Monday, November 29, 2010

Trust Part II: The Other End of the Leash

In today's America, we seem to place a great deal of importance on efficiency and technology, while getting to know another person is an almost forgotten art. It takes time, and time is at a premium. I'm as guilty of it as anyone else. The last time I made a friend whom I hang out with outside of scheduled work and dog events was about four years ago. Trust goes two ways; in order for someone to trust us, we must show them something of ourselves. And we must take the time to get to know them.

I am regularly amazed by the amount of faith and trust Maus has in me. We've been in some genuinely scary situations - like the time we were ambushed by a dozen kids, or the other day when I got attacked by a dog, or pretty much every vet visit. Through it all, Maus looks first to me for direction when he's unsure of his surroundings. His trust in me is incredibly handy, a little bit of a burden, and a humbling honor. Of course, the basis for this trust is how well we know each other. Maus knows that if he isn't sure of what to do, I will have a plan to keep him safe. For my part, I know Maus well enough to understand how he will react in any given situation. I know Maus's mind better than I know my own some days – particularly if I haven't had my morning coffee yet.

Maus and Rubi are almost complete opposites. Both are reactive, but that's pretty much where the similarities end. Their main triggers are different: Maus reacts to people, Rubi to dogs. Maus's reaction is fear-based. Rubi is more excited than worried. Maus retreats, Rubi advances. Remember that reactivity and aggression are different qualities. Reactivity is the over-the-top response a dog has to a trigger. Aggression is how badly a dog wants to hurt the trigger. Rubi is more reactive than aggressive, whereas the two are about equal for Maus. The only way to learn these things? Trial and error. I had to get to know both dogs, to see them each in different situations over time to understand how they would behave at any given moment. Once I have that information and can trust them to behave in a certain way in a certain situation, I can come up with the best behavior plan for their individual issues.

How much is enough? Is it important to B's behavior plan to know that she's rather have her ears rubbed than stroked? If we're working on relaxation, it can be the difference between calm and tense. You can never know too many details about your dog's behavior and personality. Admittedly, I may be a little more anal than most. In competition obedience, my sport of choice, a single misplaced foot can be mean a non-qualifying score instead of best in show.

I could eliminate B's reactive behaviors without getting to know her so well. Traditional, punishment-based training is very good at eliminating behaviors, and Rubi has the exact sort of resilient mind that traditional trainers thrive on. Then again, getting rid of B's reactive behaviors is not my main goal (don't get me wrong, it's on the list, it's just not the big one). My main goal is to give Rubi the highest quality of life possible. If I cannot take her places because of her reactivity, then her quality of life is not as high as it could be. In the same vein, if I cripple her enthusiasm by correcting her over and over again for inappropriate behavior, then I have made her life (and mine) poorer for it.

The balance between the two is in a concept called "LIMA." LIMA is the least invasive, minimally aversive option. Take, for example, traditional training: Rubi offers an incorrect behavior and is punished for it. Compare it to counter conditioning, where she sees a dog and I shove lots of treats at her. Both are pretty invasive and require a lot of effort from me. Which is less aversive, though? In the end, I should get the same response either way: the elimination of her reactive behavior. The way I go about eliminating that behavior does matter. Take a look at it from a human perspective. Which would you prefer: a boss who nags you constantly, or one who lets you do your job and rewards you when you've done it right? Behaving around dogs does not come naturally to B. As her "boss," it is my responsibility to make sure that she has the guidance and motivation to do her job.

That's not to say that Rubi gets to do whatever she wants and I just reward her for the behaviors I like. Positive is not the same as permissive. In addition to rewarding her for the behaviors I want, I limit B's opportunities for behaviors I don't want. For example, Rubi gets a treat for coming inside promptly the first time I call her. There's also a fence around my yard so she can't go running off. In class, I reward B for calm, relaxed behavior, but I also keep a leash on her so she can't go charging off after the other dogs. By managing the amount of mistakes she can make and rewarding her when she makes the right decision, I create a comfortable, happy working relationship. Which works well for when we're not working.

Here's a point of interest which I find extremely important, so listen closely (or read carefully, whatever):

Dogs are not children.

