Lessons from Working with Feral Dogs

About 3 weeks ago I received 3 feral dogs from a local rescue. Here's the video of them at the pound. The three I have are in the left side kennel.

As you can see, they aren't happy at all. It's assumed that, based on their teeth, they are between 2 and 4 years old and don't appear to have had much of any human contact in their lives.

This morning, three weeks after receiving them, this is where they're at:

Their progress has been phenomenal.

The point of this particular blog is not to describe their progress and how I did it. I really want to go into what I'm learning from this process and the mistakes that I've made. I'll get to the procedural stuff about our progress when they leave for their foster homes.

They need space

I have portions of my house blocked off with baby gates. Essentially 2 living room sized areas. I have Hoakey and Kreacher on one side and Winky on the other. I separated them because Kreacher and Winky kept going at each other. Once I observed them in the backyard, with 4 times the space of any of the indoor areas, I noticed they generally kept to themselves. I decided that inside I would open the gate between the two areas for a larger space. Sure enough, they stopped going for each other. I would attribute this change to how they went from having the entirety of the world as strays to just 175 square feet. It was a mistake to try and apply my general knowledge of structure and boundaries to these dogs.

They get into everything

These dogs are used to digging things up, pulling stuff, and generally searching and scrounging for everything. These behaviors do not change once they are in the house. I use this to my advantage by hiding treats under blankets or bowls or in between couch cushions.

They aren't house broken

Pretty simple. They're used to relieving whenever and wherever they want. It's a huge pain and requires a lot of diligence and watching and treats to break this habit.

Boundaries, both physically and emotionally, are important

I do not allow dogs in my kitchen. I expect dogs to wait at doors and wait for permission to leave. Learning these sorts of house manners are important for maintaining order in the house.

Emotional boundaries however, are harder to set and maintain. I always invest fully in whatever dog comes into my house. If I don't invest and treat a dog like it is my own, then that dog does not get the best of me. I wouldn't ever give less then everything to any dog in my house. The problem is in maintaining the emotional attachment that these dogs have to me. Teaching dogs to be alone is very difficult, as demonstrated by all the problems I see with crate training and destructive chewing.

Their community structure is complex and deserves time to understand

Being isolated from human contact, they've developed their own structure. While calling it a pack is technically correct, I really want to get away from all of this wolf comparison nonsense. I'll call it their community.

The complexities and intricacies of how they interact are worth understanding because that is how I learn who these dogs are. Their personalities come through. What really comes through is the type of techniques I can use. There are 12 million ways to teach sit. For Hoakey, he likes to sit with his head held proud while the other two swarm around him. A hand placed sit won't work here. For Winky, he never sits and does not mind being handled. A hand placed sit could work for him. Kreacher does not like to be handled, but takes really well to a food lure. If I did not see how they interact with each other, my approach to training would not be fully informed.

Safety is the most important consideration

I installed baby gates to keep dogs I don't know away from dogs that I do know. I keep leashes on dogs I don't know at all times. I kennel train them for safety reasons. I will never stop being concerned for every dogs and my own safety. Bites can condemn a dog to death and cause nasty infections. Bites also cause irreparable damage to your relationship with that dog. Prevention is the best way to approach safety.

Document everything

I take all sorts of videos of everything I do and upload them to youtube. It's fun because I love seeing the progress I've made. Documentation also allows me to rewatch what I do and improve.

A lot of these lessons are not just for feral dogs. They really apply to any dog I've worked with. These lessons came to the foreground very seriously because of these dogs.

Selling Vindication: How eCollars Became Popular

As anyone who reads this blog knows by now, I like to use force-free and positive training methods. One of the topics I don't ever want to write a blog about is why positive reinforcement works better than other methods. I won't do that for three reasons. 1) So many other people have written about it. 2) If you're reading my blog, you already believe in how what I do works. 3) The people who don't believe it will only become more entrenched in their positions. See how religion, politics, and even food choices work when you challenge what someone believes to be the right way of doing things. They'll generally become more confident that their belief is correct.

I'm not here to change to minds.

With that said, the actual topic at hand. Today, one of the rescues I'm working with gave me an ecollar that was donated to them. The first thing I looked for was if it was a shock or vibration collar. I love vibration collars for off leash training and for training deaf dogs. Though when I looked at the packaging this is what I found:

Does this shock my dog?

I have 1-60 levels. But do I shock my dog?

No where on the packaging does it say anything about the actual sensation being delivered to my dog. It says "proprietary 'blunt' stimulation" and and it has "1-100" levels of it. But blunt isn't shock. Blunt really just means force applied with a flat object. "Blunt force trauma" for example, like being hit with a baseball bat. Blunt definitely sounds better than shocked, right?

My next qualm is the implication that a shock collar and the dog's IQ are related with the dog studying from the book and the actual question "what's your dog's IQ?" In terms of tools of communication, how does the shock collar more effectively communicate anything that the clicker cannot? The problem with relating a dog's response to a shock to its IQ is that a response to pain is instinctual. IQ and instinct are not even related. The only thing this collar is testing is your dog's pain tolerance. It's not teaching anything.

