Study: positive dog training more effective than use of shock collars

DBB

I thought ye might like to read this recently published article, which links to the primary research showing the greater efficiency and efficacy of positive, reward based dog training, compared to the more traditional methods...
https://www.companionanimalpsychology.com/2020/07/positive-reinforcement-is-more.html?m=1

Tilikum17

Can’t understand how anyone would use a shock collar. Neighbours got a trainer in to train his two German Shepard’s. Two beautiful pups. As soon as I saw this particular “trainer” I knew it wasn't good. He doesn’t use positive reinforcement.

Now the poor dogs have shock collars on them.

dashoonage

thats shocking.....

Richard tea

I couldn't find it in the article but was the recall done under distraction?

63 dogs tested over 5 days? Not alot of time I would have thought.

Did I read it right that all dogs were fitted with an ecollar? Even the positive only trained dogs?

deadlybuzzman

I've only ever used reward based training, however I have read and listened to trainers of other methods so get as well informed view as possible.
That said this entire thread has an agenda, even the article calls them electronic collars not a shock collar which isn't accurate as these collars can also have
1. A vibrate setting
2. An audible beep that can be used at distance for a dog to come
3. The so called shock part can have settings goin from 1 (so low it can't be felt) to 99 and the whole point is to use the lowest effective level otherwise the dog shuts dog with fear

I'm by no means a fan of them but while reward based training is great for INCREASING the frequency of desired behaviors, I'd be curious to know what other methods there are that REDUCE the frequency of dangerous behaviours eg. Charging towards a busy road to get a neighbours cat etc. If the dog is driven enough it will blow off any reward to get the cat

DBB

deadlybuzzman wrote: »

I've only ever used reward based training, however I have read and listened to trainers of other methods so get as well informed view as possible.
That said this entire thread has an agenda, even the article calls them electronic collars not a shock collar which isn't accurate as these collars can also have
1. A vibrate setting
2. An audible beep that can be used at distance for a dog to come
3. The so called shock part can have settings goin from 1 (so low it can't be felt) to 99 and the whole point is to use the lowest effective level otherwise the dog shuts dog with fear

I'm by no means a fan of them but while reward based training is great for INCREASING the frequency of desired behaviors, I'd be curious to know what other methods there are that REDUCE the frequency of dangerous behaviours eg. Charging towards a busy road to get a neighbours cat etc. If the dog is driven enough it will blow off any reward to get the cat

It'll blow off punishments too. That's why dogs escape from gardens with radio fences.
Frequency of dangerous or unwanted behaviours is reduced via training! A dog that's charging off after a cat hasn't been trained to a level that it won't chase things, whether the trainer has chosen reward or aversive training.
As for shock collars, or electronic collars, or whatever you want to call them, the thing is that aversive training cannot work unless it's aversive. It doesn't have to shock the dog for it to be aversive... the vibrate mode can be quite aversive for a number of dogs, even when used for well-intentioned elements of training, such as, as a way to gain the attention of a deaf dog. Many trainers have fallen foul when deaf dogs shut down even in vibrate mode. I've also trained my fair share of dogs that find a beeping noise quite unpleasant... it makes them uneasy, restless, and takes their focus off the job at hand.
Additionally, it's been well accepted by now that in dogs that do not find the vibrate setting aversive, they're inclined to not respond to it at all. It's no good for teaching them anything, because it'snot aversive... so the only option now is to start shocking the dog.
It has also been shown in research going back many years, that if you want to use aversives to stop/reduce unwanted behaviours, you've got to go big or go home, or it won't work... by that I mean, the punishment has to be big for it to be effective.
I posted a research paper which indicates that aversive training isn't as effective as positive training, for people to make of as they wish. Many would say that, like a lot of research, it lends weight to stuff we already knew but didn't necessarily have the hard research to back it up. If posting it means I have an agenda, so be it!


@Richard Tea, as per the write-up, all dogs wore ecollars. The collars for 2 of the 3 groups were deactivated.

Richard tea

No issue with anyone having an agenda. I would rather read real world studies on such topics. Recall in your back garden is fantastic. But recall when Bruno the high drive rescue spots sheep in a close by field is a different story.

DBB

Richard tea wrote: »

No issue with anyone having an agenda. I would rather read real world studies on such topics. Recall in your back garden is fantastic. But recall when Bruno the high drive rescue spots sheep in a close by field is a different story.

