Last week, I was out of the office to attend the Games for Society and Learning conference (GLS) in Madison Wisconsin. The convention was comprised of educators, researchers, and game developers for the purpose of discovering and discussing the cutting edge in educational games and technology. At the convention an interesting thing happened, several of the Q&A sessions and much of the lunch table conversation turned into heated debates about the role of failure in education. In fact, how to handle failure showed up so often in presentations and keynotes that it became the unofficial theme of the conference. Interestingly enough, these conversations bore a striking resemblance to the debates that broke out of this year’s Game Developers Conference (GDC) on the role of failure in games. Even outside of games and learning, in an age of ‘too big to fail’ and the Greek economy, failure and its treatment seems to be on everyone’s mind.

At GLS the failure debate settled into 2 essential stances. The pro-failure group, comprised largely (but not exclusively) or educators; they argued that failure is a part of ‘the real world’ and that it is essential for teaching students about the consequences of poor performance. The other side of the debate was supported largely by game developers who sought to minimize or remove the sense of failure all together, believing it to be discouraging and counterproductive to play and learning alike. Essentially the topic of failure came down to two distinct viewpoints; failure as the cause of frustration and discouragement, and failure as a learning tool. While I have a background in developing educational games, I am no expert on the educational system; so, I will explore the idea of failure through the lens of video games.

The state of failure in video games

Summary: As games have become more sophisticated they have moved closer into a no-risk no failure model, this has been resisted and counteracted by the retro gaming movement.

Video games began as simple interactions between the player and a few blips on a low resolution screen. The rules were simple and could be described and learned quickly. As games and their players have become more sophisticated, so have their rules. Most games find ways to teach their players in a low-risk environment often referred to as a tutorial level, others still have taken this idea and extended the supporting framework of the tutorial level to the entire length of the game. Game developers want to avoid their players feeling frustrated or helpless, they want their games to feel challenging without feeling like the game is ‘cheating’. In recent years many developers have begun asking if we have gone too far in supporting the player. Many blockbuster games focus on entertaining the player rather than challenging them.

Gone are the days of lengthy games without save systems or checkpoints, lives and tokens are largely a thing of the past. Instruction manuals contain little information on how to play the game and instead tell you the game with walk you through it. While this ‘no fail zone’ attitude of popular games has many educators taking a closer look at how games teach; many popular independent games are directly questioning this trend. Super Meat Boy and VVVVVV have created punishing platformers that delight in their difficulty, Braid and Sword & Sorcery are mind bending puzzlers that encourage discovery and experimentation without giving the player much information to go on. For many players, modern games have stopped being challenging, and completing tasks in these games has become a tiresome chore rather than an invigorating challenge.

Failure as discouragement

Summary: When a player fears the consequences of failure, they are likely to give up and reject the challenge

If I were to ask a random person on the street if failure was a good or a bad thing, most people would likely say it was bad. In life, in school, and in games we are encouraged to succeed and discouraged from failure. It makes sense, failure cannot bring you all of the good things in life that success does, and otherwise it wouldn’t be failure. A popularly cited study from the European Journal of Social Psychology found that students who had their papers corrected by a red pen scored lower than peers who were corrected by other colors. When people experience failure, they often quit or give up. It’s the reason I can’t bring myself to play StarCraft2, I just experience so much failure that I feel I can’t improve or succeed at the game. Fear of failure is often more paralyzing than the failure itself, how many of us have avoided doing something for no other reason than that we thought of the consequences of failure. This kind of failure, where we fear the consequences of failing, yet are not yet equipped to succeed prevents us from learning or growing.

Failure as a learning tool

Summary: When we remove the long term consequences of failure it becomes a powerful tool for correction and improvement.

At the same time, failure can also be an ideal learning tool. Super Meat Boy encourages the player to fail continually in a low consequence environment as the primary method of learning how to navigate their punishing courses. In game development iterative design is seen as the most important tool for improving a game, watch for mistakes, and then correct those mistakes. When hiring artists, most studios want to see work in progress and sketches of final products; it is more important that you can solve problems, than that you don’t make any mistakes. This is the part of the failure curve that many modern games fall on. New Super Mario Bros Wii allows the player to die over and over again, even if you run out of lives you have an infinite number of continues. Forza 3 lets the player rewind the game at any time, so that the player can instantly redo a badly made turn. Ample save states and regenerating health means a player can go back and try a challenge over and over again with no long term penalty for making mistakes.

Failure as a challenge

Summary: Providing a player with consequences for their actions is what makes a challenge thrilling

The problem with using failure in this way as a learning tool is that we remove all the challenge from it. When I was recently playing the otherwise excellent Rayman: Origins, I found myself becoming quickly bored, and acquiring the various collectables tiresome; the problem was with 4 players there wasn’t really any way to lose. When playing a recent Call of Duty game, I found myself continually running mindlessly into the melee because the respawn time was instantaneous and I was reinserted back into the game right next to the action. Have you ever played a favorite arcade game on an emulator, or used a cheat code to give yourself infinite health, the sense of power we first get can be fun, but playing the game quickly becomes a chore rather than a challenge. I remember playing Steel Battalion for the original Xbox, the thing most people remember is that It came with a giant controller with 40 buttons; the thing I remember is that if your mech was destroyed you had a scant few seconds to press the eject button. If you failed to press it in time or if you ejected 3 times or more; your profile was deleted and all of your progress was lost. Needless to say it was amazing, I remember panicking to flip up the protective cover and press eject in time; and when I was out of lives and in a one on one duel with a battleship sweat beaded on my brow as I dodged, fired, and ultimately died.

How do we treat failure?

Summary: This whole paragraph is a summary

To my mind the only difference between failure as learning and failure as discouragement is how we treat failure. When we withhold all of feedback till the end of an exercise, the player can be discouraged by the number of errors made; they can have a difficult time processing all of the behaviors that need to be corrected and often they are told it is too late to fix them. In standardized testing, we spend hours filling in circles and working with information and processes that might be wrong, what’s worse we often don’t find out what mistakes we made until weeks or months later. When failure is short term, and quick feedback is given, the player hasn’t invested too much into that failed effort, they are more able to learn and adjust because they know exactly what they have done wrong and they are able to fix it right away. When we too securely protect a player from failure, the game becomes consequence free and the challenge loses its intensity; and while we all would love to be more supporting of students, who wants a surgeon who thinks it is okay to make a few mistakes here and there as long as they are improving?

Conclusion

So is failure good or bad? Does it belong in the classroom or in video games? Should we give players and students consequence free environments or teach them that consequences are the realities of the world? Like so many things, both answers are correct and an effective compromise can be reached. When a player or student is learning, they should be encouraged to make mistakes, have those mistakes pointed out for them immediately, and give them the opportunity to correct those mistakes instantly. When a player or worker has mastered the subject or game however they don’t need that kind of support. When they experience failure, they are better equipped to learn and improve, and they are better prepared to identify failure before it happens. So the Axiom I propose is this, a novice should be well supported and given ample opportunities to correct their failure in real time, it is only through this process that they become an expert. Once the player or student is an expert in that game or subject, only then can they experience the thrill of overcoming a difficult challenge while shouldering the disappointment of a failure.

TLDR

When introducing something to a player for the first time, let them experience consequence free failure to learn and encourage discovery and experimentation. Once the player is experienced, consequences give their achievements meaning and overcoming the challenge becomes its own reward.