How Computer Games Help Children Learn: An Interview with David Williamson Shafffer (Part One)

I’ve known David Williamson Shaffer for more than a decade. I was lucky enough to have him as a student in my media theory and methods proseminar back when he was finishing up his PhD at the MIT Media Lab. where he was doing work with Seymor Papert. I’ve reconnected in recent years with Shaffer through his work on games and education.

Shaffer has come out this month with a very important book, How Computer Games Help Children Learn. A colleague of James Paul Gee, Kurt Squire, and Constance Steinkuehler at the University of Wisconsin-Madison, Shaffer has long contributed to our conversations about the pedagogical potentials of computer and video games.

He has especially promoted the idea of epistemic games, which he discusses at some length, in the interview that follows. He is interested in the ways that we can use computer-based games (including games that involve interacting with real people in real spaces) to introduce children to the basic conceptual frameworks that govern various professional practices. For him, this is the most powerful aspect of games-based learning.

His new book makes a powerful case for this mode of teaching, including detailed case studies of games he has developed to cover a range of different professional contexts and academic disciplines and drawing parallels to commercial games already on the market. The writing is accessible and engaging, driven by his own experiences as a classroom teacher and his own passion for helping to reinvent American education.

Over the next two days, I am going to be running this interview with Shaffer. In the first part, he lays out the book’s core premises and in the second, he addresses the debates around serious games more generally.

Your biography in the back of the book lists one of your titles as “game scientist.” So, I suspect the readers might be interested to know what a game scientist does and how you train for such a position. The cynic in me wants to know what the implications are of using scientific language to describe what is essentially a position in the humanities.

There are a few different ways of explaining where the title “Game Scientist” comes from. The most superficial answer is that as we were founding the GAPPS (Games and Professional Practice Simulations) Group here at the University of Wisconsin Advanced Academic Distributed Learning CoLaboratory, we needed to decide what members of the group would be called. The title “Research Scientist” is often used for appointments in research labs that do not grant tenure, so given that we were all studying games someone (I think it might have been me) suggested that Game Scientist would be an appropriate title.

So originally the term was something of an historical artifact.

But I do think that there is some value in referring to the work I do as game science. Games are, as you point out, a forum of human expression, like books, movies, and other things that are studied as “humanities.” But it is also possible to ask scientific questions about books: to study, for example, how people read, or to study the social, economic or psychological impact of a particular kind of book. So we can ask scientific questions about games and peoples’ experiences with them.

In using the term “scientific” here, of course, I am making a statement about research methods, not values. By “scientific” I only mean asking questions that can be answered with empirical data, which can be quantitative data (surveys, brain scans, and the like) or qualitative data (like interviews and observations).

In truth, though, I am not sure that drawing explicit distinctions between the sciences and the humanities is actually all that productive. Nelson Goodman made a strong case decades ago that the similarities between the two are more striking than the differences on a philosophical level: both try to warrant claims about phenomena in the world. This is a point I have made in some of my own writings as well.

All of that having been said, I am a game scientist because the work that I do uses methods of the field of psychology, which is a form of social science.


As a graduate student, you worked with Seymor Papert, among others, at the Media Lab. Papert has written about “hard fun.” In what ways is your new work a theory or application of this concept of “hard fun?”

There are a lot of connections between Seymour’s work and my own. The concept of “hard fun” is one that I talk about in the book, but there are others as well.

Hard fun is, of course, the idea that we take pleasure in accomplishing something difficult: the joy in meeting and mastering a challenge. As a result, when someone is doing something that it hard fun, moment by moment it looks more like “work” than “fun,” but the net effect is pleasurable overall.

The concept is certainly one that applies to almost any good game–not just the games I work with, or games for learning in general. I make this point in my book, and Steven Johnson talks about it in Everything Bad is Good for You as well. Jim Gee talks about the games that work have to be pleasantly frustrating. Good games require a lot of work.

What makes hard fun valuable from an educational point of view is when the challenge you face is worthwhile in some context beyond the game itself. In Seymour’s work, kids who used Logo had to solve problems in differential geometry and computer science to build things they thought were interesting and exciting.

In my work, the challenges are the kinds of problems that professionals face in the real world: engineering design, graphic design, mediation, urban planning, and so on. The games are hard because the problems are hard. But they are fun because it is fun to solve difficult problems that matter, that have no right answer, and that give you a chance to see what it would be like to run the world–or at least some part of it.

