It's obvious that learning games should work. Not only as we hear in prognostications from James Gee, Marc Prensky, and others, but our own intuition tells us (as well as watching kids), that when learning is done right, it can and should be fun. Learning should be hard fun.
The question has always been, how do you systematically, and reliably, design fun learning? Previous attempts have been pretty hit or miss, at best. Just cramming game designers and educators in a room hasn't worked. What can we do?
A lot, I want to suggest, but to do that, I have to tell a wee bit of a story.
I saw the potential connection between learning and computers as an undergraduate, and ended up designing my own major specializing in that. Back then, it was known as Computer-Based Education. My first job after college was designing and programming educational computer games for the Apple ][, TRS-80 (Model 1, shudder), etc, first in Basic, then in Forth (FaceMaker and Spellicopter were two of the games I created). It set a theme that has continued in my work.
I went back to grad school to get a Ph.D. in applied cognitive science, and then pursued the academic route for a while. Though my passion was now called instructional technology, I was hired to teach interface design. I continued to look at what led to learning, but was also looking at what created engaging experiences (e.g. Brenda Laurel's Computers as Theatre).
At that time, I was approached by the Australian Children Welfare Agencies (I was teaching in Sydney) about designing a game to help kids who grow up without parents learn to live independently. We did so, successfully enough that it was released (and ported to the web as another student project, albeit in the original CGI implementation with all the limitations that implies: http://www.quinnovation.com/quest).
|Web implementation of Quest.|
The interesting part of this is that in the course of designing the game, a realization coalesced. I had already begun to identify the significant components that contributed to an effective learning experience. Extracted from a wide variety of theories of learning, these were the elements that surfaced again and again across the theories.
At the same time, I had also been looking at the elements that led to engaging experiences. Trolling a wide variety of ideas across theatre, writing, games, etc, I again found a series of repeated elements.
The important thing that emerged was an incredible synergy between the two lists, in fact a perfect alignment: the things that make an experience engaging are also those that make for effective learning! Learning can be hard fun!
Which should raise in your minds the question: just what are those elements? Here is the learning list, expecting that you will know the counterparts (I use Quest as an example, but the same would hold true for any game you'd use):
Contextualized - the learning should be in a setting where the learners actions make sense. A story, if you will. Learners learn best when it's in a meaningful context. In Quest, you are released into a city, and you have to survive.
Clear Goal - the learner should have an end state that they are motivated to achieve. (Note that the goal doesn't have to be obvious at the beginning, in fact some research suggests the contrary.) Learners are better able to take action when they have an outcome they know they're trying to achieve. In the Quest game, you're trying to achieve a stable existence: sufficient money, a decent job, and a place to reside.
Appropriate challenge - the level of difficulty has to be beyond the learner's capability, but not so far that the learner can't accomplish the task; learning happens best in the space just beyond the learner's capability where, with some effort and support, they can accomplish the task. Learners learn fastest when the challenge is significant but not impossible. In Quest, you can't easily get a job; there are few, you need credentials for the good ones, and if you lie you'll eventually get caught.
Anchored - the actions that the learner takes have to have a meaningful effect on the outcome. There can't be meaningless actions by the learner after which the story proceeds, but instead there have to be real consequences in the story line of the actions they take. Learners learn best when they're operating in ways they recognize are meaningful. The actions you take will use, or provide you with, money to survive on the streets of Quest.
Relevant - in addition to the actions taken being meaningful to the story, the story and actions have to be meaningful to the learner. We need stories that appeal to their interests and motivations. Learners learn best when the setting is one they viscerally care about. In Quest, learners are very interested in finding ways to survive in the streets they'll soon be facing.
Exploratory - the environment has to have a wide variety of possible choices (or at least a perception of same), and the ability to try different things and explore the internal relationships. Learners learn best when they have to make choices and face the consequences of those choices. In Quest, you can go where you will, and there are contingencies to discover (e.g. you can sign up for the government's support while you search for a job, but you can't actually get the money till you open a bank account).
Active manipulation - a related facet is having the learners active in exploring those relationships, and operating on the world in ways that are similar to the way you operate in the real world and that reflect the story setting. Learners learn best when there is minimal overhead between their intentions and the actions taken to achieve them. In Quest (albeit with its limited interface circa 1995), you have arrows to navigate, and icons to represent basic actions like asking for a job or buying food and drink.
Appropriate feedback - the feedback from the world has to come in a way that makes sense in the world. They need to know they've acted, even if they don't immediately get to know the final outcomes of their action. Learners learn best when they get feedback about how they're doing. In Quest, there's an explicit representation of the state of your wallet, and your health (hunger and sleep). There's also a small coaching agent that can provide hints, and if things get desperate will recommend actions.
Attention-getting - the action can't be totally deterministic, there needs to be some randomness and probability. Total determinism isn't desirable. Learners learn best when their attention and curiosity is maintained. In Quest, random events can happen like opportunities to take drugs or have sex, or getting mugged or having a friend/relative buy you a meal.
None of this should sound new, it should sound very like the elements you know go into a game. What should be of interest is that this list suggests that what you do to design an effective game is also what makes an effective cognitive (or attitudinal) change!
I've now been testing that design framework in a series of applications, mostly for corporate training or higher education (and smaller scale than full console games) such as a project management game. The lessons learned include that the principles work, but that there's a lot of "finesse" behind the application. I've also been finding that the design approach is learnable, though it takes some time to get up to speed with the nuances.
|Project Management Game.|
James Gilmore has suggested that we've moved from an information economy to an experience economy (we pay to have interesting experiences; e.g. theme restaurants, computer games.), and that the next stage is the Transformation economy, where you want experiences that change your skill set or beliefs or understandings of the world. That's what computer games can be, if we have and apply the tools.
Of course, it's not quite that easy. There's depth behind each of these, and an associated design process that systematically integrates these elements. However, knowing that these elements exist and align gives you an understanding of why: Learning can, and should, be hard fun!