Aren't you relieved? It is possible to be both your dog's friend and their leader. I've heard that you shouldn't be your child's friend because it interferes with being their parent. I don't have children, but I do know that most of the best show dogs I see are out there in the ring with their best friend. These are the teams that have fun, whether they win ribbons or not. Friendship isn't a big thing, it's all the little things. It's using the least aversive training method possible. It's knowing just where to rub you're dog's ears. It's playing together and working together and relaxing together. Relationship isn't a big thing or a little thing.

It's everything.

Thursday, November 25, 2010

I Am Thankful For . . .


- Author Unknown

I asked for strength that I might rear her perfectly; 
I was given weakness that I might feed her more treats. 

I asked for good health that I might rest easy; 
I was given a "special needs" dog that I might know nurturing. 

I asked for an obedient dog that I might feel proud; 
I was given stubbornness that I might feel humble. 

I asked for compliance that I might feel masterful; 
I was given a clown that I might laugh. 

I asked for a companion that I might not feel lonely; 
I was given a best friend that I would feel loved. 

I got nothing I asked for, 
But everything I need. 

Thursday, November 18, 2010

The Art of Screwing Up

At the Minnesota Renaissance Festival, there's a guy who spends all day at a booth playing chess. He wins. Or more accurately, he loses about once a week. My understanding of chess ranks up there with rocket science and trying to figure out what doctors are thinking, so one day I asked him what the secret was to becoming a master chess player.

"Well, I've won more games than anyone else here," he said. Then he leaned forward, face serious. "I've also lost more games than all of the people here combined."

Of course, there's more to being an expert than spectacularly failing at what you do. Mistakes are part of any learning process; it's what you do with your mistake that marks you as a professional or an amateur.

I'm being cliché again, aren't I? Okay, here's what it looks like when I screw up:

Yesterday, Rubi was in a weird mood. She was crabby at the other dogs, extra hyper, and generally rebelling against the house rules. I'm not blaming her for anything. I have days when I don't feel like following the rules, and they usually coincide with the days when I wish everyone would just leave me alone. Rubi's been spayed, so I think I'll blame it on the fact that none of us get to see the sun anymore. Hmmm, maybe she could use a vitamin D supplement . . .

Anyway, I'm rambling. I got to class with B all excited to do more work on our new pseudo-forced relaxation. We got in the door, settle onto our mat, and then B was having NONE OF IT. Roll over on her side? Sure. Stay there? Um, no not today.

I am only human, and being only human, I got frustrated. This worked so well last week! What is wrong with you, you horrible beast?!? I got more irritated and impatient, until finally Rubi, after giving me a final, pointed glance, decided the wall was more interesting than me. I was no longer worth paying any attention to at all. Screw you, lady.


I resisted the urge to storm out of the class, toss B in the car, and go home. Grow-ups shouldn't throw temper tantrums, and I'm trying very hard to be a grown-up (most days, anyway). Instead, I took five seconds to take a deep breath and try to figure out what happened.

It's funny how a deep breath often brings a different perspective. B wasn't saying "screw you" when she refused to acknowledge me. She was saying, "You screwed-up, crazy lady; I'm going to stop giving you attention until you calm down."

Boy, it's irritating when your own training techniques get used against you. Being part of a team means you have to communicate and compromise every once in a while. So the next time B glanced at me, I marked and rewarded her (thanks for giving me a second chance, dog). We did a few no-brainer exercises: sit, down, shake left paw, shake right paw, touch (my sanity has returned, please work with me again). Then, we worked on stuff that was a little more interesting to B, and a little less important to me. Could I have forced the issue on the relaxation exercise? Sure, but it wouldn't have been any fun for either of us.

Instead, we played 101 Things to Do With a Traffic Cone. We worked on sending B to her mat from a distance. We practiced left finishes and down on recall. We played crate games. It didn't take me long to figure out that B didn't really care about the other dogs. She wanted to work. She just didn't want to work on what I wanted to work on. It's not the end of the world. There will be plenty of opportunities to practice on relaxation in the future (and you can bet that we will). Dog training should be fun. If it's not a life or death issue, and it's not fun, it might be time to reanalyze why we're doing what we're doing.

One of my best friends is big into horses. Her world revolves around her horses the way mine revolves around the dogs. I once sat in on one of her horse-back riding lessons. During the lesson, her instructor told her, "You are not the important part of the equation; you are there to make your horse look good." I've tried to remember those words in dog training, particularly in training for competition. If I can make my dog look good, everything else will fall into place. Part of making Rubi look good is finding out what works for us. Trial and error means that occasionally something we try will be wrong. That's okay. It's how we recover from being wrong that's important.