I'm not sure what that has to do with the findings of the study.
If a dog hasn't been trained to not chase sheep, and the owner walks him through sheep without a lead on, then neither treats nor shock collar are going to stop him.
The point of the study was to show that when you DO take on to train your dog, you'll get the job done more quickly and effectively if you use positive reinforcement, than if you use a shock collar or if you use the style of training that often involves some level of coercion.

deadlybuzzman

DBB wrote: »

I'm not sure what that has to do with the findings of the study.
If a dog hasn't been trained to not chase sheep, and the owner walks him through sheep without a lead on, then neither treats nor shock collar are going to stop him.
The point of the study was to show that when you DO take on to train your dog, you'll get the job done more quickly and effectively if you use positive reinforcement, than if you use a shock collar or if you use the style of training that often involves some level of coercion.

I agree with almost all of what you're saying and I only do reward based training because I have a dog (like most of us) that like most responds well to it.
However even the likes of Ian Dunbar with behavioral qualifications coming out his ears will say reward based training increases the frequency of desired behaviors not that theyre necessarily 100% reliable and thats fine if you have to repeat a sit or come command a couple of times.

If its a situation like what I seen in a park before where a greyhound charged over to a tree and ragdolled a squirrel until it lay there alive but paralysed or if you have any hard, stuborn, driven dog that decides whatever it wants to chase/kill beats whatever its being called back for then you may well need an emergency brake to snap them back out of prey drive.

What I'm saying is they are an extreme choice for very limited situations but I've seen examples of scary dangerous dogs being brought back into line and being handleable partly with the likes of ecollars so while I'd understand calls to ban them it would also mean dogs being killed or people being hurt needlessly

Knine

deadlybuzzman wrote: »

I agree with almost all of what you're saying and I only do reward based training because I have a dog (like most of us) that like most responds well to it.
However even the likes of Ian Dunbar with behavioral qualifications coming out his ears will say reward based training increases the frequency of desired behaviors not that theyre necessarily 100% reliable and thats fine if you have to repeat a sit or come command a couple of times.

If its a situation like what I seen in a park before where a greyhound charged over to a tree and ragdolled a squirrel until it lay there alive but paralysed or if you have any hard, stuborn, driven dog that decides whatever it wants to chase/kill beats whatever its being called back for then you may well need an emergency brake to snap them back out of prey drive.

What I'm saying is they are an extreme choice for very limited situations but I've seen examples of scary dangerous dogs being brought back into line and being handleable partly with the likes of ecollars so while I'd understand calls to ban them it would also mean dogs being killed or people being hurt needlessly

A greyhound or terrier grabbing a squirrel or rat is not being scary or dangerous but is doing what comes natural to them. I'd not be using a shock collar for that. I put mine on the lead where they are likely to get themselves into such a situation, you know? Like a responsible owner! You don't get a terrier or sighthound & expect it to act like a Poodle!

DBB

Your post still misses the point behind the paper, and that is, that they found that positive reinforcement delivers quicker and better results than using ecollars, or coercive training does. So, even taking Dunbar's points on board (and he's only quoting the laws that underpin learning theory), this research is not claiming that it's 100% effective... it's claiming that it's more effective than ecollar or coercive training. The fact that they used sit and recall as the exercises taught is by the by... the point they're postulating is that no matter what you're teaching, positive reinforcement works better.

The argument about drivey dogs running off to do harm is, in this instance, somewhat pulling away from the point of this particular piece of research, but I will say that there seems to be a widespread mental block about the necessity to use ecollars to control problem behaviours.
It is effective to train a drivey dog not to run off/chase etc using positive reinforcement. I've seen it done multitudes of times, and I've done it myself, including with working dogs. I've also seen it done with ecollars. Both work, but (a) there's now research to indicate that positive reinforcement will work better on more dogs to achieve the desired end result, and (b) there's also significant research to indicate that the welfare issues, and unwanted behavioural side-effects associated with use of ecollars and coercive training that aren't an issue with positive training.
Im surmising now, but perhaps the issue is that the ecollar users and coercive trainers seem to be far more vocal about how they saved x dog's life by stopping it chasing... indeed, i often see them publicly rubbing stories like this into the faces of positive trainers, who are less inclined to be as... outspoken... in their own defence (perhaps because, unlike coercive trainers, they're bound by professional codes of conduct?!) I think people just aren't used to seeing drivey and/or problem dogs being trained positively, because we've been fed the coercive stuff via TV and media for so long, and let's face it, the gradual and unremarkable (to the casual onlooker) nature of positive training probably doesn't make for good TV.

deadlybuzzman wrote: »

I agree with almost all of what you're saying and I only do reward based training because I have a dog (like most of us) that like most responds well to it.
However even the likes of Ian Dunbar with behavioral qualifications coming out his ears will say reward based training increases the frequency of desired behaviors not that theyre necessarily 100% reliable and thats fine if you have to repeat a sit or come command a couple of times.

