We start out talking about using the clicker to teach overhead obstacles. Let me make a confession here: I still teach overheads the same way I always have. I still walk into the overhead and scare the dog with it a little. Yes, even the first time. It's because I'm good at it and I know how to do it just right, like the escalator. I am mediocre at training a lot of things. But I'm good at training overheads and the escalators. I'm simultaneously afraid to change anything about the way I teach overheads because it works, and hopeful that there is a better way I can teach it that is kinder to the dog but no less effective for the client. I care about and respect the dogs very much, but at the end of the day I do this job for people, not for dogs, and I never want to sacrifice any performance for the client just to make the dog's job easier.
GDB has the same type of overhead obstacle that Leader has. It's a vertical piece of something like PVC, about six feet tall with an adjustable, horizontal, hinged arm of PVC that you can raise or lower to any height. They have it permanently set up to the side of a sidewalk on campus, whereas ours lives in the garage of the downtown training building and has to be lugged out to the sidewalk every time we want to use it. I wish we had a permanent obstacle course set up on our campus. Maybe some day we can!
The goal of the first session with the overhead is for the dog to confidently stop at the overhead and hold its position while the handler reaches out and touches the overhead. It is accomplished like this: on the first approach, the bar is lowered just to the dog's nose level. Most dogs will, out of curiosity, reach out and sniff it. This gets clicked and treated, as many times as the dog will do it. Then you raise the bar a few inches, still with the dog on leash, back up a couple steps, and repeat the approach. The bar is higher now but the dog is still able to touch it. After a few repetitions of this, you leave the bar at the same height, but back up a little further and reapproach, holding the harness handle this time.
The thing that is stressed repeatedly throughout this exercise is something I never would have thought of. I'm putting it in all caps because it never would have even crossed my mind: DON'T CLICK WHEN THE DOG IS LOOKING AT YOU, EVEN IF IT HAS STOPPED. CLICK WHEN THE DOG IS LOOKING AT THE OVERHEAD. I am pretty sure that nine out of ten or maybe ten out of ten trainers, if instructed "Use the clicker to teach the dog to stop to indicate an overhead obstacle", would have clicked a stop no matter where the dog was looking. I know I would have. But it totally makes sense to click the dog for looking at the overhead. If you just click the stop when the dog is looking at you, I can predict that you would get a dog that does not go all the way up to the overhead because it's looking back at you in anticipation of the treat. Of course you could always hup the dog up to the obstacle, and you would eventually get there, but isn't it cleaner if the dog keeps its focus on the obstacle the whole time, and takes you right up to it? Of course it is, and that is just one more example of why this conference is totally worth it. Someone else has noticed all these little details and is willing to share them! There are all kinds of other little tips like this. For example, when you're touching the overhead to show the dog you recognize it, don't hold the food in your hand. And when you have identified the overhead and clicked, treated, and praised the dog for stopping at it, you should wait a few seconds to let it sink in before moving the bar out of the way and telling the dog "Forward". All these things are little but when combined, they probably make the difference between doing a clean training job and making things clear to the dog, and doing a so-so training job and relying on the dog to guess exactly what you want it to do.
The next step is to raise the overhead high enough that the dog cannot quite reach it, and click for going up to it and looking at it. Once it is doing that, you go on to a second overhead, the same type as the first, this one lowered to the original height, and repeat the exercise, progressing as the dog shows you it understands. Then you gradually make the overheads higher and introduce different types.
After overhead obstacles, we move on to teaching intelligent disobedience. Intelligent disobedience is when the handler gives a dog a command that, if obeyed, would put the dog and/or the handler in danger. An example would be "Forward" into the path of an oncoming car. GDB teaches this first with obstacles, as follows. They set up a couple of obstacles (like cones, or boards, I think any type of obstacle would be okay) close enough together that the dog will see them as a clearance, but with enough space between them that it's easy for the dog to see the straight line of travel continuing beyond the obstacles. When the dog stops, the stop gets clicked and the dog gets its treat. Then the handler finds the obstacle, touches it, praises, tells the dog "Forward" again, and continues on with the understanding that they are passing through a narrow space.
