General Training

Training Contacts

At "popular" method of target training on contacts is to put some sort of treat on the slats within the contact zone, so the dog has to slow down, look down and get the treat. A variant of this is to get the dog to slow up and give him/her a treat by hand. The command "easy" or similar commands work well here; when you want the dog to slow up, use the "easy" command.

The other method is to use mounted hula hoops to help with targeting. This method really works well, especially if your club does not allow food on the equipment. Make them like this: first, cut a hula hoop in one place using a Dremel tool (or a hacksaw). Then, drill holes in a piece of 2x4 at a shallow angle using a 3/4" (or whatever size is appropriate for your hoops) hole saw, and simply stick the ends of the hula hoop in the two holes you've made in the board. (Jim Hutchins)

The following is a method that can be used to successfully *start* dogs and to *retrain* dogs. The basic premise is to *pattern* the dog to go up and over the contact as fast as possible, have it ultimately pause imperceptibly in the contact zone and await release (the release is the next obstacle command, usually given as soon as a paw is in the yellow). The dog learns the performance of the obstacle irregardless of the handler's position. Incorporate body language cues and hand signals (e.g. utilize a food target with the method and signal the dog to keep his focus *down* and on the end of the plank by pointing at the end of the plank which means you bend *forward*). If the course permits you to be close to the dog, you can take advantage of the opportunity to use body cues to ensure the correct performance. Because the dogs become *patterned* to going up and over, all the way down into the contact zone, however, you should not have to be present to know that the dog will touch the contact. A "walk up" command can encourage the dog forward if it hesitates to go all the way to the end of the plank in the absence of the handler. Successful performance of Masters Gambles often requires a dog that is reliable on contact obstacles at a distance, therefore it is a valuable skill to have. Valuable but certainly not required in order to be successful.

Again, all dogs are different. Some could care less whether you're leaning forward, leaning back or standing on your head. Some dogs obviously *do* respond to their handler leaning back. Others don't. The leaning back method is most useful with dogs that didn't receive a solid foundation in contact zone training and that have learned that it is rewarding to jump off the contact and hurry on to the next obstacle. With these dogs *any* forward progression of the handler will be used as an excuse to go on.

Use whatever works.There is not any *one* method that is going to be effective for all dogs. If the dog is *rewarded* for being in the contact zone and *corrected* for leaving it, you eventually will have a dog that wants to run down into the contact and not leave until told to. (Linda Mecklenburg)

This is for completely untrained dogs and unfortunately does not work so well on those that have already done agility. You simply stop the dog at the BOTTOM of the contact, you don't say ANYTHING. Once the dog has stopped you can ask it to sit or lie down. This eliminates the urge to shout wait, stop, lie down, etc. as the dog nears the crucial area. The most important thing is to always release the dog from the contact. Even if you only stop for a millisecond ALWAYS say go on or name the next obstacle to release the dog. You mustn't say anything until the dog reaches the contact, so that you (not the dog) won't anticipate the command. (Eleanor Thomas)

Sharon Nelson's targeting method on the contacts (and otherwise) is one of the top, if not best, contact methods. Unfortunately, the term "targeting" has come to mean a number of different methods. Sharon has you teach the dog to "look", "get it" and "come" away from agility obstacles. When used with the contact obstacles, the "look" command is given as the dog approaches the contact zone. The dog has been taught that "look" means to put its head down and look for the target which is up ahead. When the dog puts its head down, it cannot jump off the contact. The dog is then given a "get it"command, which allows the dog to get the goodie off the target. To teach the dog to smoothly proceed down the contact zone, grab the goodie, and IMMEDIATELY continue working, the dog is told to ""come" the moment he starts to get the goodie.

The dog is never allowed to hop off the contact, even a little bit. Because a dog can't figure out why he is a good dog sometimes when he hops off (i.e. when he has made the contact and is praised and rewarded) and when he is a bad dog for the same behavior, hopping off (i.e. when he hasn't made the contact). So hopping off, even a little, is always bad dog, no "get it" command, no treat.

But this is only the first step of Sharon's method. After the dog is reliably doing the exercise with the target at the base of the down ramp, the target with the goodie is moved a little out from the ramp. The dog must still proceed all the way down the ramp and the handler still gives the "look", "get it" and "come" commands at the appropriate times. Slowly the target is moved out from the down ramp, as the dog continues to come all the way down the ramp. The "look" command is given as the dog approaches the contact zone, so he will drop his head while doing the contact zone, and the "get it" command is given as the dog is about to reach the goodie on the target. When the dog can do the contact zone with the target moved out from the down ramp, a jump is placed between the down ramp and the target. The dog now performs the down contact and another obstacle, with the targeting commands given appropriately.

