[jfinner1]

I'm at a complete loss on what I can do to fix this. My dog has absolutely no recall off leash. His recall in the house is perfect. His recall on a 30' line outside is perfect. His sit and wait to go outside is practically perfect. But if he gets loose, like he did today, all bets are off. Basically, I put him outside to go potty, on his lead like normal. Somehow, he slipped the lead. I had gone about my housework, and about 10 minutes later I opened the door and called him to come back in. When he didn't come, I ran outside, and sure enough, there's his lead, but there's no dog... So I grab his leash and my coat, and start walking the neighborhood calling him. I found him about 2 blocks away, trotting down the sidewalk, happy as a lark. But as soon as he sees me, he takes off running in the opposite direction. I've learned not to "chase" him, in the sense of running after him, because he can run faster, and longer, and next thing I know I'm about to fall over, and he's out of site. So instead, I calmly walk after him. But he manages to stay about half a block in front of me. And any time I get close enough to grab him, he bolts. Today, I chased him for an hour, for almost 2 miles, before I finally caught him because he had to stop to poop. By that point, I had taken back roads that I didn't know, ran through who knows how many people's yards, been chased by two teens on bicycles who were laughing at me, and jumped two retaining walls, and I had no idea where I was. Luckily, some guy was nice enough to give me directions home. But this is just insane. He doesn't get loose very often, even less so now that he doesn't try to bolt out the door. But how am I supposed to train him to come off leash, when as soon as he gets off leash, he does this? We have a fenced in back yard, but he can scale the chain-link in 5 seconds, and since the local dog park has a chain-link fence much link ours, I'm hesitant to take him. I really don't know what more I can do. Help??

Side note, Otto is a male doxie-minpin, about 2 years old.

 

[elkhound88]

You describe his recall as being excellent in all situations except for when he gets loose and starts running, so it seems that the area that needs to be worked on would be overriding his response to distractions. For example, a dog can be perfectly obedient and follow instructions off the leash, but if he happens to see a squirrel, it'll create an entirely different response because a different instinct kicks in and the instruction he received without that stimulus will no longer be present.
Typically I think it's necessary to use slight negative reinforcement for such a problem, from everything I've read, but I haven't worked on this kind of a problem with a dog before so in practice I can't give more advice than that. Running like that can be a powerful instinct and it's harder to see how other approaches would work. You could try recreating situations in which he can slip his lead, but not be able to run a great distance away uncontrolled, and try to extend his recall training to these circumstances as well, to incorporate his instinct. I don't know how effective it can be though without some kind of negative reinforcement... It's definitely something that needs to be resolved though 'cause he could easily get hit running like that or lost...

 

[3doglady]

My setter use to bolt from every opening she could find. She's very fast. I found that driving the neighborhood and opening the car door was enough to "catch" her. She could care less about treats once she's outdoors. After I started taking her to the park to run, she stopped bolting. So I did find the park was very helpful in that respect. With my neighbors' dogs, often crouching down and calling with a yummy treat is typically all that is needed. Crouching offers a more approachable body language to dogs. I think there is a chapter on that in one of Pat McConnell's books. She talks about trying to catch a stray dog on a busy road and the differences in the body language she used.

What do you use for your recall when she's on lead? Do you have an irresistible treat?

 

[jfinner1]

I use hotdogs. Lots and lots of hotdogs. He LOVES Hotdogs, lol. But I've been weaning him off of the treats for his on lead recall, and he only gets treats sporadically for his indoor recall, since both are perfect both with and without treats.

 

[Pawzk9]

You describe his recall as being excellent in all situations except for when he gets loose and starts running, so it seems that the area that needs to be worked on would be overriding his response to distractions. For example, a dog can be perfectly obedient and follow instructions off the leash, but if he happens to see a squirrel, it'll create an entirely different response because a different instinct kicks in and the instruction he received without that stimulus will no longer be present.
Typically I think it's necessary to use slight negative reinforcement for such a problem, from everything I've read, but I haven't worked on this kind of a problem with a dog before so in practice I can't give more advice than that. Running like that can be a powerful instinct and it's harder to see how other approaches would work. You could try recreating situations in which he can slip his lead, but not be able to run a great distance away uncontrolled, and try to extend his recall training to these circumstances as well, to incorporate his instinct. I don't know how effective it can be though without some kind of negative reinforcement... It's definitely something that needs to be resolved though 'cause he could easily get hit running like that or lost...

I'm curious what sort of "negative reinforcement" one would use to stop/train a bolting dog. I would find a safe (fenced) place where I could work on recall with distractions. In emergencies, I might get the car, drive past him and invite him to jump in. Or run yelling and whooping the opposite direction. Or fall on the ground and pretend to be hurt,. Or find marvelous and exciting imaginary mice in the grass. Of course, this also requires that you teach him it is really cool for you to catch him or he will only let you catch him once.

 

[embedded]

The canine experts are going to pounce on me, but I introduced "look" between his acquisition/targeting and pursuit of his innate drive. I used "look" inside with him when in the window and a stranger or other animal went by. I used it on walks and in the tennis court when playing with the ball.

