A few years ago, I decided to learn to snowboard. This was a painful process. There seemed to be multiple skills to learn. There was the balancing on a flat board that had a tendency to slide, there was the weight shifts that made the board move in the direction you wanted (and sometimes very fast in directions you didn’t) and there was even complex skills like attaching the board to your feet and then standing up. I began by getting some lessons on a dry ski slope and spent lots of time catching an edge and flipping forward onto my face, or backwards onto my (already very bruised) left bum cheek. I kept going, grittily determined to crack this annoying puzzle.
It was a little bit like trying to solve a jigsaw puzzle, but one where you hadn’t been given the picture to aim for, just a bag of tiny pieces and no instructions. The end goal of the task is fairly clear, but to start with, you have to work out a strategy using skills and knowledge you already have. You also have to gain some new skills: working out which bits won’t ever fit together with other bits; deciding what to do with those confusing pieces that have a tab on each of the four sides; trying to find the corners and the edges. All the while, you still don’t have a mental picture of what the end result will be like, although you might have seen other completed jigsaws.
Skill learning for humans and horses can be a bit like solving a jigsaw puzzle with no picture, yet both species manage very well to acquire new skills.
Imagine trying to teach a horse to step sideways when ridden. You want them to understand that although up until now, you’ve always asked them to move forward in a straight line, you now want them to cross their legs over and move forwards and sideways at the same time. You want them to understand not to do it when you don’t ask, and you want them to understand what the signal is for when you want them to start. Besides the new movement skill, they also have to work out your cues. Just as with the jigsaw puzzle, the horse has no mental picture of the end point that you’re trying to reach, they just have lots of little pieces you’ve given them. To the horse, there’s probably no obvious connection between any of these pieces!
In a (human or horse) training session, we try to practice the component skills of what we’re trying to learn. We build new skills onto a foundation of existing skills by combining the parts in a different way, or by adding a variation on an existing skill.
We try training our horses during schooling sessions to both refine and improve existing skills and to add new ones. Sometimes, these sessions can end up frustrating for both horse and human, as we try again and again, sometimes coming very close to what we want to achieve, and then finding we’ve moved even further away from the completed jigsaw. This tantalising process leads to us trying over and over to get just a tiny bit better.
There’s lots of research on learning new things that can help us with this process, and some of it is not at all what you’d expect.
The first thing to think about is that new skills involve two things: acquiring the clever knack of doing the new thing, as well as being physically fit enough to do it. We tend to plug away at both of these at the same time, thinking that they’re inextricably linked and that if we practice enough, we and the horse will be able to do it. But that’s not actually how the brain and body work together to learn new skills. You can get fit and supple enough to begin doing the new skill without ever actually doing it. Unfortunately, the repeated and not quite successful attempts actually make the learning process less efficient and more offputting to the trainee.
This is because there’s a key part of the jigsaw that’s missing. Our brains have two important things to do: they need to remember all the information about the experience of trying to do the new thing that we gain during our practice session (and these include the times we almost managed, the times we were way off and the occasional times when, by accident, we did it exactly right). One part of our brain is expert at storing all these: it’s like the cache on your computer, a temporary store for recent events that you might need to think about later. Each day, this gets all filled up with what you’re doing. If you tried to permanently store all the information you take in each day, your brain’s hard drive would end up full! The magic part is the second part of the brain… It’s the part that has stored all your skills and memories of how to learn – but while useful stuff is stored there, it needs to be uploaded again to be used.
So, picture the situation. You’ve made lots of progress today learning your new skill, and the memory trace is in the temporary cache. If it gets added in a meaningful way to the stuff already stored, you will make an amazing improvement in the skill. But your brain is busy with daily life and you can’t manage to integrate the two. So your brain waits until there’s enough processing power available to integrate the two things. This process happens while we sleep, and is most likely one of the really important reasons why we do sleep. Horses’ brains are exactly the same (in fact, we could propose that every animal that sleeps most likely uses this “memory consolidation” process to learn new things).
So that’s why you often see big improvements between training sessions that are separated by a rest period when the animal sleeps. It certainly worked for my snowboarding – I would almost always start the new lesson – a week after the previous one – looking much more competent than I had at the end of the last lesson. With horses, we sometimes laugh and say “wow, I think she’s been thinking about this since yesterday”. And that’s partly true, although it may not be during the hours of consciousness! What’s happened is that the new information has been matched with the older, stored information and pieces have fallen into place that make the jigsaw start to make sense. This jump between training sessions is called “latent” learning: it’s learning that’s hidden away until there’s a reason for it to be revealed. It can be hidden from one day to the next, or it can remain hidden for a long time until there’s a good reason to use it.
Training sessions are hard brain work. The research shows that a brain during the training process is using lots of energy, and that lots of different parts of the brain are working hard. In contrast, once the overnight storage and processing has taken place, we (and our horses) both look much more skilled while using much smaller brain effort to achieve a better result. Our brains are set up to be as efficient as possible, and to use as little processing power as possible when doing something. This allows us room to learn even more new things.
So there are a few clear tips we can incorporate in our training of new skills. Whether we’re trying to learn to sit to a bouncy trot, or whether our horse is trying to learn to step sideways at the same time as going forwards, it’s much better to focus on the new skill for only a short period of time during a longer training session. That’s because filling the brain with lots of not quite successful attempts makes the overnight process of weeding out the good attempts more difficult, and it’s also because endlessly repeating something without quite getting it right isn’t much fun. The other thing is that we can work on fitness and flexibility separately from the new skill, so we can prepare the body to be able to do the new skill once our brain has sorted out the knack involved. And the final thing is that the new (but hidden or latent) skills tend to be revealed when there’s a good reason for them to appear: the horse is more likely to try using them if they can see a benefit for themselves in the process. I suppose that must be the reason my most successful snowboard runs were the ones that ended up on a sunny terrace with a giant hot chocolate, extra marshmallows, and no bruises!