Slowly other obstacles are inserted, with targets placed randomly (to the dog) around a course. Because the dog has been taught from the beginning that he can only look for targets when given the "look" command, he should not be shopping around for targets, rather than focusing on the obstacles.

Probably the best advantage to this method is that you use the same thing in competition as in practice, to wit, the "look" command as the dog descends the ramp. You are being absolutely fair to the dog, because what you do in competition is what you do in practice. The only difference is that the target is at the end, after the dog exits the ring. (Billie Rosen)

In reinforcement/behavior modification theory, a behavior which receives intermittent reinforcement is the most difficult to extinguish. Skinner did a lot of work with his pigeons on this. In terms of targeting this means that really effective use of targeting requires that a handler will, eventually, progress to a point where the dog will not get its reward every time--in training or competition. And it certainly will not always get a reward at the bottom of a contact, etc. If the dog did the contact correctly it may get its reward several obstacles later. If the dog popped the contact, the handler may finish the course by another method in which the dog is never taken to where the target is. Eventually the reward will be intermittent (the dog will get it sometimes--but not always--for correct behavior--but never for wrong behavior). But the belief that a reward will eventually come will be so powerful that the desired behavior will be given.

There are a lot of steps between starting to use a target (or clicking) to teach or shape a behavior, for which a reward will be given every time until the behavior is strongly patterned, and using a target as intermittent reinforcement to sustain the patterned behavior. Targeting is most stunningly effective when it has been used consistently (and reasonably correctly) by a competent trainer from the time a young dog first began agility training.
(Mary Finley)

A combination A-frame/dogwalk is useful for teaching contacts. From one side it looks just like the up ramp of the dog-walk except that the top spreads out a little to match the other side which is about a 2/3 height A-frame board. The idea is that you get a dog-walk contact zone and an A-frame contact zone in a small home sized package; in the UK they are selling for about 185 pounds which is a bit less than the cost of either a dog-walk or A-frame. They are not a total substitute for the full size equipment but they do help with the contact problems.
(Tony Dickinson)

Contact zones -- the bane of existence for a LARGE percentage of agility competitors. Think of what your body posture on contact zones is saying to the dog. Many people lean forward and point to the ground or to the end of the contact zone in an effort to get the dog to focus on the zone. This CAN work, however, it can also backfire.

When you lean forward, you are saying to the dog with your body posture "Mom's gonna run!" and the dog launches. Think about it. When you run, your body is leaning forward. And when you've been going slow and are preparing to run, the first thing you do is lean forward. At a contact zone, concentrate on slowing down, taking smaller steps and STANDING UP STRAIGHT! You may even have to go so far as to lean slightly backwards.

Another problem that may result when you lean over and point to the zone is the dog sees your arm and obligingly jumps over it, missing the contact! This happens a LOT! Usually it doesn't take a handler very long to either realize this problem on his/her own or have someone point it out to him.

Early on in agility, the "accepted" method of training contacts was to stop the dog short of the contact zone with a "wait" command, then a "walk on." This method has since kind of gone by the wayside, mainly because it created its own set of problems. What happens in this technique? One thing is you are wasting a tremendous amount of time having the dog wait at the top of the A-frame, or at the beginning of the down plank of the dogwalk. If you are constantly fighting a time problem you sure don't need this!

Another thing that happens is, especially with a fast, highly motivated dog, they get so frustrated and have too much pent-up energy when stopped that when they are "released" they launch like a pebble out of a slingshot and still miss the contact zone. If you're going to stop your dog on a contact obstacle, stop it IN THE CONTACT ZONE -- not short of it.

The best approach is, when training, to teach the dog that the contact obstacle is one obstacle -- you go up from the beginning of it and you go down until the end of it. Calmly, reasonably and without making a big deal out of it. In practice, you might think about not even painting contact zones on your practice equipment -- it forces dog and handler to concentrate on correctly performing the ENTIRE piece of equipment.

Above all, in practice, don't ever let the dog get away with "one toenail in the yellow". It doesn't take very long for the one toenail to become no toenails. (Jo Ann Mather)

To teach the contacts to beginning handlers and dogs, you must teach targeting the first week. This is taught as a separate exercise away from the contact equipment.

Just for fun and initial exposure to the contact equipment, have the beginners dogs walk over planks on the ground and a 1' high A frame--no stopping at the bottom--all.

Week 2, when the dogs come back most can touch their nose to the target on cue. Now start backchaining.

To prevent having to pick up big dogs like newfies and great danes, use the pause table along side of the lowered (now 2-3') A-frame.

The dogs hop up on the table; have someone hold their leash so the dog is one rung up away from the floor. The target is positioned depending on the body length of the dog (further away for bigger dogs, closer for smaller) so that they have to step off the A frame with their front feet to touch the target but have their back feet are still up on the frame as they are touching their target.

