[Westhighlander]

Can police dogs and protection dogs be trained with positive reinforcement methods? Seems most trainers in this field use mostly negative methods. Thoughts?

 

[Marsh Muppet]

I won't say it can't be done, but there is a question WRT whether it can be done economically. Economics would be a major consideration for a large K-9 program. "Always" and "never" are terms I try to reserve for physical laws. Intelligent beings have a way of messing up absolutes.

I haven't messed with biting dogs since Reagan was in office so much may have changed. AFAIK, the Baltimore Method is still the SOP in large PDs.

I see a number of fundamental problems that would need to be overcome before purely positive training could be successful. I recently read a paper by a trainer/behaviorist from New Zealand (IIRC) who extolled the benefits of modern, positive training methods for police dogs. Very near the end of a 4 page pdf document, the author admitted as how physical correction would have to be used towards the end of the training. That may be 100% correct, but it seemed a bit misleading.

 

[wvasko]

Well I do believe parts already are, I think the biting is positive rewards in itself. It's the outing that may require other attitudes. It has been said before that a good training program should be balanced.

Yes, I too am from the Reagan era of dog work.

 

[Marsh Muppet]

...I think the biting is positive rewards in itself. It's the outing that may require other attitudes.

There's one of the big fundamental problems. To get the dog to release his bite, you'd have to find something that's more rewarding than the arm that's bleeding into his mouth. Good luck with that.

 

[wvasko]

There's one of the big fundamental problems. To get the dog to release his bite, you'd have to find something that's more rewarding than the arm that's bleeding into his mouth. Good luck with that.

That's what I'm talkin about. I don't think Spam will work.

 

[zimandtakandgrrandmimi]

there are many many police detection dogs trained using positive reinforcement.

Protection work I can see as very possible for positive methods...PROOFING those behaviors would be a different story.

 

[Westhighlander]

Why does negative reinforcement get such a bad rap then if ultimately it works more effectively than positive training in certain instances?

 

[FilleBelle]

Why does negative reinforcement get such a bad rap then if ultimately it works more effectively than positive training in certain instances?

Off the top of my head, I would say it has a bad rap because people are idiots. A trained K9 officer practicing tried and true methods as part of a police program is vastly different from next-door-neighbor-Bob incorrectly employing a choke chain on walks and whacking his shelter mutt when it does something he doesn't like.

Negative reinforcement is one of those tools that can be dangerous in the wrong hands. It's easier to advise everyone not to use it than to give people the option and risk someone screwing it up.

 

[mostlymutts]

Why does negative reinforcement get such a bad rap then if ultimately it works more effectively than positive training in certain instances?

I suspect that because most instances do not involve getting a dog to release a bloody arm. Those "certain instances" are the exception rather than the rule.

 

[Marsh Muppet]

I suspect that because most instances do not involve getting a dog to release a bloody arm. Those "certain instances" are the exception rather than the rule.

And other instances, where physical correction may be appropriate, don't necessarily require such emphatic means. Patrol dog training can get pretty rough. There is a time to sow, and a time to reap.

 

[zimandtakandgrrandmimi]

officers arent immune to being abusive towards their animals. in the city i live in not too long ago there was an abuse case involving a policeman and his k9 partner.

i do think there is room for reform in those arenas but its a topic that positive training has yet to fully explore. notice that "yet".

 

[Curbside Prophet]

Why does negative reinforcement get such a bad rap then if ultimately it works more effectively than positive training in certain instances?

Because negative reinforcement uses aversion, and those who use aversion don't typically balance Pavlov with Skinner, nor do they reason to.

I attended a seminar on punishment in dog training and the discussion was led by a few R+ trainers, a dominance theorist, a long time police dog trainer, and they all agreed the methods used by the police dog trainer were necessary for the type of work police dogs are expected to do. I'm not talking about SAR work or tracking, but rather biting the bad guy type stuff.

But to answer your original question, yes, police dogs can be taught some behaviors through positive reinforcement, and some police dog trainers do. I don't believe we can generalize police dog trainers as trainers who exclusively use negative reinforcement and/or positive punishment.

 

[Spicy1_VV]

Why does negative reinforcement get such a bad rap then if ultimately it works more effectively than positive training in certain instances?

I've no idea. Watching dogs interact and teach I don't see plenty negative. Negative does mean bad, it doesn't mean its a negative thing to do I mean.

Off the top of my head, I would say it has a bad rap because people are idiots. A trained K9 officer practicing tried and true methods as part of a police program is vastly different from next-door-neighbor-Bob incorrectly employing a choke chain on walks and whacking his shelter mutt when it does something he doesn't like.

Negative reinforcement is one of those tools that can be dangerous in the wrong hands. It's easier to advise everyone not to use it than to give people the option and risk someone screwing it up.

That is very valid. Reminds me of this chic the other day. She was wanting to buy a prong collar and was with her man friend.

