"all possitive" question

Well, there's also redirecting or breaking focus using a verbal cue. Some positive trainers use a "no reward marker", a word that tells the dog "What you are doing right now will not be rewarded." Usually that word isn't "no" but something more like "oops!" or "try again!".

I've found that regardless of how I feel about the issue, once I saw all the options available to me when it comes to dealing with incorrect behavior I pretty much never have to yell or apply any kind of aversive punishment. The opportunity just doesn't even arise.

But when it comes to keeping my dogs safe from an immediate threat, and I think any positive trainer would agree with me, I'll do whatever it takes. That includes grabbing collars, tugging on leashes, and getting physical. But I'm not kidding myself that I'm training in that moment. I'm not. I'm averting disaster, which is very different. And likewise when I lose my temper and yell, I'm doing it for me not for them. That's not training, it's just venting and I try to keep it to a minimum.

Can you maybe give some examples of the situations you were thinking of?

But when it comes to keeping my dogs safe from an immediate threat, and I think any positive trainer would agree with me, I'll do whatever it takes. That includes grabbing collars, tugging on leashes, and getting physical. But I'm not kidding myself that I'm training in that moment. I'm not. I'm averting disaster, which is very different. And likewise when I lose my temper and yell, I'm doing it for me not for them. That's not training, it's just venting and I try to keep it to a minimum.

That was basically what I meant, while I agree that in the long run convincing the dog rather than coercing him/her is better, sometimes the logistics of particular situations don't allow to do so. I guess the difference between training and "the rest of the time" that you just pointed out makes it clear to me. Although there is one thing that ALWAYS pops in my head whenever I do anything around my dog, the behaviourist we're consulting said to keep in mind that "every time you're around your dog, you're training him/her, whether you mean to or not".

IMO, you can't argue with learning theory. You can only argue the correct application of the theory in the real world as to when, where, why, and how it is applicable, which parts of it are applicable in a given situation, and when another a more social (yet less defined) approach might be needed.. Learning theory is easily defined. The nuances of interspecies communication through "feeling",information, and understanding, is not.

You can discuss "love" from a biochemical, hormonal, and scientific viewpoint 'til the cows come home, but there are no words and scientific specifications which can describe a feeling. It is beyond absolute definition and human words alone. It is a connection which must be felt to be understood.

Er...uh...what was the question again? []

I think you should understand that there are "all positive" and there are "positive" trainers.  I think that houndlove has described very accurately the way in which most people apply positive training.  Day to day, in teaching my dogs the behaviors I want to install, I do not use any punishment.  And, even in situations when they might fail - for example, if I introduce distractions too quickly, and they backslide - I use NRM's or lack of reinforcement.  Only in a life or death would I probably feel the need to correct, and it would be fast, humane, and over instantly.  But, the one thought I'd like to leave you with is that if you learn to train correctly, the number of times when you would need to feel that need to use correction will be very, very few.  Most people resort to correction far too early.  And, mostly, when they do, it was never really necessary - it was the human that messed up, not the dog.  

Some positive trainers use a "no reward marker", a word that tells the dog "What you are doing right now will not be rewarded." Usually that word isn't "no" but something more like "oops!" or "try again!".

 
I do that - actually I do say "no" because it's most natural to me mentally, but not in a scolding tone of voice. Most commonly I use that NRM when I'm asking for one behavior - say, a down - and he keeps offering the wrong one - a "play dead" - because he's too excited, flustered, confused, whatever. I lower my voice and quietly tell him "no" to try to break through that cycle of WhatdoIdo and get him to listen.
 

I've always been curious about something, in the operant conditioning world, when, if ever, and how are corrections used?
The way it's been explained to me, if the dog is showing a non-acceptable behaviour the options are(as far as I know): ignore; eliminate whatever triggers the behaviour; or turn whatever is triggering the behaviour into something good as far as the dog is concerned. I can think of a couple of situations where none of those options are viable, what happens then?

 
Another option would be asking for an incompatible behavior, like telling a jumping dog to sit or down.
 
I can't think of any situations where I would purposely use a correction (I say purposely because I'm only human and have been known to have some knee-jerk reactions I'm not proud of). In the situations that Cressida described getting physical with the dog to keep him safe - I agree, but even then I'm not sure those are corrections if they are performed while or just before the dog puts himself in danger. I guess you'd have to describe the specific situation where you'd use a correction....

