Clicker Training, A Good Idea?

[Santo66]

No, it doesn't. It depends on whether you are actually reinforcing the dog or just offering a reward and hoping "drive" will get you through a litany of poor training decisions.

It depends on the intensity of desire the dog has to earn the reward above all else. That's why high level dog sport is won by lines of dogs that possess genetically adaptable drives towards a reward based training system. That's why breeders of competition lines purposely breed abundance of genetic drive into their lines to aid in the ease of training and reliability. You don't simply throw a training method at any dog and win because of the method utilised, you utilise a training method that extracts the genetics of the dog.

Here is an example a couple of weeks ago:

4 year old working line GSD of good blood lines desexed female pet out of control surrendered to a backyard, owner can't handle the dog, dog fights off correction on a choke chain and attacks the leash. She had never seen a tug toy in her life within 15 minutes, dog is biting the tug and loving it. 15 minutes later, dog is outing the tug and re-biting on release, that's 30 minutes into the session. A futher 30 minutes later, dog is out the front on leash focus healing.

The following day, GSD with dog aggression 3 year old desexed female pet, good basic training with treats, ok until within 30 metres of another dog. Dog performs with perfect obedience in a sterile environment free of distractions and other dogs, add distractions and other dogs, there is no obedience whatsoever, all flew out the window. Dog is as flat as a biscuit in drive, won't chase a ball, won't bite a tug, dog will only food focus if not under stress.

The dog has desire to earn food reward in a non stressful situation......that's it.....the first dog went crazy over the tug toy which provides the means for something to focus on and the desire to want.......that's in the dogs genetics to begin with over the second dog who had no interest in it. With the second dog, you are behind the 8 ball to begin with in a motivational training system as the dog doesn't have the genetic motivation or drive for that training system where the first dog did.......big difference!!

The second dog was put on a prong collar and taken for a walk up the street. When she went to have a lunge at the fence up the road where a dog lives who barks at her, she copped a correction....wow did the stop the lunge in it's tracks immediately.....3 years of learned behaviour to lunge at that fence stopped in a heartbeat!!. Prong collars don't work hey??

60 minutes later the dog is learning boundaries and consequence and bugger me......the stress levels are dropping and the dog is starting to take food rewards in the street.......never has the dog taken a food reward in the street before when under stress, now she is after copping a few prong corrections. Isn't that mazing hey......one of them terrible and cruel despicable prong collars has just taught a dog consequence of the wrong behaviour and lowered stress levels when the handler took control of the dog and situation with some ground rules and the dog begins to take a food treat in her former stressful zone where before she wouldn't eat.

The point is this: If the treat trainer had of put a prong collar on that dog 3 years ago instead of messing around trying to extract drive the dog didn't have in the first place to suit their motivational training method, the massive improvement achieved in 60 minutes on a prong would have been well and truly established.

The moral is this: Had the treat trainer been motivated to train dogs instead of selling training methods and used the right training system best suitable to the particular dog, the owner of this dog wouldn't have messed around for 3 years with a dog that was uncontrollable in the face of other dogs and distractions.

The dog yesterday was walked through a dog park on leash and prong in the face of other dogs without a flinch and took at treat at the end for being a good girl maintaining her composure. That's one week on a prong versus 3 years of treat training.....I rest my case :D

But we can train heaps of animals in precise and complex behaviours using marker-reward training, not just dogs. So breeding in drive or handler focus can't be a necessary condition.

In a sterile environment yes absolutely.....where it falls over is in the face of distraction, the more drive a dog has, the more "want" a dog has to earn a reward, the less the dog cares about the environment and it's surroundings which is handler focus.....the handler means more to the dog beyond all else :)

Edited September 30, 2013 by Santo66

[Vickie]

It depends on the intensity of desire the dog has to earn the reward above all else. That's why high level dog sport is won by lines of dogs that possess genetically adaptable drives towards a reward based training system. That's why breeders of competition lines purposely breed abundance of genetic drive into their lines to aid in the ease of training and reliability. You don't simply throw a training method at any dog and win because of the method utilised, you utilise a training method that extracts the genetics of the dog.

Maybe not in the world that you live in...but in the world of obedience, agility and a number of other the sports that the majority of this forum have access to, events are often won by damn good trainers who understand how to motivate and train dogs from a number of different genetic backgrounds.

