"what is wrong with clicker training"

What an emotive and completely off the point article.  I was gobsmacked and almost laughing to read it.  For a start, to admit that a particular technique won't work with some wild mammals, does not mean that the same technique won't work with domestic animals who have problems.  I am sorry but you can't equate an aggressive GSD with a shark.  It's LAUGHABLE.

Sure, we shouldn't "coddle" our dogs or reinforce fearful behaviour.  But what has that got to do with operant conditioning????  Since when did operant conditioning equate to coddling and reinforcing fearful behaviour??  If it does you're doing it wrong! 
 
Sure maybe clicker training might not work with every dog.  But to say that a positive only trainer would give up on the animal if it wasn't working and to imply that they're "in it for the money" and care more for the money than the animals they work with is disparaging and unfair.  Sure some might.  There are some truly awful "trainers" out there, of all disciplines.  What a load of rubbish.

You could just as easily argue that correction trainers say that they can cure any dog of anything, even the most severe aggression.  But, there are also correction trainers who will take your money, and possibly take your dog to their facility (it doesn't look nice to brutalize a dog in front of its owner), and turn your dog into a puddle of mush, too afraid to move, much less exhibit aggression.  Is that really a "cure"?  A dog in that state of mind is probably better off euthanized than to live in so much fear that it cannot emote, behave, or offer anything but learned helplessness.  So, we see that there are fanatics and idiots at both ends of the spectrum.  Clicker training, in and of itself, is not the be-all end-all.  It is still necessary for the trainer to have skill, timing, and common sense.  But, that doesn't mean that such trainers don't have the right to speak out against the almost automatic subjugation of dogs by cruel, outdated, and inhumane means.  If we truly are the ones with the opposable thumbs and the huge brains, we ought to be able to figure out how to train a dog without yanking, cranking, hanging and rolling.  I don't always use a clicker, I don't always shy away from correction.  But, I try to always be humane, first do no harm to the dog, and train in ways that make sense to dogs, not to macho humans.  I do not advise anyone to "kill their dog" EVER.  That is a decision for owners to make, based on how they feel about the liability to them, their own ability to train or work with the dog, and their ability to pay for vet care, training, muzzles, new foods, testing, etc.  If I can't handle a particular case, for whatever reason, I refer to another professional, or a veterinarian behaviorist, but I do not give my clients false hope either.  Not all dogs can be helped, and not all owners are all that committed, believe it or not.  They just want the problem to go away, and if they need to euth to do that, some of them opt for PTS.  I've seen some dogs get PTS that didn't deserve it, or could have been managed.  But, I've seen some people work endlessly with dogs that really are dangerous, and yet they can't bear to give up on them.  The dogs aren't happy, the people are always treading on eggshells, and once the dog is gone from natural causes, they exclaim relief, rather than the sadness they thought they'd be overwhelmed by.  We need to maintain some perspective on methodologies, but I never think it's OK to start with the most harsh method and work down.  Start with gentle and work up.

Any training methodolgy can be executed poorly or applied to the wrong thing at the wrong time. And as we all know, anyone can call themselves a dog trainer with little training, skill or even natural talent. This is not the fault of the method but of the trainer. Truthfully my main problem with more "traditional" methods is that JQP tries to employ them thinking that if they just get tough enough or watch the right TV show they'll be able to do it themselves. Some methods themselves I do take issue with, but mostly it's people screwing up and thinking they know more than they do that is my biggest issue.

But I also think it's unrealistic to say that every single dog with every single problem known to dogdome can be saved. It's a great goal, but not reality. Some dogs simply have a screw loose and given that all dogs are born armed and dangeorus and with the mental capacity of a two year old, it's irresponsible in the extreme to suggest to desperate dog owners that if they just find the right method their dog will become normal and happy again. And let me guess, this guy here will be the one to do it for you, for a small fee of course.

equating clicker training with permissiveness is ridiculous. Some of the semi-professional clicker trainers I know are incredibly strict with their dogs.

My impression of the article is that it revolved around all the great myths, stereotypes, and straw men of clicker training. The old "all positive" booby trap, the idea that anyone really believes operant conditioning is the sum total of life with a dog, the "if you can't train 'em kill 'em" straw man... he really covered all the bases. Too bad they're fictional.

Personally I had to try my hardest not to laugh out loud in a public computer lab at university while reading this. The obvious ignorance and lack of real knowledge on the subject is astounding.
 
I REALLY love the first statement:
[font=arial]I know many of you are fans of operant conditioning and clicker training… but as many of us found out when we took this into the real world and used it as dog trainers, we saw it didn't work as promised in many real world situations.[/font]
 
This right here shows his true ignorance of anything to do with operant conditioning. To say that he's not using operant conditioning, and that it doesn't work in the real world, is just....wow. I must ask then, what DOES he use to train his dogs? [] Operant and classical conditioning are how dogs, humans, apes, cats, tigers, etc learn (aside from some things like extinction, habituation, innate behaviours, but we're talking TRAINING dogs). Even using punishment is using operant conditioning! So his very opening statement just totally discredited any useful information he might have had to share, IMO. He equates operant conditioning to being clicker training, when the two are not at all the same, as any educated person would know.
 
