An Interview with Clark Aldrich
An Interview with Clark Aldrich" The Technology Source, September/October 2003. Available online at http://ts.mivu.org/default.asp?show=article&id=1034. The article is reprinted here with permission of the publisher.
Clark Aldrich spent 2 years leading an international team that built an example of simulation learning software; the software has the look and feel of a modern computer game, but with a central emphasis on leadership content. He is the author of Simulations and the Future of Learning, a new book that helps educators understand, design, construct, and use computer simulations for an upcoming generation of students.
James Morrison [JM]: Clark, you argue in Simulations and the Future of Learning that simulations will shape the learning revolution. Can you summarize this argument for us?
Clark Aldrich [CA]: Educational simulations will be in widespread use by leading instructors within 5 years and will eventually change education as much as textbooks and motion pictures.
A single simulation can teach someone in a variety of ways all at once, and for this reason the medium is actually much closer to how people often learn from real-life experiences. For example, simulations can convey stories, scenarios, and other forms of linear content, just as lectures and books can. However, they can also do much more: By allowing for user interaction and feedback, they can be used to show how complex systems work in different circumstances. Using a simulation interface, learners can also develop an almost instinctual "muscle memory" that is characteristic of business and social interactions as much as sports. It is my belief that the effective use of simulations will result in entirely new types of material, and new approaches to old material, in our educational institutions.
Simulations will be increasingly asked for by the next generation of students and supported by the next generation of teachers. Furthermore, a few simulations will go a long way—as they are built, they can be widely deployed.
JM: What are the current options for designing a simulation, and how would you assess these options?
CA: The best way to understand simulations is to start (but not end) with computer games.
Over the years, game designers have established a number of distinct genres. There are first-person shooters (such as Doom, Quake, and Castle Wolfenstein) in which the player sees the world through the eyes of his or her onscreen counterpart, often looking down the barrel of a weapon. There are real-time strategy games (such as Command and Conquer, StarCraft, and WarCraft) in which the player is a disembodied, undisputed commander of some type of military operation. There are quiz shows (such as Jeopardy) and sports games (including football and soccer simulations). There are also games that define their own genres—such as SimCity and The Sims, which situate players as inhabitants of a virtual community that they help to create.
Each computer game genre has a set of desired goals or outcomes to be achieved. In turn, each genre has an established set of input conventions, such as how to control what is going on and how to respond to different situations. Each genre also has conventions around the "mini-world" with which the player interacts, including both the physics of the environment and the intelligence of any occupants. Through its unique combination of such elements, each computer game genre allows for a variety of strategic actions within the limits of a structurally determined environment.
The bad news is that these genres, under the surface, are quite rigid. They resist significant changes. Many people who have tried to build educational simulations have used existing game genres as their starting point. But the more they got into the project and tried to change the game, the more they realized that the genre did not lend itself to educationally relevant content. A good public example of this was the team from LucasLearning that tried to make a physics simulation from the game Jedi Knight. The team members thought that they could tweak the existing game engine, only to find out that the few physics principles built into the game were not able to be expanded or refined from their simplified state.
The good news is that the three elements of simulations—story and graphics, interface, and complex interactive systems—represent genuine opportunities for changing how we teach and learn. In order to make the most of their potential, designers will need to invent new, educationally oriented simulation genres. These new genres will be both similar to and different from computer game genres, in much the same way that current computer game genres are similar to and different from one another.
JM: I understand that you have created a simulation that educators can use. Tell us about it.
To pull this off, we created an interpersonal genre. The player is in a room discussing issues with simulated characters—"bots"—whose words and behavior are controlled through artificial intelligence (AI).
From an input perspective, we designed the players' and bots' relationships so that they would allow for supporting and opposing ideas, as well as supporting and opposing people. That is, a given character may or may not agree with another character at any given moment, which may or may not involve an interpersonal problem as well as a difference of judgment. This framework for the interface becomes a lens for looking at work. One nice feature is that if a user did nothing but learn the interface, he or she would be gaining a useful perspective on almost any leadership situation.
