Correctional Training

[Loviedovie]

I have been going to puppy behavorial classes at Petsmart and they teach "reward" training. HOWEVER mostly class is 8 other dogs with playtime and lots of listening to the trainer. My 4 month old shih tzu is a little on the difficult side to train and it didn't seem to be working, so I hired an IN HOME dog training company that does "correctional" training instead. So far so good. It's all on leash right now but it's alot of "lightly tugging on the leash and using No with it. I have to say it's working. We no longer use treats as praise, just high pitched exciting "good Lovie"

Anyone else use this method of training?

 

[tessa_s212]

You won't find many people here that praise correctional training.

I used it. I hated it. I educated myself on dog training, now I use the better methods.

 

[PWCorgi]

Ditto Tessa...

 

[Doberluv]

Motivation and reward methods based on operant and classical conditioning have been shown through science and practical experience of most trainers and researchers to be the most effective, while keeping the dog happier and better at learning HOW to learn.

Most trainers today are using operant conditioning based methods, from exotic animals; dolphins and other sea mammals, like at Sea World, zoo animals who need to be moved from one pen to another, fed etc to movie actor dogs, circus dogs, service dogs of all kinds.

The old school correctional/punishment based training methods are being left behind more and more as clicker training and emphasising what the dog is doing right, reinforcing correct responses and not so much correcting incorrect responses are being used. Punishment will stop behaviors if they're harsh enough, but they are stopping behaviors because the dog wants to avoid an unpleasant consequence rather than complying in order to earn a reward. That's how I want my dog learning....enthusiastic for learning instead of making it a drugery. Correctional based training tends to stop not only the behavior you're wanting to stop, but any other behavior which is happening at the same time. In other words, the dog tends to shut down, tone down and lose a lot of his spirit and desire to learn. I'm not talking about an occasional "correction." I'm talking about correction based training....where the emphasis is placed on what the dog is doing "wrong." He probably doesn't have a clue what he's doing wrong anyhow.

If a dog has not had a strong history of reinforcements for a behavior, he will not likely repeat the behavior because this is how dogs learn, how all mammals learn. It is a law of learning behavior. Also, cues/commands are not what drives behavior. Dogs are not obedient to cues. They merely learn to associate them with a certain response. Reinforcement is what drives behavior. So, when the dog doesn't do a behavior you want and he gets a collar correction, he may be confused, still in the guessing stage. It isn't fair to punish a dog for a trainer's mistake for not giving the dog ample reinforcements and expecting results.

There are some side effects of punishment techniques and too many corrections; collar yanks etc... and that's why most of today's top trainers are not using these methods.

 

[lastkid]

Never on a four-month-old puppy. Reason being, the puppy's still a baby. They haven't learned anything that needs to be proofed yet. There's really no need to be tugging on a puppy's leash... they're so malleable at that age that it's easier (to me) to encourage the behavior I want instead of the behavior being offered, rather than just correcting the "bad" behavior being offered.

That said, I do use corrections, and I believe in a balanced approach to training. Not just 100% positive, and not that traditional "let's jerk the dog into compliance" either. I've worked with both styles of trainer, and my personal preference is for the middle ground between the two. In my experience, doing 100% positive in the learning stages and proofing with appropriate corrections later on works the best for my dogs. "Later on" could be two months, could be two years, depending on the behavior and how well the dog comprehends what I want. Obviously, I'm not going to start giving even a light correction if the dog doesn't fully understand what I'm asking. That's not fair at all. Other people have wonderful success with pure positive, other people don't.

I'd encourage you to try positive training again. It's hard for a dog of any age to learn around a bunch of new dogs, in the middle of a store with every kind of distraction possible, let alone a little puppy. But if what you're doing is working, that's great also. Just make sure learning is still fun for your pup, no matter how you train. If it stops being fun, try something else.

 

[Doberluv]

I don't understand what is meant by "balanced" training. Scientifically, operant conditioning is how all mammals learn. This excludes yanking on a collar. There's nothing balanced about confusing a dog with punishing him for something he isn't responsible for. If a dog is not giving a correct response, it is the trainer's or owner's fault, not his. He has not been reinforced enough for the wanted behavior. If he had not been reinforced for the unwanted behavior, (inadvertantly though it may be) the unwanted behavior wouldn't exist in the first place. Incorrect responses are not the fault of the dog...not ever. So, yanking on a collar is not rational.

