The way I think about it, almost all of the behaviors that we teach dogs are actions (such as sit, shake, down, roll over, spin). They all involve movement. But “Stay” is the absence of action–it means don’t move. With everything else, we teach “move a certain way!” and with stay, we teach “do not move at all!” Think about how confusing that must be for the dog– it’s not something concrete that the dog is supposed to do. Every other word means “do something!” and “Stay” means “don’t do anything!”
So it’s much easier for everyone involved if you teach the dog that sit means to sit until you give her the cue to get up (stand/move forward). Those are both actions, as opposed to an action and a lack of action.
Also, what does “Sit” really mean to you? Usually, when told to “sit,” a dog will sit for a couple of seconds and then get right back up. This video will help you to teach your dog that sit means sit until you say otherwise. Basically, the sit is automatically a stay. In my opinion, this is the very best, kindest, and clearest way to teach a stay. I have typed out the following instructions based on Emily’s commentary in the above video.
The Easiest, Fastest, Most Reliable Sit/Stay
Step 1: Lure the dog into the sit position. Click and treat (or use a marker word like “Yes!” and treat) the moment that the dog sits.
Step 2: Use a rapid-fire treat delivery to reinforce the dog for staying in position. After luring the dog into the sit position, you’re going to continue clicking/treating as fast as you can (while the dog stays in position). When the dog is doing well, you can put more time in between marking the behavior as correct. If the dog gets up, you just lure the dog back into position, wait, and then reinforce.
Step 3: Add the release cue. Once your dog can stay in the position for 5-10 seconds, then you add the release cue. You say your cue (such as “Release!”), and if the dog doesn’t get up, you pat your legs/move backwards, and encourage your dog to get up. Click and reinforce your dog for getting up. You are hoping that the dog will start anticipating that “Release!” means get up.
You do not want to move and say “Release!” at the same time, because you want the dog to learn that only the release cue means to get up, not any kind of physical motion. So you say the release cue, wait a couple of seconds, and then encourage the dog to get up.
Step 4: Remove the lure and add the cue for sit. The next step is removing the lure and adding a hand signal. Pretend to get a treat, lure your dog into position, and then reward with a treat from your other hand or from your bait bag. You do not want your dog to be reliant on the treat in your hand. Once you can get your dog to sit without having a piece of food in your hand, you can then turn that gesture into a hand signal to tell your dog to sit. You do this in approximations.
If your dog gets stuck, go back a step and make it easier. You don’t need to constantly up criteria and make it more difficult.
Step 5: Proof the release cue. The most important part of teaching a sit/stay is proofing the release cue. You have to make it very clear to the dog that the only thing that allows them to get up is the word “Release!” (or whatever word you use as a cue).
If you don’t proof for it, your dog is going to get up when distractions happen, when she’s been sitting for too long, if you move too far away from her, etc.
To proof your release cue, you simply have to add distractions and click AS the distraction is happening. When your dog is successful with that, you can start delaying your clicks (so the distraction happens, you wait, and then click/reward). The result is that you can add distractions and the dog remains sitting through it all, waiting for the release cue.
If the dog were to get up, you simply have to lure her back into position, don’t treat, and make things easier. You can start slow, by just touching your bait bag, then pulling treats out of the bag, and so on. You can use different types of distractions. The more proofing you do, the more reliable your sit/stay will be.
If the dog can’t even sit through the beginning of a simple distraction, then you need to go back and work on clicking the dog for staying in the sit position and finding that very reinforcing.
- Emily presents a Frisbee (click/reinforce).
- Emily moves the Frisbee around (click/reinforce).
- Emily clicks AS she throws the Frisbee. She says that if this is very difficult for your dog, you can have someone throw the Frisbee from a distance and you can reward the dog for sitting and staying with you (while on leash).
To start incorporating distance away from your dog, you might simply shift your weight forwards. Then you can start adding different kinds of movement. I find this part to be a lot of fun, because I get to dance around, do jumping jacks, and look utterly ridiculous.
Don’t constantly increase criteria. Don’t take three steps away from your dog, then five, then ten, then fifteen… Instead, do five steps, click and reinforce. Then three steps, click and reinforce. That way the dog never knows when she is going to get a treat.
Emily says that it’s very important to proof the release cue before attempting to add duration to the stay. If you just ignore your dog [without proofing the release cue], she may start offering up other behaviors as she tries to figure out what it is that you want. But if she knows that she can only get up on the release cue, you can work on duration without her breaking the stay.
The same thing applies to duration as distance. You’re not going to do 20 seconds, then 30 seconds, than a minute, than five minutes, then 10 minutes and only give a treat after 10 minutes… don’t constantly make it more difficult. Switch it up. The dog should never know when she’s going to be rewarded.
*If you’d like, you can also check out Emily’s 18+ minute long tutorial on teaching stays. In the video description, she explains that “By [clicking] as distractions happen you are using classical conditioning to change your dog’s emotional response to being calm and relaxed around those distractions. [This will lower] a dogs stress levels when asked to stay, which in turn will make the behavior more reliable.” She also mentions that “By clicking the release as a behavior it becomes a secondary reinforcer. So every time you release your dog from a stay, you are actually rewarding him for staying.”
**If for some reason your dog does not want to sit and seems uncomfortable doing so, you always want to rule out medical issues (like problems with his joints and limbs). If your dog is a rescue, it may also be that he previously had a bad experience during training. It is important to be patient and gentle.
If your dog does not seem to understand the exercise, you can work on “shaping” the behavior (split up the behavior into smaller actions and slowly build up to the entire sit).