This simulation requires the player to perform a number of tasks in order to be an effective leader of group activities. For example, the player must exercise judgment on when to introduce new ideas, when to support a speaker, when to refocus on a key idea, when to bring in a quiet or disengaged person, and when to take an idea off the table. From a leadership "systems" perspective, these choices involve a number of underlying questions: Who has the power at any given point? When and how should the player gain more power as the leader? When should the player worry about the tension in the room, and when should the player raise or reduce this tension? What ideas are out in the open, and what ideas appear to be hidden? When is it the right time to introduce a potentially controversial or divisive idea? When should the group focus on brainstorming, and when should it focus on getting work done? Where does the player draw the line between his or her work and the higher goals of the group? (See Exhibit 1 for further illustrations.) Such questions are essential to any leadership role; the purpose of the simulation is to help students to address these questions within an interactive context that demands adaptation to changing circumstances.
JM: What issues or challenges did you face while creating the simulation?
CA: Designing a new genre is very difficult; it took a lot of work to get Virtual Leader right. Creating a modern computer game today costs $4 million and takes about 2 years. Creating simulation genres might take a quarter as much money and time, but each one is a high-risk activity. Some genres will work and some will not.
More challenging is the fact that simulation design requires the ability to step outside of a traditional, linear approach to content creation—a process that is counter-intuitive to many teachers. I wrote Simulations and the Future of Learning to help that process.
Over time, I believe that the academic community will build up a library of simulations, representing many genres as well as many examples within each genre. Once simulation libraries exist, the process will be less difficult. It will be easier to update (or "mod") existing simulations to make them more appropriate for a given class.
JM: Can you give us one or two examples of simulations that are now being used in higher and/or corporate education?
CA: Simulations are used in the military, in commercial pilot training programs, and on Wall Street. Not surprisingly, these institutions care deeply about their students' ability to learn and apply what they have learned.
Most corporate and higher education programs use one of four types of early (or proto-) simulation. First, there are branching stories, similar to the "choose your own adventure books" we read as kids. Corporations use branching stories to take salespeople through mock customer encounters, or help new call center employees learn techniques. For example, a player may encounter a particularly irate customer and will have to decide how to assess and respond to the customer's complaints. Based on the player's actions, the simulation will continue the narrative along a given track in order to show its likely outcome.
Second, there are business school spreadsheets, which are used to learn how things like pricing, advertising, and off-shore development impact market share and profitability. In these simulations, students allocate resources over time intervals and then are able to see the results in various graphs and charts. More than the other types, these spreadsheet proto-simulations require mentoring and active facilitation to help the learning along. The various products recently developed by Forio Business Simulations are good examples of this format.
Third, there are game-based simulations, which use familiar game models to introduce information to a captive audience. For example, a modified quiz show format may be used to encourage learners to master a given body of knowledge, or to collaborate with other learners in order to answer a series of questions. This is what Marc Prensky (2000) ends up describing in his book Digital Game-Based Learning and on his corporate Web site, games2train.com.
Finally, there are virtual products (typically driven by Macromedia's Flash MX). These are Web-deployed mock-ups of actual products. They are often used as part of a marketing effort or a certification process. Examples might include a Flash simulation that shows you how to use a new electronic device or how to operate a control panel in an airplane cockpit. The Flash Simulation page provides some examples of these tools.
All four of these models have been used successfully in corporate and higher education, and they each contain key elements. Branching stories are first-person and immediate; they can involve some rich media. Business school spreadsheets allow for subtle interactions. Virtual products are kinesthetically appealing, and game-based simulations have plenty of zing. However, the use of simulations in higher education will need to advance beyond the respective limitations of these four models.
JM: What advice do you have for teachers who want to construct simulations for their on-campus or online classes?
CA: To understand simulations, the first thing to do is get comfortable playing computer games. Try The Sims, the newest Midtown Madness game, and Roller Coaster Tycoon. Then find an online community that builds modifications for existing games. For starters, try SimsFiles.com, X-Plane.org, or JediKnightII.net. After you have explored various options, choose a computer game that you like, and try implementing modifications that can be downloaded from community sites. This may sound complicated, but the communities are full of newcomers who also need step-by-step help.
You can also look at and modify some of the proto-simulations mentioned above, or get Virtual Leader and practice customizing it. This initial "hands-on" experience will help to stimulate ideas for the potential use of this technology in an educational context.
