Purely Positive - A Few Questions

[Nekhbet]

dont decide for the dog what training you will do. You let the dogs behavior dictate which direction you go in. Its a dog and it wants food - you are making a more difficult and convoluted job by trying to make a dog come away from easy pickings to something of higher value. Hard to proof as well.

BUT the point of being reprimanded or corrected or whatever you chose - the pup accepts you DO NOT take food unless you are told. Period. No weighing up the options it is taught snatching food from your child is unacceptable and hence you find dogs will not bother as much. IMO letting the dog know it will get something either way defeats the purpose - the dog can also decide well if I show interest in the child I'll get something really good.

Whereas a dog that tries to follow the child and gets a good "aaagghhhh NO - LEAVE" thinks "whoah OK there's a no no"

Dogs are like children they need to learn boundries and consequences to their actions and that wont happen if we keep giving them something better or constantly providing treats. Yeah I'm old fashioned too but there is nothing wrong with that if you are not causing the dog harm.

[jenz0r]

i find purely positive training quite challenging with our pup, the bully breed in him makes him very stubborn and he pushes his naughtiness to the limits. our trainer says just to ignore, ignore, ignore when he's not doing what hes meant to.... but i find a correction here and there makes the training quicker and eaiser.

i say do what works for you

[poodlefan]

i find purely positive training quite challenging with our pup, the bully breed in him makes him very stubborn and he pushes his naughtiness to the limits. our trainer says just to ignore, ignore, ignore when he's not doing what hes meant to.... but i find a correction here and there makes the training quicker and eaiser.

i say do what works for you

So if he was jumping up or nipping you're just supposed to ignore it?? As I said before, its nigh on impossible to use such methods to extinguish behaviour the dog finds self rewarding. If he finds it enjoyable, he'll keep doing it. I don't think its coincidence that most 'problem' dog behaviors fall into the self rewarding category.

It's only IF he finds the behaviour brings consistent negative consequences that you'll see it diminish or extinguish. You don't have to be abusive to make behaviour unrewarding. Removing yourself or him could be enough.

I always smile to myself when purely positive trainers suggest that putting a behaviour on cue can cure problems. Barking is the classic example. My dogs sit on cue. That hardly means they don't sit at other times though now does it.

Ask a purely postive trainer how to cure animal chasing.. and don't hold your breath on a practical response.

Edited October 1, 2008 by poodlefan

[leopuppy04]

Ask a purely postive trainer how to cure animal chasing.. and don't hold your breath on a practical response.

hmm I'm curious - PF - what would you do in that situation and also what do the 'purely positive' trainers recommend?!

Never really asked others that one before

[Guest Tess32]

I've still yet to meet a purely positive trainer. It doesn't work and I don't see the point. People would spend more time figuring out HOW to do something than doing it.

There's a reason there are FOUR quadrants to operant conditioning, not one.

[Luke W]

Ask a purely postive trainer how to cure animal chasing.. and don't hold your breath on a practical response.

hmm I'm curious - PF - what would you do in that situation and also what do the 'purely positive' trainers recommend?!

Never really asked others that one before

I've lost my copy of "Don't Shoot The Dog"

- Train an incompatible behavior.

- Put the behavior on cue

There was a couple of others.

[poodlefan]

Ask a purely postive trainer how to cure animal chasing.. and don't hold your breath on a practical response.

hmm I'm curious - PF - what would you do in that situation and also what do the 'purely positive' trainers recommend?!

Never really asked others that one before

Leo I'm not a purely positive trainer.

Luke W is right .. .they would try for an incompatible behaviour like a recall. No one in their right mind is going to put stock chasing on cue in the hope that it can be prevented at other times. Luke did you see what I read about cueing barking.. same deal as cueing sitting don't you think?

My view is when a dog has a history of highly rewarding stock chasing aversives are called for. The other option much loved by the purely positives is to manage the behaviour to ensure that the opportunity to chase stock never occurs - that's not addressing the issue directly and frankly not always possible.

Some pragmatic positive (not purely positive) trainers consider this one of the few justifiable reasons to use an electronic collar aversively. The reason is simple - the dog's life and the safety of stock are at risk. The behaviour is so rewarding to the dog that strong aversives are needed to deter it.

That sure beats a bullet.

If you've tried to make yourself more interesting and offer a reward than prey, you can see why its so hard to deter chasing.

