jump to navigation

Scot Osterweil – Designing Games that Engage and Educate October 2, 2007

Posted by ficial in conference, games, NERCOMP, NERCOMP20071001Games.

Conference Report: Nercomp – Learning From Video Games: Designing Digital Curriculums

ETA: a copy of Scot’s presentation can be found at: http://www.nercomp.org/data/media/ScotO_10.01.07.pdf

The first presentation of the day was by Scot Osterweil, inventor of Zoombinis video game and director of the Education Arcade at MIT.

Scot is a self-described game designer. Also, he says “I will assert things I haven’t proven” – in some cases he will be presenting his [educated] opinion.

Education arcade designs learning games. The games they design are aimed at upper elementary through secondary. Scot believes that what applies to younger learners also applies to older learners – the same issues apply. The interesting issue is that for the 2nd or 3rd time in the history of computers people are getting very excited about games in learning. The last few years have seen lots of research and talk. The trend in generally in favor of the idea that there’s learning in games. BUT, the notion of game as an educational tool is faulty.

  1. Learning happens in play, BUT
  2. Learning (at least of the desired sort) doesn’t AUTOMATICALLY happen

Just as it’s easy to write a bad book it’s easy to design a bad game. Also, calling an activity a game doesn’t guarantee that anything worthwhile will happen.

Play is prevalent through the animal kingdom, from kittens and puppies wrestling to kids playing with dolls, or discovering gravity by dropping a sippy cup. He quotes from Johann Huzinga (Homo Ludens), who posited that civilization emerged from play. Worth reading the full text, if one is into that sort of thing.

He then gives an interesting example from The Children’s Machine by Seymour Papert 1993. In this example children use rods and clamps to get a ball from one end of a table to the other [I think that’s what the goal was, but it’s not really relevant in any case]. There were three group

  • group 1, instructed on how to use rods and clamps, then given a goal
  • group 2, plays with rods and clamps, then given a goal
  • group 3, control, just given goal

Group 2 got the goal most quickly. The point of the example is that sometimes explaining things doesn’t give as much benefit as simple playing with them.

He gave another example from his own childhood of discovering the principles of addition (though he didn’t know what it was at the time) by playing with block. The revelation was that two small squares put together were the same size as a rectangle, and two rectangles put together were the same size as a long piece, and finally that was the same size as four of the small square pieces. That understanding was later helpful when he was being taught addition in school.

The larger principle at work here is that through informal play we create scaffolds for the things we later learn more formally in school and life.

Play has no agenda – players’ motivations are entirely intrinsic and personal. He posits [possibly backed up? Though up front about presenting some of his own thoughts he wasn’t exactly clear about which parts were opinion and which not] that for play to be play there are four necessary [but not sufficient, I think] attributes, which he calls the four freedoms of play. These freedoms are:

  1. 1. freedom to experiment – players must be able to do (or not do) what they like
  2. 2. freedom to fail – players learn in the process of failure
  3. 3. freedom to try on identities – players need sometimes to be serious and sometime less so, play is a safe environment to try different ideas
  4. 4. freedom of effort – (see Peter Iope, studied schoolyard play, kids culture) players need the choice to try hard or relax our effort, they can’t enter into play with the idea that they must play hard all the time (even though they often do).

Important point: fun isn’t non-stop mirth and giggles, it can be hard work. Example in golf: It’s very hard, especially at first. Miss the ball, then barely hit it, then hit it badly, then don’t hit it well enough. People voluntarily accept difficult strictures for ‘fun’, BUT only if we can be playful. Part of the play is figuring out what club to use when, etc. (there are a continuous regime of challenges as ones mastery increases).

He goes on to say the Four Freedoms of Pay == The Four Freedoms of Learning, BUT, not the four freedoms of school as currently implemented. We haven’t figured out how to give students enough space in their learning to fail or try out different identities. There’s not much encouragement on any level of education to be an experimental learner.

Play has no agenda (motivation is intrinsic and personal). So, how do we channel play into learning activities? That’s where games come in. The promise of games is that through real play the player will build new cognitive structures and ideas of substance. With the new mania for games there’s a tendency to stuff facts/content into a game – e.g. Grand Theft Calculus. [meet the new educational game, same as the old educational game…] Without playfulness a game is just going through the motions. They look like a game but they’re really just dressed up quizzes,

Spelling Bee vs Scrabble – spelling bees aren’t really a game, they’re a contest based on memory. In scrabble it helps to know arcane words, but it’s not the whole game. You’re shuffling tiles, and your PLAYING. Discussion happens on questionable plays, etc. There are little victories inside the game (good triple score, good word, personal best, etc.). There’s lots of ways to have some success, and there’s room to fail. [though it’s worth noting that some people dislike scrabble because it feels too much like just a spelling contest… different folks like different things].

