been extended to the design of engagement. One element that arises from both perspectives emphasizes designing the "action" so that there is a thematically coherent development of the experience over time. Further, the action should be meaningful to the theme. Finally, control should be in the hands of the user, with a perception of choice and frequent opportunities for action.
A further source of input is a broad investigation of the affective experience of fun. Explorations have included the experience of the "flow" state (Czikszentmihalyi & Czikszentmihalyi, 1988), and considerations of what makes computer games "fun" (Malone, 1981). Malone indicated three elements of computer games: fantasy, the scenario in which the activity is embedded; challenge, the level of difficulty; and curiosity, the introduction of new information and non-deterministic outcomes. Czikszentmihalyi expands the concept of challenge, indicating that the level of challenge needs to be matched to skills, and should be greater than average. Another important element is having clear goals for the activity. Finally, the flow state is highest when the individual is the locus of control.
It should be noted that many elements are repeated in the different areas. Feedback is highlighted in several, as are goals and control, challenge, thematic coherence, and the need for direct action. It is this conceptual overlap that raises hopes for a productive synergy.
Combining these approaches yields a convergent and emergent model of engaging learning. I propose that engagement comes from "interactivity" and "embeddedness," and that the elements that constitute these two components match with good learning design as well.
Embeddedness includes thematic coherence, meaningfulness of action to the domain of representation, and meaningfulness of the problem in the domain to the learner (which is implicit but not explicit in much of the above). Note that embeddedness differs from immersion, which to me is the physical presentation directly and exclusively to sensory inputs, such as seen in virtual reality environments. People can get just as engaged in a non-immersive environment as they can in an immersive one.
Interactivity includes having an appropriate level of challenge through a variety of choices of action, effected through direct manipulation of the world of interest, with quick and clear feedback from those actions in ways that reflect the semantics of the world and afford further action choices, and the presence of novel information and events that contribute to those choices. (Note that this is an attempt to elicit the elements that contribute to the aesthetic of interactivity and therefore a somewhat different enterprise than that which Rod Sims has recently discussed.)
Moreover, we see that the elements of engagement (from studies of user interaction) also reinforce the elements of good learning. The degree to which the learning is "situated" is the degree to which the situation represented, or the theme, is meaningful to the learning domain. Good learning design will also propose that feedback should be rapid and linked to the problem and the action. Finally, the challenge required to maintain engagement is just the zone of difficulty where learning occurs. This is a rather sketchy treatment of what I see as a richer sharing of common ground.
I will maintain that it is the cognitive challenge that distinguishes the truly engaging activity and the truly effective learning. However, if I am arguing that we can embed cognitive challenges in games, I must also clearly describe the cognitive outcomes. If we first distinguish between knowledge "how" to do things, and knowledge "about" things, we can then characterize the use of cognitive skills as the application of "how" knowledge to use "about" knowledge to solve a problem. Then, computer games are practice opportunities for cognitive skills. As a further outcome, exploration can also help learners internalize the relationships underlying a dynamic system. As Michael Spector [ITForum #9] discussed in this forum earlier, understanding dynamic systems is a desirable learning outcome, and can fundamentally be enabled through interactive technology.
In contrast to Michael Spector, however, I have not discussed computerized "construction kits." These certainly can have the appearance of games (e.g., Cocoa), and they can have powerful outcomes. However, while the ability to model a system is a useful skill, not all systems need to be modeled to be successfully understood. I believe exploring and discovering contingencies and relationships can build a sufficient understanding of a system to meet certain learning goals. In the game Quest, developed to help youth learn how to live independently (Quinn, in press; see also http://www.cse.unsw.edu.au/Quest/), I maintain that there is no need for the player to explicitly develop an external model as internalizing the contingencies is sufficient.
I recognize that there are many unanswered questions and remaining challenges. For instance, I have not addressed the empirical result that the games that appeal to one person do not necessarily appeal to the next. The easy way out would be to argue that with some support (like it being required in the classroom), almost any game would be used in preference to some less engaging activity. Instead, I would rather argue that we may need a variety of games to illustrate a point, both to further facilitate transfer and to recognize the diversity of our students. This is particularly relevant to gender issues, where the vast majority of computer games appeal far more to males than females. I believe that cognitively challenging games can appeal equally, and in general that well developed games will have a broader appeal than games focusing on other forms of challenge such as motor challenge.
