I'll Just Throw This Into The Mix -

[corvus]

I

I am sure many people don't realise for example that Victoria Stillwell was sued for fraud by the network who produced her TV show because she is not actually a dog trainer like she claimed she was. When you watch her show (and parts are highlighted in the video for example) it is extremely clear she does not understand the four quadrants of operant conditioning and is exceptionally unprofessional, likely through lack of real experience. Yet she is held up by many in the training community as a spokesperson for "positive only" training. It's ridiculous and I think people like that should be questioned.

I am a bit over arguing about Victoria Stilwell but I can find zero evidence that she was sued by Channel 4. Channel 4 still has her on their website too. There were 7 series so I imagine that there would have been a fair bit of press around about it.

She also states that she was working as a trainer prior to the show on all official bios.

Thank you. I was going to say that, too, but maybe I shouldn't spend all my time on social media contesting people's comments. Victoria can call herself a dog trainer regardless of her background precisely because it's an unregulated industry. Can we all just grow up and get on with things rather than wasting all this time and energy trying to discredit people, now? If someone is breaking a law, then report them. If you're not going to report them, don't be accusing them and then saying you'll totally stand up in a court of law and back your claims. I don't care. If you think they are criminals, that's your problem. Go do something about it. Quietly. Because that is the professional way of dealing with someone doing something unlawful in your profession.

[Willem]

every time we use a leash (and by law we have to) to control the dog or pull him into the direction we decide to go, we use force and negative punishment: removing the stimulus (the direction / obstacle the dog aims for) to decrease behavior (dog pulling in this direction). By using a prong / choke collar it even adds some positive punishment.

...or when using a crate: just another removal of appetitive stimulus (freedom) to decrease behaviour (there is a reason why crates are made from solid material and not made from paper), hence its negative punishment.

For training like agility, yeah I can't see that anything else than positive reinforcement would give the same result, but the day has 24 hours and there is a lot of time left without an official obedience or agility training - still the dog needs to be managed and - by law - most dog owners who e.g. walk their dog have to use negative punishment from time to time for safety reasons...otherwise you could just use a string to walk the dog instead of a leash.

ETA: yes, I'm aware that the conditioning where a dog responds to the pain inflicted via a prong collar could also be negative reinforcement = escape (pain / stimulus gets removed after the dog follows the owner)...

Edited April 6, 2016 by Willem

[Mrs Rusty Bucket]

Michael D'Abruzzo wrote

There needs to be true regulation in this industry as to who can call themselves a professional dog trainer. There needs to be government regulated educational programs that teach the real science behind the aspects of training and canine behavior. The correct application of all training tools needs to be taught through training labs

I agree with this part of his article.

But then he gives no evidence of what he's talking about. Where is his record keeping of rewards vs aversives.

Mostly what "positive" trainers (and I think that term is incorrect way to describe what they're doing) are about is avoiding aversives like slip chain collars, prong collars, loud or severe reprimands, things old school dog trainer call "corrections"...

It's not that they never use +p. +p - applying something that the dog doesn't like to reduce an undesirable behaviour - that may be as simple as covering up the cup of food with your hand to stop the dog from stealing it. The dog doesn't like that - so hopefully (in combination with well timed rewards) it learns to leave the food alone unless it has permission to eat it.

But that's different from smacking the dog over the head with a newspaper for sniffing (thinking about stealing) the food. Or yanking it by the neck and firmly saying "leave it".

It doesn't help that the reward based trainers - (get reward/don't get reward) use the word "positive" when in the science jargon that can also be paired with punishment or aversives designed to discourage a behaviour.

What Michael doesn't seem to acknowledge is the increasing pile of scientific evidence that shows that using aversives in training is counter productive. It sets the training back. More so than just withholding a reward does. Tho you should see the way my dog carries on when she fails to earn the treat. But she does work harder next time and has way more joy when she gets it right than when I pay her for "participation".

PS I think he may well get into trouble for pairing a picture of Victoria Stilwell and a bunch of what look like police dog trainers - with the words "criminal acts".

