My family got our first dog, a long-legged sable Sheltie named Duncan, when I was sixteen years old.

As a brand-spanking-new trainer, I had lots of enthusiasm but very little actual skill.  We went to classes and dabbled in a few different dog sports – including, most notably for today’s topic, competitive obedience.

In the novice obedience ring, one of the exercises that dogs are required to master is the formal recall.  This involves leaving your dog in a sit-stay, walking to the other end of the ring, and calling your dog on the judge’s signal.  The dog should come directly to the handler and sit straight in front of them, then go smartly around to heel position on command.

Pretty straightforward, right?

As it happened, Duncan and I struggled with this exercise.  It wasn’t that he wouldn’t come – in fact, the beginning of his recall was a thing of beauty.  He held his stay perfectly, and came running to me as fast as he could when I called… then took a flying leap as he approached, and jumped up happily to say hi before finally settling into his nice straight sit in front.

Every.  Single.  Time.

This was a significant point deduction in the ring, and great source of frustration for me as a handler.  After all – I never rewarded him for jumping up, always withholding the treat until he finally sat.

So why wasn’t the problem getting any better?

******

A behavior chain, scientifically speaking, is a series of simple behaviors that are always performed one after the other in a particular order.  The reinforcer (a treat, a game of tug, or whatever you might be using as a reward) comes at the end of the chain.

For example, let’s say you have a service dog who is trained to retrieve items for you.  When you point to an item on the floor and say “fetch it,” your dog goes to the item, picks it up in his mouth, turns around, returns to you, and places the item in your hand.  You then reward with praise and a treat.

This is a behavior chain.

Each successive step in the chain is contingent on the step that came before.  The treat at the end is effectively rewarding the entire chain, rather than just the final behavior of placing the item in your hand.

Okay, you might say.  So what?

I realize that, at first glance, this might seem like a bit of pedantic training geek minutiae.  But in fact, it has some pretty significant implications for how our dogs learn things in the real world.

Here’s why:

Most of us, at some point in our dog’s life, have accidentally trained a behavior chain that includes some behaviors we don’t want.  It’s easier to do than you might think!  And once you’ve done this, figuring out how to get rid of the unwanted “links” in the chain can be quite a challenge.

To better illustrate how this problem can occur (as well as how to fix it!), let’s look at another example from the world of dog sports – the popular “2-on, 2-off” contact behavior in agility.

A standard agility course includes three contact obstacles – the dog walk, the A-frame, and the teeter.  Each of these obstacles has a yellow contact zone at both ends.  The dog is required to touch the yellow area as he exits the obstacle, which prevents him from getting excited and launching himself off while he’s still too far up to be safe.

Because dogs don’t readily grasp nebulous concepts like “make sure you touch this yellow area somehow”, many handlers opt to train a specific behavior called the 2-on, 2-off position at the base of every contact obstacle.  This means that the dog should run to the bottom of the contact area and stop with his two front paws on the ground, and his two rear paws still on the obstacle.

Properly done, it looks like this:

A solid 2-on, 2-off behavior ensures that your dog will never be disqualified for jumping off an obstacle too early, and gives him a specific, concrete task that he understands.  Makes sense, right?

For many of my beginner agility students, here’s where the problem comes in.

When a dog is first learning this skill, he may be quite good at hitting his position when he’s calm and focused, concentrating on the task at hand.  But… add in some speed and momentum, and he forgets.  He leaps off too early, or overruns his 2-on, 2-off spot without remembering to stop.

His handler, wanting to get on with the rest of the course, stops and reminds him – “Hit it!”

Whereupon the dog does a neat little circle to find his position, or scoots backward until his rear paws hit the obstacle again.

The handler rewards this.  Or gives the release cue (“okay!”) and takes off to continue the course, which amounts to the same thing.

Now, in the context of today’s discussion, here’s question we want to ask:

What is the dog being trained to do?