Kim MacMillan
 
 

Good article, especially where he points out how some folks (even those with academic credentials) will refer to their beliefs as "scientic", yet those beliefs are based only on the fragments of a specific iece of research which support their own personal agenda and position, and dismiss the rest because it doesn't fit their osition. 
 
Sam seems to be a balanced trainer who does not embrace the extremes. He makes good points of reference and arguements of where the shaping of behaviors break down due an animal's natural instincts. He clearly uses and endorses positive reinforcement.
 
Clicker training is just one tool in the "tool-box", and one he has personally decided is not of great value. I don't personally think it has no value. But it seems when we are dealing with living beings who share our homes as family members, operant conditioning alone will not give us all of the answers or the tools we need. Using only a portion of operant conditioning, will give us even less.
 
Training, social, and behavioral studies work best when combined when we live with dogs as family members. IMOAE Unfortunately, many trainers believe training is all you need. I'm glad some of them also acknowledge programs such as NILIF are of value to fill in the gaps.
 
"Positive Only" is a fantasy. Although it is a very nice marketing (lying) phrase to use by those who are less than honest, and seems to appeal to clients (through emotional blackmail) who certainly would never want to do anything "negative" to their dogs by using "pain, force, and fear" or the "abusive and barbaric" techniques used by all the other trainers who are not "Positive Only".
 
It's amusing when you stumble onto an arguement on the net between two trainers who are attacking each other because one claims to be more positive than the other one because one of the trainers admits to occasionally using a physical correction, and the other trainer believes no physical corrections are needed at all. Also interesting to note that it is usually the more "positive" trainer who behaves the most "negatively", and is the one doing most of the attacking. Fascinating.
 
Again, good article and points being made by a balanced method trainer, who clearly sees the damage of extremism at the owner level and will never run out of messed up dogs to work with.
 
Worth another read, I think. []

The rhetoric about killing dogs instead of working with them has absolutely nothing to do with my actual experience of working with a dangerously aggressive dog. I started dealing with my dog's aggression using strong corrections, and that made the behavior much, much worse very quickly. I got much, much further using operant conditioning.

The rhetoric about permissiveness has nothing to do with my own experience with my dog, nor does it have anything to do with the "positive methods" trainer I used. I set high standards for my dog and expect him to live up to them. I do this because my trainer was a total hardass.

The idea that operant conditioning doesn't work is kind of a logical fallacy. The whole idea is that all animals are having their behavior shaped by external stimulii all the time.

Example: I post on this forum because it is very active and the chances of someone replying in a short period of time is excellent. This recognition of my opinion feels good, so I post.

Trainers simply use this innate tendency for animals to do what feels good to shape behaviors the way they want.

Here's what I buy:

I am certain that there are many trainers who have no grasp of how to use operant conditioning to train dogs. And I am sure that many of these trainers are using the lingo anyway. One of my co-workers has reinforced some of her dog's very nasty habits using operant conditioning--basically she's used it to make bad behaviors worse instead of better. It's a technique that can be misapplied like any other.

Oh, I forgot one thing:

"Positive only" is a fallacy--that's true. Any positive trainer worth listening to will use as much -P as +R. Here is how my trainer explained this:

Dogs have a lot of desires. They want to chew, chase, pee, eat, drink, get attention, etc. etc. etc. Your job as a trainer is to figure out what your dog wants at all times (not so hard once you've looked at your dog for awhile) and strictly control access to it. The dog has two important lessons to learn in all this:

1. Owner controls access to all the good stuff.
2. Owner often releases good stuff in exchange for certain (good) behaviors. Not always, because that would make good behavior optional. But often enough to make that good behavior always worth a shot. You don't want your dog to think of you as a "treat dispenser". You do want your dog to look to you and try to please you every single time he wants something. Not just treats... everything.

My dog sits when he wants something because his chances of getting it when he sits are excellent, and he has never gotten anywhere by jumping. My dog goes to his bed when I eat because he gets a nibble if he lays on his bed, and has never gotten anywhere by coming and sniffing around the table.

So yeah, there are treats involved, but the part where the dog learns the good behavior is in what I do all the times he tries to come over and beg at the table.

ORIGINAL: Chuffy

I am sorry but you can't equate an aggressive GSD with a shark.  It's LAUGHABLE.

I think you miss his point.  The reason he brought this up is because the argument that you can suposedly clicker train a massive whale, therefore you MUST be able to clicker train a dog, is ALWAYS brought up by positive, clicker extremists.  I also don't think this guy is totally dismissing operant conditioning, he's just pointing out that it's flawed to think it's the only way.