The reason operant conditioning, motivation/reward works so well is in part, because dogs are opportunists, scavengers, hunters, predators. They do what works. They don't do things to please their owner unless there's something in it for them, either directly or indirectly. Rewarding directly works better. Getting a payoff, getting reinforced with something they like a lot really makes them want to do it again. We all work that way. What behavior can you think of that you repeat where there's not one thing in it for you? Not much, I bet. What boss would you rather work for, one who is noticing all your faults and punishing you in some way, not giving you a paycheck... or one who notices and praises your attributes and gives you a fat paycheck? What would you do? I bet you'd try really hard to do just those things which he liked, which brought you the paycheck. And you'd stop doing the things which did not bring you anything good.

A young puppy should never be "corrected" with a collar.

I started my Doberman with no collar or leash, other than to get him accustom to wearing it. That's how he began his basic training; heel, sit, stay, come, down. All enticement, fun, encouragement, luring, showing and rewarding. He is very well trained and well mannered and has been for quite some time.

 

[RD]

Corrections have no place in training, IMO. What is the point of training if it's all about avoiding a yank on the leash? The dog doesn't learn easily that way.

If corrections are to be used, they are best used as proofing for commands that the dog already knows.

 

[Roxy's CD]

Yes, it does work.. But you'll run into problems down the road...

My trainer is all about "balanced training". Leash correction/praise.

It did work! Fast! Roxy breezed through basic obedience. And now, we're having issues with motivation, Roxy shuts down...

Don't do it!

I wish I wouldn't have from the get go but am happy I know now. The difference that motivational training makes in your dog. How enthusiastic they are about the tasks at hand. THEIR enjoyment out of it, which is what its' all about!

Really, from someone that's done it and is now paying the price ditch corrections, purely motivational training IS the way to go.

 

[Doberluv]

If corrections are to be used, they are best used as proofing for commands that the dog already knows.

__________________



But, if a dog already knows a command and has been rewarded and reinforced amply, why wouldn't he do it? If he's not doing it, it's because there is a competing motivator in the environment. Something else is making it better for him to not do it than to do it. So, it's still up to the trainer to find something which out performs whatever motivator the dog is getting elsewhere....a higher value treat, a somewhat hungry dog who isn't given his meal in it's entirety in one sitting, but where it is doled out by smaller amounts as the dog performs. If a dog has some deprivation, it increases the value of the reward. So, when practicing skills to get more reliability, it is good to have a dog who wants....who wants treats, who wants affection etc. So, if he's not given everything for free, he will appreciate those things more when used as reward. We waste a lot of chances for reinforcing behaviors. And some rewards are not just food. If a dog wants you to let him outside, that can be a reward. If a dog wants to cuddle, that can be a reward, something he earns. When a strong enough history of reinforcments have been given, the behavior will be proofed. If it is not, then it's back to more practice....not a good reason for punishment. (imo)

That does't mean I don't get cross with my dogs sometimes. But not when training "skills" persay. That's because I'm a primate. That's a whole other story. LOL

Roxy...I edited this. You had already typed your good testimonial post while I was still typing this. I've used both ways too...back in the olden days. Yes, you're right....a big difference!

 

[elegy]

I don't understand what is meant by "balanced" training. Scientifically, operant conditioning is how all mammals learn. This excludes yanking on a collar.

operant conditioning very much includes yanking on a collar. it's called positive punishment.

operant conditioning in and of itself doesn't put these value judgements on aversives. it looks at what increases or decreases the likelihood of a behavior to repeat itself.

i'm not huge into aversives for training- i'd never leash pop a puppy certainly- and i don't use them when teaching new behaviors, because, like roxy said, i want training to be FUN!

but i absolutely use punishment in life with my dogs, and it's recommended in these behavior forums all the time. dog puts teeth on human during play, stop the game in walk away. dog barks at you for attention, withdraw attention, walk away. dog jumps up, withdraw attention, turn away. negative punishment all. at least in operant conditioning speak.

 

[GSDlover_4ever]

Scientifically, Doberluv, operant conditioning involves BOTH positive and negative reinforcers.

