My Clicker Retrieve
|
This clicker trained retrieve is not entirely my own work. It is based loosely on Shirley Chong’s retrieve with some adjustments that I have found useful when training my German Shepherds for Schutzhund. I have included a few short videos of the work with my youngsters, Arawen and Gana. Click on the pictures to view the video. See the videos page for a download link for RealPlayer Index of videos:
First, find a dumbbell the appropriate size for your dog or puppy. Starting with something smaller and lighter than a SchH1 dumbbell will help to increase the speed of learning for a small dog or pup. I wrap the bar of my dumbbells in Vet-Wrap so that the dog has something he can push his teeth into as this encourages a strong grip on the dumbbell. As far as rewards, I use the dog’s regular meals; kibble or chicken necks make excellent rewards and allow for several repetitions. If your dog is not familiar with the clicker, it may be necessary to "charge the clicker" (in other words, teach the dog what the clicker means). This can be done by taking an exercise the dog already understands (e.g., sit) and doing several repetitions with the clicker. The clicker is a marker of the correct behavior but can also be used to mark the end of the exercise (the click is immediately followed by the reward). Alternatively, the clicker can be used to mark the correct behavior and not the end of the exercise - in this case, the reward should only follow the release command. My dogs have usually done sit, down, stand, articles and targets with the clicker before I start the retrieve. I use the clicker marking the end of the exercise. I usually place the bowl out of reach and we race to the bowl once the click has occurred. This movement also helps to free the dog up after each repetition. I do not do any drive work with the dumbbell before I start - the goal at this stage is to get the dog to think about what is being trained rather than to perform the exercise in drive. The reward at this point is the food, not the dumbbell itself. This is particularly effective when training possessive dogs which will not want to bring the dumbbell back if they think it is a toy (much less give it up voluntarily). To begin with, place the dumbbell on the ground, while the dog watches, making sure you have your clicker ready. Your dog will probably look towards the movement, and you can click him very quickly. Walk back towards the dumbbell and watch to see if the dog looks at it. At this point, any dog that has been trained to target or articles, will probably go straight back to the dumbbell. You can reward him for either a look or touch. Try to keep the rewards close together so that the dog gets lots of successful repetitions in close succession. In effect, what you are training is that the dumbbell is a very important item. Note: In the beginning you are looking for very small behaviours towards the full retrieve. Looking at, moving towards, pawing, bumping, licking or biting at the dumbbell all earn a reward. Once the dog begins to offer new behaviors, he is actively learning (using his memory and mental reasoning to work out how he got the reward). This will help you when you continue to the latter parts of the retrieve Watch for your dog getting tired. 5 to 10 successful repetitions are plenty at this point. Often you will see the behavior start to change, the dog will look, then may take a couple of steps towards the dumbbell, if by the 5-10th repetition, he walks all the way up to it by himself, click, and give him the rest of his dinner. This “jackpot” is useful to mark the large steps towards the behavior you are training.
Depending on the dog, having the dog go up to the dumbbell and touch it with his nose should take 2-10 sessions. Dogs with previous targeting experience often are doing this in the first session. Next when the dog has gone out to the dumbbell reliably for at least 2 sessions, we start to ask for more. Spend a couple more sessions only rewarding the dog for touching the bar of the dumbbell, the ends do not get him a reward. Next we are looking for a more determined bump with the teeth, or lick. If your dog starts to get “stuck”, where he spends time offering a behavior he already knows (like sitting in front of you), just move back a step (if he was licking, but is now stuck, go back to rewarding a nose bump or even a look if he is really stuck) for a session or two. I have found it ok to ask for a little more sometimes, and taking a step back does not seem to be detrimental. Obviously your goal is to keep the rate of success very high so the dog does not get stuck, but sometimes it is fun to wait a while and see if your dog will offer anything different.
Once you are getting a reliable lick or teeth bump (I have used peanut butter on the bar to encourage the lick if it is taking too long), you are looking for the dog to try to take the dumbbell in his mouth. This is not actually very far from the lick and teeth bump, and I have not seen too much sticking at this point, but you do want to make sure you can do a session where the dog gets 100% success on licking or teeth bumping. I tend to do 2-4 sessions at one step to get the behavior solid, before I move to the next. Only your dog can really tell you the speed he wants to go. Getting the click timed well also affects your learning speed, but a couple of late ones don’t seem to hamper progress too badly.