Another example: Most college physics major can enumerate all the forces of motion, but when asked to list all the forces affecting a ball in flight they can’t – i.e. they don’t realize that what they’ve learned applies in the real world. A football player doesn’t know it formally either, BUT they do know it intuitively. This isn’t to say that everyone should play football, but take that physics major outside and throw a ball around while talking about gravity and parabolas, etc. It’s not a game per se, but it introduces experiment, experience, and playfulness into learning.

How should we think about learning games

  • Games should engage players with reasoning and process relevant to their studies
    • logic
    • ethics
    • design
    • scientific inquiry (“all games are about scientific inquiry” [I’m not yet convinced, but I could be..])
    • historical inquiry
  • Games should engage with places, ideas, and themes that matter
    • discussion of Huck Finn – it’s entertainment, BUT it’s also a vehicle for literature, history, ethics, etc. It engages the readers with ideas and is useful even though the exact content may not match
    • Civilization, SimCity – has a problem in that the game is a black box and people might get the wrong idea (i.e. mistake the model for reality, also the model incorporates elements meant to increase play value), HOWEVER, they still are useful despite their limits. Dissecting the black box can also be useful

Then Scot walked us through a quick(-ish) demo of a game with educational components. The game is one he designed about 10 years ago. It was published and has sold over a million copies. It’s called The Zoombini’s Logical Journey, though that was a decision on the part of the publisher and not at all the name Scot would have chosen (he’d have preferred something more poetic and fun). It is marketed (and themed) to kids, but it’s really 8-adult. Through out the demo we were to keep in mind: Narrative, Activity, and Structure

The frame tale is that the happy Zoombinis were exploited by Bloats. The Zoombinis escape and the player takes up the game at that point. The lead in story makes the players become invested in the Zoombinis. Also note: Zoombinis are small, plucky, and determined – matches target audience. Through the process of making little Zoombinis players learn about data structures, DB design, basic combinatorics, etc. HOWEVER, it’s not explicit! Players just play around, and they figure it out implicitly.

The first puzzle in the game involves crossing a chasm. The puzzle is presented with out direction, just an image and users can mess around. The image suggests that the Zoombinis need to get across this chasm using one of two bridges. The UI is very intuitive and has lots of cues. The player has to figure out which Zoombinis can safely cross which bridge. This is a general principle/form of the game: the player is presented with unknown, arbitrary rules and needs to figure them out. This is a good analogy for life in general. This promotes strategic thinking.

Some key features/points:

  • The game lets you fail – partial success, there’s some reward for effort and making progress, but doesn’t encourage failure.
  • Plus the game progresses in difficutly as the players improves.
  • There are rewards for high success.
  • It differentiates between effort reward and success reward.
  • It’s VERY easy to get invested. The demo was only about 5-10 minutes, but even in that short time pretty much the entire audience was drawn in.

NOTE: Players may learn unintended strategies and meta gaming – e.g. make all the Zoombinis the same and many of the puzzles get a lot easier (though some become impossible).

The structure of the game requires clarity of understanding for successful completion. This games does get complex enough to be quite useful at the college level. The relevant part is the structure and design of the game, not the specific content.

A big criticism of games in education – does transfer [from skills/ideas in game to school or the rest of the world] actually occur? No (formal) idea, no studies have done.

“The best place for games is as preparation for formal learning”; Games are a scaffold.

Then discussion (the talk had shifted towards a more interactive format at this point) moved more to game design. Scot posited his theory/views on play styles:

  • m-style, very cometitive, score/progress driven
  • f-style is more character and story driven

HOWEVER, though styles are different the quality of play is largely the same. Being competitive means lots of different things to different kids. Studies on girls not liking competition may be flawed. E.g. when any/many characters die as a necessity in the course of play, that puts of many f-style players. The m- and f- of course correspond to male and female, but it’s not a strict division.

No one likes setbacks, BUT that [alone] doesn’t mean difficult should be eliminated. “I didn’t like the part where I couldn’t get them across the bridge” doesn’t means that part should be removed. It’s a communication failure/mistake between the person asking the question and the person answering. If it was all easy and all always successful then that wouldn’t be fun either.

The game was designed largely in paper and pencil. Designer acts as the computer. Can use some simple tools for prototyping. Can use flash for actual game these days. Flashy 3D is a requirement for a very specific set of games (NOTE: not even gamers, just the games themselves). 2D is fine for many/most.

Then, we ran out of time. Too bad, I would have enjoyed hearing more from him, on games in education, on game design, or both. The MIT Education Arcade seems very cool.



1. Regina - October 2, 2007

Very informative. It is much easier to learn if it is fun. I like that point on fun can be hard work. It is just that it does not feel so.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: