You can’t always get what you want
But if you try sometimes . . .
Leash walking has been one of my biggest training frustrations with my dog. I’m not alone, of course. It’s a hard behavior to train — along with recall, one of the hardest.
One reason it’s so hard is (like recalls) we’re trying to get the behavior we want in a high-distraction environment — i.e., outside.
Any time you’re competing for your dog’s attention it makes training harder.
But there’s another reason it’s a hard behavior to train, and that’s because we tend to make a mistake in how we set our leash training goals.
Figuring out what behavior you really want is a huge step in dog training. It’s also easy to overlook, because we tend to define our training goals in terms of how our dogs are impacting us, instead of looking for objective behaviors that we can more easily shape.
Take a simple example. Suppose you have a dog who begs during mealtimes. If you don’t like this behavior, chances are you’ll define your training goal as “get Fido to stop begging.â€
That goal sets you up to encounter difficulties, however. You’re thinking of the problem in terms of how it impacts you, which suggests you can solve it like you’d solve a people problem. With a person, for instance, you could fallback on abstractions, like telling Fido that the old pattern of rewards has ended: “Okay, Fido, I know I’ve tossed you tidbits from the table before but it’s over! Never again!†You could appeal to Fido emotionally: “your begging totally stresses me out, please! For the sake of my sanity! Stop it!â€
Strategies like that sometimes work with people. They never work with dogs. Sure, if you are absolutely certain that nobody will ever ever ever toss a single morsel to Fido from the table, the begging behavior might fade in time. But the odds are against you — partly because, perversely enough, uncertain rewards can sometimes make a behavior more persistent than reliable ones do – partly because even if you don’t intentionally feed your dog from the table, mistakes happen, and a scrap dropped by accident is as rewarding to a dog as one tossed on purpose.
But what if you redefine your training goal along the lines of “I want Fido to go to his mat and lie down while we eat meals.†Now, instead of talking about what you don’t want, you’re taking about what you do want. You have something to work toward: a behavior you can start shaping and reinforcing. Figure out how to reward Fido for going to his mat and lying down while you eat – make that more rewarding than it would be for Fido to sit next to you, drooling on your jeans and boring holes in your head with his eager brown eyes — and you’ll have the begging problem licked. Not by getting rid of the begging behavior, but by training a different behavior to take its place.
I’ve discovered that the same principle can be applied to leash walking.
With leash walking, it’s become fashionable to refer to the desired behavior as a “loose leash.†The term is understandable; certainly “loose leash†is great shorthand for the experience I want when I walk my dog: I’m walking; my dog is walking; the leash is slack between us.
But it’s not the leash you’re trying to train. It’s the dog.
It was a lightbulb moment for me when I realized this.
Who cares about the leash?
What I want is a particular behavior from my dog.
At first, I tried thinking of this behavior as how the dog’s body is positioned relative to mine. If we’re walking with a four-foot leash, for instance, then as long as my dog is standing somewhere within three and a half feet of my left wrist, give or take a few inches, the leash will presumably be loose. Assuming it’s not wrapped a couple times around my ankles.
But this conception was another dead end. The reason it didn’t work: when my dog and I are walking, we’re moving. It’s not position that matters – it’s the pace of my dog relative to my pace.
To understand the significance of this, it’s important to bring in another key fundamental about training, which is timing. With dog training, timing is everything. The more agile you are at matching your response to a dog’s behavior, the faster and better your dog will “get it.†(That’s why clicker training can be such a useful tool: by precisely marking a behavior you can more quickly shape it.) (That’s also why people who have more practice at training dogs are better than we lay people who may only train one dog our entire lives, or at most two or three.)
Timing comes into play, with leash walking, because you have to time some combination of rewards and corrections (punishment) to shape the behavior you’re trying to get.
