Clicker = training???

ORIGINAL: Kim_MacMillan

So if your child was talking loudly in a library, would you repeatedly smack her...

Dogs do things for a reason - to get what they want. I have found that the easiest way to teach dogs that they can get what they want (when it's allowable), is to teach them HOW to do it appropriately.

 
For the first statement, I wish the sensationalism and exaggerations would stop.  Anyone doing such a public act would be arrested.
 
For the 2nd part of the quote.  Please demonostrate the "how to do it appropiately" without introducing a negative.  JQP here, so interpret my words not in the scientific technical sense.

For the 2nd part of the quote. Please demonostrate the "how to do it appropiately" without introducing a negative. JQP here, so interpret my words not in the scientific technical sense.

Err...

So if the dog wants attention from people, you teach it to sit, or four-on-the-floor, or to jump up if that's what you wish, or whatever you want.

Is that not what you meant, DPU?

ORIGINAL: Cita

Is that not what you meant, DPU?

No, my post stands.  This is my home situation dilemma and I think Karen MacMillan is the only person on this board that can address this properly.
 
So sorry Kim, I meant you but I had Karen Pryor on my mind.

ORIGINAL: Xerxes

So the way I see it is that clicker training is used to train dogs to perform behaviors on cue.

Other highly touted disciplines are used to train dogs to stop performing behaviors-with or without cues. 

Maybe I'm right, maybe I'm wrong.  Isn't training your dog supposed to be fun?  Do I, the human, always have to be right?  What's the price that I, the owner of a well behaved dog, pay when my dog finds a way to perform a requested behavior in a manner different than I envisioned?  (For example I give the "finish" command and my dog doesn't circle me, but rather leaps up in the air, does a pirouette and lands, immediately going into a sit.)  

Well, I certainly find that dog training is fun and rewarding for me, and apparently for my dogs. But, I don't think that "fun" is the only purpose to training, and for the vast majority of dog owners, it is probably not the primary purpose. Training is also about making it reasonably safe for dogs and humans to live together, and minimizing the intrusion of our dogs into the quality of life of others.

In another thread one of the posters said that the worst thing that can happen if a clicker trained dog is confused is that it will "sitstaydownrolldancetouchspinhighfiveleaveit" in rapid succession. I respectfully disagree. If a dog is confused about not jumping on people, however he was trained, the worst thing that can happen is that he knocks 75 year old Nanna off her feet, and she breaks a hip. And since something like a third of the people over seventy who break a hip never return to living in their own home, I'd say that those consequences are a lot more dire than "sitstaydownrolldancetouchspinhighfiveleaveit." Ok, so maybe that's not the most likely consequence of a dog failing to learn not to jump on people, but I think that owners becoming less inclined to interact with their dog, isolating the dog, and even eventually dumping the dog before he does injure someone are not exactly rare consequences of a dog failing to learn not to do certain things.

To stick with the jumping on people example, some trainers are successful in teaching a dog not to jump on people by teaching him to sit instead. But I have had many many people in my training classes who have failed using that strategy. What quite a few are actually successful at is teaching the dog to sit immediately after he has jumped on someone. So, they've been terrifically successful at teaching the dog what to do, but they've failed to teach the dog what not to do. To use the child talking in the library analogy, yeah, I might first offer the chatterbox a coloring book to work in instead, but if she continued to blather on, you bet I would tell her to hush. My first warning would be not unpleasant, but if she continued to babble, my tone of voice and overall demeanor would certainly project to her that continuing her behavior would result in unpleasantness (and no, I would not "smack her", and yes, I too am tired of the frequent exaggeration that accompanies any discussion of the use of aversives).

I have taught my own dogs to sit, down, take a bow, paw at their nose, touch my hand, fetch, shake, look in the direction I'm pointing, find it, .... using clickers. I once taught a chicken to do an agiltiy "chicken walk" using a clicker. I think it is a powerful tool to teach animals to do specific behaviors or sequences of behaviors. But I've found that clicker/R+ is not always effective in training dogs not to do certain things. So, I've also used some mild aversives to teach them not to engage in certain behaviors. Amazingly enough they still greet new people enthusiastically, they seem pretty darned comfortable and relaxed and trusting in my presence, they are eager to interact with me and the world in general, and they regularly offer new behaviors. And they are rewarded for offering fun new behaviors, and they accept my letting them know if a few specific behaviors are not to be repeated. None of them have turned into a psychologically and emotionally battered puddle of goo.

Maybe I've just been lucky, but it does get tiresome to have the occasional use of a collar correction, verbal correction, or poke in the neck equated with cruelty, abuse, and physical violence, and to have training to curb certain behaviors characterized as repressing their sentience and ability to make decisions on their own.