At this stage, your dog will probably be taking the dumbbell in his mouth, and maybe lifting or tossing it in the air. Be very careful not to reward the toss, but click when the dumbbell is in the mouth. Then you can begin to click the dog for picking up the dumbbell and taking a step towards you. Don’t expect a sit in front of you, but you can try to catch the dumbbell when you click and the dog releases. This gives him the idea that he is returning the dumbbell to your hand.
At this point I break off from Shirley’s retrieve and concentrate on the hold. This is very important for Schutzhund, we want a nice calm hold. So at this point we are starting the retrieve from the end point – back-chaining. Make sure you have a good straight Hier / Come position in front of you. Hold the dumbbell out at the height of the dog’s mouth, and start at the beginning where you click for looking, then nose bumping, then taking the dumbbell in the dog’s mouth. Keep a hold of the end and begin rewarding for longer and longer time spent with the mouth around the dumbbell. This stage can take a while, but be patient, it WILL happen! Once the dog is reliably holding the dumbbell for 30 seconds with no chomping, you can take a finger off the end, then 2 fingers and so on until you can remove your hand from the dumbbell. Move slowly and don’t take your hand too far away. Gradually you can move your arm to the normal position for competition.
To improve the hold, tap the ends of the dumbbell, switch hands, gently try to pull the dumbbell from the dog’s mouth. If he drops it at any time, start over. Build the strength of the pull slowly. You are looking for the dog to clamp down and pull down on the dumbbell. I have also found that introducing the heavier Schutzhund dumbbells improves the hold. Now we do the last part of the retrieve. You will need a short sit stay. Sit the dog in front, ask him to hold the dumbbell and take a step back, call him to you and as he arrives, tell him to sit. Don’t worry about straightness right now. The dog will probably drop the dumbbell a few times, if he does not have a single success in the first 2-3 tries, you need to go back to the hold.
Once this is going smoothly and you can leave the dog on a sit stay with the dumbbell, move away 10 feet and recall, you can insist on a straight front, and only reward the straight ones. Now we are ready for the full retrieve. Sit your dog beside you and throw the dumbbell, you may want to hold his collar if he has good prey drive, this increases the drive and teaches him not to go until you say. Let the dog go before the dumbbell stops. Again you may have to go back to rewarding a few bumps, but my dogs nearly always pick the dumbbell up at this point, I then take a step or two backwards and encourage them to return. You can’t expect perfect fronts and pick-ups at this point. Throwing the dumbbell so it lands next to a wall will help stop the dog from going around the dumbbell. If the dog is not straight in front, I often take another step back and encourage him in straight before I reward. The complete retrieve needs to be reliable before you start insisting on straight fronts.
Your dog must learn to generalize this behavior to many places, you will need to take a step back when doing a session at a new place, but you can start doing this right from the beginning, as long as there aren’t too many distractions. Begin adding distractions after you have done 10-20 new places and surfaces. Or you may prefer to complete the retrieve at home before you take it to a new place. When you graduate to a larger dumbbell, take a step back also. When you go to a new place, do a couple of basic holds and recalls with holds before you start the full retrieve. Teach your dog all his commands at new places and soon, you will find he will generalize better. If you have already generalized the sit in 50 places, generalizing the retrieve will be easier. Generalization with reward based training is generally a little slower than with other methods. The more times you can work your dog in as many places as possible, the more reliable your retrieve will get. Don’t expect perfection at a new field for at least 6 months. Click Retrieves taught at home in one place typically take 6 weeks with 1-2 sessions per day with a dog which has no prior experience with dumbbells. A dog which has had previous experiences may be a little more challenging and require emphasis on different parts of the retrieve they are having problems with. Often, going all the way back to the beginning and proceeding in the same manner as above will help, sometimes you may have to focus specifically on one part - the hold for instance. Many dogs are sensitive about the handlers hands being close to their face, you can begin without the dumbbell, by touching the dog on the neck, clicking and rewarding, then moving your hand up to the muzzle, then work on the hold, get a good solid hold, and do the same exercise with your hands touching the dog's neck and lastly head and muzzle. Even with dogs without previous experience, I introduce a lot of different things while they are holding the dumbbell, including touching their heads and nose, tapping the dumbbell and lifting it while the dog continues to hold.
|
