Clicker Training for Dogs

Herding

Dry-Work

Don't allow a stock shortage to keep you from training flanking commands, stops, downs, and walkups on your herding dog. Just because you don't have sheep in your front yard doesn't mean you can't train a dog in the basics of herding.

Herding "dry work" is fun for dogs of any breed, and for future herders it bridges a training gap when stock availability is limited. I started using dry-work several years ago to gain more control on my dogs even though I had sheep and ducks of my own. Stock are a major distraction for beginner dogs.

The key benefit to dry-work is that the dog learns without the interference of stock. "What? Stock interferes with herding training?" Of course! Livestock can be a major distraction when trying to teach a dog to perform stops and to take flanking commands.

Training away from stock can decrease training time by a measurable amount. The benefit of dry-work becomes evident when a beginner dog that is experiencing sheep for the fifth or sixth time can perform behaviors he learned in dry-work lessons and obey flanking commands.

I have a 10-month old Aussie who has benefited greatly from playing herding games away from stock. On stock his natural style of working was far too close, he was a sneaky biter, and was highly distracted by wool, "flossing" at every opportunity. He had another common beginner dog training problem - he ran around so much before finally coming under control that he was too worn out to work! He needed a break before any serious training could take place.

I started playing dry-work games with him and saw obvious improvement the next time he was on sheep. He gave ground (removed pressure), downed, stayed, and called off the sheep. After a few more dry-work games, he also stopped on command and readily changed direction.

The most significant benefit had to be his sudden ability to flank stock in both directions. He started out as a confirmed go-bye dog. It always took several attempts to redirect him to the away flank in the round pen, while I had to run to reposition the balance point. Then he wouldn't hold the away for long, crossing over to return to a go-bye at his first opportunity. Dry-work quickly took care of that problem. The next time working sheep he took the away flanking command without a problem and it was wider, rounder, and more stable than the go-bye.

No Cure-all

Before working the go-bye with dry-work games my dog's clockwise flanks were shallow. He would go wide at the top but then dive in toward the flock as he passed balance point, a common behavior in beginner dogs. I'd like to be able to say that's all solved now due to dry-work games, but it isn't. Dry work greatly improved his go-byes, but we still need to work on some shallowness when passing balance point. This will not be done with dry-work games however, but on actual stock, because it is a problem that the games won't fix.

Dry-work is not a panacea for every herding problem and can't prevent future problem, but is good training tool for behaviors that don't require the presence of stock.

Dry-Work Game

One version of a dry-work game that I have used with a lot of success is one that I can also take into the round pen with beginner dogs. The finished game has the dog performing wide outruns in both directions, stopping, backing up, downing, and walking up. Since most beginners are working on fetching stock with the dog facing the shepherd and stock between them, this game stimulates that picture quite well for the dog. Downs and backing up should be taught separately to give the dog a concept of these behaviors before they are added to the game.

I do not use the clicker a lot in playing dry-work games because most of what the dog does is self-reinforcing (bring back the toy and I'll throw it again or play tug for a minute and tell you in a high voice what a clever doggie you are).

I do use the clicker with treats and toys when working real stock, but fade it and use the stock as reinforcement as soon as I can to prevent creating a herding dog that is focused more on the shepherd than the stock.

From the game's introduction to the dog's ability to perform the game on stock is a step process like herding itself. I've tried to include everything I do in the following steps and listed them in the order I use them.

Catch

Teach the dog to catch - socks are the best thing to use for this purpose. Click at first for making any effort to catch, then for a good catch, and then fade the clicker as soon as he can catch the sock with any consistency.

Teach him to catch a "Frisbee." I use a fabric disc toy and start with short and slow tosses until he figures out how to catch. As he becomes proficient at catching it, toss it a little ways past him until he starts to follow it before it hits the ground. When he starts trying to intercept the toy he's ready for farther and faster throwing until you have built up some distance.

Flanking

When the dog is anticipating the toy by running out to intercept it, start throwing it to the side and give a flanking command. The dog is probably facing you and it's good to just leave him alone at this stage and not get insistent about stays and downs.

Catching the toy can be self-rewarding behavior that needs no additional reinforcement. I'm not teaching a retrieve, so it's up to him whether he wants to give it back. But I want to play this game again and again, so I add R+ for bringing it to me. I click and feed, play tug, or rub my dog's tummy and praise him as PR for returning with the toy.

The dog will start to anticipate the throw by running out a ways before the toy leaves your hand (he will run in the direction you normally throw). When I see this behavior I know it's time to add criteria for distance and commands - one at a time, of course. Since this is a flying disc toy, it can go high, far, and fast. Do keep in mind that jumping puppies could become injured puppies and keep your throws low enough to keep jumping to a minimum.

