|
Post by ripley on Oct 1, 2004 14:43:14 GMT -5
Yes you have. Maybe I'm weary of the purely positive methods because all my past dogs have been dominant, strong-willed and weakly motivated. (I seem to attract those types) I'd love to be able to use those methods exclusively, but they just haven't proven effective on ANY of my dogs. (Buddy and Shiner were strong willed and Micki was a spaz who wouldn't take treats or toys. And Ripley hates everything) I guess it comes down to this: whatever works for the dog and owner in question. If the OWNER feels uncomfortable giving the dog a correction for blowing off a command, the dog will probably sense that discomfort and take advantage of it, and the correction will be useless.
|
|
|
Post by CampWoof on May 11, 2005 6:54:09 GMT -5
Interesting that no-one in this thread has spoken yet to what barking behavior is. Firts of all, barking is a natural behavior. Second, what is being called problem barking is being selected/elicited by the environment in some way. Of course you can't train the stopping of a behavior. The positive way to deal with problem barking is to do two things simultaneously: 1. extinct the problem barking by finding out how it is being reinforced and removing that; 2. train in a mutually exclusive behavior. Sometimes both can be done sometimes only one. For example, a dog that barks at the mailman is being reinforced when he walks away. That is, the dog "thinks" his barking made the mailman leave. You can't remove the reinforcer of the mailman walking on by. But you can train the dog to say, grab a certain toy and run with it. He can't bark while doing that. Then you bring that under the stimulus control of the mailman's presence. In other words the dog learns good things happen when he grabs that toy and runs but ONLY when the mailamn is going by. In effect, the mailam becomes a cue for the dog to engage in the alternate behavior. Easier said that done? Yes, most worthwhile things are. But to say a purely positive method doesn't exist is just silly. How do dolphin and whale trainers accomplish what they do? They don't jump in the tank and administer a correction!
|
|
|
Post by Am on May 11, 2005 15:39:16 GMT -5
Hi Campwoof, You recommend that owners of barking dogs... I think the problem is that for some dogs barking is self-reinforcing. Since the barking is its own reward, you can't rely on it extinguishing in the absense of other stimuli. That's why a lot of people here recommend negatively reinforcing the barking - so that the dog can be taught that barking isn't always fun. A Karen Pryor fan, I'm guessing? I think you've got to remember that captive cetateans tend to live in fairly boring enclosures. That's why they're so amenable to PP training, because even the attention of a trainer (let alone a fish) becomes rewarding in the absense of other stimuli. I bet if you made their enclosures 100 times larger and filled them with schools of fish you wouldn't find the dolphins nearly so obliging. You can see the same with many PP trained pet dogs - put them alone in a room with you and most of them are quite happy to perform for a dog biscuit. Take them to an area full of distractions (like a busy dog park, or a field full of rabbits) and they'll rapidly find something else more interesting to do than work for a dog biscuit! Just my 2 cents - and I admit that I am a very amateur dog trainer, lol.
|
|
|
Post by CampWoof on May 11, 2005 16:30:04 GMT -5
I agree that barking "can" be self-reinforcing, but that when it is it is almost always in a dog that is bored. Otherwise I think barking is usually being reinforced by something else that can be identified and possibly removed. I've heard people say that when their dog barks the only way they can stop him is by taking him for a walk Who's training whom?
|
|
|
Post by Laura on May 11, 2005 21:23:15 GMT -5
Am, after reading your quite astute post, I'd say you're pretty far from amateur status ;D.
|
|
|
Post by Am on May 11, 2005 23:14:03 GMT -5
Campwoof - It's true that some dogs bark simply because they're bored, and I agree that providing these dogs with additional stimuli is probably the easiest and kindest way of dealing with the problem. I also agree that if a dog has learned to bark at a stimulus that can be easily removed then that should be done (for example, a dog that barks at postmen could be kept in the house instead of the yard). However, IMO some dogs that bark incessantly do so simply because they want to. Ignoring these dogs won't work, because they actually enjoy barking. Entertaining them might work, but it's not practical for an owner to entertain their dog 24 hours a day - nor should they have to. I think it's completely reasonable to expect an adult dog to wait quietly when left alone for short periods of time. So, how do you discourage a dog like this from barking? As you say, it's certainly possible to train a dog to perform a mutually exclusive behavour instead of barking. But this isn't going to be very useful if the dog decides to barks when you are out of the house, so you are not there to give or reward the mutually exclusive command. To me it just seems more sensible to use programme of immediate and consistant negative reinforcement to deter a dog like this from continuous barking. Laura - thanks!