The actual performance that the handler wants would look like this – dog hits the obstacle at full speed, slows down and shifts his weight back as he approaches the bottom, and stops neatly in his 2-on, 2-off position.

However…

What the dog is practicing, and being consistently rewarded for, is an entirely different behavior chain – hit the obstacle at full speed, charge forward with wild abandon, leap off halfway down, then back up or circle on the ground.

These two things are emphatically NOT THE SAME.

Sad to say, the second version in this example won’t work in the competition ring.  So if you’re an up-and-coming agility student, it behooves you to pay attention to what you’re really training.

So how do we fix it?

In a nutshell:  by paying attention to the entire chain and making sure it’s correct, rather than focusing on just the end behavior.

If your dog overruns his contact position in agility class, you’re better off to go back and have him repeat the entire sequence – or at the very least, re-do the obstacle from the beginning – rather than trying to do a “quick fix” at the end.

This way, he gets the entire chain right.

(If this is happening regularly when you run sequences, then it’s likely your dog also needs some more foundation work to make sure he really understands the behavior and how to find his position at full speed on the first try.  So you can do this too!  A full discussion of how to troubleshoot contact issues in agility is beyond the scope of today’s post – but you get the general idea.)

******

With Duncan, if I could go back in time and train him again, I would approach our recurring competitive obedience issues very differently.

There are lots of training strategies I could have used to help fix a problem with jumping up on the recall.  I could have started with shorter distances, to keep the excitement level down.  I could have lured him into a sit as he approached, to ensure that he didn’t jump.  I could have taught him to sit squarely on a platform as a separate skill, then incorporated this into the recall to provide some extra insurance that he would make the right choice.

What I would NOT do, is continue to practice the “wrong” version of our recall – jumping up, and all – over and over again.

I didn’t know these things then, so we muddled along as best we could.  And that’s okay.

When we know better, we do better.

******

Lest you think that accidental behavior chains are only a problem in the competition ring, here are a few real-life examples that might sound familiar:

• Counter surfing

Dog jumps up and puts front paws on the countertop.  Owner commands “off” or “leave it.”  Dog gets down, and is rewarded with a treat.  This happens every time the dog enters the kitchen.

• Jumping on people

Dog gets excited and jumps up to say hello.  Owner commands “down” or “off.”  Dog eventually calms down enough to put four paws on the floor, and is rewarded with treats or attention.  This occurs with every visitor, as well as friendly strangers encountered on walks.

• Loose leash walking

Dog runs ahead and hits the end of the leash.  Owner calls him back and rewards with a treat.  0.2 seconds later, dog charges forward and hits the end of the leash again.  Repeat ad infinitum, for the entire walk.

• Barking for attention

Dog is bored with downtime in the house, and pesters owner by barking or whining incessantly.  Owner commands “quiet” or “settle,” and rewards with a stuffed Kong after a short period of silence.  This sequence of events occurs every evening around the same time, like clockwork.

If you’ve accidentally taught your own dog any of these behavior chains, don’t worry – you’re not alone!  We’ve all done it at least once or twice, without meaning to.

So what do we do about it?

In all of these examples, you’ll notice a running theme.  It’s not particularly complicated or profound, but it can be a real game-changer once you see it.

Accidental behavior chains happen when we continually allow the dog to make the same mistake over and over again, before stepping in to try and correct it.

That gives us this recurring pattern:

Wrong thing -> right thing -> reward

No matter how many times you do it, you’ll never get rid of the wrong thing this way.  NEVER.  In fact, with every repetition, you’re cementing it more firmly as a necessary part of the chain.

So.

Don’t wait for your dog to make a mistake, then try to fix it after the fact.  Instead, be proactive!  Arrange the environment to set him up for success.  Reward right away for the behavior you want, before things have a chance to go south.  And if the same problem pops up more than once or twice, change something in your training plan so that it doesn’t happen again.

Perfect practice makes perfect, as the saying goes.

So do your best to practice perfectly.