Then I used it reguarly when doing off lead tasks in the field. So now off lead say in the woods with squirrels, foxes, & deer, my GSD simply looks and will even look at me waiting for direction.

He will only pursue under any condition if I say "go".

 

[elkhound88]

I'm curious what sort of "negative reinforcement" one would use to stop/train a bolting dog.

I wasn't referring to a dog that has already bolted in an instance, but rather preparation work for not bolting. I had in mind something like using flags at different distances from a work area to help teach the boundaries around the yard, to walk them around and everything, and to scold them when they pass that boundary, rewarding of course if they stay within the boundary, perhaps extending the scolding to include a time-out if it persists. This of course is more appropriate for situations where it's a matter of the dog staying close to home, and I imagine a different technique would need to be used in the case of walking with a dog off the leash in various areas. From what I had been reading, it might need to incorporate these types of negative "punishments" because the reward to a dog of running like crazy or chasing a squirrel or cat might be more valuable than the type of rewards that they normally get. This is, of course, assuming that the dog has already been trained a lot in general but they still don't listen in those circumstances.

 

[Pawzk9]

I wasn't referring to a dog that has already bolted in an instance, but rather preparation work for not bolting. I had in mind something like using flags at different distances from a work area to help teach the boundaries around the yard, to walk them around and everything, and to scold them when they pass that boundary, rewarding of course if they stay within the boundary, perhaps extending the scolding to include a time-out if it persists. This of course is more appropriate for situations where it's a matter of the dog staying close to home, and I imagine a different technique would need to be used in the case of walking with a dog off the leash in various areas. From what I had been reading, it might need to incorporate these types of negative "punishments" because the reward to a dog of running like crazy or chasing a squirrel or cat might be more valuable than the type of rewards that they normally get. This is, of course, assuming that the dog has already been trained a lot in general but they still don't listen in those circumstances.

I'm not sure you really understand what negative reinforcement (or the other quadrants are). Scolding (which isn't that useful when the dog crosses the boundary would be positive punishment.

 

[petpeeve]

This dog now has experience with a viable (and apparently unattended) escape route. Unfortunately, I can see it happening again and again if precautions are not taken. So, first thing I would do is shore up the fence ... somehow... if for no other reason but the dog's general welfare.

Further, ... without a secure place to practice in, it'll be difficult if not impossible to even begin to train a solid, safe, and checkline-free outdoor recall.

 

[ArlosMom]

I would also look into a martingale collar for when he's on his lead, or a harness so he can't get off of the lead. Otherwise, you'll have to watch him the whole time. Arlo is a doxie/chi mix and when we first got him, he easily backed right out of his lead. It was on there pretty good, too. he just happens to have a thick neck that is not all that smaller than his head

 

[Finkie_Mom]

Just as an FYI, I NEVER stop rewarding for recall. My dogs get something every single time they come to me when called, even in the house. It may not always be food, but it can be praise, toys, pets, whatever. I actually have a bowl of treats on a table inside the door so that they can get a reward for coming back in the house after a play/training session or potty time or whatever. Coming to me is ALWAYS a good thing. My recall work has proven invaluable three times now with two different dogs. (I wish I wasn't that many, but things happen)

 

[JessCowgirl88]

I agree with arlos mom: look into a harness or martingale collar, so he cant back out of it. We had a bigger dog long ago who we had to get a harness for because he would back out of his lead.

and i also agree with Pawk9:

would find a safe (fenced) place where I could work on recall with distractions. In emergencies, I might get the car, drive past him and invite him to jump in. Or run yelling and whooping the opposite direction. Or fall on the ground and pretend to be hurt,. Or find marvelous and exciting imaginary mice in the grass. Of course, this also requires that you teach him it is really cool for you to catch him or he will only let you catch him once.

is a really good idea as well, like said though make sure its fenced but has distractions.

Hope you find a place you can work on it safley and look into something he cant slip out of. Its always scarey when a pet gets out.

 

[Amaryllis]

Just as an FYI, I NEVER stop rewarding for recall. My dogs get something every single time they come to me when called, even in the house. It may not always be food, but it can be praise, toys, pets, whatever. I actually have a bowl of treats on a table inside the door so that they can get a reward for coming back in the house after a play/training session or potty time or whatever. Coming to me is ALWAYS a good thing. My recall work has proven invaluable three times now with two different dogs. (I wish I wasn't that many, but things happen)

This times a million. "Come" must ALWAYS be a pleasant experience or you are providing a reason not to come. You also cannot use recall to call the dog to you to do something he doesn't like, such as washing, brushing, etc.

As to the escape itself . . . It's really not a good idea to leave an escape artist unattended, even for a minute. Muggsy was like that and he was never outside without a human at the end of his leash. I came to that rule after he nearly died because I left him unattended in the back yard even though I knew what he was like.

 

[elkhound88]

I'm not sure you really understand what negative reinforcement (or the other quadrants are). Scolding (which isn't that useful when the dog crosses the boundary would be positive punishment.