Another complex problem is developing a story-line within a game. As indicated above, the story develops out of an embedding theme, but this does not address capturing the characteristic buildup of tension and release (Laurel discusses this). Without too heavily presaging Rob Moser's Ph.D. thesis (and I will acknowledge here both his contributions to my thinking and his valuable comments on this paper; the same for Andrew White), he's finding that the structure of learning task hierarchies maps nicely onto the requirements for maintaining an appropriate level of challenge (which maps nicely onto the requirements of structure for plot development).
A further question involves the educational effectiveness of the games I am proselytizing. We do not yet have a systematic evaluation of the benefit of even the most hyped examples of "edutainment." Certainly, we need to consider how they are used, but we also need to know how to use them effectively. I believe that with materials to help teachers or parents facilitate the learning in ways coherent with the convergent learning model proposed above, these environments can be effective (our anecdotal evidence from Quest supports this), but this is still in need of empirical validation.
We may also be able to enhance the ability of games to, on their own, facilitate learning. We can embed scaffolding and support tools into the environment (Hedberg, et al, 1994). To facilitate reflection, we can record traces of the learner's activity or exploration (Schauble, et al, 1993). There is considerable potential for exploration here.
I am optimistic that, with good design and good guidelines for support, computer games can be engaging and educationally effective. I would go so far as to say that they could be engaging enough to be played without educational goals at all. Certainly examples exist. The problem is to educate the game designers to move past their easy reliance on sordid rehashes of the latest technical advance, and to make the effort to design challenging games that are based on valuable cognitive skills. We learn the things needed to play games. We should therefore strive to design games that require the things we wish to learn. To effectively engage learning, we can and should make learning engaging.
Anderson, J.R., Reder, L.M, & Simon, H.A. (1996). Situated learning and education. Educational Researcher, 25(4), 5-11.
Barrows, H.S. (1986). A taxonomy of problem-based learning methods. Medical Education, 20(6), 481-486.
Collins, A., Brown, J.S., & Newman, S. (1989). Cognitive apprenticeship: Teaching the craft of reading, writing, and mathematics. In L.B. Resnick (Ed.) Knowing, learning and instruction: Essays in honor of Robert Glaser. Hillsdale, NJ: Lawrence Erlbaum Associates.
Csikszentmihalyi, M., & Csikszentmihalyi, I.S. (1988). Optimal experience: Psychological studies of flow in consciousness. New York: Cambridge University Press.
Hedberg, J.G., Harper, B., Brown, C., & Corderoy, R. (1994). Exploring user interfaces to improve learning outcomes. In K. Beattie, C. McNaught, & S. Wills (Eds.), Interactive multimedia in university education: Designing for change in teaching and learning (pp. 15-29). Amsterdam: Elsevier.
Hutchins, E.L., Hollan, J.D., & Norman, D.A. (1986). Direct manipulation interfaces. In S.W. Draper & D.A. Norman (Eds.), User centered system design: New perspectives on human-computer interaction. Hillsdale, NJ: Lawrence Erlbaum Associates.
Laurel, B. (1991). Computers as Theater. Reading, MA: Addison-Wesley.
Laurillard, D. (1993). Rethinking university teaching: A framework for effective use of educational technology. London: Routledge.
Lepper, M.R., & Cordova, D.I. (1992). A Desire to Be Taught: Instructional Consequences of Intrinsic Motivation. Motivation & Emotion, 16(3), 187-208.
Malone, T.W. (1981). Towards a theory of intrinsically motivating instruction. Cognitive Science, 5, 333-370.
Quinn, C.N. (in press). Designing an instructional game: reflections on Quest for Independence. Education and Information Technologies.
Schauble, L., Raghavan, K., & Glaser, R. (1993). The Discovery and reflection notation: A graphical trace for supporting self regulation in computer-based laboratories. In S. Lajoie & S. Derry (Eds.), Computers as cognitive tools (pp. 319-337). Hillsdale, NJ: Erlbaum.
Shneiderman, B. (1983). Direct manipulation: A step beyond programming languages. IEEE Computer,, 16 57-69.
Tognazzini, B. (1993). Principles, techniques, and ethics of stage magic and their application to human interface design. Proceedings of the International Joint Conference on Human-Computer Interaction. Amsterdam, April.
The buttons that appear below will be found at the top and bottom of each page of the discussion. The first button will take you back to the previous page (in this case, to the beginning of paper #18). The middle button will take you to the ITForum home page. The last button takes you forward into the discussion as it progressed on-line.