What percentage of this forum have access to, experrience or interest in high drive working shepherds for Shutzhund? Or the skills to train them? I would guess not even 1%. So it doesn't matter how many times you explain the same thing, it's simply not relevant to most of us.

[Kavik]

Also, all dogs get distracted, we have to teach them how to pay attention in the face of distractions. My current pup was distracted by a plastic bag blowing in the wind the other week :laugh: . The top trainers have a program to work through distractions.

[Santo66]

It depends on the intensity of desire the dog has to earn the reward above all else. That's why high level dog sport is won by lines of dogs that possess genetically adaptable drives towards a reward based training system. That's why breeders of competition lines purposely breed abundance of genetic drive into their lines to aid in the ease of training and reliability. You don't simply throw a training method at any dog and win because of the method utilised, you utilise a training method that extracts the genetics of the dog.

Maybe not in the world that you live in...but in the world of obedience, agility and a number of other the sports that the majority of this forum have access to, events are often won by damn good trainers who understand how to motivate and train dogs from a number of different genetic backgrounds.

What percentage of this forum have access to, experrience or interest in high drive working shepherds for Shutzhund? Or the skills to train them? I would guess not even 1%. So it doesn't matter how many times you explain the same thing, it's simply not relevant to most of us.

The sport type makes no difference whether it be agility, Schutzhund or scent detection for narcotics in a working role and scent detection is a good example......the dog isn't indicating to smoke a joint as reward, they are indicating on the scent quite often for a throw of the ball or tug toy.....the point I am making is the drive has to be in the dog in the first place for training to extract it. The theory that great trainers use shit dogs unsuitable for the job to showcase their great training system is rubbish. Sure they like us to see it that way that some super training system took a genetically unsuitable dog into a champion but we all know in reality it's not case. A training system is only as good as the dog's limitations. I once thought that my own lack of performance was all attributed training error until I got a proper dog bred for the job and the difference in result was unbelievable.

Speaking of Schutzhund, a two times world champion trainer who would never finish outside the top 5 in any competition entered a showline dog in the world title a few years ago......it was the only showline entry out of 120 working line dogs. He finished 53rd...was that the dog didn't measure up or he couldn't train it properly?

Also, all dogs get distracted, we have to teach them how to pay attention in the face of distractions. My current pup was distracted by a plastic bag blowing in the wind the other week :laugh: . The top trainers have a program to work through distractions.

Of course, but are you implying that regardless of drive level that all dogs exposed to the training program respond exactly the same?

Edited September 30, 2013 by Santo66

[Aidan3]

The following day, GSD with dog aggression 3 year old desexed female pet, good basic training with treats, ok until within 30 metres of another dog. Dog performs with perfect obedience in a sterile environment free of distractions and other dogs, add distractions and other dogs, there is no obedience whatsoever, all flew out the window. Dog is as flat as a biscuit in drive, won't chase a ball, won't bite a tug, dog will only food focus if not under stress.

That's what I would call waving around a reward and hoping that "drive" will get them through a litany of poor training decisions. It's not training using the science of positive reinforcement, it's training as a craft with "rewards".

The point is this: If the treat trainer had of put a prong collar on that dog 3 years ago instead of messing around trying to extract drive the dog didn't have in the first place to suit their motivational training method, the massive improvement achieved in 60 minutes on a prong would have been well and truly established.

Or they could have trained the dog properly with +R. So could you, for that matter. You might be surprised at what happens with some good instruction.

But we can train heaps of animals in precise and complex behaviours using marker-reward training, not just dogs. So breeding in drive or handler focus can't be a necessary condition.

In a sterile environment yes absolutely.....where it falls over is in the face of distraction

If you're waving a reward around hoping that drive will get you through, that's what happens and it's entirely predictable.

I'm not sure how sterile the environment is when wild dolphins are trained within 3 months to perform complex missions over large distances in the open ocean.

[OSoSwift]

I think there is a difference in a reaction to stop something like stealing to setting up and constructing training sessions though. In the latter you have more choices and the option to plan to ensure maximum efficiency and minimal stress and fallout.

Yes you are right. When training specific behaviours I have a plan as to what I would like to do to get it to happen or not. I also consider if the safety of my dogs or kids is on the line I do what is required, if that is negative then so be it, I deal with any fall out later.

