Computers Can Support Learning by Doing


The notion of learning by doing has been with us for a long time, at least as long as the time-honored tradition of apprenticeship. Nevertheless, learning by doing is underutilized in today's educational system. There are an amalgam of reasons for this, ranging from outdated status distinctions (apprenticeships are for the working class, books are for the upper class) to economics.

When businesses look to hire someone, one of the first questions they ask is whether applicants have relevant experience. In doing this, they are expressing a need for people who have already learned by doing. Schools, however, do not help students get experience. Instead of having students do the tasks we want them to learn about, schools often just tell students about such tasks. They then break down those tasks into pieces that they teach in isolation. When teaching a foreign language, for example, they do not typically start by throwing students into a situation in which they must understand and use that language. Instead, they teach them pronunciation for one unit, how to conjugate verbs for another unit, and how to form grammatical sentences in yet another unit. Students then become enmeshed in the mechanics of the language without really getting a sense of how to use those mechanics to communicate. When it comes time to actually use the language, they have difficulty putting the pieces together.

The Dustin program illustrates that tasks that are currently approached through such piecemeal book-learning may be made both more effective and more fun by being implemented through coached Learning by Doing simulations where students learn a tough task by actually trying to perform that task and then getting support as they need it. Such an approach may be effectively delivered by computers, but is certainly not solely based on them. For example, the work of Alan Collins, John Seely Brown, and Susan E. Newman on "cognitive apprenticeship" shows how basic cognitive skills such as reading may be profitably learned though an apprenticeship to a skilled practitioner.

The major obstacle to implementing Learning by Doing is one of economics. Learning by Doing works best when a student has a bevy of coaches looking over his shoulder, ready to offer suggestions when asked. It is clearly not feasible to assign five or ten (or, really, even one) experts to look over each student's shoulder in the mass education system. Computer-based Learning by Doing systems have the ability to relieve this economic bottleneck. To a useful degree, they are able to "put the expert in the box." The GuSS system illustrates how a range of coaches may be packaged with a complex social simulation to offer a rich educational environment.


Next Story How You Know Things Without Trying

Outline Where am I in the content of the book?


Give Me An Example

Give Me Alternatives

What Led To This?

What Can Be Done?

Give Me Background


Start Over Who Built Engines? Contact EFE Team ILS