"I'm wrong all the time. It's how I get to right."
                        -Gil Grissom, CSI

Tuesday, November 16, 2010

Relationship Is Important

Well duh, right?

Get this: relationship is so important that EVERY speaker I went to at the three day APDT conference made some reference to it. Pia Silvani, CPDT-KA, who taught the Feisty Fidos workshop, even refuses to admit reactive dogs who have been in their home for less than two months into her reactive dog classes.  Relationship is that important.

But what did they mean by relationship? After all, people who beat their dog still have a "relationship" with the animal - it's just not a good one. On the other side of the spectrum, I'm met a lot of dogs who clearly loved their owners, but just as clearly didn't listen to or respect them. I sat down and I thought about it, and I decided that what these wise and brilliant trainers meant by "relationship" was actually "trust." It is possible to train a dog without him or her trusting you. Heck, that's a lot of what being a dog trainer is. But ask that dog to walk into a situation he or she feels is scary and dangerous? That takes trust. And trust takes time.

Think of your dog's trust like a bank account (thanks, Vera, for the analogy!). Every positive event that happens to your dogs around you - every treat you give, every snuggle on the couch, every time you throw the ball - that's making a deposit into your trust account. Every time you do something your dog doesn't like - each correction, each time you trim nails or give a bath - that's a withdrawal from your trust account. Like with any bank account, the goal is to have a high balance.

I'm not saying that you can never let anything bad happen to your dog in your presence. Nails need to be trimmed and shots need to be given. There's no such thing as a withdrawal-free life, for people or for dogs. However, if your trust account has a high balance, then it can handle the occasional withdrawal. Here's a real life example: Maus and I work hard together, and he trusts me a great deal (ie, I have a high balance). On the other hand, Maus hates the vet. The people there have negative fifty points as far as Maus is concerned. I have all my dogs' blood drawn once a year - definitely a withdrawal from the trust account. I know that if someone tries to forcibly restrain Maus for his blood draw, there's a good chance that person will get bitten. So I restrain him myself. Because Maus trusts me, he lets me do it. I've never trained him to allow me to hold him like that. Although come to think of it, I probably should have.

Maus lets me restrain him for blood draws for the same reason he believes me when I say a stranger isn't going to hurt him - he trusts me. This kind of trust takes a long time to build. Rubi and I aren't there yet. Maus and I still occasionally disagree about people. But the Maus of three years ago was very different from the Maus of today.

In contrast, I am not implying that you should shower your dog with attention and treats when ever you're together. You're dog will love you for it, I'm sure. I'm also sure your dog will get spoiled. One way to increase your dog's trust in you is with consistent, positive training. What your dog learns then is that good things happen when he or she does what you say. Remember when I was talking about NILIF way back in the day? That's a great way to help establish trust. So is training with positive reinforcement. I've taught B a fair number of tricks. She doesn't need to know how to roll over or shake or beg, but it's another way for me to show her that when she listens to me, good things will happen. It's fun.

At the same time, it's important not to confuse trust with hunger. There's more to teaching trust than just passing out treats. One of the reasons it takes so long to build trust is that you need to get to know someone in order to trust them. Think of someone you trust. How did you come to trust them? Did it come instantly? More likely, you spent time together. You learned the person's likes and dislikes. You came to know what they would do in a given situation. You did activities you both enjoy. It's the same with dogs. You can't establish trust if you keep your dog chained out in the backyard and never do anything with him or her. Personally, I'm a big believer in play therapy. What better way to get to know someone than have fun with them?

Dogs are ever honest creatures - creatures of action. Tell a dog, "trust me," and you'll likely be met with a blank stare (and a wagging tail; dogs are funny like that). In contrast, dogs pay a great deal of attention to our actions. Act consistently and positively, and over time, your dog will trust you to act that way all the time - even if he or she doesn't trust the situation.

Friday, November 12, 2010

Relaxation: Conditioned and Forced

Have you ever had someone tell you to smile when you're down because it will make you feel better? Did you know that it works? For humans, we've know that form follows feeling for a century or two. Smile and you feel better; frown and you feel worse. Humans and dogs are more similar than we'd like to think, and I've seen enough anecdotal evidence to believe that this physical feedback loop also applies to our canine companions. Basically, get a dog's body to form the shape of a calm dog, and it will calm the dog.

