Tuesday, 1 August 2006

Why Most Off the Shelf Commerical Games Will Not Work in Education? And What Is The Alternative?

[Draft 2 of an earlier essay. Do not cite before contacting me for update. Comments welcome.]


There is a lot of hype on using commercial games in education lately1. James Paul Gee2 shows us that everyone can learn something from games. Clark Aldrich3 talked about how to select, research, build, sell, deploy, and measure the right type of educational simulation for the right situation. John C. Beck, Mitchell Wade4 argues that gamers glean valuable knowledge from their pastime and that they’re poised to use that knowledge to transform the workplace. The premise is that commercial games are interesting because they are immersive and offer the possibility of a higher fidelity learning environment. Commercial games are engaging and hence by using commercial games, educators can both adapt to the learners and engage them. Commercial game is the modern Torjan Horse for conquering the minds of our next generation – so many of us hope!

Unfortunately, this is not an informed and balanced expectation. There are millions of failed commercial games. So not all commercial games are engaging. Different genres of games attract different gamers. So even successful commercial games are not universally engaging. Commercial game designers single-mindedly create engaging games. When we try to add additional requirements, such as delivering learning outcomes, the result is disastrous. Game designers acknowledge the problem as evidenced by all the conferences about “serious games” lately5.

Successful commercial games cost millions to develop. Education does not have that money. Even if education does have that money, one would question the wisdom of investing such good sum of money in commercial game design without the evidence of effectiveness.

A commercial game may cover a topic in a subject domain. Education covers hundreds of topics in thousands of domains. Again, in order to cover most topics in most subject domains, we need thousands or millions of games. Learners do not have the time to play all these games. Education will not be able to afford the cost of building such games.

Angela McFarlane, Anne Sparrowhawk & Ysanne Heald6 pointed out also that many of the skills valuable for successful game play, and recognised by both teachers and parents, are only implicitly valued within a school context. Teachers and parents both valued the conversation, discussion, and varied thinking skills demanded by some of the games employed. However, these alone could not justify the use of the games within a crowded school curriculum.

However, game developers and gamers do tell us one very important message. Games can engage some learners and can be a vehicle to provide learning. Quoting Angela McFarlane, Anne Sparrowhawk & Ysanne Heald again, Games provide a forum in which learning arises as a result of tasks stimulated by the content of the games, knowledge is developed through the content of the game, and skills are developed as a result of playing the game. The argument here is that commercial games, off the shelf, are unsuitable for use. But we can find valuable lessons and design concepts from games to better do our job.

Using Commerical Games For Education Purposes

SimCity7 is one of the most successful commercial games and has been used quite extensively by educators. Player assumes the role of the chief architect and mayor of a city. Player is responsible for building and managing the city. As the player builds the city, in-game SIMS (simulated citizens) would come to the city, engage in different activities and generate a tax base for the player to continue developing the city. In the paper Playing With Urban Life: How SimCity Influences Planning Culture8 Daniel G. Lobo explained that the work of Forrester, Alexander, and Rybczynski have served as SimCity’s foundational ingredients. It seems to be an endorsement of the use of SimCity for urban planning courses where Forrester, Alexander, and Rybczynski’s models are examined. However, Lobo pointed out that the SimCity’s game goal is to build a megacity and the player is given absolute power of being the “god” in the simulator. “Unwieldy growth and megalomaniacal, destructive behavior are the two poles of city operation and the player’s most likely courses of action.” This is not the kind of message we want to give to our learners. As a commercial game and limited by the demand of engagement, the game goal cannot be modified. These undermine the usefulness of the game in serious educational settings.

X-plane9 is a very interesting flight simulator. Firstly, it is developed and marketed by a one-man company called Laminar Research based in the owner’s home. Secondly, people have been asking how he can compete in the flight-sim business, make money and survive since 1994. The answer lies in the way x-plane has engaged a wide community which supported the development. Thirdly, this simulator allows users to create planes - different type of planes including Space Shuttle10 and simulated flight in the atmosphere of Mars11 to amazing accuracy of the Physics involved. Fourthly and most incredibly, x-plane received FAA approval to train pilots towards their commercial certificate12!