Have them touch the target several times, including twofers and threefers etc. prior to being released. At first it is easier for the handler to stand in front of the target with their feet stradling it Charlie Chaplin style--but fade this prior to starting to backchain across the frame.

For homework have your students work on this behavior on a flight of stairs. The student should sit on the second to last step about a dog's body width away from the wall. They encourage the dog to start on the first step with their target away from them on the landing and on cue have the dog go and touch.

By sitting next to the wall you create a barrier which stops the dog's butt end from swinging around so you only get their front feet on the floor not the back feet as well. This homework works best if the landing (where the target is positioned) is a different floor covering than the stairs as you want to simulate an environment where the dog's front feet are on a different surface then their back feet (i.e. grass and the a frame or matting and the a frame).

If the stairs and carpet are both carpet they can simply put a piece of floor matting or cardboard down to (secure it so it does slide out from under the dogs feet though).

Remind the handler not to touch their dog during this exercise. The dogs need to figure things out for themselves.

Gradually reduce the help of the handler sitting next to the wall. First they move further away from the wall and eventually are standing in front, beside and anywhere so the dog doesn't rely on their body position as a superstitious cue to complete the behavior.

The handler can backchain up the stairs at the same rate they are backchaining up the A frame in class.

Use two commands; "frame" which directs them to the obstacle and "target" which makes them drive with great speed to the position they know best--their target position. The word "frame" is not introduced until the whole behavior has been completely backchained.

(Susan Garrett)

A soft, whispered, positive "wait" command can be used to break or shorten stride before an up contact or a too fast approach to (or slippery) table. It is used as a half halt with a horse would be used. The "wait" would be given and then an immediate continue on type of command. So you're not bringing the dog to a stop that would be judged as a refusal but, rather, creating a brief shortening or collection of his stride. (Pati Hatfield)

For a flyball trained dog, the "hit it" command can be used to get the dog to touch the up contact.
(Georgia Thomas)

In addition to the "easy" command, some groups find that a hoop on the upside makes the dog think about the ascent a little more and touch the yellow rather than leap into the middle of the obstacle. Train the behavior "go through the hoop" away from the obstacle, as the springy dogs might jump the obstacle too- and use that command to signal the ascent of the A-frame. If the hoop is shoulder-height on the dog and the dog has to lower his/her head to pass through it, this may be enough of a posture change to get the upside contact. You can build hoops out of plastic tubing (Builder's Square, Home Depot, etc) striped with tape for visibility. Attach the tubing to a scrap of lumber -- 1x4' X 2'. The idea is to make it so soft and light that if the dog gets tangled up in it the hoop will give and not hurt the dog.
(Liby Messler)

Any variation of the pause-in-the-contact method requires a *lot* of repetition in order to really solidify the concept in the dog's brain. While the dog may voluntary stop in the zone after a few sessions and we think that they "get it", it really requires 4 to 6 months to really build confidence and ensure that the dog knows what the job is and exactly *where* he is supposed to stop. With confidence comes more speed.

Many dogs new to agility typically don't have this same problem whereas dogs that have previously been doing contacts may slow down and creep in response to some confusion on their part about "the job". Some of us induce the creeping problems by turning the job into too much of a control exercise -- for example, asking for very long waits and such in the zone. While it performing contacts needs to be done with control, you can push the dog into being too careful and worried about making a mistake. Here are a couple of things to help the process along:

1) For people who are working on speeding up the contacts in training, but are still competing with the dog and concerned about time spent on contacts, teach your dog to hit your open hand with his nose on command. This can be taught easily with a clicker. In competition, put your hand at the bottom of the contact obstacle and tell the dog "Target". The dog will rush to "bop" your hand with her nose, but withdraw it just before she gets there and then release to the next obstacle.

2) In training, be very consistent with where it is you're asking the dog to stop on the obstacle. Regardless of whether it's actually in the contact or two feet on the ground as long as you're consistent. Identifying the exact spot you want him to stop.

3) Try using two food targets. Place one treat on the bottom rung of the contact if you're having the dog stop there, or at the base of the A-frame if you're having the dog stop there. Place a second food target on the ground about 4' away from the A-frame. Make sure that you put the food on something the dog can see. As soon as the dog gets the first treat, immediately release him with your release word and then send him to the second target *with a command* -- Get It, Take It, whatever. As the dog speeds up, start moving the second target further away and using it intermittently.

4) Get out the dog's favorite toy. The second he gets to the spot on the contact you want him to hit, give his release command, and toss the toy.