Edited October 1, 2008 by poodlefan

[Luke W]

poodlefan

Saw the comment about barking...

I think I agree, but I'm not experienced or knowledgable enough on conditions to say for certain.

Putting my dogs barking on cue was the start of my barking problem

...added

PS - I've read the struictly speaking "under stimulus control' means the behavior only occurs in the presence of the cue - never any other time.

That would mean to have 'sit' under stimulus control - the dog would never sit unless told...hmmm..... ?!?!?!?!?

Edited October 1, 2008 by Luke W

[poodlefan]

poodlefan

Saw the comment about barking...

I think I agree, but I'm not experienced or knowledgable enough on conditions to say for certain.

Putting my dogs barking on cue was the start of my barking problem

...added

PS - I've read the struictly speaking "under stimulus control' means the behavior only occurs in the presence of the cue - never any other time.

That would mean to have 'sit' under stimulus control - the dog would never sit unless told...hmmm..... ?!?!?!?!?

A healthy dose of common sense is all that you need to see that naturally occuring behaviours will occur whether or not you put them on cue. If you want to extinguish them except in the presence of the stimulus, then you're going to have to use aversives or control what stimulates the behaviour completely (that's "management")

Dogs will do more of what they find rewarding and less of what they find you don't. For puppies, they tend to offer behaviours they know you will reward. That's why you'd be getting a spike in barking I reckon. Of course you could simply have a barky dog.

I don't generally encourage my dogs to bark.. I spend enough time and energy discouraging them from doing it in the first place.

Lily will do bark to the cue word "cows". That's because she'll do it when she sees Herefords (only Herefords make her bark.. long story) and I added the cue. She'd do it whether or not I gave the cue though.

Edited October 1, 2008 by poodlefan

[leopuppy04]

Ask a purely postive trainer how to cure animal chasing.. and don't hold your breath on a practical response.

hmm I'm curious - PF - what would you do in that situation and also what do the 'purely positive' trainers recommend?!

Never really asked others that one before

Leo I'm not a purely positive trainer.

Luke W is right .. .they would try for an incompatible behaviour like a recall. No one in their right mind is going to put stock chasing on cue in the hope that it can be prevented at other times. Luke did you see what I read about cueing barking.. same deal as cueing sitting don't you think?

My view is when a dog has a history of highly rewarding stock chasing aversives are called for. The other option much loved by the purely positives is to manage the behaviour to ensure that the opportunity to chase stock never occurs - that's not addressing the issue directly and frankly not always possible.

Some pragmatic positive (not purely positive) trainers consider this one of the few justifiable reasons to use an electronic collar aversively. The reason is simple - the dog's life and the safety of stock are at risk. The behaviour is so rewarding to the dog that strong aversives are needed to deter it.

That sure beats a bullet.

If you've tried to make yourself more interesting and offer a reward than prey, you can see why its so hard to deter chasing.

I don't see how teaching an incompatible behaviour (such as a recall or a down) is not being positive.... (not that I care... I will do what works...)... basically I am trying to say is how can some trainers find this not suitable?!

I don't disagree with anything you have said here PF.

My only argument 'for' putting barking on cue should someone want to recommend it is more on the opposite side - eg: teaching the dog a quiet command - so if a dog barks when people come to the door, giving the dog a 'quiet' or 'enough' command which you can heavily reward, in my mind would work.

[Luke W]

poodlefan

Saw the comment about barking...

I think I agree, but I'm not experienced or knowledgable enough on conditions to say for certain.

Putting my dogs barking on cue was the start of my barking problem

...added

PS - I've read the struictly speaking "under stimulus control' means the behavior only occurs in the presence of the cue - never any other time.

That would mean to have 'sit' under stimulus control - the dog would never sit unless told...hmmm..... ?!?!?!?!?

A healthy dose of common sense is all that you need to see that naturally occuring behaviours will occur whether or not you put them on cue. If you want to extinguish them except in the presence of the stimulus, then you're going to have to use aversives or control what stimulates the behaviour completely (that's "management")

Dogs will do more of what they find rewarding and less of what they find you don't. For puppies, they tend to offer behaviours they know you will reward. That's why you'd be getting a spike in barking I reckon. Of course you could simply have a barky dog.

I don't generally encourage my dogs to bark.. I spend enough time and energy discouraging them from doing it in the first place.

Lily will do bark to the cue word "cows". That's because she'll do it when she sees Herefords (only Herefords make her bark.. long story) and I added the cue. She'd do it whether or not I gave the cue though.