I'm currently a voluteer for a 5 day seminar on behavior for vets and vet techs and we just talked about why punishment/corrections may be detrimental due to not giving the dog an alternate behavior to the one being punished/corrected.

The bottom line is that the lack of information on what to do means that the dog may, and often does, experience conflict which can turn into anything from learned helplessness (it's safer to just do nothing) to aggression (I don't know what you want, so I'm going to growl to make you stop because that works) - it creates a lot of stress on many dogs.

Traditional methods teach the dog to not move because that's the only thing they know won't get them punished/corrected whereas methods like clicker training teach the dog specific behaviors for certain situations (it's better to offer something than do nothing/keep performing the undesired behavior).

Some clicker trainers do use "no reward markers" which, if implemented properly, tell the dog "try something else"; unfortunately they can also become conditioned punishers for some dogs and thus create some of the same problems as you see with "normal" punishment/corrections.

Think of it this way: if you tell the dog not to do something but not what *to* do instead, he has about a million other things he can choose from to do, and without knowing which behaviors are "good" and which are "bad" the dog might choose another wrong behavior and get another punishment/correction, still not know what to do, try again and may or may not do what is "right" for the situation. Dogs don't speak "human" unfortunately and thus I find it much easier to show the dog the right thing to do in a given situation (i.e. set up for success) than to let him try and try and try and possibly fail numerous times and thus get stressed.

Clicker trainers do tend to use "negative punishment" (i.e. taking something good away) in many situations, but it has fewer repercussions than positive punishment and/or negative reinforcement as it is less liekly to induce conflict, especially if implemented consistently.

eta: In response to "buster the show dog" - the jumping situation you describe is not one in which a clciker trainer will be training the animal not to jump. Teaching no jumping in my world = management to prevent the problem when we can't actively reinforce good behavior and prevent "bad" behavior (i.e. the dog is leashed when gma is over) and working with a number of able bodied people to teach the dog that jumping=no attention, 4 on the floor=attention such that the behavior becomes generalized. Once i see that the behavior has generalized, only then is the dog allowed in more uncomtrolled situations, though still managed as needed to allow for success (i.e dog is leashed on a short leash when he approaches gma for the first time and until I know he's reliable around her).

Of course the above generally only takes a few minutes to teach and generalize, though older dogs may take a few extra sessions, especially if they have a strong reinforcement history related to the jumping behavior. Consistency is key, just like with punishment/corrections.

ORIGINAL: stardog85

Some clicker trainers do use "no reward markers" which, if implemented properly, tell the dog "try something else"; unfortunately they can also become conditioned punishers for some dogs and thus create some of the same problems as you see with "normal" punishment/corrections.

Think of it this way: if you tell the dog not to do something but not what *to* do instead, he has about a million other things he can choose from to do, and without knowing which behaviors are "good" and which are "bad" the dog might choose another wrong behavior and get another punishment/correction, still not know what to do, try again and may or may not do what is "right" for the situation. Dogs don't speak "human" unfortunately and thus I find it much easier to show the dog the right thing to do in a given situation (i.e. set up for success) than to let him try and try and try and possibly fail numerous times and thus get stressed.

The above quote nicely describes my dilemma.  My foster Marvin is a true SA hounddog and I have successfully modified his destructive panic behavior over 5 months, but I have always wonder if I have had an affect on his psychy.  For this particular dog I am committed to Truly Dog Friendly Training and wish not to introduce any aversion or negatives, be they verbal, visual, or psychological.

SA is a truly difficult set of behaviors to address ime. Talking to the vet behaviorist here at Purdue about SA, he said in most cases he's found the following things most important when addressing SA (to the point at which he says that if the following doesn't work, medication is likely indicated): Walks off the property daily, devaluing departures and arrivals (i.e. ignoring the dog before leaving and after returning home), providing a treat dispensing toy upon departure (or slightly before), and allowing the dog more freedom in the house. he says he rarely uses the planned departure method for the dog's benefit, rather he uses it with owners that are afraid to leave the house with the dog not crated.

(I have to say that not using the crate when I leave the house is the one thing that completely changed Maggie's SA around - she went from a dog that shredded everything in a crate, barked nonstop, drooled profusely, and bent wires/opened crate doors to get out to a dog that appears completely normal if left loose in my house.)

Stardog85, thanks.  The set of behaviors of the SA hound has been addressed and modfied so the dog is calm.  What I think about is the psychy of the dog.  Just because the dog's behavior has changed the nerve firing within the brain that cause the panic can still exist but not be surfaced in the dog's behavior.  I know in going forward with formal training I can not use aversion because of the risk of the SA behavior surfacing.  I chose Clicker training and attending classes to accomplish this but in that process the dog would exhibit peculiar behavior at withholding the treat.  I am coming to the conclusion that the Truly Dog friendly approach and positive only training is non existence. 