Watching this, I feel ashamed of myself. Even though I myself always make a point of acknowledging the obstacle with either my hand or my foot before continuing on, I am 100% sure that I have not always taught either apprentices or clients the importance of LETTING THE DOG KNOW YOU KNOW THE OBSTACLE IS THERE. Because, think about if you don't. Say the dog stops and you don't know why. You say, "Good dog!" thinking, he must have stopped for a reason. Then you say "Forward". The dog has already been praised and to the dog, you know the obstacle is there. But when the dog starts walking again, you crash into the obstacle (or stumble off the curb, or whatever). The dog feels bad like it has made a work error, and you also feel like the dog has made a work error because, hey, it let you hit something! Really, if the dog has stopped for something and you haven't acknowledged the reason for the stop, the dog SHOULD refuse the Forward command. I resolve to do better with this from now on.
After a couple repetitions of clicking the stop, you then respond to the stop with "Hup-up" instead of a click, but you click immediately afterwards, before the dog has a chance to continue. After clicking, treat, touch the obstacle, and praise before asking it to move on again. This is the beginning of telling the dog that if you encourage it to move on before you have identified the obstacle, standing still is the right thing to do. It is important that you don't try to get a second intelligent disobedience once you have identified the obstacle with your hand -- that would only confuse the dog, and muddy the waters as to what you are actually trying to teach it.
After this step, the next steps just consist of increasing the duration between the initial stop and the click, and getting progressively more insistent with your verbal and hand signals to continue forward. By the end of this exercise, you have a foundation of the dog recognizing that sometimes a handler will give a command that shouldn't be obeyed.
All of the afternoon is about puppy raising. The last time I raised a puppy for GDB was in 2009-2010, when I was living in Arizona. I remember that at that time they were experimenting with having puppy raisers teach one exercise -- the go-to-mat exercise -- using a verbal marker of "Nice". Since then they have expanded the number of things puppy raisers can use the verbal marker for. They don't have raisers using the clicker for a couple of reasons: one is that the puppy raisers wouldn't always have the clicker with them, and another is that this way the clicker is reserved for formal training. Also, the client can use the verbal marker too. This makes sense to me although I have to say that I have gotten a few Leader puppies in that were familiar with the clicker and I really don't feel like it interfered with their learning process at all. If anything, it helped. But then I don't know anything about which puppy raisers are allowed to use the clicker at Leader either.
When asking raisers to switch their training methods to a focus on positive reinforcement, there is a lot of emphasis put on explaining the end goals, the behaviors the dogs will need to do when matched with clients. In other words, they tell them why these things are important. I remember when I was a puppy raiser as a kid I didn't have a clue why anything was important. In my head the trainers at GDB had magical abilities and if my dog was bad at anything, they would just fix it. I probably would have followed the rules better if I understood the reasons why they were in place. (Or maybe not. I didn't like rules that much when I was a kid. Or now.)
One of the primary jobs of puppy raisers is to create a dog that wants to work for and with people. It is easy to forget that these dogs don't spring from the womb waiting to do our bidding, and don't automatically think that being petted and praised are the best things in the world. One thing about using food is that it is so, so easy to make it all about the food. I know I have been guilty of doing this ever since I started using food to train guide dogs. Raisers and GDMI's (and clients) all have to develop and nurture the dog's acceptance of praise, both verbal and physical, as a secondary reinforcer. It's easier for me to understand and describe the science behind primary reinforcers changing behavior than it is to understand and describe the way that praise as an example of a secondary reinforcer builds and deepens the relationship between a handler and a working dog. I think that some aspects of the relationship will always be outside the bounds of science.
We spend the afternoon in the puppy kennel, watching six- and seven-week-old puppies get their first lessons in taking food gently and also in collar yielding using a ground tether. Who else remembers the old, old days when the dog training books said that puppies shouldn't begin training till they were six months old? Probably that was because that was the age at which their necks were strong enough to tolerate a chain collar. There were a lot of good things about traditional training, but personally I'm glad that training methods continue to evolve and improve and that I'm living and working at a time when there is a different way.