I think Santo doesn't actually understand exactly what clicker or positive reinforcement training is or entails. IT is not a matter of waving a piece of food under their nose and hope they follow it.

By the way I do not think prongs are evil, nasty or should be banned. I would like to think with my own dogs at least I would never have my dogs get to the level of behaviour that they required one. I have also helped people with dog training issues and so far haven't needed one. That is not to say I never will, just haven't needed one so far.

[Aidan3]

I also think that that label is a bit of a misnomer due to things already discussed such as with holding a treat is negative as is a nonreward marker.

It really shouldn't be aversive, though. I'd feel like I'd failed to explain what I wanted, or asked too much, if a dog I was training actually got stressed about a reward being withheld, or didn't use a NRM as an opportunity to try something different.

[huski]

I also think that that label is a bit of a misnomer due to things already discussed such as with holding a treat is negative as is a nonreward marker.

It really shouldn't be aversive, though. I'd feel like I'd failed to explain what I wanted, or asked too much, if a dog I was training actually got stressed about a reward being withheld, or didn't use a NRM as an opportunity to try something different.

Do you really think there is no stress at all involved with removal of a reward? Sorry, but I just can't buy it and science doesn't either. We know removing a high value reward is stressful to a dog. It's stressful to people too. IMO there is stress in all learning and I don't think that's a bad thing.

If removing a reward or giving a NRM created no stress in the dog at all it wouldn't be effective.

[Santo66]

Or they could have trained the dog properly with +R. So could you, for that matter. You might be surprised at what happens with some good instruction.

For what purpose or advantage over the prong collar work which took 3 corrections to stop the dog lunging from a 3 year old habit in a timeframe of 30 minutes. R+ is faster than that, more reliable what?

[huski]

Or they could have trained the dog properly with +R. So could you, for that matter. You might be surprised at what happens with some good instruction.

Again it's not about whether it can be achieved with rewards only, it's about what can be achieved in a reasonable amount of time within the skill limit of the owner. There's no surprise or question IMO about what can be achieved with the use of rewards.

[Aidan3]

Or they could have trained the dog properly with +R. So could you, for that matter. You might be surprised at what happens with some good instruction.

For what purpose or advantage over the prong collar work which took 3 corrections to stop the dog lunging from a 3 year old habit in a timeframe of 30 minutes. R+ is faster than that, more reliable what?

You make it sound so simple, has it stopped your dog getting upset at the vet?

[Aidan3]

Or they could have trained the dog properly with +R. So could you, for that matter. You might be surprised at what happens with some good instruction.

Again it's not about whether it can be achieved with rewards only, it's about what can be achieved in a reasonable amount of time within the skill limit of the owner. There's no surprise or question IMO about what can be achieved with the use of rewards.

Didn't the OP have this exact question?

[huski]

Or they could have trained the dog properly with +R. So could you, for that matter. You might be surprised at what happens with some good instruction.

Again it's not about whether it can be achieved with rewards only, it's about what can be achieved in a reasonable amount of time within the skill limit of the owner. There's no surprise or question IMO about what can be achieved with the use of rewards.

Didn't the OP have this exact question?

And as I said in my OP what is best or right for each dog depends on many factors.

We cannot make a blanket statement of "x method is always best for every dog and owner" as there are far too many variables involved. There's no question rewards are a powerful tool as corrections are also a powerful tool.

Edited September 30, 2013 by huski

[LisaCC]

Not a pro, but I love clicker training, especially for fiddly stuff. I don't think it's just "high drive" dogs that can benefit from it for the following: I've clicker trained a range of animals from tigers to wallabies and chickens. If a chicken can learn with clicker training I think most dogs would be able to.

[Aidan3]

I also think that that label is a bit of a misnomer due to things already discussed such as with holding a treat is negative as is a nonreward marker.

It really shouldn't be aversive, though. I'd feel like I'd failed to explain what I wanted, or asked too much, if a dog I was training actually got stressed about a reward being withheld, or didn't use a NRM as an opportunity to try something different.

Do you really think there is no stress at all involved with removal of a reward? Sorry, but I just can't buy it and science doesn't either. We know removing a high value reward is stressful to a dog. It's stressful to people too. IMO there is stress in all learning and I don't think that's a bad thing.

If removing a reward or giving a NRM created no stress in the dog at all it wouldn't be effective.