I agree with lastkid about the balanced "diet", lol. I use the WHOLE idea of operant conditioning, rather than ONE element. My dogs get corrected and praised. Learn from both negative and positive experiences. Personally, I wouldnt pop a leash on a dog of that size, too much stress on the neck, but a 100lb GSD is a different story, .

Never TEACH a dog through compulsion. You want a happy and energetic dog. The time will come when it is appropriate to give a correction but a dog so young has not built enough confidence in itself to accept a correction and move on. Correcting at this learning stage will create a fearful, standoffish dog.

 

[lastkid]

I don't understand what is meant by "balanced" training. Scientifically, operant conditioning is how all mammals learn. This excludes yanking on a collar.

By 'balanced' I meant, not Koehler but also not relying entirely on positive. Balanced to me is like, 99.9% positive, and 0.1% corrections for proofing purposes only. Which isn't really (literally) balanced at all, lol. I really believe that positive reinforcement is the best way to train behaviors, but I also believe there's a point where, for some dogs, it's useful to use the other three parts of OC, including positive punishment. Not all of the time, not for every dog, and certainly NEVER during the learning phase.

If a dog has been taught to sit in a room with no distractions at all, and has been rewarded for responding correctly 10,000 times, but doesn't respond on the 10,001st time, it's not irrational to me to give a verbal or light leash correction. Because, and maybe I'm not thinking through this all the way, it doesn't make sense to me that a dog hasn't learned something after 10,000 repetitions, in a room with zero distractions.

Like I said, my personal preference is corrections during proofing ONLY. They're not "yanking the dog's head off" corrections, they're very light - more attention-getters than anything. Like a tap on the shoulder. "Nope, not that, try something else" or "Hey, would you please do what I just told you to, instead of staring off in the distance? I know that's really interesting, but if you do it, something even MORE interesting will happen." I never correct and end the behavior there. It's always, "Nope, not that." and when the dog does what I wanted, there's an immediate "YES!!!!!!" I've never, ever had a problem with motivation or a dog shutting down/getting aggro on me. If I did, I sure as heck would back off corrections for that dog immediately.

To me, it really depends on the dog, on the behavior, on the response. Using corrections in the (very) limited way that I do is a preference, that not everyone shares (nor do I expect them to). I've done a lot of reading, a lot of hands-on research, and a lot of talking to every trainer I can get my hands on from as many disciplines as possible. It works for my dog, with no ill effects to her or to our relationship, which is the most important thing to me.

 

[Roxy's CD]

all on leash right now but it's alot of "lightly tugging on the leash and using No with it

Elegy, your right. Corrections are important. Well, I find verbal is good to give them "guidelines". Ah ah, wrong etc.

But this seems based on leash corrections than the verbal...

Which is exactly what I did, and EXACTLY why we're having problems now with the activities that were trained with these methods.

Ex) I can go from doing a retrieve. (Roxy's tail is wagging, couldn't be happier) to some heeling work. (She shuts down. Tail stops wagging, her bounce in her step disappears... nothing could motivate her)

 

[Brattina88]

I'd say try positive training again... with someone a little more qualified than a Pets Mart trainer

 

[Doberluv]

Technically, of course, you're right Elegy. I understand about positive punishment and that punishment can be as simple as stopping a game if the teeth hit you. Actually, it can be that you're removing something good rather than adding something bad. Four consequences: good thing starts (positive reinforcement), good thing ends (negative reinforcement), bad thing starts (positive punishment), bad thing ends (negative reinforcement)

However, my point is that with most trainers these days who advocate the use of as little punishment as possible for the reasons I mentioned. For example, trainers like Karen Pryor. I agree with her ideas and those of Jean Donaldson. They explain why punishment, aversives often will produce undesireable side effects. And they explain why a dog doesn't do something and it's not because he's being stubborn or defiant or rebellious.

I use "eh, eh" as a no reward marker, quite different from a collar yank.

So, perhaps it's a matter of a difference in interpretation of ideas.

At any rate, to get back closer to the topic....my recommendation is that you don't rely on collar corrections and other aversives to train your dog. And instead rely on rewarding and reinforcing with something the dog loves for behaviors or approximations of behaviors you want. Ignoring in certain cases, distracting and giving alternative, incompatible behaviors which can be rewarded, rather than punishing with aversives unwanted behaviors is most effective.