A reward is something your dog likes, such as a treat, or praise, or a toy. For leash walking, trainers might suggest that you praise the dog, or hand the dog a treat, when the leash is loose. This is fine and good, except that in a high-distraction environment like the outdoors, your dog may not care about treats. (This gets into the hierarchy of reward value — one implication being that the more competition you have for your dog’s attention, the more desirable the reward has to be.)
A correction is when you give the dog some sort of unwanted experience as a consequence for not walking nicely — i.e., when the dog pulls and the leash becomes taut. In theory, any collar could do this – in theory, the dog won’t like the sensation of a taut leash as it is transferred to her collar — but for whatever reason, dogs often seem to become oblivious to pressure on their necks/throats, and if they pull too hard (or worse yet, throw themselves to the end of the leash when you’re not ready for it) there’s a risk of tracheal damage.
(Some people use tools like choke collars to make the pressure-on-the-collar sensation more acute, however, I’d recommend against using any tool that relies on hurting or frightening your dog. This is outside the scope of this article, but in practical terms alone, raising your dog’s stress level can create more problems than it solves.)
I have found a tool that I find works well: the Easy Walk Harness. I learned about this harness from Terry Ryan, the trainer with whom I wrote Outwitting Dogs
Perfect, right? Except for one thing: the harness only does its thing when the leash goes taut. From a timing perspective, therefore, the interruption of the pulling behavior happens after the unwanted behavior has already happened.
So yeah, switching to this harness helped. From the first time I clipped it on, my dog wasn’t pulling constantly when we walked. But she still wasn’t the model dog, by a long shot. For much of our walks, every time something caught her attention she’d run ahead and hit the end of the leash, then slow down or take a step back toward me, then when the leash was loose again she’d run ahead and hit the end all over. It still wasn’t the behavior I wanted.
But notice what I’ve just described: it’s a pace thing. My dog hadn’t learned to match the pace of her walk to mine.
What I needed was to reward my dog for walking at the right pace.
The harness is a great tool, but it’s only a tool. I needed to use it right.
I needed to use it to interrupt her behavior when she picks up her pace, instead of when she hit the end of the lead.
So I combined it with two other old training standbys: “be a tree†and “penalty yards.†You can find articles that elaborate on these terms here, under leash walking.
“Be a tree†means that if your dog pulls, you stand still. When the leash is loose, and only when it’s loose, do you walk. Since most dogs find the walk itself inherently rewarding – they want to move ahead – this technique uses walking itself as a reward you control. Pretty ingenious, eh?
Penalty yards is a variation on “be a tree.†With penalty yards, you go back a few steps if your dog pulls. If your dog finds “exploring†to be rewarding (and what dog doesn’t?) then penalty yards help reinforce the consequences of pulling by putting the as-yet-unexplored part of today’s walk that much further away.
The trick, for me and my dog, was using “be a tree†and “penalty yards†to control her pace, instead of worrying about the state of the leash.
I published a digital clip of this on Youtube so you can see what I mean.
Here’s what I do: I watch my dog’s feet. Any time she starts to walk too fast, I stop walking. The leash becomes taut, and the pressure on the harness interrupts her behavior. She then slows down or takes a step back toward me so that I’ll start to walk again.
Over time, she’s begun to realize that the harness pressure – the consequence I’ve set up – happens because of her picking up her pace. As a result, she’s begun to match her pace more to mine. She checks on me more often — looks up to me to see what I’m doing. This is huge, especially considering that she’s a fairly tightly wound dog. Remember what I wrote at the beginning of this piece, about outside behaviors being hard to train, because it’s high distraction? I want my dog, despite all the other sounds and smells and sights around her, to be also paying attention to where I am, the direction I’m walking, and the speed I’m walking. Using this technique, I’m getting her to check with me – it’s a major breakthrough for us.
Notice what I don’t do: I don’t pay any attention to the leash. The leash takes care of itself. I pay attention to my dog’s pace.
I started using this technique a couple of months ago. I’ve been practicing maybe 5 minutes a day since, and in that time, I have really turned my dog’s leash walking behavior around.
Let me know if you try it yourself and how it works for you.