Be careful to make the distinction that OsoSwift and I were talking about withholding a reward, not removing a reward. Withholding a reward is an extinction procedure, and whilst extinction can be aversive, it usually isn't and is easy to manage so that it isn't.

Punishing a response by removing a stimulus or access to a stimulus is negative punishment, and that's a different kettle of fish.

[Santo66]

I also think that that label is a bit of a misnomer due to things already discussed such as with holding a treat is negative as is a nonreward marker.

It really shouldn't be aversive, though. I'd feel like I'd failed to explain what I wanted, or asked too much, if a dog I was training actually got stressed about a reward being withheld, or didn't use a NRM as an opportunity to try something different.

Do you really think there is no stress at all involved with removal of a reward? Sorry, but I just can't buy it and science doesn't either. We know removing a high value reward is stressful to a dog. It's stressful to people too. IMO there is stress in all learning and I don't think that's a bad thing.

If removing a reward or giving a NRM created no stress in the dog at all it wouldn't be effective.

I agree totally. I mentioned previously in the discussion that my working line GSD has bitten me twice from withholding reward and the dog was stressed and getting angry and decided to help himself. Because someone hasn't experienced a dog of that intensity and the stress isn't as obvious, doesn't mean it's non existent........on more extreme dogs these cracks IME are greater highlighted to see the effects more vividly.

With my dog withholding a bite reward, the first stage is intense calm focus.....then the body trembling the anticipation......then high pitched yipping, feet bouncing.....then the bark deepens then he will fight you for it with aggression. As his handler/owner I don't want to take it the past the yipping in fact I reward the calm focus, but you can see the stress building in full drive and arousal to point of overload and he could nail me for that reward.

The after effect of that is submission, he knows he went too far, the ears are flat on his head and he sooks and licks me......to say that reward withholding is not stressful for the dog......I have never seen my dog as stressed as that scenario when taken too far.

[Aidan3]

...to say that reward withholding is not stressful for the dog......I have never seen my dog as stressed as that scenario when taken too far.

That's clearly not what I said.

Have you sought some assistance from your local dog sports club for this problem?

[Santo66]

Or they could have trained the dog properly with +R. So could you, for that matter. You might be surprised at what happens with some good instruction.

For what purpose or advantage over the prong collar work which took 3 corrections to stop the dog lunging from a 3 year old habit in a timeframe of 30 minutes. R+ is faster than that, more reliable what?

You make it sound so simple, has it stopped your dog getting upset at the vet?

That one departed in 2005 at 15 years old......he was run off old guard dog lines, sharp and defensive......taught me a lot by necessity but never again would I have a dog like that.....then again perhaps I would make a better job of a dog like that second time around?

That's clearly not what I said.

Have you sought some assistance from your local dog sports club for this problem?

That behaviour isn't a problem, it's a crucial element in a dog for protection training except withholding reward is best done against a decoy where you give the dog a sleeve bite which is the reward when training in fight over prey....you just don't ideally work your own dog to that point which another lesson learned I guess.

Edited September 30, 2013 by Santo66

[Aidan3]

Or they could have trained the dog properly with +R. So could you, for that matter. You might be surprised at what happens with some good instruction.

Again it's not about whether it can be achieved with rewards only, it's about what can be achieved in a reasonable amount of time within the skill limit of the owner. There's no surprise or question IMO about what can be achieved with the use of rewards.

Didn't the OP have this exact question?

And as I said in my OP what is best or right for each dog depends on many factors.

We cannot make a blanket statement of "x method is always best for every dog and owner" as there are far too many variables involved. There's no question rewards are a powerful tool as corrections are also a powerful tool.

But some people would have us believe that it's the tool that's broken, not the application :laugh:

[OSoSwift]

My male does not find the withholding of a treat particularly stressful, he obviously in my opinion would find it even to a small degree negative as he wants it, that means he works harder to get it. No worries at all. My bitch will find withholding or reward or NRM more stressful and will worry but she was my first fully clicker trained dog and she became that way because of me. I didn't cause that issue second time around. Let's hope third time around I get it all even better :)

I love having a little 8week old pup sitting near me and watching its little cogs turn over figuring out what I am doing/wanting etc. all the while relaxed happy and wagging its tail. I never get tired of that :)

Edited September 30, 2013 by OSoSwift