As for verbal reprimands - what is the difference between telling your dog "no treat for that" and saying "ah ah". Tho personally - I don't like "ah ah" - I have been known to say "don't even think about it" and "OI" if she starts rolling in something stinky. Mostly she gets "ah hmm" if she's doing something I don't want her to be doing. Or I just give her something else to do (like drop or get the hat).

Edited April 6, 2016 by Mrs Rusty Bucket

[Salukifan]

I honestly don't give a toss about how trainers label themselves or how they train these days.

Just get out there and help people get results with their dogs. Provided the methods aren't abusive, I don't care what they are called.

Too much focus for me on those in the profession (I use that term loosely because I dont' think it qualifies) seeing who can pee higher and worrying who is training "right". What's needed is more focus on what matters - well behaved dogs and happy clients. How about less competition and more sharing.

Edited April 6, 2016 by Haredown Whippets

[corvus]

every time we use a leash (and by law we have to) to control the dog or pull him into the direction we decide to go, we use force and negative punishment: removing the stimulus (the direction / obstacle the dog aims for) to decrease behavior (dog pulling in this direction).

Only if the leash pull actually influences future behaviour. I think you will find that is a big 'if'. My dogs walk nicely on leash until they detect bread/chicken bones/cat poo nearby, and then they throw themselves in unison towards the ground lollies, creating heaps of leash pressure that I instantly contribute to by pulling them back. Yet, if anything, they do it MORE now than they did a year ago. What does that tell us? It tells us a) in this context pulling is not punishing; and b) in this context, the possibility of accessing the ground lollies is highly reinforcing and they have figured out the most likely behaviour to achieve this.

Guys. Quadrants are only confirmed AFTER behaviour change. Declaring anyone uses any quadrant without knowing what the effect on behaviour was is putting the cart before the horse.

[corvus]

People who practice positive dog training are actually using negative punishment all the time. Every time they withhold an expected reward they are using negative punishment.

No, they are not, actually. For that to be true, the behaviour that precedes "no reward" would have to be suppressed. And in fact, "no reward" itself would have to be punishing, which would be challenging, because often the dog doesn't know when that is going to happen. Many of us do not even signal when there will be no reward, and many of us are using variable reinforcement schedules, so the dogs are not especially bothered by "no reward" in any context. What is actually taking place is DIFFERENTIAL REINFORCEMENT. i.e. some behaviours are reinforced and some are not. If I reinforce lifting a paw to chin height but I don't reinforce lifting a paw to chest height, then lifting a paw to chest height is not punished. It declines simply because it's not being reinforced as much as higher lifts are. Otherwise, I would have a hard time getting any paw lift, let alone the high ones I am after.

Thank you for taking the time to actually write that out as I couldn't be bothered

The misinformation rage... So strong...

Ironically, I'm not going to back up my assertions with credible sources this time. I'll just pers. comm. my behaviour analyst friends. :D

[Mrs Rusty Bucket]

I think you will find that is a big 'if'. My dogs walk nicely on leash until they detect bread/chicken bones/cat poo nearby, and then they throw themselves in unison towards the ground lollies, creating heaps of leash pressure that I instantly contribute to by pulling them back.

So glad I'm not alone.

[JulesP]

I honestly don't give a toss about how trainers label themselves or how they train these days.

Just get out there and help people get results with their dogs. Provided the methods aren't abusive, I don't care what they are called.

Too much focus for me on those in the profession (I use that term loosely because I dont' think it qualifies) seeing who can pee higher and worrying who is training "right". What's needed is more focus on what matters - well behaved dogs and happy clients. How about less competition and more sharing.

Any trainer that starts entering into a peeing contest or bad mouths another trainer is immediately taken off my list of people I would use.

I am particularly impressed if they refrain from peeing even if given the opening to do so ...

I also use that rule for tradies etc though. It is just not professional behaviour.

[Willem]

every time we use a leash (and by law we have to) to control the dog or pull him into the direction we decide to go, we use force and negative punishment: removing the stimulus (the direction / obstacle the dog aims for) to decrease behavior (dog pulling in this direction).

