Eight Rules for Punishment (and why we shouldn't use them)

Liesje

Do you agree with all 8 rules?

After you mentioned that in your type of training rule number 5 some times has to be related to the trainer i guess there are exceptions to the rule. I dont agree much with rule 7 if the alternative needs to be given by the owner. If i correct Chuck he actually can choose the alternative himself, i dont care what he does after as long that is not the behavior he was corrected for

Liesje

espencer

Back to the original topic, if people cant apply these 8 rules for punishment to perfection then those people should not use them

 

Do you agree with all 8 rules?  Just curious.  I don't.  I also find it interesting that those who use no punishment or an absolute bare minimum are the ones advocating for these "rules".  Should we let a "traditional" crank and yank type trainer make up our rules for free shaping with a clicker?  I don't think so.

The "rules" for free shaping with a clicker are based on principles of operant conditioning, as is punishment.  The "eight rules" seem fair to *me* and are scientifically sound as well.  The type of trainer you choose has more to do with your own determination about how you wish to treat your dog, and not whether the methods work or not.  But, again, we are back to the premise of whether you want your dog working toward a reward, or you want him working to avoid a penalty.

spiritdogs

The "rules" for free shaping with a clicker are based on principles of operant conditioning, as is punishment.  The "eight rules" seem fair to *me* and are scientifically sound as well

I agree. Rule 7 is a good rule with strong scientifc support.  Those familiar with the studies into punishment understand why it's there and why the having the alternative behavior is a much better option for the animal.

The "eight rules" seem fair to *me* and are scientifically sound as well.

What is the scientific basis for #5?  It seems partially correct.  I agree that the punishment must be associated with the behavior, but I think there are times where the dog shouldn't know where the punishment is coming from and times that they should.  I think dogs are smart enough to take int account what is being punished, how they are being punished, and who or what if anything is punishing them.  #5 statement is far to vague for me.  I can't outright disagree with it because obviously the dog should be learning to cease the behavior and not fearing the handler.  There are so many different forms and nuances for punishment whether one agrees with using it or not does not mean you can over simplify it.

we are back to the premise of whether you want your dog working toward a reward, or you want him working to avoid a penalty.

Even if I use a punishment, the dog is always setup to be rewarded.  Rule #7 - the dog MUST understand the alternative behavior in order for the punishment to be fair or effective, IMO.

espencer

And you say this even when you live at the other side of the world. Ok

This is a new low. You might not know it but Australia is one of the US best Allies. But this statement really gets to us.We here it over and over again.And we are heartlily sick of it. May be electrons spin the opposite way in our country or something like it..

Now if you had done some research you would have looked up the AKC web site and noted that ONE US poodle got it's TDX title last year. You would have noted that Luci has these funny letter in front of her name TCh . They stand for Tracking Champion. If you were really interested in checking out facts about  you would have checked what was required for either title. You would have found out the a Tch tile has no restriction on obstacles, that the off track requirements are tighter, that the length is longer, that there are at least two acute turns....

Oh and she has CDX too, just an easy little number between tracking seasons.... and is trialling in UD ...oh and there is video evidence of her doing these exercises...

 Oh and that a fair bit of my spare time goes into coaching others in tracking... right up to advance levels 

******content deleted - rude and personal attack***********

poodleOwned

espencer

And you say this even when you live at the other side of the world. Ok

 

This is a new low. You might not know it but Australia is one of the US best Allies. But this statement really gets to us.We here it over and over again.And we are heartlily sick of it. May be electrons spin the opposite way in our country or something like it..

Now if you had done some research you would have looked up the AKC web site and noted that ONE US poodle got it's TDX title last year. You would have noted that Luci has these funny letter in front of her name TCh . They stand for Tracking Champion. If you were really interested in checking out facts about  you would have checked what was required for either title. You would have found out the a Tch tile has no restriction on obstacles, that the off track requirements are tighter, that the length is longer, that there are at least two acute turns....

Oh and she has CDX too, just an easy little number between tracking seasons.... and is trialling in UD ...oh and there is video evidence of her doing these exercises...