If its a situation like what I seen in a park before where a greyhound charged over to a tree and ragdolled a squirrel until it lay there alive but paralysed or if you have any hard, stuborn, driven dog that decides whatever it wants to chase/kill beats whatever its being called back for then you may well need an emergency brake to snap them back out of prey drive.

What I'm saying is they are an extreme choice for very limited situations but I've seen examples of scary dangerous dogs being brought back into line and being handleable partly with the likes of ecollars so while I'd understand calls to ban them it would also mean dogs being killed or people being hurt needlessly

Richard tea

Getting back to the research or paper. As I suspected after a little reading.


The research contains ''serious errors and bias''


''academics at Lincoln had campaigned against e-collars prior to conducting the research'' To me this is the equivalent to the tobacco industry or alcohol industry sponsoring research into the health benefits of both vices.


''the dogs Lincoln selected for training with e-collars had more pre-existing behavior problems'' These are massive points. Whats that all about?


I assumed when academics carried out research the norm is other academics are ready to pounce and pull apart the research. Why did Lincoln publish such a biased study?


https://joinardo.com/lincoln-research/ Short read but contains some good points.

DBB

Ah jayney RT
Every research paper you will ever read has its detractors. The convincing detractors will go and set up their own research project to try to show they've evidence to support their contrary view, or at the very least will quote research to support their view... but the more normal default position, as is the case here, is for detractors to shout criticism from the sidelines... in this case, we have detractors on a website that supports and encourages the use of ecollars, who themselves come up with no peer-reviewed research of their own to support the use of ecollars, or what effect they have on training reliability. For this reason, I'm not at all surprised that they have issues with the findings of the Lincoln paper! It's perfectly normal, but such criticisms, in the absence of their own peer-reviewed evidence to back it up, are guilty of the very things they're accusing Lincoln's (peer-reviewed) evidence of!
That they detracted from the paper by saying it is the work if an MSc student is reprehensible... that MSc student, like all students, researched and wrote the paper under the supervision of some really well respected scientists in the behaviour world. That's how a huge amount of research gets done. To get a paper published in a scientific journal is really not easy... there are a lot of hoops to jump through before it will be considered by the journal's external and independent editors for publication.
I'm also unsure, because they don't say, as to how they drew a number of the other conclusions that they drew. Lincoln runs a dog behaviour clinic... if they used clients' dogs as subjects, I'd be beyond surprised if ALL dogs in the study didn't have some form of issue that resulted them attending the clinic in the first place!
[Edited to add] Indeed, the paper does actually say that ALL dogs in the study had known issues with recall and obedience around livestock, not just the ecollar dogs.
I will also tell you that Lincoln has produced a huge amount of research into the effects various tools have on training and behaviour over many years. So widely accepted is their peer-reviewed work, criticisms and all, that it has been included amongst the body of research used by UK legislators to frame their welfare legislation... legislation which is lauded and copied the world over.
Lincoln carried out a chunk of the research (that does exist, despite the quoted detractors' claims) that indicates the welfare issues, and unwanted behavioural side-effects, of using ecollars. It was THIS research, and presumably their own experiential information from their behaviour clinics, that informed their lending their voice to the campaign to ban ecollars. They were far from alone in this campaign by the way... they were joined by the RSPCA, Dogs Trust, the Association of Pet Behaviour Counsellors, the UK Association of Pet Dog Trainers, Nottingham University, the Animal Behaviour and Training Council, the Kennel Club... and more... these are the ones I remember off the top of my head.
So, I suppose my message is to beware of detractors who don't or can back up their criticisms. The strength of research is that it CAN be criticised and rebuked, but such criticisms can only be taken seriously when they put their money where their mouth is, and put their own research forward for the same level of peer-review that the research bodies do, rather than lobbing criticisms in from the sidelines!