In this training video, I'm working on something called "Generalization." If we say that a dog has "generalized" something, we mean that the dog understands something in all different contexts and environments. For example, sit means sit not only in the living room but also on the beach, in the car, at the vet, if your back is turned, if you are lying down on the sofa, if your spouse is giving the command instead of you, etc.
Future sports dogs like Halo must have extremely solid generalization skills because they are asked to perform at high levels in brand-new locations with all sorts of distractions. So, what we are working on is generalization.
Here's how generalization goes, at least for a sports dog (there are some slight variations for pet dogs and service dogs.) You train the actual behaviors (front, down, heel, etc) in the house or at your "home training grounds" -- you never try to train a new behavior in a new location.
THEN you go out to a novel environment and initially you do allow the dog some acclimatization. For the sports dog, you want the dog to choose YOU. So you'll see in the video how I'm not asking Halo to work with me, and certainly not requiring it. I'm just making it a really, really good idea. The leash is preventing her from blowing me off completely, but I'm careful to try to use it as neutrally as possible.
Halo actually does start showing some engagement and wanting to work almost right away. This is great. Away we go. The work is easier out here and the rewards are higher. For a sports dog you always want them to be successful; that way they think they can never lose. Halo looks great and everything's fun and jazzy until suddenly a friendly man on a bicycle rides by and Halo, thrilled, gallops to the end of the leash and leaps up enthusiastically!
Well! First off, at the end of the day, I work with so many aggressive and fearful dogs that I take this naughtiness with a grain of salt. I'm HAPPY that she's friendly and enthusiastic to the man on a bicycle. Do you notice that he's also wearing a hat? Friendly dog, that Halo! But obviously I don't actually like the jumping-up behavior. So how would I deal with it?
When you're working out in the environment, there's some things that you can't control. I can't control whether a friendly bicyclist appears, but I CAN control whether I allow Halo to get reinforced by it. If she'd actually jumped all the way up on the man and then he'd petted her, that would have reinforced the jumping up. In training, what you DON'T reinforce is often as important as what you DO reinforce.
You'll see that as the man rides away, I'm backing up several steps with a visible treat lure. I rarely lure dogs while doing distraction work, but in this case, Halo needed some extra help as she'd never really done distraction work with bicycles and friendly strangers.
Generalization work (which was the point of today's lesson) is different from distraction work, so once the distraction was gone I made a mental note to go ahead and start distraction work any day now. Keep watching the videos and soon you'll see some bona fide distraction work training!
Comments