 Oh and that a fair bit of my spare time goes into coaching others in tracking... right up to advance levels 

******content deleted - rude and personal attack***********

I may be wrong, but I don't think he's in the US - he's in Mexico.

spiritdogs

I may be wrong, but I don't think he's in the US - he's in Mexico.

oh well, got it wrong again...

I thought that it would be useful to look at two case studies of dogs and punishment. They are almost extremes and as such serve as useful limits.

My now deceased Lab was a very high drive dog, very affectionate on his own terms and with a quite high pain tolerance. For quite a long while i wa a fan of Koehler really based in my case of knowing very few alternatives that actually worked. For my dog, actiaving the correction collar without a fair bit of force was a waste of time.

Rule one /two .. I never gave him warning when a correction came and made sure that it was harsh enough that the behaviour ceased  I was sometimes accused of being heavy handed, but nagging puniisment is ineffective

Rule Three This is a super thin line. With a dog this insensitve to pain it is very hard ot get that perfect intensity and in his case he became quite flat. I also at one point used an ear pinch for cornering with tracking. It became a real dissaster and was an issue with years. Yet a lot of the texts suggested it, i guess that many of them had not noticed the huge difference in styles between breeds.

Rule Four I did get this mostly right with some help. Most novice handlers correct way too late. I.e the dog is in full lunge where as the dog should have been corrected when it was thinking of lunging.or thinking of leaving an LLW posiiton.

Rule Five  Cadbury certainly didn't act like he was happy to be with me sometimes.

Rule Six: I was absoultey meticulous with this... 

Rule Seven : I don't think that the alternatives were clear. Sometimes when you combine punishment methods with dominane stuff you are on to a real loer training wise.

Rule Eight : What positive reinforcement? This dog had his first treat at four :(

When i changed to a regime of R+ and P- (yes i do agree you have to have some element of punishment) this dog turned around and i developed a reationship with him that was wonderful. I guess i can't understand why any one else doesn't want the same... iN reality he would have had significant safety related P+ amount 1 a year. I don't muck around with these. He was an obedient wonderful companion and working dog under this regime.

I guess that i fiest learnt about training in the zone, where training is as highly rewarding for you as for the dog (we hope) with him.

 Well Luci my poodle

She has had four P+ corrections in her life for saftey related issues. 

She has had two sessions with up to five TOs to extingish a residual behaviour from her noise phobias

I have applied P- twice for prey hunting when she was asked to do something else.

She has consistently worked well under crowd enviromental and space pressures using trianing (proofing).

Now i agree that many R+ dogs (and balanced dogs and others ) sometimes work soft. I don't think IMHO that the use fo corrections fixes these issues. I think that good prooffing does. 

I think that arguing the priniciples is arguing pretty sound empirical data, and is arguing wheter teh sun iis up or whether my Rugby Team will win .Of course they do. Filling it out with some interesting case notes can be helpful but we have to watch what and hwo we are assesing.

poodleOwned

I think that arguing the priniciples is arguing pretty sound empirical data, and is arguing wheter the sun is up or whether my Rugby Team will win .Of course they do. Filling it out with some interesting case notes can be helpful but we have to watch what and hwo we are assesing.

Another way is to design trials together to prove or disporve the disupted points . What is the consensus view of the disputed points? We have some awesome professional talent on this board who can look after the various issues. So lets make a difference. Can the various protagonists just list the various points??

espencer

Liesje

Do you agree with all 8 rules?

 

After you mentioned that in your type of training rule number 5 some times has to be related to the trainer i guess there are exceptions to the rule. I dont agree much with rule 7 if the alternative needs to be given by the owner. If i correct Chuck he actually can choose the alternative himself, i dont care what he does after as long that is not the behavior he was corrected for

So, if I understand you correctly, if you correct your dog for chewing the cushions on your couch, you then don't care if he trots off to raid the garbage can?  Sorry, but your argument makes less sense than if the dog defaulted to a deliberately trained behavior such as "sit" as an alternative to jumping up, for example.