Only if the leash pull actually influences future behaviour. I think you will find that is a big 'if'. My dogs walk nicely on leash until they detect bread/chicken bones/cat poo nearby, and then they throw themselves in unison towards the ground lollies, creating heaps of leash pressure that I instantly contribute to by pulling them back. Yet, if anything, they do it MORE now than they did a year ago. What does that tell us? It tells us a) in this context pulling is not punishing; and b) in this context, the possibility of accessing the ground lollies is highly reinforcing and they have figured out the most likely behaviour to achieve this.

Guys. Quadrants are only confirmed AFTER behaviour change. Declaring anyone uses any quadrant without knowing what the effect on behaviour was is putting the cart before the horse.

it tells us that by avoiding negative punishment in the past you reinforced this behaviour - I used negative punishment with our dog (did a lot of crazy walking) and she became a good loose leash walker responding to the slightest increase in leash tension...conversely to you I don't have to use 'heaps of leash pressure' now to control the dog.

Eta: so actually 'lead pulling' always affects the future behaviour: if you don't stop it respectively let the dog get away with it you reinforce this behaviour (positive reinforcement)...

Edited April 6, 2016 by Willem

[The Spotted Devil]

People who practice positive dog training are actually using negative punishment all the time. Every time they withhold an expected reward they are using negative punishment.

No, they are not, actually. For that to be true, the behaviour that precedes "no reward" would have to be suppressed. And in fact, "no reward" itself would have to be punishing, which would be challenging, because often the dog doesn't know when that is going to happen. Many of us do not even signal when there will be no reward, and many of us are using variable reinforcement schedules, so the dogs are not especially bothered by "no reward" in any context. What is actually taking place is DIFFERENTIAL REINFORCEMENT. i.e. some behaviours are reinforced and some are not. If I reinforce lifting a paw to chin height but I don't reinforce lifting a paw to chest height, then lifting a paw to chest height is not punished. It declines simply because it's not being reinforced as much as higher lifts are. Otherwise, I would have a hard time getting any paw lift, let alone the high ones I am after.

Thank you for taking the time to actually write that out as I couldn't be bothered

The misinformation rage... So strong...

Ironically, I'm not going to back up my assertions with credible sources this time. I'll just pers. comm. my behaviour analyst friends. :D

:laugh:

[Simply Grand]

The quadrants and the terminology used in them weren't developed to be used as training methodologies, they were developed based on observation and experimentation to explain animal (including human) learning and behaviour.

They didn't come about by someone saying "how can I get a dog to do what I want them to do?", they came about when scientists observed what animals do actually do when put into situations and given the option to do what they would.

I think it's important to understand that in this debate, so saying certain quadrants should not be used or are not effective in dogs training, or dog learning really, is not accurate. The fact is animals, including humans, are influenced by all four quadrants all the time, whether someone is actually trying to teach something specific or whether we are just living in our environment.

Also, because the quadrants were developed based on observing how behaviours come about, not every single thing fits neatly into one of the boxes. Some things are just...things, not a punishment or a reinforcement or a positive or a negative. So when my dog sits and looks out the window, as she sometimes does, it doesn't fit into a quadrant, it's just neutral. It's a behaviour she already knows, it's not increasing or decreasing, it's just a thing she does sometimes, therefore by definition it isn't in any of the quadrants.

ETA same with putting a leash on a dog. It is not a case of every time we put one on we are applying negative punishment. If the dog is standing or sitting or walking as they would be whether or not the leash was on, then the leash isn't in any of the quadrants, it's just there.

Edited April 6, 2016 by Simply Grand

[Two Best Dogs!]

Wouldn't head halters and front clip no pull harnesses fall under aversive? Aside from looking friendlier, am not sure the difference between those and a properly fitted prong and the like.

[huski]

Also, because the quadrants were developed based on observing how behaviours come about, not every single thing fits neatly into one of the boxes. Some things are just...things, not a punishment or a reinforcement or a positive or a negative. So when my dog sits and looks out the window, as she sometimes does, it doesn't fit into a quadrant, it's just neutral. It's a behaviour she already knows, it's not increasing or decreasing, it's just a thing she does sometimes, therefore by definition it isn't in any of the quadrants.