I also don't think this guy is totally dismissing operant conditioning, he's just pointing out that it's flawed to think it's the only way. 

 
Soooooo, what other ways are there to train a behaviour (note: classical conditioning doesn't apply since we are talking overt, non-reflexive responses)? If you aren't using operant conditioning, please, let me know, what ARE you using? []

The reason he brought this up is because the argument that you can suposedly clicker train a massive whale, therefore you MUST be able to clicker train a dog, is ALWAYS brought up by positive, clicker extremists.

I think the idea behind the "clicker training a whale" line isn't so much "if you can clicker train a whale, you can clicker train anything!" but rather to point out that clicker training and other sorts of positive reinforcement techniques can be effective even when punishment-based techniques are not, or are simply impossible. You can't strong-arm a whale into jumping over a stick, but you can make it want to do so by equating the behavior you want with a reward the animal wants.

Okay, so maybe you can't clicker-train a Great White Shark. But could you train it with another method? Probably not.

I was thinking this morning, there are some things "wrong" (for lack of a better word) with clicker training. Definately terminology is an issue. To begin clicker training really effectively you have to learn all the technical terms well enough to be able to follow your trainer's or a book's instructions. All that + and - and R and P stuff confuses people, though I don't really see a way around that. But that's definately a problem with clicker training.

You also have to do some serious learning of how to fade the clicker and treats, how to properly apply the cue, and how to get stimulus control. Again, these are not easy concepts to grasp at first and another problem with clicker training is you have to learn this stuff or it simply doesn't work. I know I've told a number of people, "Oh, it's easy!" and I must confess that I only say that to be a cheerleader, not because it is easy. I mean, not at first. Once you learn it all and get the hang of it, it really is quite easy, but the learning curve is steep.

It also goes against a lot of our instincts as humans and what we consider "common sense". I know from my experience as an educator that when the kids are acting up it is the rare teacher who can actually have enough self-control to not yell at them. All the research says you shouldn't yell, that remaining calm and addressing students in a normal (or even lower than normal) tone of voice is much more effective. Yet, when the kids are being little monsters, instinct kind of kicks in and that yell just wells up inside you, almost totally out of your control. And if you didn't read the research, you'd tell me that I'm crazy, of course you have to yell when your kids get rowdy, how else can you get their attention? We almost always automatically assume that you just need to punish more and have more rules and be more forceful and if you can just strike enough fear into the hearts of your students, they'll behave. It just seems like common sense to do that, most teachers don't even try to ease up and calm down and take the pressure off to see what would happen. It's terrifying to do that, and visions of students running amok and throwing paper airplanes at you dance through your head.

Like trying to tell teachers that they need fewer rules, not more, clicker training butts up against what we consider to be "common sense". We kind of have to retrain ourselves first before we work with a dog. And that is very difficult for a lot of folks and I don't think that difficulty should be discounted when we say that clicker training is "easy".

I was thinking this morning, there are some things "wrong" (for lack of a better word) with clicker training. Definately terminology is an issue. To begin clicker training really effectively you have to learn all the technical terms well enough to be able to follow your trainer's or a book's instructions. All that + and - and R and P stuff confuses people, though I don't really see a way around that. But that's definately a problem with clicker training.

You also have to do some serious learning of how to fade the clicker and treats, how to properly apply the cue, and how to get stimulus control. Again, these are not easy concepts to grasp at first and another problem with clicker training is you have to learn this stuff or it simply doesn't work. I know I've told a number of people, "Oh, it's easy!" and I must confess that I only say that to be a cheerleader, not because it is easy. I mean, not at first. Once you learn it all and get the hang of it, it really is quite easy, but the learning curve is steep.

Wow, Cressida! You READ MY MIND! I have been playing around with Clicker training. I got "Clicking With Your Dog" and read tons online. It's harder than it looks! I do have a tendency to way over-research and over-anylize so I thought it was just me. Thank you for making me realize I'm NOT crazy (at least on the clicker thing - the rest is debatable)

All the research says you shouldn't yell, that remaining calm and addressing students in a normal (or even lower than normal) tone of voice is much more effective. Yet, when the kids are being little monsters, instinct kind of kicks in and that yell just wells up inside you, almost totally out of your control. And if you didn't read the research, you'd tell me that I'm crazy, of course you have to yell when your kids get rowdy, how else can you get their attention?

I am always AMAZED by my son's first grade teacher. She was also my husband's first grade teacher. She's been teaching almost 40 years and speaks barely above a whisper. And the kids LISTEN to her. And the way she deals with the squirmies while they are getting ready to go is fabulous. After the kids get their coats and backpacks they are supposed to sit quietly and wait for the bell. Usually though, kids want to chat and goof off. So Mrs. Miller watches and silently gives a piece of candy or a small toy to the first kids she sees who are sitting quietly. She says the noise turns off like a switch when the other kids notice that someone got a treat and they all race to their seats to see if they can get one too!