So, let’s get to the heart of the matter. What are epistemic games and what value do you think they bring to education?

Simply put: Epistemic games recreate in game form the things that people do in the real world to learn to think in innovative and creative ways about problems that matter.

They are, in other words, role-playing games where players take on the role of being a professional in training–where “professional” in this sense refers not to so-called white collar professions, but to any kind of work in a complex domain that requires the exercise of autonomy and judgment.

Professional training is based, for the most part, on professional practica: times and places where professionals-in-training do supervised work, and then talk with their peers and mentors about what they did and why. Think about internship and residency for doctors, moot court for lawyers, the design studio for architects, capstone courses for engineers and journalists, and so on.

These repeated cycles of action and reflection create a particular kind of professional thinking that Donald Schon (also at MIT, as you know, before he passed away some years ago) characterized as “reflection-in-action”: literally the ability to think and to work at the same time, and thus to do work that requires constant evaluation of the situation and adjustment of the work plan in order to solve non-routine problems.

So epistemic games give players a chance to work on simulations of real problems, and to think about what they are doing–to debrief, if you will–the way professionals do when learning to solve those problems.

The games are “epistemic” because any professional practice has a particular epistemology: a way of justifying actions and warranting claims. To be a professional of some kind means you solve problems in a particular way, and you accept some kinds of solutions as legitimate and not others. The way a doctor argues that removing a patient’s spleen is the “right” thing to do is different than the way a lawyer argues about it. If you’re in the hospital, you probably want to go with the doctor’s way of thinking. If you’re in the courtroom, stick with the lawyer–assuming, of course, that you have both a good doctor and a good lawyer.

Put another way, practica are where new professionals learn the epistemology of their chosen profession–along with the skills, knowledge, and values they need to put that epistemology into practice. Epistemic games recreate those practica in game form so players can learn to think like professionals who solve non-routine problems.

The point, as I emphasize in the book, is not for players to become professionals, but rather to have innovative and creative ways of thinking about real problems as part of their intellectual toolkit.

You discuss a number of these epistemic games in the book. Can you pick one of them and describe how it might contrast to existing school practices in this area?

As you know, the book has two chapters that look at this very question. One chapter looking at history and what it means to think about history–in school, as a real historian, and in a game called The Debating Game. Another chapter looks at mathematics as it is learned in school and in a game called Escher’s World.

I think the history example is an interesting one because the differences are so clear. Sam Wineberg at Stanford University did a lovely study comparing how graduate students in history and high school history students evaluated a collection of historical documents.

What Wineberg found (and here I’m summarizing from my book, which summarizes Wineberg’s study) is that what distinguished the high school students from the historians was not the number of facts that they knew about the American Revolution. Instead, the difference was in their understanding of what it means to think historically. For the students, history is what is written in the textbook, where “facts” are presented free of bias. For the historians, historical inquiry is a system for determining the validity of historical claims based on corroboration of sources in conversation with one another rather than an appeal to a unitary source of truth–it is a way of knowing based on using specific evidence to support claims rather than trying to establish a set of facts that exist without bias.

In the same chapter, I describe a game–The Debating Game–that asks players to think about historical evidence the way historians do… or at least more like the way historians do. The game is described in more detail in the book, but basically in the game players compete in a debate over whether the actions taken by some historical actor or actors were good or bad, selfish or public-spirited, constructive or destructive.

To win the debate, they have to convince the judges of the debate that their interpretation is better than their opponents’ interpretation. To do that, they have to find specific pieces of the historical record to support their position: they have to argue, as Wineberg suggests professional historians do, for the validity of historical claims based on corroboration of sources in conversation with one another rather than an appeal to a unitary source of truth.

The kinds of things that players of the game do are very different than what happens in most high school history classes. (The game has been played by middle school students as well, and there the contrast is even more striking.) Players in the game (debaters and judges) have to write essays where they defend a point of view, rather than take tests where they remember facts or recite received interpretations of events. They work with primary and secondary sources with conflicting viewpoints, rather than a text with one point of view. They make their own interpretations and judgments about arguments and evidence, rather than trying to decode and remember some canonical interpretation. And so on.

So the differences are quite striking: the game is about learning to use the “toolkit” of historical analysis to think for yourself; the class is about learning to give the right answers for a test. Thus the game is more realistic, in a sense, than class is.