It may be necessary to use a correction for certain self-rewarding behaviors with certain dogs in certain situations.

Nobody here uses time outs?

I don't really. I may use a "chill out" time out if things get too rambunctious, but if I do send the dogs to their places as a time out, I always give them something nice to chew on while they're there. It's not a punishment really, just a management tool if things get too rowdy.
 
I do use negative punishment occaisionally: removing something the dog likes from the scene until they can behave. But I do not use positive punishment if at all possible: applying something the dog does not like to make them stop doing something. I redirect, refocus, ignore if possible (sometimes you can't for safety and general annoyance reasons), use a verbal interuption "Hey!" or "Ah ah!" but I don't cause discomfort or pain (at least not on purpose) in order to train.

I guess you might say I use time-out although I tend to think of it as simple a down-stay... if we have company (or are visiting friends) and Russell's getting in everyone's hair I might put him in  down-stay until he's more chilled out.

I think that quite a few of us use time outs, but not for punishment.  As others have said, my time outs are "Ok! Enough!  Mommy needs time away from furkid!  Go to your kennel and chill!" and I chuck a bone in there with Strauss and he either chews on it or takes a nap.

time-outs, as in put the dog in a kennel, are not a good training tool. In order for a dog to connect his behavior with consequences you need the consequences to occur instantly after the behavior. By the time you get the dog into his time-out he has no idea why. The only "time-out" that may be effective is an immediate withdrawl of attention-- example: your pup ties to elicit play by biting your arm, you freeze and stare off into space.

That's the kind of timeout I am suggesting, MP, not going to the crate. I have successfully used this kind of immediate action as negative punishment for a lot of different behaviors, including whining and jumping up on tables. Yeah, timing is critical, but when isn't it?

I use "time outs" for Emma. She's always wanted to pull stuff out of the trash can. I should really just go buy a trash can with a lid, but I'm cheap, and love trying new training techniques.

I started using a one minute down stay, the second her feet left the ground (which is really just using an incompatible command) and ignoring her for the whole minute (TORTURE!!!) when she tries to look in the trash can. It worked. I started out using it six or eight times a day (used it three times in a row, the first time), and now I only need it once or twice  a week. I've done this for a couple of months, so that's good progress, IMO.

I am one of those people that doesn't think there IS such a thing as an "all-positive" trainer. It just doesn't happen, even in us clicker-only people. I do NOT use positive punishment at all in my dealings with my dogs. Ever. I have learned, from experience, and from research, that it's simply not necessary, and everything else you need to do can be done via other means. I have not yet used R- in my training at all, as I haven't found a need, since most R- also requires and element of P+ to work, but I keep my mind open about it's use in very select few situations in which aversives were not applied.

MOST of what you would call "positive" trainers actually DO use punishment indeed! But they use negative punishment! And some use it a lot! Of course this does not involve any aversives being applied, if you know your OC terminology, but it's still punishment nonetheless! If you turn your back on a barking dog, it's negative punishment (IF the dog lessens the frequency of jumping). If you remove yourself from a room because of a bitey puppy, it's negative punishment (well, it is IF the puppy lessens the frequency of biting). If your dog wants to go for a walk and it won't sit at the door, you leave. That is negative punishment (IF the dog lessens the frequency of "not sitting" in the future). The point is, most of what people call "positive" trainers do indeed use punishment in a lot of their lives. It is, in some cases, unavoidable.

However, if you understand OC terms, you understand there is a BIG difference between negative punishment and positive punishment! And of course, whether a consequence is a punishment or not lies in the eyes of the dog, not the handler. I could turn my back on my dog all I like when she jumps, but if she keeps jumping she obviously doesn't find it punishing. I can jerk on the choke chain as much as a I want, but if the dog still continues the behaviour it's NOT a punisher - it's simply rough annoyance by handler.

And that's not to say that a negative punisher isn't aversive. Some negative punishments can be VERY aversive to a dog, depending on the dog and the situation. For dogs who crave, LIVE for, attention, including those dogs with behavioural issues such as separation anxiety, removing attention to try to lessen a behaviour can be very hard on a dog. Dogs face aversives in every day of their lives. However the type of stress often occurring is very different, and the fallout from negative punishment is not at all the same as the fallout from positive punishment. There are a lot more risks of side effects from P+ than there are from  P-.

