A real learning curve
WE'VE trod the maze of error round,
Long wandering in the winding glade;
And now the torch of truth is found,
It only shows us where we strayed:
"It's a steep learning curve" you hear that a lot don't you. It's the most basic kind of business speak. A nicer way of saying - "this is actually very difficult to understand." Or - "We don't give our staff any training."
But how often do you see a real learning curve? How often do you get to see any kind of graph that shows you how people actually learn? I can't actually remember ever seeing one and I was in full time education for 20 years - supposedly all that time I was learning stuff. Nobody ever showed me any way, realistically, of checking my progress. Nobody ever told me what to expect. I do remember about how to actually remember the kinds of things that you get taught at school, including the importance of "overlearning" in Kenneth Higbee's book on memory.
I produced this learning curve by playing a video game. Quite an old video game which I've been having a lot of fun playing recently - Gran Turismo on the Sony Playstation. GT has a series of what I believe are called "Challenges." These are various courses that you have to drive around under a certain qualifying time. They start off easy (once you're realised the difference between the hand brake and the footbrake - which took me a while) and guess what? - they get harder!
I've briefly been reading about the idea of gamification - I'm only starting to get the basics. But one of the things that I think is really important is a fast-feedback loop. In a racing game like GT, you definitely get this, because at the end of each attempt, you get your time - as well as a nice friendly message (until you get the qualifying score) - "You've failed"!
As a trainer, I am very interested in learning. So, out of interest, I thought I'd keep track of my progress from being an absolute novice on a particular course to the point where I qualified. Here are some of my thoughts looking at this graph.
I think in future, this graph is the final slide I'm going to put up in my training courses.
In my day job, when I'm not playing games, I'm an Agile project manager. Looking at this graph makes me think that I've probably done sprint/iteration planning (a crucial part of Agile project management) about 89 times. I'm starting to think that maybe after another 89, I'll really have the hang of it.
1) Even when you're getting better, a lot of time you're doing badly. Even when you're doing badly, you're getting better.
The graph is showing an overall pattern of improvement, but you can't necessarily see this from the individual scores. Even when things are getting better, and I'm getting nearer the qualifying time, there are a lot of scores that are nearly as bad as when I started.
2) After periods of progress, for a while, things seem to (and actually do) get worse
Learning for this task seems to come in three phases. There is an initial period where performance seems to be mostly getting better, this carries on until around turn number 29 and then, for about another 31 turns up to about turn 60, things don't really improve, in fact, at some points, they seem to get worse.
3) Conscious use of techniques comes later (and doesn't necessarily bring immediate improvement)
Even though I've driven a lot of the other courses and come up with a technique, which I'm certain everybody comes up with, for getting around corners really quickly (I won't go into detail, but it involves breaking and turning at the same time) I didn't really use this in the "first phase" up to about turn 29. The second phase, up to about turn 60, I spent trying to consciously apply this technique, which I knew would be quicker, if I could only get it to work.
The final phase - from about turn 60 to turn 85 I'd pretty much "got" the simultaneous breaking and turning technique and I was trying to add on to it a refinement - aim for the apex of the corner as you brake and turn. Again, this refinement too about 25 turns before I got it to pay off.
4) Recording poor performance is embarrassing - even if it's only a game
When I was driving around the course and I knew it wasn't going to be a good time, I was tempted to crash (I didn't record that crashes - maybe I should) so that the score wouldn't show. And this was just in a game. Imagine what it would be like in a real work environment!
5) It takes longer than you think it's going to do, and there's a lot more variation
How often are you going to need to do a moderately complicated task before you "get it" - before you can do it to the required standard? Would you have thought 89 times? How often before you can reliably do it to that standard? My guess now, after doing this task and recording it, would be another 89 times.
6) When you get things right - you're not quite sure why, and you're certainly not sure you could do it again
At around turn 60, I had a turn that took 2m 14.5s. I was please, but I also had a queasy feeling that I wouldn't be able to replicate it. And I was right. It was going to take another 25 turns to get near that again, let alone better it.
7) Failure has to be an option (otherwise there's no success)
You have to be able to make mistakes - crash into things, spin round, set off in completely the wrong direction, this is part of having lots of goes. If I had been too scared to make mistakes, I literally wouldn't have gone anywhere. Almost all the time when I made those mistakes, I had a genuine, positive intention.