ETA same with putting a leash on a dog. It is not a case of every time we put one on we are applying negative punishment. If the dog is standing or sitting or walking as they would be whether or not the leash was on, then the leash isn't in any of the quadrants, it's just there.

The other thing to consider is that something has to be motivating to the dog to create behaviour plasticity. If the dogs behaviour doesn't change when you add or take away a reinforcer then it wasn't reinforcing. What is reinforcing for one dog, may not be for the next. Dogs are only motivated by two things - avoiding correction or gaining reward.

[huski]

Wouldn't head halters and front clip no pull harnesses fall under aversive? Aside from looking friendlier, am not sure the difference between those and a properly fitted prong and the like.

You can't really label a tool as aversive just by name, it depends on how the dog perceives it. If using a head halter creates discomfort for a particular dog and this discomfort motivates them to change their behaviour then it would be aversive. They are avoiding that sensation by changing their behaviour.

[Willem]

Wouldn't head halters and front clip no pull harnesses fall under aversive? Aside from looking friendlier, am not sure the difference between those and a properly fitted prong and the like.

ask yourself what is the purpose of a collar, harness etc...is it only for fashion?...then it is not aversive...and could be made from paper; however in most cases the purpose is not fashion, but for facilitating a tool that allows to apply a force if required.

[Simply Grand]

Wouldn't head halters and front clip no pull harnesses fall under aversive? Aside from looking friendlier, am not sure the difference between those and a properly fitted prong and the like.

This is were I think the terminology becomes confusing. People across the whole world of dog training use different words and mean different things. Terms like aversive, correction, force, interuption are all used and can mean different things to different people. That's why I like the simplicity of focusing on does the behaviour increase or decrease, and what is the dog's emotional response.

I personally think (and my thoughts do change as I learn and reflect more) that all of those things - flat/martingale collars, harnesses, head collars, prong collars, e collars - are tools that provide a handler with different levels of physical control over a dog IN ORDER FOR the handler to effectively deliver the reward or punisher needed to effectively change behaviour. I believe that the tool that delivers the lowest level of intervention needed for that particular dog is what should be used.

ETA Huski said it much more simply than I did!

Edited April 6, 2016 by Simply Grand

[Two Best Dogs!]

Thanks all. Mostly I was wondering in terms of them being used to teach a dog not to pull - but have been having slight confusion as "positively only" people using it in this method and I'm not seeing it as any more or less aversive than prongs, martingales etc (the dogs reaction depending as you guys say)

But yes most is the confusion around terminology and meanings etc

Sorry to detract, it's been something I ponder about sometimes!

[JulesP]

Thanks all. Mostly I was wondering in terms of them being used to teach a dog not to pull - but have been having slight confusion as "positively only" people using it in this method and I'm not seeing it as any more or less aversive than prongs, martingales etc (the dogs reaction depending as you guys say)

But yes most is the confusion around terminology and meanings etc

Sorry to detract, it's been something I ponder about sometimes!

The argument vs halters and prongs is usually over the amount of damage that they can do if used incorrectly. And let us not get into that again!!!

I wouldn't tell someone to use a halter to teach a dog not to pull. I would (and have) suggested one as a temporary safety device - tiny female, big dog.

[Papillon Kisses]

Wouldn't head halters and front clip no pull harnesses fall under aversive? Aside from looking friendlier, am not sure the difference between those and a properly fitted prong and the like.

It depends on how they're used and how the dog responds. For my particular dog a head collar is a grade A aversive, while a front-attaching harness is calming (I suspect it's a trigger point thing).

This article might help you:

Gentle Leader Head Collar: Reinforcement or Punishment?

[The Spotted Devil]

I only ever use a front attach harness when I walk Em around bird life as sometimes she completely loses her head and nearly hurts herself on a regular collar. The predictive nature of the harness means she goes off her NUT with excitement when I get it out because it means "oh my Dog! BIRDS!!!!"