She says its not inhumane or they wouldn't sell it. Then she say well I used the shock collar and that didn't bother her so this will be fine or something along those lines. and because "she is going to learn to walk without dragging me" Then she was saying about DA, that she (the dog) tried to get a dog (chihuahua I believe she said) through the fence. That she (the lady) was kinda proud of her (the dog). That for 6 months old she is nasty already. I wonder why, idiot owner the reason, yes I'd say so. Senseless people shouldn't own dogs let alone try to train them.

 

[Xeph]

Positive methods are LARGELY used for all sorts of training...especially the obedience. However, the reason so many remain in negative reinforcers in the bitework is because they are necessary.

When a dog is in high drive, especially the dogs bred specifically for K9 units, these are dogs with relatively high pain tolerance and a certain hard headedness.

They have clear heads in "level" obedience, but get them ramped up for the bite and while they still have a clean bite with good grip, the immediate relief of endorphins from the excitement of the reward of the grip can make the dog "Grip stupid". The dog cannot out because he cannot "hear". Nothing registers...like tunnel vision.

Out is EXTREMELY important in any kind of bitework. When you're sending your dog after a bad guy, you are sending him ONLY to apprehend and hold...NOT to tear the bad guy apart. When the dog takes first bite, the department is still in the clear, but the moment you tell that dog to out, if he doesn't and causes more damage on the bad guy...he could very well turn around and sue the department, because he "got more than he deserved".

On top of that, if the dog misdirects to an innocent person (it does happen....people showing up in the wrong place at the wrong time), that dog needs to out immediately #1 because it's the WRONG person, and #2 the REAL Perp is still on the loose!

Dogs are not infallible...they make mistakes, but there can be NO mistakes made when it comes to somebody possibly losing a limb or muscles in their body...even if they deserve to burn in Hades.

 

[Westhighlander]

Can I conclude that done properly negative reinforcement methods can be more effective than positive methods?

 

[Inga]

Can I conclude that done properly negative reinforcement methods can be more effective than positive methods?

I don't think that is true in general. Maybe for the "out" command during bite work yes, but not in general training. I think that positive reinforcement not only can get the job done but it can do so with a happy willing dog at the end. Negative reinforcement, especially poorly timed or extremes can actually make your dog shut down. It can hinder your training by creating all kinds of negative behaviors along with what you might have been trying to train.
As Xeph stated, in police dogs that were bred for that activity, they are "hard" in the sense of being less reactive to pain. More of a pain threshold if you will. Training your general household pet is different then training a police dog to do bite work.

 

[trumpetjock]

Can I conclude that done properly negative reinforcement methods can be more effective than positive methods?

No, you cannot. Setting up loaded questions like that doesn't get you anywhere.

What you can conclude is that for one specific behaviour in one specific area of dog training it is necessary for the .000001% of dogs that are trained for it.

Nothing else.

 

[Marsh Muppet]

I've now seen yet another well known advocate/practitioner of positive training methods use a moderate-to-severe negative correction on a dog. I won't mention his name, but it starts with "D" and rhymes with "fun bar". It was not a physical correction, but there is a not-insignificant number of dogs who would tolerate an appropriate collar correction--delivered without emotion--better than if you bellowed at them threateningly. To his credit, he left it in the DVD/video presentation, when it would have been a simple matter to edit it out and pretend it never happened.

I'm just sayin'......

 

[Westhighlander]

No, you cannot. Setting up loaded questions like that doesn't get you anywhere.

What you can conclude is that for one specific behaviour in one specific area of dog training it is necessary for the .000001% of dogs that are trained for it.

Nothing else.

It's only loaded if you are reading too much into it. I train only positively btw. I am always willing to learn which is why I am asking the question. I am just wondering if positive reinforcement has its limits.

My friend has a herding dog that is/was dog aggressive. Tried some positive methods which had some limited effects but then went to a negative trainer who uses a prong collar for little corrections. This seems to have worked for him, although I don't know how consistent he was trying to use the positive methods I suggested. Now this trainer suggested that for some dogs, positive methods just doesn't work as well. This trainer has won some awards for training police dogs and such.

Someday I would like to get a bigger dog when my lifestyle allows it so this got me thinking about what this trainer said.

 

[trumpetjock]

It's only loaded if you are reading too much into it. I train only positively btw. I am always willing to learn which is why I am asking the question. I am just wondering if positive reinforcement has its limits.

My friend has a herding dog that is/was dog aggressive. Tried some positive methods which had some limited effects but then went to a negative trainer who uses a prong collar for little corrections. This seems to have worked for him, although I don't know how consistent he was trying to use the positive methods I suggested. Now this trainer suggested that for some dogs, positive methods just doesn't work as well. This trainer has won some awards for training police dogs and such.

Someday I would like to get a bigger dog when my lifestyle allows it so this got me thinking about what this trainer said.

There's a pretty key part to that. Your friend had a dog with a psychological problem and tried to treat it themselves with positive training and failed. They then went to a ***trainer*** that used negative training and got results. The key part there doesn't look to me to be the method, but the person employing it. They more than likely would have got much better results, and just as quickly if they had gone to a positive trainer instead of "trying some positive methods". DA is not something the average (or even the above average) dog owner is equipped to handle, through any method.