When I think about effectively using video games to teach, I start thinking about RPGs (role playing games). This framing partially comes from personal bias -- I play RPGs more than anything else these days -- and partially from half-conscious thought about what RPGs have to offer compared to other types of games.
RPGs offer a number of advantages over other game models, for teaching. First, as Jess pointed out to me the other day, they require minimal dexterity with controllers to play. It's not like, say, Super Mario Brothers (to name one even people who don't play video games should know), fighting games like Tekken, or shooters like Doom, in which hand-eye coordination is required to win. RPGs generally offer a long time in which you can consider and prepare for your next move.
Gee notes that in general, good games work on this kind of an "amplification of input principle," "designed so that they adjust to different levels of play and reward each sort of player, if the player is putting in effort, with some appropriate degree of success." (Gee p 64) This encourages rather than discourages beginning players. Gee also suggests that "in the real world, science often operates by the amplification of input principle;" he sees the very visible reward one gets for tinkering in a lab (new strains of fruit fly with weird characteristics, explosions if chemicals are mixed the right way) as motivating in itself. (Gee p 64) I think current RPGs, with their extensive, diverse opportunities for trial-and-error tinkering with equipment and strategies, suggest possibilities for particularly good amplification of input.
RPGs also allow for gameplay within a narrative, which not only can be compelling in and of itself but also makes for more opportunities for putting knowledge in context, which all of the books I'm reading say is important for recalling knowledge and keeping it from becoming inert (look for the topic to come up repeatedly in both Bransford and Gee.)
Of course, game type is increasingly fluid. RPGs in particular tend to offer a range of games-within-a-game these days. The Final Fantasy series, for example, has added animal raising, animal racing, treasure hunts, puzzles, card games and other games of chance, dexterity side-games, and army battle tactics simulations to the small-party battle simulations of a traditional RPG. RPGs could easily be used to frame simulations (let me know if they have already), offering more opportunities for the player to identify with the characters onscreen than a simulation alone would.
This could be an excellent way to develop the identification with academic communities that Gee wants to see among students. (Whether we all agree that encouraging students to identify with academic communities is a good idea, or not practical, or displays a bourgeois bias -- or whether Gee is even correct that a student in an ideal science class takes on the identity of a scientist -- is a question for the flamers in the peanut gallery ;)) Imagine a game, for example, in which you role play as a scientist, studying microbial development in a lab simulation, and then put what you know to practical use in healing a community (or heck, germ warfare if you prefer) in an RPG scenario.
The limitation of RPGs that I keep bumping up against, however, is that in them, "skills" are only metaphors for code. Gee talks briefly about how one's skills contribute to one's identity (p 52-53). The identity a character builds up in an RPG can be comprised of a range of physical skills (lockpicking, swordsmanship, martial arts, the ability to tame animals, knowledge of how to raise particular plants, and so forth), socially relevant skills (charisma, orientation towards "good" and "evil"), and personal attributes (dexterity, ability to heal). I think in general this is a promising model for encouraging kids to think about themselves and their abilities along multiple dimensions and at multiple levels. I've also been messing with the idea that one could, say, create a game in which players would gain the ability to "see" like a historian, an anthropologist, an architect, a physicist, etc, and use those "lenses" to demonstrate how one might notice and interpret things differently as those different kinds of experts. (See Bransford, Brown, and Cocking's How People Learn, Chapter 2 for more on why an understanding of expertise is important to teaching.)
But the problem is, like I said: skills in RPGs are no more than code. If your character gains skill in lockpicking -- to use one of Gee's examples -- in a game, doors simply open for her when she approaches, rather than remaining impassable. There's no manipulation with the controller, even; no presenting of a lock model on the screen. The player doesn't learn anything about the mechanical principles behind a lock, or about categories of locks, which she can apply to locks she encounters in real life. Game-specific strategies and hand-eye coordination are still the only skills that an RPG really develops in the player,, as far as I can tell.
What a player might even develop in a game model like this is a superficial appreciation for the benefits of lockpicking -- or knowledge of plants, or charisma -- without ever trying the skills out herself. Games continuing to work in this model could end up developing attitudes in students which are profoundly unhealthy for future learning, rather than encouraging positive attitudes towards learning, testing hypotheses, and other benefits which Gee suggests games foster. (I should note that Gee generally does not seem to be suggesting ways in which games can be used for educational purposes. Rather, he is talking about the learning that goes on in games, as they stand, to point up the ways in which contemporary schools fail to present good learning opportunities. It's my own interests which are leading me to apply his ideas to the possibility of developing better educational games.)