|
|
|
Post by CampWoof on May 12, 2005 2:03:35 GMT -5
Dogs that bark incessantly don't do it because they want to or because it pleases them. Dogs that bark incessantly are in distress and need something other than punishment which is what you mean when you say negative reinforcement. Punishment and negative reinforcement are two very very different things. Trainers who use punishment do so because it is easy for them and requires little effort or knowledge on their part. The use of a learning theory approach is hard work. In fact even learning the terminology is hard which this thread has proven.
|
|
|
Post by Am on May 12, 2005 7:39:00 GMT -5
Well, as regards the punishment/negative reinforcement thing, I was using the terminology that I was taught - which was that reinforcement occurs simultaneously with the unwanted behaviour whereas punishment occurs after the fact (both being variations of P+). If you'd prefer different definitions for these concepts, I'm quite happy to use yours - as long as we both understand what each other is saying! Probably some do. However I have met some dedicated and knowledgable animal trainers who do use 'punishment', and also some completely ignorant and lazy trainers who only use PP. But back the original question... Well, I guess we've met some different dogs then. Like I said above, I agree that many dogs bark because they are distressed or neglected. However I've met several dogs that bark just because they like to, and because they can. If you haven't, then you're lucky. But it doesn't mean they don't exist! A neighbour of mine a while back had two dogs that would bark incessantly at everyone walking past the house. They weren't distressed at all - they had food, water, shelter, toys and company. They merely thought it was fun to bark at people, even people who completely ignored them. Now, once barking has got to this point it is obviously not going to extinguish in the absence of a response, since the barking has become its own reinforcement. (In fact, in most of the learning theory articles I've read, a dog barking is given as the classic case in which extinction is ineffective!) Would the barking stop if the dogs were rewarded for keeping quiet? If the reward for not barking outweighed the fun of barking, maybe it would. But bear in mind that the owner is not going to be there to reward the dog all the time. What will happen when the owner goes out and the dogs are faced by the choice of either keeping quiet (now without reward), or continuing the self-rewarding game of barking?
|
|
|
Post by CampWoof on May 12, 2005 8:08:06 GMT -5
Let's delve deeper into this dog that barks incessantly because he likes to and he can. And let's remember that we are not talking about all barks, but incessant barking. People can laugh, and they like to. But incessant laughing would be obsessive-compulsive and almost certainly not indicative of a person who was happy but one who is sick. Much has been written about O/C behavior in animals and I think it all concludes that these animals have a problem and need our help. So, what are we doing if we conclude that they do it because they like it and they do it in definance of what they know is our wish that they stop, and so we punish them. There is also a body of knowledge on the effectiveness of punishment in stopping O/C behavior in humans - the effectiveness is zero. Fortunately, dogs are a lot less complex than humans and with dogs we can pretty much conclude than an O/C behavior, like any other behavior, was selected by the environment. In humans these behaviors can be treated with behavioral approaches but they might be coupled with other psychopathology and the behavior therapy might not work well. Again, dogs are fortunate to have too small a brain to be afflicted with these complex big brain illnesses and behavior therapy does work well. Another whole read on this is that you are not talking about really O/C barking at all but just more barking than you would like. Then you would be talking about a normal and natural behavior that you want to get rid of. Punishment might actually be more effective there than with an O/C case. I just find it better to channel natural dog behaviors that don't work for me into more constructive outlets than to try to beat it out of them. By the way, negative reinforcement is when a good thing stops or is removed and punishment is when a bad thing is started or imposed. Also, I didn't make my final comment on a dog barking O/C because he "likes" to. Why does he like to? What does he get out of it? Where does it come from? When these questions are answered (and they must be) then you are back to just the dog and the environment and the laws of learning. Saying he likes to is ok for casual conversation but not for training where precision is important. Edited later: So now you have edited your response aftr I had responded to it? A strange ability on this board, I don't like it. Now you are describing barking that is not at all incessant. I'll say 2 more things only then I'm done with this and you can have the last word. !. You don't have to offer a more powerful reward to change a behavior. That is bribery. 2. I don't approve of leaving dogs outside where people walk by when no-one is home.
|
|
|
Post by Am on May 12, 2005 8:40:41 GMT -5
Well yes, obviously I am talking about more barking than I would like. If you are talking about a dog that barks literally every minute of every day of his life, then I'd agree that a visit to a canine psychologist would be in order. However, I've never met a dog that does that, and I hope you haven't either.
