I was training a new behaviour with Gabby yesterday – an agility handling technique that was new to me, too. I had done all of my homework: studied the technique, watched a variety of people execute the handling move on video, and practiced the footwork without my dog. I set up the video camera, warmed Gabby up and started training with her.
Here’s the part where my reality takes a sharp left turn from actual reality. I did the handling sequence a number of times – sometimes Gabby got it right, but most of the time she got it wrong. I stopped and reviewed the video and my notes to make sure my handling was good, the set up of the jumps was good…check, check. So that left Gabby. Why was she only getting my desired terminal behaviour less than 50% of the time? Why was she taking. So. Long.
I videotape almost all of my training sessions because one of my matras is “communicate as clearly as possible to the dog.” Video allows me to see all of the gaps in my communication to my dogs. Sometimes it’s a poorly delivered reinforcement, sometimes it’s weak body language and usually it’s missed clicks. When I reviewed the video, it turned out that I had only been working with Gabby a total of six minutes and forty-three seconds. How did I have had the expectation of introducing a completely new – and complex – skill to Gabby, and have mastery in less than ten minutes! Head shake. It took me longer than ten minutes to learn the skills myself, and I was looking at the finished product. I knew what outcome I was working for. Gabby was just trying to follow my arm cues with no idea what we were doing.
What to consider when training a New Behaviour
Know what you’re doing. I see this all the time in class. I start a sentence with “We’re going to shape your dog to go to his mat” and out of the corner of my eye, someone has tossed the mat down on the floor, and is rewarding their dog for getting on the mat with a triumphant look of “See? We’ve got this one already!” on their face.
All that person heard was “go to mat” and completely ignored “we’re going to shape…”
Whether it is listening to instructions in class, or at a seminar, or working through something yourself – know what you’re doing. Where should your head be? Your eyes? Your arms? What do your feet do? What criteria must the dog meet in order to earn reinforcement? Have you broken this training down into teachable steps, or are you going for the whole enchilada at once?
Reinforcement. Reinforcement is how you pay your dog for playing your game. Pay your dog well for the work he’s doing.
Use a training plan. People often resist doing this – I’m guilty of not having a training plan myself – but I guarantee, if you put together a training plan, your training will go more smoothly. To start with, it tells you where you’re going. If you don’t use a plan when you train, it’s like trying to get to a destination without your GPS.
A training plan also becomes a record for you. It tells you if you are working the left turn more than the right turn. That’s ok if you are trying to improve the left turn, but if it’s because the left turns are better or easier than the right turns (and therefore, more reinforcing for you) then it’s good to have a record of that.
Splitting and Lumping. Think in terms of small progressions. Can you break this behaviour down further to make it clearer for the dog. A perfect example is novice owners who are teaching the Down. Instructions are to hold a piece of food and lure the dog down so that the food ends up between his front paws. For 75% of the dogs in the world, that will get you a Down. But what about the other 25%? You need to reinforce when their elbows bend a few times, and then reinforce body weight shifting backwards, and then look for elbows on the ground.
A skillful trainer is able to see what they have, know what they want, and create “splits” to get them to the terminal behaviour.
Training is fun, or it should be. There should be equal parts of joy in achieving the desired behaviour, and working together to get there.