Method: Capturing
If you use clicker training, you probably know what I mean by capturing. Capturing means to literally capture, as if you were taking a picture, a desired behavior and marking it with either a "yes!" word or a click from a clicker. Once marked, you give a reward, such as a treat. The dog will begin to do the captured behavior more often as he/she figures out what you want, and you can then add a word to the behavior by saying it as they are in the process of sitting, then marking (click or "yes"/other verbal word) and giving a reward. Gradually, you will be able to say the cue word, and the behavior will follow. This method is usually good for cute or good behaviors that you want your dog to do more often, such as head-tilting. You can capture the behavior you want while watching TV or just sitting around. You see what you want, mark and reward it, and your dog will learn in no time.
Method: Shaping
This is a fairly advanced clicker training method usually reserved for complex tasks, though it can be used for any trick or behavior. Shaping is a way to gradually mold a behavior into what you want. For example, to teach "sit" with shaping, you would mark (click or verbal marker) any small progress toward the dog actually sitting. Maybe his rear dips just a little bit. Mark and reward it. Then for dropping a little further. Then a little further, until eventually he is sitting on the floor. The problem with using this method for sitting is that sitting is a natural behavior that dogs do. You don't need to click gradual steps to get them sitting as they will usually sit on their own and then you might as well be using the capture method. This method is useful for things like closing doors, spin, and turning off the lights. Training these things also employ targeting, which I may write about another time. This method is also very powerful because it makes your dog think about what you want. They have to figure it out on their own and are rewarded for using their brains, making for a very smart and connected dog who is an active participant in learning a behavior.
Method: Luring
This is by far the most popular method for teaching "sit" and for good reason. Most dogs respond well to it and it makes things go much faster than capturing or shaping. It's very simple. You essentially stick a treat in front of your dog's nose and lure it to do what you want. To lure a dog to sit, bring the treat in front of its face and them move it straight back toward the dog's tail over the center of its head. If the treat is too far forward or to either side, you will probably just prompt your dog to continue standing or to follow it. You don't need to lift the treat up, as this will invite the dog to jump for the treat. Just move it back right over the dog's head and the dog will probably plop its rear on the floor. When this happens say "yes", click, or just reward with the treat and praise. Dogs can be lured to do all sorts of things. Although this training is very easy to start with, you must remember to phase out the treat or they'll think they only have to perform when you have a treat in your hand. Do a few repetitions to make sure your dogs knows what you want (no more than 5) and then lure them without a treat in your hand. When the dog sits, show them that you didn't have a treat (open your hand) and then produce a treat from a pouch or from behind your back. This way, they learn that if they do what you say and follow your hand they could get a treat even though you aren't holding one. You can also teach luring behaviors if your dog knows targeting, which I mentioned in the shaping method.
Method: Modeling
This method is probably the least used in clicker training. That's because, unlike the methods above, the dog is touched physically and is less of an active, thinking participant in the training. If you were to push your dog's rear onto the floor to teach "sit", you would be modeling. In modeling, the dog is physically compelled into a certain behavior. You essentially model your dog into whatever position you want by placing them there, much like a doll. This method is slow and does not teach your dog to problem-solve. It can actually hurt your relationship with your dog. Sometimes trainers will physically place a dog into position, but this is usually done rarely and only under certain circumstances. Modeling also doesn't teach your dog to want to learn. The other clicker training methods above foster a desire in your dog for learning and it becomes a fun game.
Other thoughts
What if my dog doesn't do what I ask? Well, the easy answer would be to correct them. However, corrections are starting to go the way of the past. There's a difference between correcting a dog who is just learning a behavior and one who knows the behavior very well and chooses to disobey. If done properly, training and learning can be so rewarding to a dog that they want to obey and will 99% of the time do what you ask. It takes work, but it's possible. The general rule for getting to this point is to set your dog up for success. If they've just learned sit yesterday in your home with great, high value treats, don't expect them to be able to sit in the middle of a busy park tomorrow when you ask (with no treats). It's just not fair. It also wouldn't be fair to ask a kindergartener to write a two page essay after she just learned to write the letters in her name. You have to have the basics before you increase the difficulty. If your dog doesn't do what you ask, you must consider if the criteria you're asking for is too hard. If it is, it's not your dog's fault. You have to continually train and proof and work on the behaviors to be able to get that consistency. If your dog doesn't do what you ask, take a hard look at yourself and ask "what do I need to focus on in my training". If you need to focus on duration (how long your dog can stay), or distance (how far away you can be for your dog to comply), or distractions (how busy the environment is) then you know how to approach your training sessions to take things at slow increments in order to set your dog up for success. Add those criteria little by little. For outdoor distractions, for example, first learn in your home, then in your back yard, then in your front yard, then at a quiet park, then at a busy park, with all sorts of variations and planned practice curve balls in between. Don't be discouraged. Just keep working and it will improve.
As always, if you have any questions, please leave a comment or message.
No comments:
Post a Comment