It's quite possible that I had the terminology wrong, as I wasn't referring to any kind of system in particular. I just used the words that seemed to describe what I was referring to, negative because it would be perceived that way by the dog and reinforcement because it was reinforcing the sense of boundary. If you have a good link so I can learn more about the classifications you're talking about I'd love to read it. Also, I can see how scolding might not be useful in some circumstances but I think it would depend on the relationship between the owner and the dog and what that scolding would mean for the dog coming from that owner - if the dog looks up to and respects the owner a lot it would probably carry a lot of weight. I didn't mention what I've read others specify, like shock collar usage, because I personally wouldn't use it.

 

[Snoopy11]

It's quite possible that I had the terminology wrong, as I wasn't referring to any kind of system in particular.

I always have to stop and think about the operant conditioning terms when I use them, because they're easy to mix up. Put simply, a reinforcer increases the likelihood of a behavior occurring again, while a punisher decreases the likelihood of a behavior occurring again. The words "positive" and "negative" here do not mean "good" and "bad" as they do in common useage... it helps to think of them as "adding" and "subtracting."

Therefore:

Positive reinforcement = you add a rewarding stimulus.
Negative reinforcement = you remove an aversive stimulus.
Positive punishment = you add an aversive stimulus.
Negative punishment = you remove a rewarding stimulus.

Bah, I hope I did that right.

 

[hast]

How often do you take your dog outside of the yard? And if you take her out every day, what do you do when you go outside? If you want her to come when you call they need to have enough exercise and mental stimuli not to be bored to tears with their life.

Just as an FYI, I NEVER stop rewarding for recall. My dogs get something every single time they come to me when called, even in the house. It may not always be food, but it can be praise, toys, pets, whatever. <snip>

Ditto to this. I have as close to 100% recall as you can have ... so far she comes when I call every time I call, even from mid chase of deer and squirrels. I recall several times a day, often with a treat, a tug toy, or the ball ... and NOT to catch her, she's free to go after her reward. I only catch her after a recall if absolutely necessary, usually I go to her and just tell her to come along if I need her to follow, but recall doesn't mean "getting caught and all the fun stops".

I'm curious what sort of "negative reinforcement" one would use to stop/train a bolting dog. I would find a safe (fenced) place where I could work on recall with distractions. In emergencies, I might get the car, drive past him and invite him to jump in. Or run yelling and whooping the opposite direction. Or fall on the ground and pretend to be hurt,. Or find marvelous and exciting imaginary mice in the grass. Of course, this also requires that you teach him it is really cool for you to catch him or he will only let you catch him once.

Ditto to this ... and not use recall for catching the dog unless you have to, but always be happy to see the dog and fuss over her when she comes.

 

[Pawzk9]

I always have to stop and think about the operant conditioning terms when I use them, because they're easy to mix up. Put simply, a reinforcer increases the likelihood of a behavior occurring again, while a punisher decreases the likelihood of a behavior occurring again. The words "positive" and "negative" here do not mean "good" and "bad" as they do in common useage... it helps to think of them as "adding" and "subtracting."

Therefore:

Positive reinforcement = you add a rewarding stimulus.
Negative reinforcement = you remove an aversive stimulus.
Positive punishment = you add an aversive stimulus.
Negative punishment = you remove a rewarding stimulus.

Bah, I hope I did that right.

A+ I do think if one is going to use the terms. one should attempt to get them right. You described it very well.

 

[Snoopy11]

A+ I do think if one is going to use the terms. one should attempt to get them right. You described it very well.

Oh good, thank you. Glad it makes sense.

 

[lil_fuzzy]

Another way to fix it is to make sure the collar is properly fitted. Either get a martingale so it tightens when he tries, or make sure the flat collar isn't too loose. You should be able to move it around his neck freely, but when you try to pull it over his head you shouldn't be able to. I see a LOT of dogs whose collars are way too loose, it almost seems like owners are scared to tighten them.

 

[lil_fuzzy]

I always have to stop and think about the operant conditioning terms when I use them, because they're easy to mix up. Put simply, a reinforcer increases the likelihood of a behavior occurring again, while a punisher decreases the likelihood of a behavior occurring again. The words "positive" and "negative" here do not mean "good" and "bad" as they do in common useage... it helps to think of them as "adding" and "subtracting."

Therefore:

Positive reinforcement = you add a rewarding stimulus.
Negative reinforcement = you remove an aversive stimulus.
Positive punishment = you add an aversive stimulus.
Negative punishment = you remove a rewarding stimulus.

Bah, I hope I did that right.

Actually not exactly right. Right most of the time, and I guess for most people that definition would work fine. But the definition technically doesnt say that R+ has to be nice or rewarding for the dog. If you give an aggressive/reactive dog a correction it's likely to wind them up more, so technically you have then used R+. It's the result that determines if you're punishing or reinforcing, and reinforcement isn't always pleasant for the dog. Also, you can reward a dog without reinforcing behaviour, which happens if you give the dog a treat but it didn't meet the dog's expectation, e.g. if you give the dog a piece of dry food when he expected a piece of the juicy steak on the kitchen counter.