Then begin to sketch out a preliminary design for the simulation you would like to create. In considering the design, don't be satisfied with papers and speeches as acceptable output for the work of your students. Build simulation models using spreadsheets. Practice putting in as many hard variables as possible to describe potential relationships or scenarios. And visualize as much as possible: Draw charts and annotated pictures to indicate the features you want to incorporate into the simulation. Because effective simulations involve a range of overlapping skills or activities, it will be necessary to adopt a flexible approach during the early stages of development.
To those who are ready to dig in, begin the process of finding or adapting educational content for the simulation. For example, if the simulation requires its users to diagnose a medical condition, it would be necessary to research a range of ailments along with their characteristic symptoms and treatment options. If the simulation involves a mock trial, it would be necessary to research various legal precedents or procedural rules that may be pertinent to the outcome of the trial. Once you have prepared the content, then consider how it may be incorporated within the simulation—that is, how it may be correlated with the user's input options, the artificial intelligence of virtual characters, and the feedback elements of the system.
Then I would suggest researching third-party graphic engines, the software applications that convert computer code into what look like three-dimensional images on a screen. At one time, every game had a unique a proprietary graphic engine. Increasingly, in part due to the cost and time frame of development, third-party graphic engines have been licensed to game developers. Graphic engines tend to be built and optimized for a given game genre. Common metrics around graphic engines are the number of polygons they can present on the screen at a time at a given frames-per-second rate. LithTech, Doom, and Quake are three popular engine brands. I expect that open-source graphic engines will become more available and powerful over time.
After that, bring in programmers, and get ready for a lot of testing and tweaking. Finally, if the simulation needs an instructor to facilitate the experience (which will be common with more complicated, robust simulations), develop a supporting "teacher's guide."
JM: Philosophically, is the long term goal to create a "perfect simulation," a Star Trek holodeck-like experience indistinguishable from real life?
CA: I do not believe in a "perfect" educational simulation. Most practically, once something can be perfectly simulated, the activity can be automated.
With this in mind, I realize that many educators may be inclined to question the pedagogical value of simulations. If a simulation cannot reproduce every variable in a given situation, does this not inherently compromise its ability to foster a meaningful learning experience? As a general point, this objection is misguided. Even in traditional learning environments, instructors often provide hypothetical scenarios or role-playing situations for their students—despite the fact that such scenarios typically lack certain qualities of the "real experience" they are supposed to evoke. These activities can still have the value of helping students make careful discriminations and informed judgments based on their knowledge, as long as the limitations of the scenario are also taken into account by the instructor. Moreover, precisely because these fictional scenarios filter out much of the distraction and noise of real life, they allow for more focused thought or activity on the part of students. They do not perfectly replicate the "messiness" of life, but they approximate a particular situation in order to elicit a certain set of knowledge and skills.
Much of the same can be said for the pedagogical potential of simulations. The goal of any simulation will always be to focus on some relationships, but not all; the technology will never be "perfect." It is true that some learning scenarios lend themselves more readily to simulation than others: A flight simulator or gross anatomy lesson, for example, may be designed much more easily than a mock trial. Nevertheless, the number of variables that a single simulation can replicate will go up with time, and I believe that this will allow for increasingly complex and nuanced learning experiences. The challenge for educators in upcoming years will be to take full advantage of this potential.
JM: Do you think that the majority of teachers will be using simulations in their classes by the end of the decade?
CA: Yes. While it may not be a universal practice, I think that this will be a predominant trend by 2010. Through the widespread inclusion of emerging simulations, students will be able to learn much more complicated skills, such as project management, negotiation, problem resolution, time management—or even how and why to eat well.
There will be a transition period when someone can go back and forth between being a teacher and being a simulation designer. But over time, the two roles will separate based on very different skill sets. Then the challenge will be, as with textbooks and videos, for teachers to evaluate and adapt simulations rather than build them.
I do not believe it is possible to overestimate two things—first, how different and richer simulations are than other learning content, and second, how much they will eventually dominate the way classes are researched, designed, and implemented. Anyone who wants to understand and contribute to education in the near future must develop a working knowledge of the philosophy of simulations.
Prensky, M. (2000). Digital game-based learning. New York: McGraw-Hill.pc gameshidden object gamesplatform gamescard gamesadventure gamespc game downloadsmanagement gameskids gamesword games