I recommend the books, Culture Clash by Jean Donaldson, Don't Shoot the Dog by Karen Pryor, The Power of Positive Training by Pat Miller. These will explain in detail how it all works.

 

[DanL]

Corrections have no place in training, IMO. What is the point of training if it's all about avoiding a yank on the leash? The dog doesn't learn easily that way.

If corrections are to be used, they are best used as proofing for commands that the dog already knows.

I'm not trying to be smart, but didn't you have a recent experience where your training methods didn't work with one of your dogs when it was placed in a new environment? Wasn't his behavior something that is normally in control in his usual environment, but you were forced to have him in a new situation and he was acting out? This would have been an ideal place to use corrections. You were proofing an already learned behavior.

In this situation, with a young pup, then I agree, unless the pup knows what is expected of him when he is given a command, then correcting him is not doing a lot of good. That's not to say that using a leash and directing him to the behavior you want can't work, but you have to be careful.

 

[dr2little]

I'm not trying to be smart, but didn't you have a recent experience where your training methods didn't work with one of your dogs when it was placed in a new environment? Wasn't his behavior something that is normally in control in his usual environment, but you were forced to have him in a new situation and he was acting out? This would have been an ideal place to use corrections. You were proofing an already learned behavior.

In this situation, with a young pup, then I agree, unless the pup knows what is expected of him when he is given a command, then correcting him is not doing a lot of good. That's not to say that using a leash and directing him to the behavior you want can't work, but you have to be careful.

Physical correction would absolutely not be appropriate for a dog who does not comply in a new environment. Behaviors don't automatically transfer seamlessly, they do have to be trained first in new environments before one can assume that the dog is choosing not to comply, especially in a young dog such as the one RD was talking about.

I agree that leash corrections are totally inappropriate for a pup.(although I don't use physical "stuff" for any dog with any issues)
Not only is it ineffective but can cause life long problems for a dog during a fear period not to mention with the particular breed you're talking about (Shih Tzu).

I'd ditch the trainer and find someone who knows what they're doing. Ask for not only their background but where they learned how to train dogs. It sounds like this trainer needs some education...a little compassion for your pup from the trainer wouldn't hurt either.

I have to add, I just came home from teaching class 4 of a 6 session puppy class. The behaviors learned and demonstrated by ALL puppies reliably tonight were: Watch me, Sit, Down, Leave it, Stand for Vet exam, Hold (muzzle) for Vet exam, Stay, Drop it, By me (loose leash walking), Wait, Go to place, and all problem issues such as jumping up (off), bite inhibition (no nipping)...I know I'm missing some... without "leash pops". Pups in this class range from toy breeds to large working breeds.
All of these were done WITHOUT physical correction and I don't allow any correction collars in any level of class. We are also fading lures, rewards and clickers for the first 6 behaviors with great success.

 

[Doberluv]

Thank you Dr2Little. My point exactly.

 

[Bailey+Ralph]

Really, from someone that's done it and is now paying the price ditch corrections, purely motivational training IS the way to go.

Really Roxy knows what she is talking about........she is now having problems with Roxy and needing to go through alot more training to get her where she needs to be.

Please listen to someone that thought along the same route your going.

 

[silverpawz]

I agree with lastkid about the balanced "diet", lol. I use the WHOLE idea of operant conditioning, rather than ONE element. My dogs get corrected and praised.

I very much agree with this. I trained my own dogs and client dogs the same way. In the learning phase there is no corrections at all, once the dog knows a command and has shown consistantly that he understands then I will correct for not responding.

I don't care if there's a competing factor in the enviroment. Sit means sit, and come means come. A lot of obedience training is life saving skills for the dog. For example, let's say a dog gets out of the house and refuses to come when called and instead goes after that squirrel crossing the road. Dead dog.

I'd rather make it absolutly clear that obedience is not optional even if they're distracted. But that's just me. If someone doesn't want to train that way and they're having success their own way then go for it. I'll do what works for me, and everyone else can do what works for them.

For a puppy just learning? Nope. Wouldn't correct at this stage, not fair to the dog and possibly damaging to the relationship. I'd find another trainer that has more experience working with young puppies.