There have got to be ways around this problem. One of them is probably incorporating simulations into games, as I suggested earlier. I'm pretty much finding it tricky, though, to come up with ways in which a game might reward or even assess academic ways of thinking. Getting students to transfer skills used in games to real life is a whole 'nother can of worms...
Cited in this essay:
Bransford, John, Ann Brown, and Rodney Cocking. How People Learn.
Gee, James Paul. What Video Games Have To Teach Us About Learning and Literacy. New York: Palgrave MacMillan, 2003.
TrackBack URL for this entry:
http://www.twistedmatrix.com/~gus.twistd/movabletype/mt-tb.cgi/107
I'd say, as a first knee-jerk response, that it is a very bad mistake (and one made by Gee throughout his book) to presume that identification is the mode in which players relate to game characters, even ones they control. Since the parallel idea of identification with literary characters is universally condemned (okay, except for a certain kind of resurgence in psychoanalytic contexts, but that's more complicated), it's surprising how much weight this presumption seems to carry in writing about games. Happily, though, the game literature is not so inadequate here: there is an article which precisely and convincingly refutes identification from the second issue of Game Studies:
http://www.gamestudies.org/0102/newman/
I would argue, in fact, for a stronger version of your claim that "a player might even develop...a superficial appreciation for the benefits of [skills]...without ever trying the skills out herself." I'd suggest two additional arguments which might be added on to this:
1. You can't do anything "yourself" in a video game except play a video game, that is, manipulate a simulation. The degree of transfer between "skill" in manipulating a simulation and skill in manipulating a problem/object/task/concept as a person in the world is not clear, and probably has to do (among other things) with how abstracted and regularized the "real-life" domain is already -- how similar it is to what can be simulated.
2. Conventional RPGs do something even worse to players' thinking about "skills" than just encourage an abstract appreciation for them. They encourage us to think of skills as property, to translate the game-economics of enforced scarcity and their value assumptions into the realm of personal qualities. They encourage us to think of talents and personal variation on the model of commodities. Anyone interested in a more intelligent analysis, and lucid condemnation, of this way of thinking (of which the games are nothing more than a particularly obvious symptom) might see Adorno's _Minima Moralia_, in particular sections 106 (English pages 166-7), 126 and 127 (pp. 196-9), 147 (pp. 228-231).
Posted by: Roger at February 9, 2004 11:39 PM
"2. Conventional RPGs ...encourage us to think of skills as property, to translate the game-economics of enforced scarcity and their value assumptions into the realm of personal qualities. "
James, any comment on that one? This sounds like it touches on some stuff you've covered, if I understand what you've told me correctly...
Posted by: gus at February 10, 2004 12:29 AM
Learning is the opposite of fun.
To professional game designers, the context this discussion is framed in is "player skill" vs. "character skill". A character skill is a commodity which you can buy and sell within the context of the game's rules, as Roger points out. (In some skill systems, you can actually "sell" skills to re-gain experience points, e.g. skill money, to buy different skills with.)
A player skill is something that the actual player has to learn to do. This is near-universally considered a bad thing, although unavoidable. Requiring player skills is especially frowned upon in the console-game market, though there are PC gamers who still want games that are actually intellectually difficult to play.
Just as modern user-interface designers seek to minimize learning how to use the interface, modern game designers seek to minimize learning how to play the game. If a player has to learn something to play the game, that's time that they've wasted which could have been spent just playing. A large contingent of RPG game players are students who are frequently cognitively exhausted from spending a good deal of their time studying and learning; they would actively avoid entertainments that were attempting to force them into more learning before they can relax and enjoy themselves.
Since interactive simulations are sold almost exclusively for entertainment, there is very little incentive to build anything that is meaningful beyond a satisfying story-arc and fun buttons to push.
As long as the real-world economics of game development drive things in this direction, it will be impossible to build a game with high production values which is also educational in a meaningful way. "Educational software" is designed and sold based on existing curricula, which makes it difficult to do anything new there, either.
I'm pretty sure it's possible to build a game, that would involve role-playing, that would be educational and interesting. However, what you're calling "RPGs" here are really "CRPGs" - so called by pen-and-paper gamers because of the vastly different emphasis of computer vs. real RPGs. It would be more accurate to call them statistics-manipulation exercises, although many are attached to a plot to associate a role with your particular statistics. In a good educational game about locksmithing, for example, one might play the role of a locksmith (or a thief) and the gameplay might even be quite abstract and slow, but you'd be manipulating locks rather than numbers. Little of what a typical CRPG fan would consider an "RPG" would be left, although a pen-and-paper gamer might recognize the genre if there were characters which interacted with the player in a somewhat realistic fashion (or, better yet, other players who played other roles).