Amazing as it may sound, I downloaded a demo copy and tried to “play” with x-plane to fulfill one of my childhood dreams of becoming a pilot. After installing the software, I looked at my monitor and my brain went blank. I knew I wanted to fly, but to where and why that particular city. Suddenly, I did not know what I wanted to do with the simulator! I lacked a “game goal”. May be I should have a friend with me and compete to land the plane in bad weather or do some flying upside down tricks.

These two stories tell us two things.

First, educational use of game will require the game’s underlying black box to be opened. SimCity’s failure of being a good educational game is because of its underlying black box does not allow educator to either modify the underlying models or to adjust the game goal to match the curriculum.

Second, an amazing game like x-plane is not engaging to me (even I have a childhood dream of being a pilot) because I have not formulated a game goal. X-plane does not have an implicit game goal, making it more like a simulator than a game. X-plane is successful enough to be endorsed for FAA certification because of the accurate underlying Physics-engine modeling of the aerodynamics of the air-foils. It is basically a simulator. Learners can create their own game goal. The other power of X-plane is that people can create different planes for x-plane. Hence if you in the process of obtaining a commercial pilot license for a particular plane, you can either get that plane from someone or create your own. SimCity’s failure as a simulator is exactly because there is an implicit game goal in the game – and the game goal does not necessarily match (in some cases, actually contradicts) the learning objective!

The second half of my last statement is actually very interesting. We can put on an education outfit to a suitable game.

Simple Games

Sivasailam Thiagarajan13, I called him Thiagi as most people who know him would, started his consulting business in 1976 from his basement. Now, 30 years later, he continues to operate the same business in pursuit of the same mission. During these years, he has designed hundreds of training games and activities. His games do not have complex graphics, no animations, simple, usually run for a few minutes. However his games are engaging, inspirational and sometimes very funny.

Here is a link to an email game called DEPOLARIZER: http://thiagi.com/email-depolarizer.html. It is a bit too long to copy here. Please review the game before reading on.

In this Thiagi’s email game, there are 5 rounds. Participants are asked to give their positions on a 9-point scale on an issue. They are then asked to predict the average of the group, then role play to provide statements at the extreme positions. The final step is to predict the average of group after revising their personal positions based on reading of the extreme position statements.

This game is applicable to almost any subject/issue. It is collaborative, can be played without much interference to the normal work schedule of the participants.

I played another Thiagi’s email game (called half-life) for a group of online role play simulation users in higher education institutes about 3 years ago. During the 2 weeks of playing the game, I did not lose even one participant throughout the 4 rounds of email exchanges. During the face to face debriefing, there was a heated debate of the outcome of the game too.

Thiagi’s game is a shell (he calls it a frame) whereby we can adapt to different subject matters. It is engaging not because it has high production value. Like chess, it is engaging because of nature of the game is engaging. It is engaging because these games give the player a real sense of freedom to do whatever they like and bear the consequence of the player’s action. It is engaging because there is a meaningful game goal. It is engaging because it is challenging.

There are deadly dull boring content which one must master in a subject matter, such as the names of bones for a medical study, the chemical symbols of the elements in Science etc. We can sweeten the learning process using a competitive game, like “Who wants to be a Millionaire” or “Trivial Pursuit”. We make use of the competitive nature of the players. We can make use of the pacing (as in shorter time for each higher level) to ensure mastery and fast recall of facts. (Hint: how many times we keep playing the same console game just in order to get a higher score?)

Here, we can put on a game outfit to an education activity.

The Role Of Game Design

I started this essay by proclaiming that commercial, off the shelf, games will not work in education. I also admit there are critical elements of game designs which are useful and informative in our endeavour to create engaging learning. Next, I will explain what I have learnt in the last 5 years as Fablusi14 creator. Fablusi is an online role play simulation platform on which educators can create and deliver their own role play simulations. I am in a privileged position to be able to see many engaging role play simulations being prepared and delivered to hundreds of learners in a broad range of subject areas.

Game Goals And Learning Objectives

Here, I want to make explicit the distinction between game goals and learning objectives. Game goal is the position a player wants to achieve at the end of the game within the context of the game. It is the “winning” position. Learning objectives are things an external institute (one who provided the games in the first place) wants the players to acquire during and/or after playing the game.

