These easy “tricks” were taught with a clicker. But this kind of training is really interesting when you start working with people.
I thought it might be fun to do some posts on games you can play with kids, only they aren’t the ones you usually think of. The first one that I’m thinking of is not really suitable for very young children, although precocity can make all the difference. It also happens to be a really intense way to experience a small part of what it must be like to be an animal and deal with people!
This game is called “The Training Game” and I got it from Karen Pryor. Pryor, it has to be noted, wrote what I think is probably the single most influential book in the history of animal training (at least for the general public – hey, you can use it to train people), Don’t Shoot The Dog. It’s not a dog training book, per se. The Training Game’s purpose is to get humans to try to understand how to train animals better, but it offers some incredible insights into our own behavior, and not just with animals. And it’s a hell of a lot of fun. Because my audience here is not composed of hard-core animal trainers I’m modifying the game somewhat to avoid all sorts of explanations that would be important in the context of Pryor’s instruction. By all means, read her book, but you don’t need to for our purposes here.
To play the game, here’s what you need:
Your family. Or a group of people. That’s it. Well, let’s add a little signaling device, like a bell, or a horn, or a hand clap or anything else that makes an abrupt, noticeable sound.
We need that item because in this game you don’t get to speak. We are going to train the animal only with positive reinforcement. We will use the bell/horn/clap as a replacement for speech, which is why I specified the abrupt, noticeable sound. Trainers are used to using clickers (a little plastic box that sounds like a cricket), but hardly anybody is going to have one of those around the house and if they do, they probably don’t need this game anyway.
So here’s how it works: we take turns being the animal and the trainer; everybody gets a chance at both roles. We start by having the first “animal” leave the room so we can discuss what Teacher #1 is going to teach her. We pick something simple for our first attempts, maybe on the order of “Let’s have her turn on the light” or “Let’s get her to pick up the telephone”. Seems dumb (and easy!) but it is neither. After we decide what #1 is going to teach the animal, we have the animal come back in the room, and remember: NOBODY TALKS.
Teacher #1, meanwhile, has the bell and is watching the animal. Let’s go with the pick-up-the-phone behavior; as soon as the animal makes even a tiny move in the direction of the phone, Teacher #1 rings/blows/claps/whatever the signal. Once. Then Teacher #1 waits for the animal to get closer to the phone; if the animal moves in the wrong direction nobody says anything. It’s dead quiet. When the animal heads back towards the phone, reward. (In this context, the sound of the signal is a reward; it tells the animal “you did good”.) By now, you can see how this works; it is a series of approximations, with the animal getting closer and closer. A simple task like this example might be very easy but it could also surprise everybody and be a nightmare, and that’s when the game becomes interesting and useful; I think the most valuable lessons are learned by the animal, and they have to do with how lousy humans are at communicating what they want. You really start to feel for the poor dog when you have had to try to figure out what a lousy trainer wants you to do, and that empathy can make you a much better trainer.
Just look what can happen if the trainer signals at the wrong time or if the animal simply misinterprets the signal! The only way to correct the behavior is to wait until a non-error happens and reward it. I promise you, this can be frustrating, but the people who get it, really get it.
As Pryor points out in her book, this kind of technique can be used to reinforce all sorts of behaviors, not just train dogs. But in order to use it you have to understand how it works, and that takes a little bit of practice and some fooling around. And some mistakes along the way. Some people seem to have an innate gift for this and others struggle with it – some of those will probably just quit, which is too bad – immediate buzzkill for the group. So try not to let that happen, but don’t stress over it, either.
You can play this with just two people, of course. It still works but I find it less interesting because of the missing group dynamic. Remember, everybody needs to experience both ends of the process, and nobody gets to talk during the game.
(I must add that when doing this with a real animal you first teach the creature to associate the signal with a food treat or some other thing it likes. With people, when you play this game you don’t have to do that because they can have the concept explained to them and they’ll automatically respond to the “reward” without prior conditioning.)
Okay, holding my car keys isn’t that impressive, but Coulaine learned to do this when she was 12 weeks old. I taught her with a clicker. The Training Game pays off. If you saw the movie Babe, you saw clicker training at work, at least on many of the non-animatronic animals. (The duck carrying the clock on the stairs, for instance.)
I taught her to wear the mask the same way.
Update: you can read about the game on Karen Pryor’s own site.
And finally, here is a very short video of Karloff weaving as I taught him to with the clicker. We could have done this a lot faster if I hadn’t had to take video and avoid stepping on him at the same time.