In any case, why ever would you want a dog to fear you as the trainer, instead of merely recognizing that his behavior is what warranted the correction?  IMO, that's when you are likelier to get a shut down dog, or one that only behaves when the owner is present.

spiritdogs

So, if I understand you correctly, if you correct your dog for chewing the cushions on your couch, you then don't care if he trots off to raid the garbage can?

Seriously, what are the chances of that to happen? He can choose to sit, he can choose to laid down, he can choose to eat a sandwich, i dont really care, as long as he is not "chewing the cushions" anymore. So far he (or other dogs that i have corrected) have not chosen to do another "bad" thing instead as alternative

spiritdogs

In any case, why ever would you want a dog to fear you as the trainer, instead of merely recognizing that his behavior is what warranted the correction?  IMO, that's when you are likelier to get a shut down dog, or one that only behaves when the owner is present.

I dont want the dog to fear me (which also depends a lot from rule # 3 and # 6). I was answering Liesje question, i agree with number 5 but i guess that if in some special training like the one Liesje does the dog needs to fear the trainer (like she said before) then there are exceptions. I think you have to ask that question to her.

So far he (or other dogs that i have corrected) have not chosen to do another "bad" thing instead as alternative

See, that's where you and I think so differently.  I think that dogs don't do "bad" things, they just do "dog" things, and our obligation is not to "correct" them, but to explain that we prefer them to do behaviors that we like, rather than the ones that could get them into difficulty (such as jumping on someone and knocking them down).

Here's something to train your dog to do.  Put the dog in a stay, walk ten feet away and drop a little pile of milk bones on the floor, then walk another ten feet, turn and call your dog as you tell him to "leave it" when he gets to the bones.  He must ignore them and come to you instead.  Try doing it without correction.  C'mon, just once - try it.  Then video.

Oh, and go read Sillysally's thread about her dog's new trainer;-)

spiritdogs

I think that dogs don't do "bad" things, they just do "dog" things, and our obligation is not to "correct" them, but to explain that we prefer them to do behaviors that we like, rather than the ones that could get them into difficulty (such as jumping on someone and knocking them down).

That's why i used the " " between "bad", because i knew you were going to go in that direction, i know dogs dont do """"bad"""" things.

Let me put it more clear:

So far he (or other dogs that i have corrected) have not chosen to do another behavior that i dont want him to do instead, as alternative

I think espencer has said plenty of times that there are no bad dogs, just bad owners. I gather that goes for behaviour as well.

Although I have to say, just before Chrissy I came home from the field and Erik was so overjoyed to see me after ten days that he spent most of the afternoon discovering just how many objects I would take off him before I gave in and went outside to play with him. It was somewhere in the order of about a dozen.

Guess my dogs are on the inventive side.

The moral of the story is if you don't want to punish it's a good idea to look after your dog's needs before they start telling you what they need! 

espencer

So far he (or other dogs that i have corrected) have not chosen to do another behavior that i dont want him to do instead, as alternative

Espencer it is not at al unusual that this occurs. I have seen one of my own dogs grab an instructor's arm that was correcting him. I have seen dogs escape in ways that were never thought of in correction. I need you to know that i come from a training position that certainly didn't spare the horses as far as correction was concerned . Correction is not some new fanged toy that i am either scared about or don't understand.

I rmemeber very clearly this BC that was in a class that i saw very early on in my dog training. we were by the railway line. Now it was true that the women who owned the dog  was a veyr gentle soul, but of course had been warned off reinifocement and the dog was running rings around her. BCs are very birght and will get what they want. So the train rolls by and the dog nuts off, and the person who was running the class ripped into the dog it sure stopped, that time. Next time the train came past, the women jangled the correction collar and the dog bit into her casuing 13 stiches. Guess you know the end of the story.

I wish i could say look i know you have you view but it just isn't backed by research and hasn't been backed for many years . You can look at  a much more temperate source than me, Lindsay in Applied Dog Behaviour and Training. I reccomend this book if your are dead serious about knowing where and who "invented" all thiis codswollop we talk about.Linday is no card carrying liberal.

So lets have some fun and have a look at some scenrios as to whether we should correct or not.We all need to keep our cool, but i think if you don't back it up you should be prepared for some pretty hard questioning.