Posted by: Glyph Lefkowitz at February 10, 2004 1:38 PM
As far as Glyph's point about whether or not there is a potential market for intellectually simulating games goes, I personally find that many current computer games have learning curves that are too steep for me -- given that the skills are totally inapplicable outside the game domain. I can't be bothered spending hours perfecting my arcade game speed reaction skills.
I'd probably be more willing to play games that taught me something interesting that wasn't speedy thumb movements.
But I spend about AU$200 per year on comnputer games, so I'm not even in the running as far as influencing the market goes. As best I can tell, the market is devoted to people who spend hours every day gaming, and buy something like a game a week.
Posted by: Mary at February 10, 2004 5:17 PM
Ignoring market issues from here on in, would it better more feasible to attempt to teach more abstract concepts such as the impact of crime, the (sometimes) competiting and valid demands of group and society or even media literacy through some type of roleplay game?
I'm not entirely convinced by the lockpicking story -- using simulations to teach people about the mechanics of a fairly cheap and common object seems like an expensive way to teach it, especially given how good the simulation would need to be to teach something that a real lock couldn't.
Hence, the power of a simulation might be better spent presenting something that can't be easily physically presented: like the mechanics of society. In a game world, in particular, the powerless could explore what it would be like to have power, and vice versa.
You need to solve the usual problem here of presenting economic or social problems in a sufficiently manipulatable form so that you're not presenting a transparently obvious lecture on politics and economics.
Posted by: Mary at February 10, 2004 5:22 PM
"the power of a simulation might be better spent presenting something that can't be easily physically presented"
That was the motivation for "SimForest," an educational software simulation of forest ecology (disclaimer: I was co-developer of this project for quite a while):
http://ddc.hampshire.edu/simforest/
Forest ecology is difficult to deal with in the real world because "experiments" (cutting down trees and planting them, comparing different locations, soils, etc.) often take decades and/or lots of travel to produce interesting data. This software allowed a fairly rich computational model to be experimented with instead, using student-set parameters (climate, soil quality, available species, etc). And the "glass-box" version that I worked on also exposed the *mathematical model* itself to manipulation by the students, which I think no other educational software does at all, sadly. (Admittedly the software doesn't exactly make it easy... it's very difficult to create a good UI for editing complex equations.)
That, I think, completely avoids the problem of didacticism: if the students don't agree with the model, they can write a new one. Our goal was to allow any formalizable theory of tree growth to be substituted for the one we had (though this one was taken by a Ph.D. botanist from a leading text in the field).
(end digression, end transparent shill for Hampshire research)
Posted by: Roger at February 10, 2004 7:35 PM
More fantastic discussion! Yay!
Mary, yes, it does seem that simulations of more complex systems are the way academia (well, Columbia and Hampshire at least) is heading when it comes to "games" (they start to lose their gaminess when they are just simulations, I think?) I guess these are mostly intended for use in classrooms, and as such will be used by the teacher in such a way as to promote transfer of observations about the simulation to "real-life" material and other situations.
I have this unhealthy fixation with a "slipping it into their food" model, though. The kind of thing Sesame Street does: puts heavy research on how kids learn from media into what its writers think of as a comedy show. Without teacher help, would this model make for effective transfer of, say, sociological ideas like "poverty leads to increased crime"? Probably not. I imagine players would consider that knowledge part of the fantasy world of the game...
And as Glyph (and also my advisor, Chuck) is claiming, doing these kinds of things to games may also make playing not fun.
Which is why we need the Primer ( http://www.amazon.com/exec/obidos/tg/detail/-/0553380966/qid=1076466707/sr=1-1/ref=sr_1_1/104-0254878-6899122?v=glance&s=books , try reading that five times fast). Which Frank ( http://www.columbia.edu/cu/news/media/00/frankMoretti/ ) says he wants to make happen ( http://gus.protest.net/MT2archive/000558.html ), but obviously it's all a long way off.
Posted by: gus at February 10, 2004 9:40 PM
"(they start to lose their gaminess when they are just simulations, I think?)"
