|
Shaping Behaviour in Small Steps Clicker training was first developed by marine mammal trainers who shaped performance exclusively with positive reinforcement. In shaping you take a small tendency to perform in a desired way, and by reinforcing that behaviour you gradually shift it towards a more complex behaviour. Dolphin training is the easiest way to view this. You have a dolphin swimming in a tank. You want it to swim through a hoop you have hung in the middle of the tank, so you blow a whistle and throw it a fish every time it turns in the direction of the hoop. By gradually delaying the whistle, you can train the dolphin to swim through the hoop. This is shaping in its pure form, but it is not the only way to use the clicker. The clicker is a BRIDGING signal. It says "yes! Thats exactly the behaviour I wanted. Now I'm going to give you a reward." It doesn't say anything about how that behaviour was created in the first place.
You can wait for the behaviour to occur, or you can use shortcuts that trigger the response you want. For example, in dog training, you don't to just wait for a puppy to sit down and then click it. You lure the behaviour by holding a bit of food above the puppy's head. When the puppy looks up, his haunches sit down. Click! He gets a treat. The food lure is very quickly faded out, and what you are left with is a hand signal that triggers the sit. (If you want to watch an excellent video on clicker training dogs, check out check out Karen Pryor's "Clicker Magic", or Gary Wilkes' videos "Click and Treat" and "On Target". See the links to the Clicker Resources section at http://www.angelfire.com/az/clickryder/kit.html for more information.) This kind of training uses TARGETING to prompt the behaviour. When I first taught my horse to touch a target, I thought it was just an amusing trick. I have since discovered it is an incredibly useful tool that can be applied to a wide variety of situations, including trailer loading, ground tying, leading, obstacle training, and lateral work. Targeting isn't the only shortcut I can use.In horse training we use pressure to trigger the responses we want. For example, I can ask my horse to back up by tapping his front legs with a whip. As soon as he shifts his weight even a little, I'll stop tapping. He'll quickly learn that the way to avoid the tapping is to back up. By definition I'm using a negative reinforcer: an uncomfortable or painful stimulus, which the animal can avoid by changing its behaviour. Negative reinforcers make great "shaping shortcuts", especially when you add the right answer cue of the clicker to them. With the clicker the tap becomes information the horse uses to get to his reinforcement faster. It tells him what we want. "Move away from here, and I'll click you." The horse learns that the whip is not there to intimidate him, but to give him clues to understanding us. With the clicker negative reinforcers lose their adversarial associations and become instead information providers. The backing exercise is very important in the early stages of clicker training. I'm telling the horse that the best way to get the vending machine to work is by stepping away from it. Mugging me for treats won't get it anything. If you have a pushy horse, this is a super way to teach good manners. Can you teach backing without the clicker? Of course you can, but, if you want your horse to understand how to use the clicker for more complex tasks, you have to start with simple exercises. Most horses can benefit from a review of ground manners, so this is a great opportunity to improve your horse's leading skills, and at the same time introduce him to a new tool.
The printed information contained in this fact sheet is kindly provided by JudyRyder Duffy and Alexander Kurland; author of "CLICKER TRAINING FOR YOUR HORSE" GETTING STARTED: CLICKER TRAINING FOR HORSES You can purchase Alex's books by visiting http://www.crisny.org/users/kurlanda |
||||||||||||||||
Copyright 1994 to 2024 Equiworld at Hayfield, Aberdeen, Scotland - 30 years on the web. Archived Version.