Game goal is a powerful motivation for player. If the game goal requires the learning of something, players will have a strong incentive to get that knowledge in order to achieve the game goal. The essence of “goal-based learning” can be summarised by Roger Schank’s15 question: “Why would anyone learn anything if not to help in the pursuit of a goal?”16. By creating a scenario and giving a learner a role in the scenario, Schank set up a goal for the learner. In pursuit of the game goal, the learner seeks out advice from experts (provided as short video clips in one implementation I have seen), makes decisions and eventually achieves the game goal. Although I called that particular implementation a glorified multiple choice, there is obviously a certain engaging facet for some learners.

The Scarlet Letter Simulation17, a role play simulation18, is “a psychological examination of the values, mores, and traditions impregnating American literature as portrayed by Nathaniel Hawthorne's characterization of early American Puritan culture. The simulation aims to explore the themes of sin, hypocrisy, repression, self knowledge and the fall of Puritan society. Players will experience the conflicts inherent in questioning why individuals struggle with their actions and feelings, why individuals feel the need to chastise others, and how individuals deal with the conflicting desires of nature and the demands society”. Some of the roles in the simulation are taken from the novel, some are created by the simulation designers Mary Noggle and Roni Linser. To start the game, players are asked to write a “role profile” which gives a description of the role (and the hidden agenda of the role). This set up a game goal and the ownership of the role. Unlike Schank’s implementation, the game goals for each player are not the same although there is a common learning objective. Furthermore, as the game progresses, the game goal can and may change. We19 call this dynamic goal-based learning.

Again in the Scarlet Letter Simulation, the game started as the moderator released a “kick-start” episode – a compelling reason for the roles to act. In fact, in this simulation, two kick-start episodes were used. The second one brought the players to a time 15 years later.

Engagement is about engaging the mind. When the players own the role they play and have the freedom to do what they wanted to do – and enjoyed the consequence of the actions, the engagement is overwhelming. When the Scarlet Letter Simulation was used, all students read the text thoroughly, some several times in order to understand the characters, the circumstances surrounding the story and the historical context. This was the first time in the history of that course that all students did extensive research beyond the basic reading in order to play the role. This is the power of game goal.

White box game engine

The games which can be used successfully in education, such as x-plane, Thiagi’s simple games and role play simulation all have one common characteristic – the game goal and the context can be easily modified and adopted for different learning objectives.

Both Thiagi’s game and Fablusi online role play simulation platform can be considered as a shell where educators can easily add content, context and set up game goals. The Scarlet Letter Simulation, and many other simulations, ran on Fablusi™ online role play simulation platform which is based on a role play simulator generator20. There is no need to have any technical knowledge except the ability to visit a web site to create, moderate and deliver a role play simulation using Fablusi.

Similarly, Game engine can provide an engaging framework or as a simulator. The role of educator is to create compelling game goal which when achieved will require the mastery of the material that was covered by the learning objectives.

Designing games for use in education requires a shift of the focus of the creation process. Traditional instructional design calls for identification of learning objectives, the material that support the learning objectives and then formulate the sequence of the material. In designing games for education, after identifying the learning objectives, we look for a scenario in which we can set up compelling game goals for the players. At the same time, in order to achieve the game goals, the players need to have progressive mastery of the material covered by the learning objective. Learning is the by-product of playing.

Collaboration & Minimalist Design

Thiagi’s email games and online role play simulations share two features which are very desirable in education – collaboration and minimalist design.

Both of these designs require players to be competitors and collaborators at the same time – an essential requirement of survival in today’s hyper-competitive world. Thiagi’s game is player inter-dependent. The output of a group of players becomes the input for another group or the next stage. Role playing, in a sense, is a co-creation of the acts in the scenario. Players act out a persona, without a script, to investigate, understand and experience the circumstances the persona is placed in the scenario. Players solve the problems, which require mastery of the material of the learning objectives, in the light of a persona. This may involve collaborating with other players. This may also involve competition with other players.

All the role play simulations on the Fablusi™ platform supports an additional layer of collaboration. A team of players is to play one role! It is required that a role to act coherently throughout the period of the simulation. As a result, players, when playing in a team, need to co-ordinate among themselves in order to ensure the role is acting coherently. That would require the articulation of tactics and strategies adopted by the role among the team members. This process of articulation serves as a learning process that we aimed for.

Thiagi’s email games and online role play simulations do not have elaborate and expensive graphics. The focus is to elicit the imagination of the players – not the designers. Without those expensive 3-D virtual environment or graphics, educators can create the email game or an online role play simulation in very short time, typically half an hour for email game and a day for role play simulation. Both email game and role play simulation runs for weeks. That is a very effective use of development time.

