Dog-dolphin analogy

ron2

I don't see any "bond" between dog and owner who uses punishment. Nor would I assume a loving "bond" between one dog that "corrects" another.

 

I've actually seen both, although I can't cite any sciencific document. Now in human behaviour I would say there is lots of science to suggest bonding with ones 'punisher'. My father actually was known to use punishment on occasion and I have a very strong loving bond with him.   

ron2

I don't see any "bond" between dog and owner who uses punishment. Nor would I assume a loving "bond" between one dog that "corrects" another.

You have said that your dog is a crossover and you used punishment before. Would you said that you didn't have a bond before with your dog whatsoever?

Again, it seems we want to say that the "bond" is either-or, and that isn't the case.  Dogs are dependent upon us for resources, and you can't help but bond with someone who is providing you the resources you need or want.  But, captives have been known to "bond" with their kidnappers, too, so I think we need to be careful of how we assign legitimacy to that wagging tail...

ron2

It seems the author is trying to find some justification for downplaying the importance of reward training so that he can justify his use of punishment and his lack of use of rewards training.

I don't think a single effective training method has a "lack of use of rewards training"  In fact - KMODT has plenty of rewards, it's just not always food.  I think there's a perception that trainers who use punishment or corrections use them far more heavily than any rewards, which just isn't an accurate interpretation.  If an individual has personal experience where a dog was punished more than rewarded, then I'm sorry, and that's likely the reason why positive reinforcement training is working so well for them when they switch - because it DOES work to achieve what you want so much better than exclusive punishment and correction.

ron2

And to the counter argument, I don't see any "bond" between dog and owner who uses punishment.

  Bonds aren't always exclusively positive - as the article points out.  I use corrections and I believe there is a bond between my dog and I based on all the other pieces of our experiences together, and that bond involves more than just love.  I realize you don't see us, but it doesn't make it any less real.  I'll choose not to be insulted since this wasn't personally directed, just a general blanket statement.

Edit to add:

spiritdogs

Dogs are dependent upon us for resources, and you can't help but bond with someone who is providing you the resources you need or want.  But, captives have been known to "bond" with their kidnappers, too, so I think we need to be careful of how we assign legitimacy to that wagging tail...

  True! Which also makes me think of the people who misinterpret tail wagging as a positive sign for "play".  Viewing dogs through human perceptions is a failure to respect the dog, imo.

espencer

ron2

I don't see any "bond" between dog and owner who uses punishment. Nor would I assume a loving "bond" between one dog that "corrects" another.

You have said that your dog is a crossover and you used punishment before. Would you said that you didn't have a bond before with your dog whatsoever?

I'm sure we had a bond before, and an even better one after. In the after, we had better communication. I've tried explaining it until I am blue in the face, sometimes, to no avail. When I have tried to punish Shadow physically before, he thought it was play, because he was raised initially by his first owners to wrestle.So most of the physical corrections seem like play to him. So, for him, they are not corrections, just because I decide that they are. Be careful what you reinforce. And the dog has to see the stimulus or aversive as such and connect it to the behavior or it is inconsequential or coincidiental. (I miss the "beating my head against a brick wall" smiley.)

And, I suppose that some owners have a bond with their dogs even if they use what they think is punishment or correction. Again, it depends on how the dog interprets it. (I miss the "beating my head against a brick wall" smiley.)

I think it's a fallacy that captive wild animals are not trained to do anything against their instincts. Temple Grandin oversaw the training of antelopes (antelopes!) to voluntarily enter a crate and allow blood to be drawn from their leg or neck. You cannot tell me that an antelope would naturally enter any kind of small, fully enclosed place in the presence of potential predators (humans) and have me believe it. And that's just one example. As well as tricks, sea mammals are often taught to allow uncomfortable husbandry procedures. And for safety's sake, they do need to be very reliable. Grey Stafford talks about a Walrus with a very low frustration threshold that would try to crush people against walls or stab their thighs with his tusks if annoyed during training. Thankfully, he had such a strong reward history with some incompatible behaviours that he was still able to be worked with as long as someone was spotting. I think the practice of keeping Orcas in captivity at all is questionable, but we do have to remember that wild animals don't have the natural inhibitions towards us that dogs do.

As others have mentioned, it's not whether you have a bond or not with your dog when you use corrections, but how good your relationship is. If you have never tried it without punishments, then how can you compare? I've tried it with and without and I will say that I will always be working towards eliminating punishments completely. I thought it would be hard, but it's not, really. I'm on my second dog since I made that decision and he's a real handful. Smart and pushy and extremely enterprising. He twists everything to his advantage and if he can't, he bites something. To stop that I would have to teach him to stop thinking about winning all the time and then he wouldn't be nearly so fun and easy to train. So I make sure everything I want him to do is advantageous to him. He's going through his adolescent period at the moment and I know this is as bad as it's gonna get, but he's still very reliable off leash, and I could count the times I've punished him on one hand. It's not that hard and reliability is only an issue if you let the dog (or Orca) get into the habit of failing in the first place.

Lastly, I don't know why a dolphin shouldn't be held up as the poster child of positive reinforcement. Because it's not a dog? Nor is a hare a dog, or a dolphin, yet I'd like to see you train one of those using punishment at all. A chicken is neither a dog or a dolphin, but they don't think much of punishments, either. My rabbit is not a dog or a dolphin, but when she doesn't like something she bites or she runs, so aversives are out with her as well. Do you see a pattern emerging? There are a lot of animals that aren't dogs or dolphins, but considering you won't get far with the majority of them using corrections, but will have no trouble at all using the "dolphin" technique, I think I'll go with the method that works on all animals rather than just dogs. Unless I feel like getting my hand bitten off or watching helplessly while my animal bolts for the hills.

ron2

I miss the "beating my head against a brick wall" smiley.

 

I'm with you there....

 

 As I said in a meeting last week, if you don't think positive reinforcement works, you shouldn't accept a paycheck for you work. If it doesn't work then it wasn't a reinforcer, or you didn't apply it correctly. Same is true for a punisher, but I'd feel much safer not correctly applying a reinforcer than a P+.

To add to it, though at this point of the night, I am rather unfocused, it seems with the dolphins, as far as I recall, there was fair use of P-. I seem to recall the trainer picking up the bucket of fish and walking off upon some inappropriate behavior. That I absolutely make use of. I don't hesitate to take the reinforcer and walk away from a dog who isn't displaying ready to work behaviors.