One way to do this is with conditioned relaxation. Maus and I started working on conditioned relaxation about two years ago. I have Maus lie down on his mat, and then I mark/reward him (quietly and calmly) each time he relaxes. So for instance, when he lies on the mat and shifts over to one hip, I mark and reward that. I mark and reward him for putting his head down, for relaxing his ears, for relaxing his shoulders, for deep breaths, and for any other behavior that leads to a more relaxed body. When we first started doing this, it would take about twenty minutes for Maus to truly relax at home. Now, I can pull out Maus's mat in even a busy area and have him relax within about five minutes. It's a handy tool to have for an anxious dog.

In Pia Silvani's workshop at the APDT conference, she discussed an exercise she called "forced relaxation." Many reactive dogs, Rubi included, tend to pace a lot. This extra movement works them up, which causes them to pace more, which causes them to work themselves up more, which makes them more reactive. Pia's solution is to have the dog lay on its side. When it struggles to get up, she has the handler restrain the dog until it relaxes again. Eventually, the dog learns that when it lies on its side, it has to relax because, well, it's doesn't have any other options. (Hopefully someone will correct me if I misunderstood Pia – one of my fellow trainers picked up Pia's book on Feisty Fidos, and I'll read it once I get through the books I picked up at the conference.)

If you know me or have read the blog for any length of time, you probably realize how much I dislike forcing my dogs to do pretty much anything. Rubi in particular has some bad associations with being restrained. She's been alpha rolled more than a few times, plus the fiasco that used to be getting her nails trimmed (hey, if someone pinned me down and came at my fingers with a set of bolt cutters, I'd probably pretty upset, too). On the other hand, what Pia said makes sense – dogs that move around a lot tend to work themselves up. It increases their heart rate up and boosts tension. Reactive dog behavior training is an awful lot of teaching the dog to be calm. I don't feel like Rubi is a good candidate for conditioned relaxation because her mat work just isn't strong enough. With Maus, Piper, and Allister, if I take the mat out, they're all over it to the point where they try to get on it before I even set the mat down. Rubi could take it or leave it. Since for conditioned relaxation, that mat is the cue for the relaxation, the dog has to be pretty invested in staying on the mat for it to work. I'm not saying Rubi will never be there, but she's not there yet.

For the past few weeks, Rubi and I have been working on a combination of forced and conditioned relaxation. I sit on the floor and have Rubi lay down in front of me. Then, I lure her onto her side. I mark and reward her for each relaxed behavior – for relaxing her legs, for putting her head down, the usual stuff. I also started counter conditioning her to restraint. I started by touching her shoulder and marking/rewarding her for that. Once she was okay with that (which took a few sessions), I put a little pressure on her shoulder and marked/rewarded her for that. Eventually, I was able to give her a massage. Massage isn't really a part of restraint, but who doesn't like a little back rub when they're trying to relax? Around this time, when she would try to get up before I released her, I would use that counter conditioned pressure on the shoulder to stop her. It's not a lot of pressure; no more than I'd say I use to hold her collar when I'm putting her leash on. If Rubi really wanted to get up, she could – and did on several occasions. After which I'd refocus her and ask her to lie down again. Rubi seemed to really dig these sessions at home. Of course, the real test came when class started again on Wednesday. How did it go?

I think the pictures speak for themselves:

Monday, November 1, 2010

One Day at a Time

Here's where we are now:

Today, all four of the dogs and I played ball for forty-five minutes. Rubi refused to give the red ball back, so she lost the privilege of playing with it and had to play with the green ball like everybody else. Then Rubi and I went for a walk, because tired dogs are good dogs. I tried using the Freedom Harness and figured out after about a block and a half that I didn't like it. It might be a better tool for dogs that aren't so heavy in the chest or so into pulling (basically, for dogs that aren't pit bulls). I might pull it out again for a dog that doesn't work so well on the gentle leader, but for now, I switched B from the Freedom Harness back into her gentle leader, since I had wisely brought it along. I also found out that while the Lickety Stick does have a high-value enough treat to use for reactivity, the ball occasionally gets stuck, and the dog can't get the treat out. No treat does not work for reactivity training at all. I'm rather disappointed.