Mine 'demand barks' during training - it's nearly extinguished though.

I also made a mistake of using the cue 'bark' for a dog who's name is 'barkly' D'OH !!!!

[poodlefan]

Mine 'demand barks' during training - it's nearly extinguished though.

I also made a mistake of using the cue 'bark' for a dog who's name is 'barkly' D'OH !!!!

Ah isn't hindsight is a wonderful training tool! .

[leopuppy04]

Mine 'demand barks' during training - it's nearly extinguished though.

I also made a mistake of using the cue 'bark' for a dog who's name is 'barkly' D'OH !!!!

OOPS!

Don't you love the frustration barkers I live with 2!

[Reddii]

Ask a purely postive trainer how to cure animal chasing.. and don't hold your breath on a practical response.

hmm I'm curious - PF - what would you do in that situation and also what do the 'purely positive' trainers recommend?!

Never really asked others that one before

Leo I'm not a purely positive trainer.

Luke W is right .. .they would try for an incompatible behaviour like a recall. No one in their right mind is going to put stock chasing on cue in the hope that it can be prevented at other times. Luke did you see what I read about cueing barking.. same deal as cueing sitting don't you think?

My view is when a dog has a history of highly rewarding stock chasing aversives are called for. The other option much loved by the purely positives is to manage the behaviour to ensure that the opportunity to chase stock never occurs - that's not addressing the issue directly and frankly not always possible.

Some pragmatic positive (not purely positive) trainers consider this one of the few justifiable reasons to use an electronic collar aversively. The reason is simple - the dog's life and the safety of stock are at risk. The behaviour is so rewarding to the dog that strong aversives are needed to deter it.

That sure beats a bullet.

If you've tried to make yourself more interesting and offer a reward

Completely agree with this. My girl doesn't chase stock, but when we were herding she was very hard on the stock and needed to be kept out using aversive (physcal) methods. We got her to the point where she was quite good and manageable on larger (in city terms -50-100) mobs, but it was still difficult on smaller numbers (<10). I didn't like how hard we had to be on her to keep her out so we don't play with sheep any more, but we don't live in the country either so it isn't really a problem.

Interestingly we've been to some shows where there has been a lot of stock and she hasn't glanced sideways at them, even walking next to a field where they are being worked. (OT)

That said I train for flyball with some purely positive trainers and I really don't believe that it is as effective as a balanced approach. They trust me with their dogs and will let me work them my way so as an interesting point for comparison - one of the dogs is VERY high drive, VERY arrogant and VERY tough to manage. The owner does a great job managing him within the bounds of what PP will allow her to do. That said when she plays frisbee with him the frisbee becomes a tug toy every time he brings it back. I had never played frisbee with him before and was at a big pet expo where we had a competition. By verbally correcting him and turning my back and withdrawing the game I had him dropping the frisbee at my feet within 5 minutes - he still does this to this day.

I'm not saying I'm better with dogs than she is, but I think it does show that by applying even gentle corrections where appropriate it is possible to get a much better result from most dogs. I think the best advice I've ever had is 'let the volume of the correction be determined by the volume of the offence'. By starting off with a mostly positive approach and being consistent I think you should be able to avoid harsh corrections on most dogs, but know there is an exception to every rule.

[Purpley]

Its all very interesting everyones different takes on it all. I am going to continue with as positive as I can, also using the tone of voice, as this seems to be working for him at the moment.

He has learnt to not take food off the older 2 kids (4 and 8yrs old), I think that with bubs he is more at his height, so it makes it more enticing. So its going to take a little longer for him to get it. It all takes time.

I haven't had puppies before, as I have a rescue, and Bronson was a show reject so I got him at 7.5 mths, so this is a rather steep learning curve for me, but I am loving the challenge.

I really appreciate everyones imput.