Adding Flanking Commands

Because I want very crisp turns when I send the dog on a flanking command, I try hard to make my throw go toward 10 or 2 o'clock. It looks like this: with the dog standing directly in front, throw the toy to the 2 o'clock position for a "go-bye" and at the 10 o'clock position for an "away." This will teach him to move to the side rather than fading forward as he makes his flanking turns (see drawing).

I add back ups here and there. By throwing the toy directly over his head so there is no direction paired with it and telling him in advance to, "Get back," you can teach a pretty good back up behavior. R+ by throwing the toy directly at his mouth when he backs a step or two. Add a down/drop here and there and send him from that position, making sure the resulting flanking behavior is something close to a 90-degree angle to his down, by throwing the toy to a point that is perpendicular to his ribcage (see drawing).

Gaining Distance

By the time I get to this phase, it's getting really fun for me and for the dog. He's becoming operant and offering behaviors like fast runs, quick downs, and prancy little backups. He's starting to look over his shoulder, watching for the toy as he runs. Now I start adding distance - not in how far I throw the toy, but in how rounded the dog's path is.

I use several ways to accomplish this, but here's one that works well and doesn't require props. Hold the toy close to your body (keeps the dog from jumping up to snatch it away) and start to slowly turn one way or the other with it. He will follow the movement, so if you're rotating to your right, he's turning to his left (moving in a clockwise circle) which is the "go-bye" direction. Wait just a moment and he will naturally move away so he's in a better position to chase and catch the toy (if he jumps for it, raise it up and wait). His movement is reinforceable behavior, which causes you to throw the toy in the direction of the flank he's taking. You can add the flanking command here if you want, and speckle them in while playing the game.

The next phase of the game includes obeying flanking command from the stand, down, and sit (I use stand and down) positions and from a distance. Dogs tend to fade in toward stock and/or shepherds when given flanking commands from a stopped position. Prevent this by making sure your throws are not in front of the dog, but to his side as explained earlier.

If he has a favorite direction, that's usually the one he will travel in. If it's not the direction you want, give the other flanking command and wait. As soon as he offers it, which he will do if you have worked both directions equally, R+ by throwing the toy.

Stop, Turn, Wait

Everything you've done up until now is an excellent segue into introducing the stop command, "There." When the dog will perform a flanking behavior and circle you widely at least once, it's time to add the stop. Just say "There," and throw the toy directly at his mouth. Do this a few times and he'll start anticipating the throw by stopping and facing you when he hears the command. The "There" command means stop, face your stock, and wait.

Redirect

While the dog is circling and before you throw the toy, stop him and send in the opposite direction by giving the flanking command and throwing the toy in the desired direction. Now start fading the toy, because you're almost done.

Stop the dog while he's flanking and send him in the same or different direction. If he takes the command, keep it simple and allow him to travel for only a few steps, say "There" to stop him again, then lots of praise, run over and pet, feed treats, and so on. Have a party for your great future herder!

I know that sounds easy, and there are pitfalls. If you've been clicker training your dog though, you already have a pretty sound idea of what to do. Wait. If you stop playing the game, he'll start trying to figure out what made you stop and "fix" the problem on his own. If you've done the ground work by playing the game in steps as outlined, he'll offer the correct behavior and, if you feel more comfortable using a clicker in these situations, bring it back out and use it. In fact I recommend it if you have one of those sensitive dogs that becomes worried when he is unsure.

From the Game to the Pen

Some parts of this game can be used in the round pen and even in the big arena or field with a lot of success. Not all dogs feel totally rewarded by the movement of stock alone; some respond better to "human" games. If this is your dog, give him what he wants and take your flying disc, treats, and clicker into the pen. It might take several clicks for the dog to hear it the first time, so be prepared to click four to five times before he realizes he's being clicked. After that, be prepared to click and throw or treat to toy to the dog if he's where he should be.

Here's where the game pays off with real stock. If the dog is not in the correct position, change balance point or give a command and then throw the toy to where the dog should be. When he goes there to catch or get the toy, click. I realize this is a form of reverse clicking, but it works. If he comes running in for a treat, give him one and send him back around. Coming for a treat isn't a bad thing; it gives you the opportunity to restart the game and resend the dog, which is just more practice for you both.

He's done with treats and toys when he starts refusing to leave his "post" for reinforcement. He'll do this fairly quickly as long as you are moving about in the pen, which allows him to see the job at hand and to realize his role in it. If you are beginner enough that you still tend to stand still and watch what's happening, it will take your dog a little longer to realize his role. Just keep practicing and it will come naturally. If your stock is heavy (they stay with you) and your dog isn't attacking them, you have all the time it takes to gather your thoughts and take the next step.

As with all training methods, use what you can from this article, evaluate your dog's progress, and then continue or change what, how, and how much are doing based on his progress and yours.