So yes, in my training I do use punishment - negative punishment. The extent to which I use it might be to remove attention from me, or to remove attention from another dog, or to remove opportunity for reward. However, depending on the situation I use a lot of extinction as well, along with Differential Reinforcement of Other Behaviour (DRO) and Differential Reinforcement of Incompatible behaviour (DRI).

Even NRM's can be very aversive to dogs. People call them NRM's, but what some people don't realize is that the DOG determines what effect that has on them, not the handler. I have a dog who will completely shut down upon the use of an NRM, and this happened earlier on in my training when I thought that "good" trainers use them. There are dogs who do not respond well to the use of NRM's, and it affects them much the same way as any positive punisher would. Now I realize that it's not the case at all, and I, and many others, do not use NRM's of any sort.

The part where "positive" (if you use the term) trainers draw the line is that they understand that they have no use for positive punishment, and often no use for R- either. That is, they do not believe that the answer to problems lies in force, pain, intimidation, etc. As well, a goal of a positive trainer is usually to do their best to use the LEAST amount of P- in their training, where other options are available (extinction, DRI, DRO, etc).

I have never physically corrected my dogs. I do not see a need, and I have much experience that tells me that, on a daily basis. I also don't believe in things like sticky tape, noisy cans, water pistols, tabasco sauce, etc. I have had great success in working with my dogs, and helping people with their dogs, using R+, P-, and extinction alone.

As for time-outs, again it depends on the context of the situation and how people view a time-out, how they define it. For me, I do not ever use the kennel as a punishment or a time-out place. I HAVE gotten up and walked away from a puppy, or stepped out of reach, or walked into another room and closed the door for 15-20 seconds as negative punishment. You could call this a time-out I suppose, if that's your definition. Or, if I know that a puppy is overtired, has gotten out of control (which some puppies can do so easily until they mature), and really needs a moment to calm itself down and rest, I might put the puppy on a tie-down or down for a nap in its kennel. But this in itself is not a time-out or a punishment, it's simply addressing the issue that the puppy needs to have time to calm down in a safe environment, for it is not yet old enough to have the self control that most new puppy owners often wish of it, and it is to meet the needs of the puppy is the puppy is acting the way it is from overstimulation or overtiredness (much like a two year old who acts out when overtired).

As for emergency, life or death situations - of course you do what you have to do, what needs to be done in that situation. Last weekend we had an awful, awful scare. Due to the power of human error, someone in our household had had one of our parrots out playing in the computer room. She put the bird on the playgym and went out to check on supper for a moment. Being the human that she is, she got onto a task and forgot about not putting the bird back in its cage. At the same time, we have a female that is in heat, so needs to be kept away from our boys. Therefore the dogs are separated at this time in the home. So some dogs go on into the computer room (which is the most-used room in the home, where people always are). It was only about 30 seconds before the mistake was realized, but you can imagine the kerfluffle of one 54 gram parrot, on the floor, in a room full of terriers. You can bet there were humans on the floor right there, flinging dogs aside this way and that to get to the bird. Long story short, we are very thankful that the bird is, three days later, alive and well, other than quite a few missing feathers, and a broken blood feather. And we will take it as a lesson in learning for us humans, that management can fail (we have three parrots in our home and have not had any problems until that day), and that one can never become too relaxed about these things.

The point? The point was, chances are some of the dogs were handled in that instant in ways we would never handle them in a "normal" situation. Chances are some were pushed aside rougher than we would ever touch them. However, this is NOT a "correction", nor is it punishment. It was simply reacting to the situation at hand. Our intent was not to "lessen the dog's behaviour in the future", nor would it, as most would see easily, be likely to lessen a terrier's behaviour of "chase and kill the bird". The point, if you will, is that the concepts of corrections and punishments apply only to training situations. It does not apply to emergency situations where life is at stake. In an emergency there is no conscious thinking of "Okay, how can I make it so that this won't happen again in the future", it is simply "Get that bird out of there before it's killed!". So there's a big difference when discussing what corrections/P+ you use on a TRAINING basis, and what things you are capable of doing when somebody's life is at stake.

So basically it's in the eye of the beholder. If you have the tools, and the knowledge, you can live your entire life with dogs and not ever apply corrections to a dog (read: corrections, not punishments). If you don't have the tools, then of course you aren't going to be as effective in trying to work without corrections, because corrections are all that you know, or at least a big part of it. The more you learn, the more tools you obtain for that magical "toolbox" of ways to solve the bumps in the road you pass as you go along.

Kim MacMillan