Failure-driven Learning versus “Coloring Inside the Lines”

In my last post, I talked about the most common challenge to the theory of failure-driven learning. In this post, I want to address another frequent challenge, which I call the “coloring inside the lines” challenge.

The gist is something like this: When there is a standard procedure for performing a task, the aim of training on that task should be to drill learners on that procedure. Exposing learners to what happens when the procedure is not followed properly is a waste of time.

Now, I think it’s safe to say that no one believes this when it comes to teaching complex, risky procedures. I’m pretty certain no one would be happy to find out that their pilot, their doctor, or even their plumber had only drilled on standard procedures and never experienced a scenario in which anything went wrong.

So I think the argument is basically that for simple procedures, simply drilling on the procedure will allow learners to reliably perform it without making mistakes.

I think this idea is wrongheaded in a variety of ways, but I’ll confine myself to four:

  1. There is no task so simple that people won’t mess it up.
  2. There is no procedure so foolproof that people won’t deviate from it.
  3. There is no better way to remember a procedure than to see what happens when you don’t do it right.
  4. When something goes wrong, there is no one more lost than a person who only knows the standard procedure.

1. There is no procedure so simple that people won’t mess it up.

Think about a really simple procedure like, say… using an elevator.  Something so simple that pretty much everyone knows exactly how it’s supposed to work. Do people make many mistakes executing this simple procedure? I work in an elevator building, so I can tell you unequivocally: yes they do. Two mistakes in particular stand out, which are sort of mirror images of each other. One is getting on an elevator that’s going the wrong way, the other is getting off at the wrong floor. I see people do one or the other pretty much every day.

Of course, I picked “use an elevator” at random. You can try this thought experiment yourself. Think of a really simple task. Then ask youself if you can think of any common mistakes people make performing that task. My guess is that you will have a hard time thinking of a task so simple that people reliably avoid messing it up.

2. There is no procedure so foolproof that people won’t deviate from it.

While there’s no official standard operating procedure for using an elevator, we can ask whether people make mistakes using elevators because they don’t know what they are supposed to do, or because they deviated from what they know they are supposed to do. And I think the answer is pretty clearly the latter. Think about people who get on elevators going the wrong way. This invariably happens when someone gets off an elevator going the other way on your floor. You make the subconscious assumption that the next elevator arriving will be the one you called, so that when the elevator arrives, you don’t bother to check. Likewise, think about people who get off on the wrong floor. This generally happens when someone else gets on the elevator at a floor before yours. You make the subconscious assumption that the next time the door opens it will be for your floor, so that when the door opens, you don’t bother to check.

No matter how carefully you define correct procedures, people will never execute them robotically. Sometimes that is to the good. But it also means that people will make assumptions, leap to conclusions and cut corners when following a procedure that would have worked fine if done properly. Any training regime that assumes that once learners know the procedure they can be counted on to follow it properly is doomed to fail.

3. There is no better way to remember a procedure than to see what happens when you don’t do it right.

People who just want to stick to the procedure in training are generally assuming that learners will be able to remember the procedure. Unfortunately, “memorization” is a terrible way to remember things. When we learn something by rote memorization, it is usually forgotten, often quickly. Psychologists have measured the phenomenon of forgetting, and characterized it in what is called the “Ebbinghaus Forgetting Curve” ( According to the curve, people forget about half of what they learn by rote in an hour, and 75% of it in a week.

But things are different when you learn from experience. An experience with emotional impact can easily be remembered for the rest of your life. And that’s why when it really matters, we like to rely on people who have actually done the wrong thing and seen what happens. Because, if the result was dramatic enough, they won’t forget the lesson. It’s one thing to remember to do something because it’s step 7a in the standard procedure. It’s another to remember it because you once forgot and caused a catastrophe. This is why we have flight simulators, for example. In flight simulator training you are placed in difficult situations in which failures are likely, and you then experience the results of those failures being played out in an extremely realistic way. Pilots stepping out of flight simulator capsules invariably have elevated heart and perspiration rates, and other symptoms of stress. They also have vivid memories that will stick around a long time, and help keep their passengers safe.

4. There is no one more lost than a person who only knows the standard procedure when something goes wrong.

Standard procedures work great, right up until they don’t. When things go so far wrong that it is no longer clear how to apply the procedure, the person who has learned only the procedure is lost. If you’ve only seen what happens when thing are done right, and not when they are done wrong, you have a fragile kind of understanding.

In general, the steps in a standard procedure are there because, one way or another, they each prevent something from going wrong. When you learn the procedure via a failure-driven methodology, you get insight into why various steps are in the procedure by experiencing what happens when they aren’t done, or aren’t done properly. As a result, you end up with an understanding that is both more robust–because you understand not only how but why–and more easily remembered over time. And that’s why, if it’s worth learning a procedure, it’s worth learning what happens when you fail at it.

Make sure to check out part 1 in the series, Failure Driven Learning vs. Success Driven Learning. And check in next week for part 3.

2 thoughts on “Failure-driven Learning versus “Coloring Inside the Lines”

  1. Hi Gregg,

    Any thoughts on how do you apply “failure-driven learning” when the objective of a training is only to cover “Conceptual information” about the various products in a department. The target audience for this training is employees of various other departments who are planning to switch from Department A to Department B.

Leave a Reply

Your email address will not be published. Required fields are marked *