I don't think so. But this may depend on where, exactly, we as players draw the boundary line around what constitutes a "game." Maxis didn't used to call their Sim software "games"; they called it "Software Toys." Some people still think SimCity is boring. But there are all kinds of games that don't fit the brain-off, thumbs-a-twitchin' picture that well. I can say from personal experience that it is at least *possible* to interact with complex simulations in a fun game-like way, though this seems to involve generating personal experimental goals which are satisfying to complete, rather than just idly manipulating the software.
Posted by: Roger at February 10, 2004 10:25 PM
I think of SimCity as a game :) And SimCity does in fact, force you to manipulate a bunch of notionally boring things like traffic and pollution. I'm looking forward to it getting to the point where it's modelling down to the individual level, where each person in the city has something like 10 personality traits, and people with a certain propensity to commit crime do so or not depending on police coverage (let's face it, SimCity is very Tipping Point). That may happen in SimCity 5 or 6. Computers must have nearly enough clock cycles.
Some non-sim games also have teaching potential. Battle simulations tend to the fantastical, but Medieval: Total War probably has some claim to realism as regards the makeup of medieval armies and appropriate military tactics for archers, cavalry and so on. It also teaches the national(/imperial?) boundaries of medival Europe, although it doesn't seem so good on the geography.
And now we've about exhausted the games I've played in the last six months.
Posted by: Mary at February 10, 2004 10:42 PM
I think you bring up really interesting points about why the "skills" you gain in RPGs don't actually teach you much, but I'm not sure I (entirely) agree. You may not be learning the actual domain-specific skill - lockpicking, riding, hunting, whatever - but if it's a well-designed game, you do learn something very important. What you learn is how to play the game in a way that the skills you have are effective, and the ones you don't have are minimized. Maybe it won't make you a better thief in real life, but it'll certainly make you better at coming up with strategies and solutions to problems using the skills you have and avoiding the ones you don't.
If anyone's played Deus Ex, that's actually a superb example of this type of learning. It's a FPS, not an RPG, but there are RPG-like elements involving developing different skill trees. You can choose to get better at shooting, at swimming, at electronics, and so on. The player doesn't actually get better at any of these things, but what the player does learn is how to go through the game world in such a way that they're never confronted with a problem they can't solve. Spent all their points on stealth skills? Well, oops, guess they'll have to find a way around the cameras that will reveal their position . . . .
A badly designed game allows players to either max out all their abilities, or doesn't provide challenges where there needs to be strategic choice-making based on the abilities the characters have. A well-designed one forces the player to plan and strategize based on their skill choices, and that's what the player ends up learning how to do. (That's part of why the Final Fantasy games, despite being really fun, are on my bad-game list, by the way: the player's skill choices are only rarely relevant except in battle strategy.)
On a larger-scale note, I've never been a fan of the idea that games are particularly good at teaching domain-specific information. Interactivity is rarely good at delivering specific facts, because the person playing the game wants to be doing something - in an active, not a passive mode. Take Mary's example of learning strategy for archers, cavalry, etc. In the game, the player actually makes strategic decisions, so learning about the strategy is something they are actively doing in the game, not just passively being fed by the game. Remembering the dates of battles, on the other hand, isn't something the interactive model supports. Who really cares if it was 1066 or 1170? The player just wants to go and fight!
So does that mean there's no hope for computer games as delivery systems for educational information? Are we stuck just hoping the meta-cognitive skills (planning, research, evaluation of consequences) will stick? Yes, Virginia, those were rhetorical questions. I do think that it's possible to use games as a content-delivery system, and that the way to do it is to take advantage of the degree to which people get obsessed with these games.
Think about the endless fan-fiction written about these games. The books detailing the fantasy worlds. The giant expensive art tomes. The character models. And it all sells! Now imagine a game set in a realistic historical world. Our Heroes run amuck through, say, Constantinople in 1463 - but the series of historical novels follows them beforehand and after, the official website posts the best stories written about the time period, and a simulation tool or two actually lets you mess with the environment your characters are experiencing in the RPG itself.
Suddenly you have a suite of educational tools with the game providing the emotional impetus to get people involved, and you have people relating to the information in many different ways: the player expects to be actively playing, the reader passively absorbing information, the writer performing, and the simulator strategizing, planning and revising. People learn from all these activities, but they learn different things and they learn them in different ways. If the game can make people excited about doing the things they might do in school, even if it's not overtly educational, it's already one step ahead of most educational games I've seen.
Posted by: Jess at February 11, 2004 3:07 PM