After the packaging, here's some pages from the instruction manual:

The first page says two things. One, the word "conditioning" is in parenthesis. This is because this is not real conditioning. They are implying that you need a certain level of intensity to condition the dog. The issue here is that, once again, it's simply finding out what your dog's pain tolerance is. 

My second issue with the first page is that it says "...some dogs are more stubborn and require more stimulation." A dog whose pain tolerance is high is now considered stubborn. The implication is that a dog will stop being stubborn if you shock it more. Pain tolerance is conflated with stubbornness. 

The second page is just silly. It says, "for example, if you want your dog to come to you and heel, you can hold the button down [for continuous shock] until your dog comes and heels..." If your dog does not know how to come and heel, why would shocking it for 30 seconds train that? In the alternate scenario, the one in which your dog does know come and heel, why would shocking him be the best method of reinforcing it? It seems like a lazy way to train. If your dog is not coming when called in all situations, perhaps spending some time proofing the behavior is what you need. 

What upsets me most about shock collars is that it is a cheap way of feeling vindicated when your dog does something you perceive to be wrong. It's this attitude that makes it hard to get rid of corrective training. When a handler hits or punishes a dog, it is rewarding to the handler to keep doing it. It's hard to get someone to stop doing something that's so rewarding. Try telling a small child he can't have soda. It's rewarding to drink soda so naturally he's going to be upset when he can't have it.

That's how I think these collars became popular. Not because they're effective at training dogs, but because their effective at relieving stress in people.

Today's blog is more of a rant than anything training related. It just really upsets me that these shock collars are touted as humane or even "stimulation" collars. They're lazy and painful. No words or marketing can change that.

If you have any questions or comments feel free to email me at Michael@concentricdog.com or find me on Facebook.

Proofing Behaviors Part 2: Criteria Setting

This week's blog is a continuation from last weeks. I'm going to get more in-depth with the details instead of just discussing the concept.

Because my clients are mostly rescue dogs, I get a lot of dogs with high prey drives, meaning the dog likes to chase down anything that moves, especially cats. It's a pretty basic instinct to chase things that move (greyhound races and fetch) just as it is basic instinct for dogs to pull on leash (Iditarod). The problem with working the prey drive is that it's fueled by a desire for food, a much stronger instinct than pulling on leash. When working on specifically cat chasing with these rescue dogs, I'm often times working against years of this dog trying to chase down cats for food. Telling a dog to stop chasing the cat is like telling him he can't eat. Even if you know your dog isn't going to eat the cat (which you can't really ever know), it's still telling him not to eat.

How does this relate to proofing? I'm going to go over the steps I use to diminish prey drive. The whole process is really taking what the dog already knows and making it work around cats.

What's hard about writing a how to is that I cannot possibly cover every single scenario for my readers. I can really only provide a generalized sense of how to approach proofing and hope that the reader understands how to fill in the gaps from there. I am available on Facebook and by email to try and fill the gaps between what I write and you're specific situation. Just let me know!

Now, I'm going to narrate this story in my experience with Polo, a pit mix that I've worked very hard on for a long time now.

Polo doing what he does best

Polo hates cats. There's no doubt in my mind that he would try to eat one. For this reason I know he will never be okay with cats, but his behavior will become manageable. You can take the dog out of the streets, but you can't take the streets out of the dog.

The first thing I had to do was isolate the cats and Polo. A traditional way of dealing with this behavior would be to collar correct or shock collar, but as you know by now I'm force-free and don't like doing those things to my dogs. A traditional way of stopping prey drive behavior would stop his desire to chase cats in front of me. I turn my back and he'd be right out. I've just taught my dog to resent my presence.

The main reason I isolate the cats away from Polo, other than safety, is that if I put Polo in a position in which he can perform the behavior I don't like, he is rewarded. What this means is that when a dog performs an instinctual behavior, just simply doing it is reinforcing. I want Polo to be successful all the time and very rarely experience failure. AS big of a pain as it is to always have Polo away or the cat away, it's worth it to me. If it isn't worth it to you and you have cats, then I would really suggest trying to adjust your priorities towards creating peace in your house.

"But Michael, we're still not even discussing proofing yet!" I know, and it really sets up my next point nicely. In order to proof a behavior, your dog must first know it. Sounds silly to say, but if your dog doesn't know stay and you try to get him to stay at the dog park you're setting your dog up for failure. 

Obedience work is the core of all dog training. Scales are the core for music. Knife skills are the core for cooking. Form is the core for weight lifting.

For a long while, I never even worked on the cat situation with Polo. I wasn't ignoring it, I was just putting it to the side for the moment. How am I supposed to teach a 60lb pit to behave himself around cats when he still jumps on people at the door? This is how criteria setting works.