“Some dogs will get this cue right away. Not so for others. If your dog seems to have trouble figuring out what you’re asking him to do and won’t sit for you even after several training sessions, you want to dissect the whole action into smaller parts. We call this breaking down a cue, or rewarding for the smallest successes, even for simply lifting his head. If his nose goes up in the air, treat him. Little by little, get your dog to move his nose a half inch higher on each subsequent repetition. Once you get his nose pointing toward the sky, his rump will automatically start to lower. Treat him as his rump starts to move down. On the next rep, reward him when his rump is a half inch lower. Eventually he’ll touch down…
If you or your dog is bored because it’s too slow, or you’re frustrated because he isn’t getting it, take a break before the dog totally loses interest. Do another quick activity that your dog enjoys… If training is not mixed with playtime experiences, you may undermine your dog’s confidence level and enjoyment of training.”
– Dawn Sylvia-Staciewitz, Training the Best Dog Ever
Quick note: For other reasons, that wouldn’t be one of the books that I’d recommend, but I thought the above excerpt was well-written out and helpful.
You can also work on “capturing” the behavior (clicking and rewarding whenever he randomly sits of his own accord, so that the sitting is reinforced and he offers up the behavior more frequently).
If sitting just doesn’t seem to be working for your dog, I personally wouldn’t worry too much. I’d just try to put him straight into a down. You can try this by sitting on the ground and luring your dog so that he has to lower himself down and go under your knee. That way you can teach a down/stay with release (instead of a sit/stay).
Warning: Mild Rant Ahead! (Or: What Inspired This Post)
I recently attended two dog seminars here in Costa Rica. Though there were a few good points, it was overall not what I had hoped for, and I’m still quite cranky about various inaccurate and outdated ideas that were presented. I may later elaborate more on specifics, but for now I will say that it has made me realize how much help Costa Rica—and the world—needs when it comes to modern day methodology and kind, benevolent dog training. One of the BEST things we can do is teach the new generations BETTER (and attempt to educate the old). Also, please don’t mention the word “dominant” (dominance is not a personality trait!) to me for at least a week, because my head still hurts.
During the second seminar, the speaker was talking about how to teach a stay. I very much disagree with the methods suggested, and I personally found it to be what I would call “sloppy training.” The speaker continuously repeated the command and scolded the imaginary dog any time that she tried to move.
I say command and not cue because the speaker and I have very different methods of training. If a dog doesn’t respond to my cue, I am not going to threaten the dog or physically manipulate her body in any way. I am going to assume that the dog does not yet fully understand what I am asking, or that the dog is overstimulated and we need to make it easier before working up to that level. However, the speaker believed that if you gave the command once and the dog didn’t respond accordingly, you needed to raise your voice and say the command again very sternly/sharply in a way that showed you were the boss–because, of course, you need to gain dominance over the dog. The speaker used intimidation through tone of voice and body movements (such as stomping).
The speaker felt that you should vary the intensity of your voice during dog training. He passionately declared himself “anti-clicker” and made fun of the clicker, calling it “the stupidest instrument that there is,” “a fad,” and “absurd.” Now, if someone prefers not to use a clicker, that’s perfectly fine–you do you, as the cool kids say. But to announce in front of a class of impressionable people, with no sense of tact or thoughtful reasoning, that a clicker is the stupidest instrument that there is, just because he thinks so is, in my opinion, limiting and childish. The clicker can be a valuable tool for teaching both simple and complex behaviors. To be fair, after I spoke with the speaker during a break, he did apologize to the class for the way he expressed his opinion.
Guess what? You don’t even have to think about the word “dominance” when training your dog. I have previously mentioned Patricia McConnell’s blog post “Dog Training and the ‘D’ Word,” but I’m going to quote from it again right now. She wrote:
I’ll start with the bottom line. I don’t use the word “dominance” when talking to people about training their dogs. There’s just no profit in it. Even given that dominance is about “priority access” and “social freedom,” but not about how to get it, I still see nothing but the potential for confusion and misuse. Given that in general parlance dominance means “total control,” and that it is so often it is equated with force (completely inappropriately), I avoid the term as if it were toxic. Which is exactly what I think it can be in this context.
Look at all the absurd uses of the concept sent in by readers. “Expressions of dominance” include: A dog sitting with its back to you, forging in front on walks, jumping up on people, pulling washing off of a clothes line (one of my personal favorites), acting scared when someone approaches, a “refusal to be potty trained,” (did the dog hire a lawyer?), using signs of fear or appeasement to manipulate their owners (no kidding), and another personal favorite–dogs who are good retrievers as youngsters should be avoided because they are acting as alphas by provisioning the pack (bringing back a chicken in this case) and are thus predisposed to be dominant. Oh my. Oh my my my.
Thank you all so much for adding fuel to my fire that we need to drop the concept of dominance in relation to dog training.
So there you have it, folks. From Patricia McConnell herself–expert dog trainer, animal behaviorist, and one of the coolest people in the universe (which, by the way, I can back up with plenty of facts & thoughtful reasoning).
P.S. I personally think that we should start referring to the dominance nonsense as Dominope. “My dog does [insert utterly random thing] because he’s trying to be domi–” “NOPE!” “My dog’s just really domi–” “NOPE!” “You have to be rough with dogs and show them you’re the boss because otherwise they’ll gain domi–” “NOPE!”
But seriously–leave the D word at the door. And remember that intimidation is not a necessary component of effective dog training. Keep it kind.