Anyway, for our walk today I decided to head for an area that I had previously avoided. There are five dogs on this route: a lhasa apso that likes to run up and down it's fence really fast, a very reactive, very angry shepherd mix, and three shepherd/huskie mixes that aren't particularly reactive - but there are three of them. We approach all three yards from the other side of the street to give B and I plenty of space. The lhasa is first: as soon as B sees it racing around the corner of it's house, I shove a handful of treats in her face. Rubi thinks this is a great idea! She happily trades me looks at the other dog for treats as we make our way by. w00t!

Next is the hardest dog: the angry shepherd mix. True to form, he charges the fence like the Tasmanian Devil himself, snarling and barking and biting the fence. This proves to be too much for my own Tazmanian Devil - she snaps to the end of her leash, but I don't give her a chance to react beyond that. I get my body in front of her and body block her back the way we came. After maybe five steps, she looks up at me in frustration and JACKPOT! I click and throw a fistful of treats on the ground.

The throwing treats on the ground trick is handy for a couple of reasons. First, it's a huge reward for B's brief moment of attention, and it will also serve to momentarily distract her from what she was doing - namely spazzing at the other dog. If I'm quick like a bunny when she gets done with the treats, I should be able to redirect her behavior into something more appropriate. Second, it gives the angry shepherd a calming signal; Rubi looks like she's sniffing the ground from where he is. Of course, he doesn't seem to care, he's still wailing at the fence, but for some strange dogs, it can help. Third, throwing treats on the ground and letting B clean them up gives me a few precious seconds to decide what I'm going to do next. I could turn around and head back home, but I hate to leave on a bad note, and we've hardly done anything yet. Or I can try to work through this.

Time's up, look alive! B's head pops up, and she starts looking for the angry shepherd again (he's kinda hard to miss). I click her and give her a treat. Oh, hi! Yes, I exist - and I have tasty food. She looks at the other dog, I click and treat. She looks at the other dog, I click and treat. She looks at the other dog, I click and treat.

And here's the part where I'm a bad scientist: She looks at the other dog, I click and treat. She keeps looking at me, so I click and treat again. Technically, you're only suppose to work on one variable at a time. For example, when you're practicing heel, your work on walking next to you OR you work on duration (how long the dog stays in heel) OR you work on sitting when you stop OR you work on pace changes. If you try to work on more than one variable at a time, you end up confusing the dog and everything can fall apart. So technically, we should just be practicing auto watches - look at the other dog, look at me, get a treat somewhere in there. Here's the thing, I've been around dogs who's owners just worked the auto watches. Their dogs get around other dogs, and the poor things look like they're going to give themselves whiplash they look at the other dogs/owner so fast. I also know that dogs repeat what you reward. So I reward B for her auto watches, and I reward B for looking at me any instant longer than she has to. I don't mind if she looks at the other dog, but I'd rather she looked at me.

Remember when B and I started CA and she would try not to look at the other dogs until she was already over threshold? This is how I build attention without losing confidence. B doesn't feel like she can't look at the other dogs, but ultimately, looking at me is going to be more rewarding.

So this is how we go forward: I mark and reward B for looking at the other dog, I turn around so we're going the same direction. I mark and reward B for looking at the other dog, we take a step forward. I mark and reward B for looking at the other dog, we take another step forward, and B's still looking at me, so I mark and reward her again and we take two steps forward. Meanwhile, there's a dog across the street making like a banshee. You get the picture. With my other dogs, no one is allowed to go forward unless they are focused, attentive, and absolutely under control. With Rubi, I have to balance that against her pacing and working herself up. The longer we stay in one place, the more anxious B gets, and the harder it is to refocus her. There's no easy way about it, it comes down to knowing my dog. This is one of he reasons we're taking CA again - to learn to be calm and still around other dogs.

The rest of the walk was relatively easy. After the angry dog, the three dogs weren't nearly as big a deal. Of course, then we had to turn around and go back home. Before we tackled that, I wanted to give both of us a break. So we sat down on the side walk and practiced sits and downs, right and left shake, and some hand touches - really simple, easy to reward stuff that would boost both our egos. Then we headed home.

Once again, the three dogs were only moderately interesting. We approached angry dog's fencing line, and once again he charged us. B looked at him, let out one enormous and heartfelt sigh, and went back to working for me.

Good girl!

After that, the lhasa apso was nothing. We didn't even cross the street to ease ourselves in. "Ha!" Rubi seemed to say as she pranced by, tail wagging. "I've pooped out things bigger than you!"

And that, I suppose, is progress.