Commercial games are technically challenging to create, partly because of the sophisticated rendering of fast moving graphics. As a commercial game, production value counts. However, judging from purely “engage-ability” point of view, the engagement of the mind is more important and valuable. Of course, if education can afford it, there is no reason why we should not adopt high fidelity rendering. When a balance needs to be made, I will lean towards engaging the mind rather than on production value of the game.


I believe that a direct adoption of commercial off the shelf game, an attractive proposal as it might look at first glance, is not the best approach we should take. Rather, I suggest that we should look beyond games and identify what really capture the minds of our students. Engagement is about engaging the mind. Like chess, the production value of the chess pieces is secondary to the engaging nature of the game.

I have been involved in an engaging pedagogical design for the past 5 years. The engaging nature of online role play simulation is unquestionable. The learning outcomes were surprisingly pleasing. Learners went beyond the traditional requirement and searched out more information in order to play the role. The learning is much deeper, more memorable and more fun.

In this essay, I have shared with the readers several key messages. The most important one is about the design of game goals which align with learning objectives. If the game is engaging and the game goal is compelling, the players will master what is required to achieve the goal.

Why would anyone learn anything if not to help in the pursuit of a goal?” – Roger Schank, 1992.

1 See for example Begona Gros, (July 2003) The Impact of digital games in education, http://www.firstmonday.org/issues/issue8_7/xyzgros/index.html

2 James Paul Gee (2003), What Video Games Have to Teach Us About Learning and Literacy

3 Clark Aldrich (2005), Learning by Doing: A Comprehensive Guide to Simulations, Computer Games, and Pedagogy in e-Learning and Other Educational Experiences

4 John C. Beck, Mitchell Wade (2004),Got Game: How the Gamer Generation Is Reshaping Business Forever (Hardcover)

5 See Serious Games Initiative (http://www.seriousgames.org/index2.html); Serious Games Summit (http://www.seriousgamessummit.com/); Game Develops Conference (http://www.gdconf.com/conference/seriousgamessummit.htm); and many other conferences just in 2006.

6 Angela McFarlane, Anne Sparrowhawk & Ysanne Heald (2005), Report on the educational use of games, online http://educationarcade.org/files/videos/conf2005/Angela%20MacFarlane-2.pdf

7 http://simcity.ea.com/

8 http://www.americancity.org/article.php?id_article=21

9 http://www.x-plane.com/

10 http://www.x-plane.com/orbiter.html

11 http://www.x-plane.com/mars.html

12 http://www.x-plane.com/FTD.html

13 http://thiagi.com

14 http://www.fablusi.com

15 http://www.engines4ed.org/hyperbook/misc/rcs.html

16 Schank, R. 1992. Goal-based scenarios. Chicago: Northwestern University Institute for the Learning Sciences. http://cogprints.org/624/00/V11ANSEK.html

17 Mary Noggle (2005) A Novel Simulation for the Literature Classroom, online: http://www.simplay.net/LOW/papers05/novel_simulation.pdf

18 http://www.fablusi.com/renderer/default.asp?simID=rps2_94osy-1jpg

19 Som Naidu, Albert Ip & Roni Linser (2000), Dynamic Goal-Based Role-Play Simulation on the Web: A Case Study, online: http://ifets.ieee.org/periodical/vol_3_2000/b05.html

20 Albert Ip, Roni Linser & Som Naidu (2001), Simulated Worlds: Rapid Generation of Web-Based Role-Play, online http://ausweb.scu.edu.au/aw01/papers/refereed/ip/

1 comment:

Albert Ip said...

Stepen Downes in his OLDaily has comment on this essay. I am adding it here as a reminder for my next revision. Thanks Stephen.
This paper is a draft, but I think it has significant issues to address. I have played both Sim City and flight simulators and would have to say that these are the most non-standard 'computer games' out there. They certainly do not resemble 'most off the shelf' commercial games, and they differ precisely in that they do not have any specific game objectives. To cit them, and to then leap directly to the other extreme, Thiagi's low-tech games, is to be misleading about the potential of games. What about games with variable objectives and mod-packs, like Civilization or even Lego Mindstorms?

There are many other comments from ITForum as well. Thanks to everyone. I will be incorporating (and addressing these issues) in a next revision.