Glad to hear the SA is resolving itself. []

One thing to keep in mind when looking for the "truly dog friendly" technique: the real world isn't 100% positive itself and thus I believe that animals, unless there is something wrong physically/physiologically, can handle some guidance using "negatives" - granted this level depends greatly on the individual (and is the reason that I stay away from positive punishment and negative reinforcement given the greater risks of side effects from those elements).

I'm guessing that what you may be seeing in your dog is some level of frustration, which can be both beneficial or detrimental depending on the animal. Have you tried clicking for very small movements? Some dogs require high levels of reinforcement to create a "buffer" for those times where frustration occurs and clicking smaller pieces of behavior will allow for an increased reinforcement rate ime.

ORIGINAL: Xerxes

So the way I see it is that clicker training is used to train dogs to perform behaviors on cue.

Other highly touted disciplines are used to train dogs to stop performing behaviors-with or without cues. 

Stardog85, in order to truly understand the effectiveness of a training method I believe it has to proofed on the exception dog, such as Clicker training on a true SA hound. 

Back to the Original Posting (OP), generally do you think the use of one specific method can create a well balanced dog that can live in harmony within a home?  By your statement that the real world isn't 100% positive I take it that negative are introduced and that means other methods come into play.  Is it safe to say that when the "experts" speak of Clicker Training, they talk in theory and not able to relate to the average dog owner and their relationship with the dog?

ETA:  the SA did not fix itself, do you know how much time and money it took to change that behavior?

buster the show dog, excellent points! []

DPU - clicker trainers often do use "negative punishment" - taking something good away to decrease behavior is negative, but it is less likely to induce side effects like the use of positive punishment (adding something to decrease behavior) or negative reinforcement (taking something away to increase behavior).

I suppose you can say we use negatives, but it's part of the method, not separate and definitely not similar to the use of "negatives" in more traditional methods. And I acknowledge that the world is not positive because afterall we can't control everything, we can just do our best to set our dogs up for success.

Clicker training is totally applicable to the real world - after all, if it wasn't my dog wouldn't have just been able to serve as a neutral dog in class when the students were working with dog reactive animals. Maggie would've been a snarling ball of fur if clicker training didn't work in applaication (she was dog reactive before we used clciker methods to change behavior), but yesterday she stood quietly while being sniffed, barked and growled at, and generally being the target of some interesting reactions by other dogs. Not to mention the behavior that has earned her a CGC and allowed her to enter her 5th year as a therapy dog via Delta Society Pet Partners.

I truly believe that clicker work has improved my relationship with my dog - I'm not the person creating stres anymore - and many of the people I've worked with have exclaimed over how wonderful the methods have been for improving their interactions with their dog. It's a pretty cool thing when people start focussing on positive behaviors rather than negative behaviors with their dogs.

I also realize the SA didn't solve itself - I'd love to know how you accomplished it - but did you use positive punishment/negative reinforcement to solve the issue?

ORIGINAL: stardog85

DPU - clicker trainers often do use "negative punishment" - taking something good away to decrease behavior is negative, but it is less likely to induce side effects like the use of positive punishment (adding something to decrease behavior) or negative reinforcement (taking something away to increase behavior).

 
BTW, glad the jogging dog made it out of the shelter system.  I did look to try and find the dog a good home and if I had the room and if I knew the dog was jeopardy, the dog would be welcomed here.  But I knew the dog was safe because you were watching out.
 
Lets add the element of time to 'negative punishment' and 'positive punishment'.  It seems to me that prolonged 'negative punishment' is more detrimental to a dog's psychy than the short time application of 'positive punishment' because of the belief a dog lives in the moment and is forgiving.  Please also consider the type of punishment to be mild or even gentle for that matter as Buster the show dog describes.

From Stardog:
I truly believe that clicker work has improved my relationship with my dog - I'm not the person creating stres anymore - and many of the people I've worked with have exclaimed over how wonderful the methods have been for improving their interactions with their dog. It's a pretty cool thing when people start focussing on positive behaviors rather than negative behaviors with their dogs.

 
I've been reading quite a bit on this practice/theory/idea lately and it's fascinating and makes great sense, once it gets through my not-so-speedy brain.
 
Very thoughtful stuff indeed. Here's to you for sharing!

Nefowler, welcome to this thread.
 
You know Marvin, you know hounds.  I am serious with my concerns in training the SA hound.  I also made a committment, a promise not to use any negatives.  Is this all nonsense to you?  Do I need to compromise my promise?