A recurring emphasis in your discussion is on the movement from abstract school subjects towards school subjects framed around specific real world professions — the difference between studying math, say, and studying accounting. What’s the case for the use of these professional categories for secondary school education?

As I point out in the book, school is organized around a set of things that are supposedly fundamental ways of knowing–the building blocks of all thinking if you will–which in the case of school are the traditional academic disciplines.

This is a very old view of thinking, going back to ancient Greece. The disciplines were organized a little differently then, but the basic idea was the same: education is about learning some basic ways of thinking out of which all more advanced thinking is formed.

The problem is that a century of study in the psychology of learning suggests that this just isn’t how it works. Complex thinking of the kind that characterizes expertise isn’t simply lots of basic pieces put together. You can’t teach a bunch of facts and skills and then expect that people will reassemble them as needed.

Expertise–indeed anything beyond rudimentary skill–is based on experience working with real problems, and usually quite a lot of experience. So if we want people to learn to think about problems in the real world, they need experience learning how experts solve those problems.

I should add that there isn’t anything wrong, in principle, with having school focus on learning to think like historians or mathematicians, if we decide that these are the kinds of problems kids will really face later in life. But if that’s what we want to do, then we should build games (and by extension design curricula) where players meet simulations of real historical and mathematical problems the way historians and mathematicians do–which is a far cry from what they are doing now.

I’d also want to see the argument made that teaching everyone to be 5 or 6 different flavors of academic is really more useful than learning to think as professionals. What, for example, would our health care system look like if everyone who went to a doctor’s office understood the kinds of questions that the doctor should ask, and the kinds of answers that she or he would use to make decisions? What would our body politic look like if everyone who read a newspaper or listened to talk radio understood how a journalist thinks about stories–and thus what makes it into the news and what doesn’t, and why stories get reported the way they do? How would that kind of education compare to what we have today–or to doing a better job of teaching students to think like biologists or historians?

Comments

  1. Alex Newman says:

    Please tell David I say “hello”. We grew up together.

  2. Alex is being too modest. He’s the one who taught me to play Dungeons and Dragons–so in a way all of this is partly his doing!

  3. I’m trying to absorb the following statement in the interview:

    “In truth, though, I am not sure that drawing explicit distinctions between the sciences and the humanities is actually all that productive. Nelson Goodman made a strong case decades ago that the similarities between the two are more striking than the differences on a philosophical level: both try to warrant claims about phenomena in the world. ”

    Is this not kind of like saying that since both oranges and eggs are roundish, can be peeled, and taste good, thus they are almost the same?!

    Science uses statistics to *disprove* a claim. The social sciences too often relies on a much more limited and potentially misleading tool: Persuasion by the written word.

    I guess I’ll have the pick up the book to find out if it’s “science” or “pseudo-science”. A great game, by the way, would be one that teaches kids the very important distinction between the two. ;)

  4. Peder,

    I do hope you’ll have a chance to read How Computer Games Help Children Learn and decide for yourself whether you are convinced by the research on which it is based. (I should add that a more traditional scientific presentation of the findings is available in the published papers, which are listed and linked to the companion website http://epistemicgames.org.)

    If you are interested more particularly in my comment about the differences (or lack thereof) between science and the humanities, you can look at a paper I coauthored with a statistician and philosopher of science here at the University of Wisconsin, What Good are Statistics that Don’t Generalize? (http://aera.net/uploadedFiles/Journals_and_Publications/Journals/Educational_Researcher/Volume_33_No_9/03_ERv33n9_Shaffer.pdf)

    The paper deals with the status of statistical warrants, and very specifically with the issue you raise here. In any event, there is a long body of work in the philosophy of science (upon which the paper builds) that suggests that the similarities between scientific methods and other methods of argument are more significant than both being roundish, peelable, and tasty. This is not to say that they are the same, but only that the similarities may be more significant in some circumstances than the differences.

    (To continue your analogy, an orange is much closer in shape to a tennis ball than to an egg. However, that fact may not be of much use when considering what to eat for a snack. Similarly, a candy easter egg and an egg are more similar in shape than an egg and an orange, but again, I perfer my children not use that information in determining what to eat for lunch. In other words, in food as in warrants for claims, what counts as important depends on the decision to be made.)

    David