Dog Training

Two years ago, my wife and I adopted a dog, Didi (picture at the end of the story). One day, I brought him to a session of dog training.

If you’ve never been to any such sessions, the name is misleading. The dog trainer doesn’t train the dog. She trains the owner to train the dog.

This makes a lot of sense! The dog will see the trainer for a few hours of his life maximum, whereas it has daily contact with their owner. The training is more effective and will produce more sustainable results if she trains the owner than if she trains the dog.

(In case you’re wondering, this is not really a post about dog training but about management consulting and people management; that said, the anecdote is real.)

During the hour I spent with the trainer, a middle-aged woman called Nadia, she taught me to teach the dog a few commands, such as “stay” and “[walk next to my] foot.” The object of the command is irrelevant. The activity is all a pretext to get me to practice training my dog so that she can give me feedback on how I do it.

For example, Nadia asked me to teach Didi to “Stay.”

The theory is very simple. I get the dog to sit (Didi already knew this command), then I tell him “Stay!”, I make a few steps back to create some distance, walk back to the dog, and hand him a piece of meat while saying “Yes, stay!”

The practice is damn hard. I get the dog to sit, he sits, I make a step back, I make a second step back, he gets up, maybe he runs to me, maybe he goes smelling the tree nearby, maybe he barks; either way, I cannot reward him (because he didn’t do what I wanted him to do) and so no learning occurs.

I was about to complain about Didi’s lack of focus when Nadia told me I was doing it wrong.

I was surprised. “What’s wrong? I did exactly what you told me: I gave him the command, I took a few steps back, and then he moved.”

Nadia told me that two steps back were too many at a time when my dog was still young and untrained. If, with two steps back, the result is that he does something I cannot reward, I should try again with only one step back.

So, I tried again. I asked Didi to sit, he did that, and then I told him to stay, I made a step back, Didi stayed, …

At that point, Nadia quickly told me to “Say Yes!, step towards Didi, and hand him the meat.”

I did that, and he swallowed the treat.

“Good,” Nadia reinforced me, “you got Didi to do something right, no matter how small, and you got the opportunity to reward that. Now he knows how to do that, and you can try again, making two steps back. Sometimes, dogs can start learning directly with two steps back; some dogs, even more. But if they can’t, it’s your job to give them smaller and simpler things to do until you get something you can reward. That will not only reinforce the behavior but also get them more engaged about listening to you and getting trained.”

Excellent! I was happy with the results, now eager to try again, this time making two steps back. (In the back of my mind, I noticed that as I was training and reinforcing Didi, Nadia was training and reinforcing me.)

This time, I asked Didi to sit, he sat, I told him to “Stay!”, I made one step back, another step back, he stayed, so I made two steps forward, showed him the meat, he got up, I gave him the meat and told him “Yes, stay!”

Then I turned to Nadia, but she wasn’t happy.

“You trained him to stand up. Now, he thinks that ‘Stay’ means to stand up, because that’s what he did before you rewarded him.”

She was right! I didn’t notice, and couldn’t have noticed unless she told me. In my mind, it was obvious that I was rewarding Didi for staying. But to him, it wasn’t obvious at all.

He didn’t know what’s important. In his mind, he’s doing all kinds of things, staying, looking left, looking at me, standing up, it’s not clear at all what matters and what doesn’t, and it’s my job as his trainer to let him know, by being very specific in what I reward – and expressing this specificity not with words, which might be ignored or misunderstood, but with actions, by being very fast in noticing and rewarding a desired behavior just after it’s done.

The same applied to Nadia’s comments to me. If she had trained me in a classroom environment, she wouldn’t have been able to give me timely and specific feedback. I would not have understood what I was doing wrong or would not have believed her that I was being slow in rewarding my dog. Only by getting me to practice in front of her could Nadia teach me how to train my dog.

Training dogs, training people

As I anticipated, this anecdote (which is fully real!) is also a metaphor for training people. Don’t get me wrong: people are not dogs nor should they be trained as if they were! That said, there is a lot to learn from dog training – because, despite our brains being very different, some learning processes are surprisingly similar.

In particular, the three main lessons are:

– The importance of training the manager. As a management consultant, clients often ask me to train their team. I usually insist on working with the team’s manager too. The reason is that I can very well train the team and teach them some skills or best practices, but what determines whether what I taught will get used in practice and stick over time is whether their manager is on board and knows how his or her actions and feedback impact the team, what behaviors they reinforce, and what they discourage.

– The importance of hands-on training. To properly manage someone, it’s not sufficient to do the right things but also to do them right. Classroom-style training and books are great for teaching the former but they can hardly do the latter. The best way to teach how to do things right is to ask the students to practice something and give them feedback on how they’re doing it.

– The importance of immediate and specific feedback. If you want someone to learn a piece of knowledge, you don’t need to reward them immediately or at all. That’s because knowledge learning happens in the cortex, an area of the brain that is not very sensitive to rewards. Conversely, if you want someone to do something, you must reward them immediately. That’s because doing takes place in other areas of the brain that are much more sensitive to rewards or lack thereof.

People are obviously not dogs, and in no way am I saying that you should train your team as you would train dogs. What I’m saying is that training behaviors is very different from training knowledge, and in a business setting, 99% of the times you’re teaching knowledge you actually want to teach behaviors, and you cannot teach behaviors as you would teach knowledge. Nor teaching the right things is sufficient for behaviors to be adopted; you also need to teach them right.

Similarly, in business settings, 99% of the times an expert is brought in to train a team, their manager should be trained too – otherwise, results won’t sustain over time.

Hopefully, this story helped pass these two concepts.

(share this on Twitter)