I'd define problem barking as barking that occurs disproportionately to a given stimulus.
For example, a dog that barks if you leave him alone for 24 hours is completely normal. A dog that barks every time you leave the room is a problem barker. Why? Because the barking is occuring disproportionately to any environmental stimulus that is being offered to the dog. Problem barking is barking that is unreasonable when compared to normal well-adjusted canine behaviour.
I think I've already answered this - it's fun. What does my dog get out of playing stick, being petted, going for a walk, or playing with other dogs? Well, it's fun. He likes it. He enjoys it. It entertains him.
These behaviours are just something that he finds instinctively rewarding. For some dogs, barking seems to be similar.
Actually, I think we cross-posted. I edited since I re-read my post and it wasn't as clear as I would have liked.
|
|
|
Post by willow on May 13, 2005 15:53:18 GMT -5
Interesting subject and responses, and even more interesting is that I recently picked up one of my many training books and under the chapter: "The Mind of the Dog", the author had this to say. And I quote: "'Psychiatry" is defined in the dictionary as 'the study of mental disorders'; "psychoanalysis" as 'the treatment of nervous ailments in which the causes are traced to forgotten concepts in the mind." "This quickly damns psychoanalysis as being applicable to dogs, and the psychiatrist as assuming that all naughty or difficult dogs are so because of mental disorders. If this is the case, I reckon that at least a million of our dogs alive today should be hastily destroyed, as psychoanalysis is impossible with dogs, since they cannot answer questions and the concepts of their minds cannot be recalled and probed and changed, cast our or anything else." End of quote. I agree with what this author said and for humans to try to analyze a dog and understand his mind as one would a human is a lesson in futility. The author goes on to say, and again I quote: "People are gullible. Invent a big word, charge high fees and with the help of a working knowledge of dogs, one can get away with murder". End of quote. At the very least, IMO, they can sure confuse the dog owner! I believe whether a dog becomes a nuisance barker or not depends more on the dogs inbred temperment more than anything. In other words, some dogs are more prone to developing problem barking than others, and it doesn't take much, if anything, to get the habit started. If exercise would stop the habit, my Aussie would not bark at noisy cars when they go by. If re-directing by shoving a toy or treat in their mouth would stop it, my Aussie would not bark at strangers walking down our street. If bringing her into the house and giving her a "time out" worked, she would not bark, and so on and so on....... What has worked? Teaching her that "enough" means "enough". I don't know if that is called a "negative re-inforcement" or not...but it is considered a "negative correction", which PP trainers are against using. This works fine when we are here, but when we are gone, all bets are off. BTW...when she is really "into" something going on outside, I can try to re-direct her all day with a toy etc. and she will just ignore it and me, and look at me like, "Play? Are you kidding??? The boogy man is out there and is going to get us!"
|
|
|
Post by CampWoof on May 13, 2005 17:04:01 GMT -5
Like I said, it "is" easier to beat it out of them than to train it out. But it is an absolute fact than any behavior in a dog's behavioral repertoire can be treained out, and any behavior of which he is physically capable can be trained in. The problem is summed up by the old adage that..... 99% of all dogs and 20% of all owners are trainable.
|
|
|
Post by willow on May 13, 2005 17:36:40 GMT -5
Like I said, it "is" easier to beat it out of them than to train it out. Where did I say I "beat" my dogs?
|
|
|
Post by Laura on May 13, 2005 18:30:54 GMT -5
Camp Woof, you're walking a very fine line in this thread.
This is NOT a PP board, which I believe is evident by our homepage that has links to Leerburg and how to fit and use a prong collar. If you cannot refrain from using phrases such as "Like I said, it "is" easier to beat it out of them than to train it out.", which implies that those of us that use negative reinforcement "beat" our dogs into the correct behavior, you will be asked to leave.
Purely Positive training only works when the trainer is present, so it's almost a moot point when discussing nuisance barking since nuisance barking occurs when nobody is home.
|
|
|
Post by CampWoof on May 13, 2005 18:50:19 GMT -5
How totally fitting that those who would threaten a dog with "enough is enough" would also threaten someone saying things you don't agree with. 1. I don't think the advocacy of positive training (or the statement of scientific fact) is limited to the PP area is it? 2. You are not using negative reinforcement you are using punishment and the fact that you don't know the difference speaks volumes. 3. You can ask me to leave all you want but you'll have to invoke some sort of force and power like you do with your dogs to make me leave.
|
|