What can I expect my dog to do? This is a judgment call. I unfortunately can't write what to expect and when to expect it for each dog. I'm trying to demonstrate to you how to have that judgment for yourself. Again, I'm always available for questions.

Once I had Polo's obedience in order, I chose to proof his sit-stay around the cats. I can't proof all his commands at once, so I picked the easiest one of them all. I also picked sit-stay because it is an incompatible behavior. He cannot sit-stay and chase cats at the same. 

My first step was to figure out his threshold of success. Where is Polo going to fail? Once I figure this out, I can work backwards into a place in which he will be successful. As I learned, he can't even handle the little bell one of the cats has on her neck. He hears it and dive bombs the crack between the door and the floor. 

Working backwards from there, I went to the end of the hall to see if he was still reactive to the cat's bell. At the end of the 15' hallway all he did was perk his ears forward and tilt his head. I'm not teaching Polo to avoid being curious. I'm teaching him to sit-stay under all conditions. I got him in a sit-stay and now I have another set of criteria to worry about. 

Polo will sit-stay for 3 minutes without moving. What can I expect him to do here? We're moving backward 20 steps to eventually move forward 50, so I dropped my criteria to a 10 second stay at 15' away from the door with the cat and bells. I'm going to work my criteria all the way up to 3 minutes again then take one giant step towards the door.

Rinse and repeat until you're at the door.

Then I get to just open the door for a second, then a minute, and so on. 

As we can see, proofing is all about proper criteria setting. I don't really want to get into every tiny minutia because the point is to set reasonable goals and work slow.

Today, Polo will sit-stay for 1 minute with cats in the same room. We've been working for 6 months. I have a ton of patience, which is what proofing requires. 

It's okay not to proof behaviors. Just don't get mad at your dog when they don't listen to something you've never trained. 

If you have any questions, find me on Facebook or shoot me an email at Michael@concentricdog.com

Proofing Behaviors: Performing All the Time No Matter What

The hardest thing to do is not getting a dog to perform a behavior, but getting the dog to do the behavior all the time no matter what. It's so easy to teach a dog to sit. No matter a person's experience with dogs, I've found that all but one dog I've trained knows how to sit before we start working. These inexperienced dog owners/handlers can make the dog sit some of the time, but the dog doesn't do it all the time no matter what. This illustrates that training is the easy part, anyone can do that. Proofing is what's hard.

Proofing a behavior is the process of getting a dog to do a behavior all the time no matter what. It is also one of the more important concepts in maintaining a dog's training. Proofing exists in two major categories: 1) desensitization and 2) obedience.

Proofing by way of desensitization means exposing your dog to new things and ensuring a positive and successful experience. Let's consider a dog who is generally dog aggressive. In my house and front yard, this dog will interact very nicely with other dogs. When at the park, this dog will tear into other dogs. What we can assume here is that this dog has the ability to play nice with other dogs. The question then becomes, how do we proof this good play with desensitization? The short of it is that we start from a distance away from other dogs in which he will not react and slowly move forward from there.

The only reason a dog performs a behavior we don't like is because at some point the dog was rewarded for that behavior. In a more traditional style training, we would essentially use aversion therapy and try to replace those positive associations with, in our case dog aggression, with negative associations, using collar corrections, shock collars, spray bottles, etc. That's one way to do it. I do not think it is the most effective, though it is the quickest to produce changes in behavior.

The slower, but more fun way is by treating your dog every time it is not trying to attack another dog and treating every time the dog is behaving how you like. When I say every time, I literally mean 1 treat per as fast as you can treat. What this does is 1) creates a positive association with being around other dogs and 2) teaches your dog what to do when faced with a situation in which the dog would normally be aggressive. Doing this slowly and making sure your dog never goes past the threshold of success should proof the behavior of playing nicely with other dogs.

One of my favorite things to do is obedience. What I love more than obedience, is going out to my favorite stores and doing obedience. My clients will attest that my college graduation level test for dogs is going to Bookmans (local dog friendly used book store) and doing all of our obedience. Sit, down, stay, loose leash, etc.

Now this graduation requires a lot of desensitization as well. The key difference here is that behaviors in desensitization are about what the dog already does, whereas obedience is about what behaviors we've taught the dog.

Honestly, I didn't need this whole length to describe proofing behaviors. It really is a simple concept. It's the precise execution of constantly changing criteria that is hard. When is it appropriate to push your dog? When should I pull back? What am I expecting? Criteria setting is going to be next week's blog. It's the next important concept in proofing because it is the finer detail of it all.

To end, I'll provide a brief list of important reasons why proofing is important. 1) A dog that listens in all situations will be less likely to get injured/injure others. It's safer. 2) A mentally stimulated dog is less likely to cause trouble. 3) It's more reason to get out a do fun things with your dog. 4) It's fun to include dogs in the family.

Let me know what you think! Comment here or shoot me an email at michael@concentricdog.com