Case 1:  Female Poodle Sunndely stops heeling for no reason and refuses to continue heeling. Has a succesful record of trialling , 15 months old Sapyed early well socialsied, been trained with rewad based methods. Correct to get going How when and where any consequences?

Case 2 : Female Poodle (same dog as above)runs across road and doesn't ocme whe called aged 2 years. Correct or not?

Case 3: 16 month old male poodle ex show. No ones and twos everywhere. Ignores human voices, jumps up 6 ft high and licks face when excited . Will not sit drop , stay or recall. Correct or not? What for ? What consequences?

Case 4: 18 month female GSD working line. R+ trained well installed boundaries. Starting to show  some idependence , some inattention.Well bonded to owner. Good trial potential. Correct or not? What for? How?

CAse 5: 7 Month old male lab (not mine). Showinf "disrespect" to owners. Jumps up showing "dominace" by pulling on lead. Won't sit drop or heel. Nearly impossible in class, real troublemaker. Coreect or not? What for? How.

Case 1:  Female Poodle Sunndely stops heeling for no reason and refuses to continue heeling. Has a succesful record of trialling , 15 months old Sapyed early well socialsied, been trained with rewad based methods. Correct to get going How when and where any consequences?

No - IMO, always, the very first thing to be sure the dog has no physical problem (hips, eyes, cut foot, anything).  Second, determine if something came in to view that might have startled or frightened the dog, i.e. consider "intelligent disobedience" before you assign blame to the dog for being stubborn or willful.  Evaluate whether the dog is ring sour.  If so, adjust your training program accordingly, but don't make the dog hate the ring even more.   There are many more possibilities, but you get the picture - the first response is not to correct, even if you resort to it later. 

Case 2 : Female Poodle (same dog as above)runs across road and doesn't ocme whe called aged 2 years. Correct or not?

Insufficiently trained recall, plain and simple, unless a case of "intelligent disobedience" (such as the dog sees the truck coming before you do).  The correction I would use is really not a correction per se.  Dog would go back on a long line for more training, especially at this age, when they are often asserting some independence from the handler.

Case 3: 16 month old male poodle ex show. No ones and twos everywhere. Ignores human voices, jumps up 6 ft high and licks face when excited . Will not sit drop , stay or recall. Correct or not? What for ? What consequences?

Clicker train the dog!!!!  Poodles are smart, however, "sit" is often NOT a default or well trained behavior in show dogs - people want them to stand, after all.  Assume that he may never have learned the behaviors you mention until you train them yourself.  Also, for safety, because he jumps high enough to injure someone, put him on a lead and step on it so that he cannot jump up without "self correcting" (an instance where we can see that the handler is not associated with the correction because you make no movement, only the dog moves for the correction to take place.  

Case 4: 18 month female GSD working line. R+ trained well installed boundaries. Starting to show  some idependence , some inattention.Well bonded to owner. Good trial potential. Correct or not? What for? How?

Vary the training routine, introduce puzzle toys.  Train some exercises specifically for attention, and put it on cue, then add distance and duration.

CAse 5: 7 Month old male lab (not mine). Showinf "disrespect" to owners. Jumps up showing "dominace" by pulling on lead. Won't sit drop or heel. Nearly impossible in class, real troublemaker. Coreect or not? What for? How.

Assume adolescence, not disrespect.  Assess the owners before you assess the dog!  Figure out how the owners might be accidentally reinforcing the jumping up (probably by saying "no" or pushing the dog away, both of which, if he is seeking attention, are reinforcing).  Advise the owners how to ignore the bad behavior, click and reward the dog once he is calm, and train behaviors that are incompatible with jumping.  I would take a dog like this out of group class and do private sessions just to install the cues so that the dog understands them and obeys them without distraction.  Then, add one or two benign distraction dogs at first, not a class full.  In other words, gradually increase the distraction level.   Also, this dog may be under-exercised as many adolescent Labs are - if so, discuss options for that (day care, fetch, etc.), and teach owners about puzzle toys, Buster Cubes, etc.  Use DAP collar.