[percyk]

just so depends on the dog doesnt it

i have one here

hes a dream

once he had a bone in his mouth - got if from somewhere and brought it into the house

it was his first offense with this kind of behaviour

i yelled NO DROP IT

and he SPAT it out so hard it literally spun across the tiles

hes that kind of dog

and no he doesn wee in submission or even show belly

then theres my friends dog who growled at her because she shooed him away from the cat food that was high up - she took him by the collar and said a firm short no and he growled

he has been growling at little things he doesnt want to do but she is using the distraction and praise method rather than confronting him head on

so he goes to chew on her lounge - she uses my trick with 'will you be having tabasco with that" and it has stopped him but generally because his recall is good she can call him off without tugging at him in most situations

how successful she will be - time will tell

she will go to obedience and work him hard and talk to someone on the central coast about getting a professional to look at him

so whether you go positive or mix it with a bit from the other school of thinking depends on the nature of your dog i think

some dont need any heavy handed techniques at all

look at how differently we housebreak our dogs from the old 'rub his nose in it' of the 60s and even 70s!

i dont even comment on their mistakes

but with some i dont even reward with a treat outside ( with some i do - the ones that are taking longer )

id try to minimise the contact your dog has with your son when hes eating ( perhaps your son could eat a snack a table and you could have your dog on a leash and do some training which involves him focussing on you while your boy is eating)

then when it's trained a bit more you can challenge it a bit more- otherwise it's pretty hard to learn so much

[corvus]

I like to say I'm a LIMA trainer, Least Invasive, Minimally Aversive. When I got my puppy I was determined to be as positive as possible with him. I believe it is more or less doable as I raised my wild hare with almost entirely positive interactions with him. With a hare, you can't ever frighten them or make something unpleasant or you'd lose them for good. I think that just because SOME dogs respond favourably to punishments or "corrections" doesn't mean that you should use it. We are human after all, which means we can fly in the face of nature if we feel like it and have the brains to figure out ways to make that work for us. Don't let anyone tell you you can't be purely positive because you can. You just have to be prepared to accept certain things your animal does as the way they are and manage rather than train. There is nothing inherently wrong with managing. I manage my pet rabbits' chewing desires because I'd rather not risk the carpet, the walls, the furniture, and the rabbits should they chew something like a live power cord. I don't know a single rabbit owner that would consider that "not addressing the problem". My point is, just because you can train something doesn't mean you should, and there's absolutely nothing wrong with deciding to manage rather than use P+ or something else aversive.

Anyway, for all my efforts I discovered it was much easier to be positive with a hare than a puppy. Puppies have their moments when they are driving you up the wall and you snap before you think of a positive way out of the situation. I think that puppies are quite resilient to this sort of thing. One thing to remember is that all animals have their good days and bad days, and I think they are all capable of understanding to some degree that you can be cross with them and still be their best friend a moment later. Don't beat yourself up too much if you make mistakes.

I think more important than whether you use punishments or corrections or not is the way you use them. My rule of thumb is that I don't want to do something that has a particularly lasting impression on my animal. If they are subdued for more than a few seconds after the event, then I was too harsh. That's my personal rule of thumb, but I'm not trying to teach my animals never to do something again. I'm trying to teach them that one behaviour is a less satisfying choice than another. So ignoring a puppy when they do something you don't like and rewarding them when they pick something more appropriate. And yeah, incompatible behaviours. I also like to use classical conditioning. You'd be amazed how far you can get with consistency alone. You can condition a punishment in a way. Like when you say "ah-ah" in a tone that isn't punishing to the dog, but follow it up with always shooing the dog away from whatever it's doing or something similar. Eventually "ah-ah" comes to mean the dog will be shooed away, and they start moving automatically as soon as you say it. Worked with my hare. I don't consider this very punishing unless you have an animal that is somehow terrified of being gently hustled away or distracted from something.

Anyway, good for you for wanting to be positive. It's a little harder sometimes, but I think you feel better for it and a lot of crossover people tell me the dogs respond better. I'm a crossover person myself, and I believe it.

[Erny]

My point is, just because you can train something doesn't mean you should, and there's absolutely nothing wrong with deciding to manage rather than use P+ or something else aversive.

I agree that +R is great to use when it is what works best for the dog and situation. But comparing a rabbit or hare with a puppy is not, IMO, the best comparison to make. I mean, you might be happy to "manage" your hare's ...... ummm, shall we say "issue", but hares or rabbits are not thrust into and amongst society as dogs are. In addition, rabbits and hares are animals of prey - with a different psychic to dogs who are hunters of prey. At some stage, "management" breaks down through sheer human error; misjudgement; or whatever. It might take a while, but it does. Consequently some behaviours that you might think are ok to merely be "managed" could in fact end up being disasterous for the dog, or other people or other dogs etc. etc. etc. So I disagree with what you've said above in (my) highlights. There is something wrong with deciding to manage some behaviours when in fact the application of an aversive could well prove to thwart an escalation of behaviour to more serious levels. Whether that be serious for the dog or serious for other animals (including people).