(Monica Percival)

You can retrain a dog's contacts using "hoops" -- which the British books call "frames". For this barrier method to work, you need to use it EVERY time your dog does a contact (except in competition). It sets up a "muscle memory" for the dog's performance. Remember, retraining works best if you can establish a new habit by having the dog perform correctly the new behavior 3-5 times as many times as the dog performed it incorrectly!

A handler can use operant conditioning (with the conditioned reinforcer, also known as the clicker!) to teach some agility skills, especially those skills which are counter to the dog's personality. For example, getting the bouncy/enthusiastic dog to stop at the bottom of the contact (or table or weave entry) or getting the reluctant dog to move out briskly or getting those dangling rear toes to be picked up going over a jump.

The bottom line is, YOU NEED TO TEACH THE DOG EXACTLY WHAT THEIR JOB IS! The handler needs to first have a clear picture of the finished behavior which they want from the dog. Only then can they use various techniques to build, back-chain and otherwise shape that behavior using reinforcements (mostly positive reinforcements).

One can get good results with a trained stop with two front paws on the ground, and rear paws on the board. This may make more sense to an enthusiastic dog than four paws on the board, so listen to your dog. Get other handlers to give you their observations after watching you & your dog work.

Don't forget the importance of the backwards pressure on the front paws on the descent of the A-frame. Make sure toe nails are short and that there is no other discomfort/arthritis in the front paws.
(Lynnda Lenzen)

If you have to coax a dog down the contacts, that is a good sign that she doesn't know what is required of her! Take a step back and think about what you want and then train for that - consistently!

So what do you want out of contact performance? You want the dog to get on the obstacle cleanly (i.e. safely, hitting the contact and being able to climb the obstacle with minimum effort), run over the obstacle as fast as is safely possible and then hit the down contact reliably.

You don't want to be coaxing a dog down the equipment at any time - in training, it is building up trouble for a trial and at a trial it is wasting time. As to the contact, think about that as well - can your dog see the contact area - don't bet on it!!!! What your dog can see is where the obstacle touches the ground and that is the other end of the contact area. So if you train the dog to run to the bottom of the obstacle EVERY time then you will have reliable contacts - yes?

There are several different methods of getting the dog to go to the bottom of the contacts EVERY time:

One method is to train the dog to stop with his front paws on the ground and rump on the contact - call this position 'bottom' and the dog gets clicked and treated for it.

Other people swear by targeting - usually by placing a food target at the base of the contact so that the dog has to run the contact to get it and then fading it to a small non-food target and then eliminating that as well. However, many dogs go sniffing for food when trained this way.

A third method is to teach the dog to run down the contact by guiding it using a short lead. Use this as a starting place for the bottom method and to speed up a dog who needs to be coaxed - you are not forcing the dog to be faster - just showing him that speed is a requirement.

Another method which is a refinement of this is to use hoops on the contacts. This forces the dog to run the contact. This method may be OK if you have your own training facility, but when you are sharing a facility with a number of other people, it doesn't work too well.

The final link in all this is very difficult to take. This is the fact that behavior that is unacceptable in training is unacceptable at trials - if that means walking off a course when she blows a contact then you have to swallow your pride (and the fee) and just do it. Call your dog to you, put her in a down, thank the judge and carry her off course. She hates it but soon gets the message.
(Tony Dickinson)

Some of the fastest teams in the world train their dogs to never stop on the contacts. This is accomplished by using hoops. Here's how it goes.

Start with good visible hoops. The important thing is that hoops are there EVERY time you send your dog over. After sending him over for many, many times, the dog will automatically drop his head down on the zone (and so hit it) even if there are no hoops just because he'll follow the behavioral pattern that he had to repeat so many times.

To make this transition from hoops during training to no hoops at competition less dangerous, use practically invisible hoops made from tiny wire that the dog can pull from the ground if he crashes into it. That's what finally convinces a dog that he has to drop his head no matter if there are hoops or not (because there are, he just doesn't see them).

Training with hoops definitely takes more time than the stopping method, but it's worth the effort. First, you gain some time, because in the stopping method the dog will begin to slow down in the middle of down ramp. You lose time even if he won't actually stop (because he'll be released before stopping). And second, it's a self-correcting method. In the dog's eyes you have nothing to do with it, so there can't be any conflict. There is no stopping, calling the dog back, placing him back in the zone and so on.

The hoops method is especially efficient when used from the very beginning. This means that the dog has never crossed a contact obstacle without being forced to drop his head. But retraining is possible. The only condition for using this method is that the dog must have a short enough trot that he definitely hits the contact if he trots over it (not jump off or something!) This is true for almost all breeds used in agility, including BCs and Belgian Shepherds. Only bigger dogs can have a problem with it. (Silvia Trkman)


« Back-Chaining  |  FAQ Home  |  Call-offs »


Main Categories