Okay, here is an overview of the steps required to ‘get animal behavior on cue.’ That is animal training. Getting behavior on cue and proofing said behavior for reliability in the real world is precisely what we are doing here.
When we initially teach a new behavior we make things easy: We work in a calm, familiar environment, we use high value treats, we factor out distractions, we shorten distances, we leverage the environment to cleverly elicit the behavior we are wanting to train.
We have to start this way to keep the game moving along at a mentally interesting pace until the behavior is learned. However, if you think about it, you will easily realize that perfect behavior on cue in the living room does not translate into anything at the dog park. In fact, you have to retrain in the kitchen that behavior which was learned in the living room, same goes for the family room, the garage, the back yard, the front driveway, the empty park, the empty dog park, and finally, the dog park. Only after many trials (and errors) and in many locations is a behavior going to be reliable for use in the real world. In addition, if there is more than one handler, each one will have to again retrain each behavior in every one of the environments before you could expect reliability. The bottom line is it takes minutes to train up behavior in a clicker savvy dog, but months to proof the behavior into a reliable one for use in the real world.
Physically and mentally engaging, good exercise, good fun, because if you don’t figure out a way to clearly communicate what behavior you want and when you want it, you will have a dog who is deciding for herself what behavior is working. This likely won’t resemble any behavior you would have chosen.
90 percent of the training I do is over breakfast or dinner. Meals go in my treat bag and are slowly (or not so slowly) negotiated for good behavior over the course of ten minutes to two hours (depending). It is so much fun for animal and handler and so much training can be done in minutes per day it is quite amazing. It is so much better then purposelessly feeding your dog. Don’t you enjoy the relevance of bringing home a pay check or making a contribution? So too does your furry little friend.
Here are the steps you want to know about before you get started.
Don’t say ‘no’, you don’t need to and there is always another thing you can say, besides no, which clarifies the matter (say ‘off,’ ‘sit,’ etc...). If you are concerned that your driver might not see a cyclist turning left, it is much more helpful to calmly say, “Watch out for the cyclist up there” than it would be to scream “Oh my God!” in a blood curdling fashion. It is the same idea. Ask for the behavior you want and be poised to appreciate it (with a reward).
Don’t touch your dog to get behavior. Of course, you can touch your dog for reinforcement like petting after a click. You will have to learn to be clever in setting the the dog and the environment up for success. Touching is not allowed (at least not at first, until we get to behavior where the point has to do with touching or yielding to touch). Don’t push your dog onto the mat. This is so important to understand that the dog, in order to learn and learn well, must make the choice! It will start to happen, but a lot of patience will be required at first.
Introduce release cue early and release every behavior you ask for so the dog learns that sit really means ‘sit until released’ and down really means ‘down until released.’ There is no need to teach stay since sit until released implies a stay.
No ‘Corrections’ of any kind. No choke collars, no prong collars, no shock collars, no instruments of compulsion of any kind. For one, they are rarely utilized appropriately, but most importantly they undermine your true goal of having a well trained dog. A choke collar would only be required if you had the opposite of a well trained dog. We notice, mark, and reward the behavior we like. The world is punishing enough without our yelling “No!” all the time.
Click and Treat the behavior you’re after. Prevent or ignore (depending on the behavior) undesirable behaviors. In this video notice that Willow leaves her post and jumps on the ottoman prior to being asked when I turn around. More importantly, notice that I do nothing about it. I complete my turn and reward the dogs that are still on their posts. Most importantly, notice what Willow then does to rectify the matter on her own accord!
Step 1 - Charging the Clicker
This is a term of art which means to Click for no reason other than to establish that a Click predicts that a treat is coming. You Click, then you Treat. Over and over. Here are some examples of canine students who are learning this very concept:
Chula’s first click:
Step 2 - Click for Behavior
This might be for eye contact or hand targeting at first...it doesn’t really matter. The concept is simply to teach the dog that she is in control of making the world click by virtue of her behavior. Behavior which is clicked and rewarded is more likely to occur again. That is the basic principal throughout.
Here Chula gets treated for touching the target stick:
Step 3 - Getting Behavior
The main ways you learn to get behavior are capturing (clicking for behavior which is already happening), luring (using food or a target stick/mat), and free shaping (clicking for incremental movements toward the desired behavior).
Set the dog up for success. Set the environment up for success. Where are you placing the target mat? Is the dog going towards the treats or away from the treats? Can the furniture be arranged in a more helpful fashion?
Here, you can see that placing the target mat in a location which features a narrower opening in the second half of the video yields a much better result:
Step 4 - Getting Behavior on a Loop
Getting your first behavior on a loop is very exciting indeed. It is the first real break through for many students. Reinforcement of behavior does, indeed, induce said behavior to repeat.
Here is ten minutes before Jake Roush and I left for the vet. The idea was to get ‘On the scale” behavior on cue so we could more easily interact with the scale at the vet’s office by practicing on our own makeshift scale:
Step 5 - Adding the Cue
Notice in the preceding ‘On the scale’ video that the cue is not added until after the behavior has been learned. Only when the offered behavior is up to snuff will the cue then be inserted. As the video suggests, the cue is initially spoken just as the behavior is happening (or 1 second before, even). It is a label, at this point, not a request or command. After half a dozen or so successful interations which are labeled with the cue, you should be ready to request the behavior and likely get it. This would involve doing something else (sit, down, whatever) for a bit then, out of nowhere, request the behavior by speaking the cue and see if the behavior is offered.
Step 6 - Proofing the Cue
As previously mentioned, proofing behavior is the most important thing to learn since you will not have reliable behavior in all circumstances without the most exceptional proofing. “I don’t understand, he did it perfectly at home” is a phrase which is heard often by surprised handlers who didn’t realize that well trained behavior at home doesn’t translate to the park. Behavior must be retrained in every environment and circumstance in which its use will be utilized. Or, at least so many that the animal finally recognizes the familiar cue regardless of the environment or circumstance. We will revisit the issue of proofing early and often.