I haven't "crossed over". My mind is open to using any one or all of the four quadrants of learning. Which one I use depends on what best befits the dog and the situation. "Crossing over" in the context you've used it only means that you have denied yourself being able to use one of those quadrants (ie P+) even though there might be an instance and a dog for which this might prove the most beneficial and in the end the most congeniel to the dog's long term welfare. This IMO poses a problem for those who profess to have "crossed over". It precludes them from applying that 'missing' learning principal even though it might be the best, most efficient and only way to curb a behaviour from becoming serious and injurious to others.

Isis - I say good for you for going the positive track. I prefer to work (especially) pups this way as I like to teach a dog what TO do (preferably) before I must show him/her what NOT to do. But I won't preclude myself from using +P if else fails or is not efficient enough for everyone's sakes. Hope little puppy is doing well. Do we have some photos? You know they are a must and I don't remember seeing any here .

Anyway, good for you for wanting to be positive. It's a little harder sometimes, but I think you feel better for it and a lot of crossover people tell me the dogs respond better. I'm a crossover person myself, and I believe it.

There are many times I have seen injurious behaviours from dogs simply because their owners did what "they felt better" about doing. IMO we need to think it from a dog's learning perspective and not necessarily always what makes us feel better.

Edited October 5, 2008 by Erny

[Tonymc]

Corvus, sorry but your post has many holes.using the logic you imply in your post,I will not use any aversive to stop my dogs crossing a busy road, but will simply manage the resulting mess. Tony

[corvus]

Well as it happens, Tony, I don't use any aversives to stop my dog from crossing a busy road because my dog has a very good recall I have taught with PURELY POSITIVE means a la Leslie Nelson. He turns on a pinhead when he hears that emergency recall and gallops into my arms. It's not perfect yet, but I'm happy to make sure he's on leash around busy roads until I'm 100% confident with it. In fact, I am happy to keep my dog on leash around busy roads even when I am 100% confident. I would never trust one of my animals around a busy road with any contol I had instilled in the animal with aversives for one big reason. I won't do anything to my animal that is bad enough to overcome their desire to do something they think is really great. If pup sees a dog on the other side of the road and wants to go play with it, I've got nothing bad enough to make him think that's not such a great idea, however, I do have something GOOD enough to make him think heeding me is a better idea.

Also, if you check my post I said that it was much easier to be purely positive with a hare than a puppy. I do not claim to be a purely positive trainer, as you'll also find in my post if you look. I do and have used punishments, but here's something important: I am not the kind of person that likes doing those things. If I can get away with not doing something to my animals that they don't like, I sure will choose that way. Are you seriously telling me that I am WRONG to want to avoid doing something to my animals that they don't like? I am a zoologist and the art of least invasive, minimally aversive has been so drilled into me that I automatically think of alternatives to force and punishments first. You've got to be true to yourself in life, and that includes training another animal. If being true to yourself means avoiding aversives, then more power to you for recognising your own limitiations.

And no, there's nothing wrong with management if that's what you choose. Management does not always break down inevitably if you account for your own shortcomings. I have a 2 degrees of separation rule for my rabbits and dogs that I don't know well. There is always at least 2 closed doors between dogs and bunnies for the moment when human failure could cause disaster. It's saved the hare maybe twice and has meant that I don't have to expose the hare to excited dogs that want to eat him to train something I will never trust anyway. I work full time and spend a lot of time in the field. I don't have time to sit down and train away every little thing that could cause trouble in the future. I manage and link management with training if it's something I care very much about.

The argument that you're denying yourself some part of the operative conditioning quadrant is, IMO, one that doesn't hold a lot of water. I don't really see why you have to use every quadrant or some such just because they exist. I do use punishments, but I don't use ones that are strong enough to actually stop behaviour forever. I'm finding that it's working quite nicely for me. I will always be able to change my methods if I find they are not working quite nicely for me, so how am I denying myself of some training tools when I simply haven't found a need for them yet? And this may come as a surprise, but I don't actually mentally classify everything I do into OC quadrants and make sure I've picked R+ and R-. I just make a habit of picking least invasive and minimally aversive methods. You've got to do what suits you, and so I do. There's no point upsetting yourself using punishments or force if you can make yourself feel happy avoiding them as long as everyone is safe and you are meeting your own expectations in training.