Dog Killing Cats...

[DutchBunny03]

but hte dog "correcting his own mistake doesn't get a reward. Just avoids a correction. There are many different parts of dog traing +P (giving a reward) -P (withholding a reward) +N (giving a correction) and -R (withholding a correction) Balanced dog training uses a mixture of all 4, depending on the situation.

I can tell you several fields that don't use alpha theory - police dogs, search and rescue, detection dogs, service dogs of various sorts including guide dogs for the blind. All give corrections, sure. But none of them use alpha theory. All of these are fields where you want the dog to be right 100% of the time but you can't use corrections once training is over. Plus, with apprehension dogs in Police work, you need a dog that thinks he is 1000% king of the world, could take down King Kong if you asked him to, and will keep his grip despite being hit, kicked, stabbed, or even shot. So physical corrections to a dog like that are nothing. The biggest punishment for many of them is the training session ending before they get to take down the decoy. Or disappointment for their handler. Also, the dog must have 100% certainty what is expected of him from a given command, must obey immediately and the handler must have 100% certainty that that will happen. He must put his life in the dog's hands and trust. The dog must do the same - face down the threat with the knowledge of what is expected of him and knowing that the handler is his partner.

Part of the "alpha theory" problem is the belief that a dog does things out of a desire to be dominant. There are actually very few dominant dogs. There are, however, a lot of insecure dogs that will take advantage of a lack of leadership and throw their weight around because they don't see anyone else as being in charge. These dogs are easy to spot because they are always doing "dominant" things like trying to hump other dogs, trying to "stand over" other dogs, being pushy with their owners and others, etc etc A truly alpha dog doesn't do any of those things except as a last resort. 99% of "problems" are solved by a simple look and the other dog offers submissive behaviors.

As I said, what you are describing isn't alpha theory. It's balanced dog training. Alpha training is something entirely different. Alpha theory is a "hammer and nails" practice - when the only tool you have is a hammer, then everything looks like a nail. Meaning that alpha theory is based on the idea that your dog wants to be the boss and you must regularly put him in his place through the use of corrections. When that is all that is in your toolbox, every interaction with a dog is you vs the dog.

Alpha training is not inherently bad, when not approached with the "hammer and nails" idea. The dog should be rewarded, but not with food. That teaches the dog that if you dont have a treat for him, he doesnt have to obey. The dog needs firm correction, with which a correction collar helps. Trying to think up your own correction method could lead to injuries. Dominance training and alpha training are not the same thing. Most alpha training has the "hammer and nails" approach, but dominance training does not. It shows the dog that you are in charge without breaking your dog's trust or hurting him. The dogs are given a correction if they do not obey after the second time a command is given, sometimes after the first time, and are rewarded with praise or petting from the owner.

 

[dainerra]

actually, a big part of treat training (skipped by many) is to wean out the treats as a behavior is perfected. That ends the "if you don't have a treat, I won't obey"

Speaking as someone who has been training dogs for over 13 years, "alpha training" is a very specific form of training which is a very much "hammer and nails" approach. It allows no rewards of any kind, be that treat or praise. Basically, it is the polar opposite of "pure positive" training which allows no corrections at all. Some don't even want you to use the word "no". A good program offers a balance that is clear and precise - every correct exercise is rewarded in some way (even if that reward is delayed till later) and every incorrect action will be marked and brought to the dogs attention.

Some dogs will happily work for praise. Some dogs will work best for play/toys. Others will work for food. Every dog will work for a specific payout and many dogs will happily work for more than 1 or willingly transfer from 1 to another.

Another important part of training is the "delayed payout" That means that the dog will do an entire routine/job and wait for later for the reward. Much like we work all week knowing that we will be paid on Friday, dogs will work knowing that after they leave the field, get home, are crated after an assignment, they will get some reward or praise.

I've met dogs that, if praised in the middle of a routine, will go so over the top excited that they blow the rest of the exercise. It all depends on the individual dog.

 

[DuckGirl77]

alpha training is much more confrontational - harsh corrections, no (or very rare) rewards, man-handling the dog into position and to correct behaviors. Often views the entire dog/owner relationship as you vs the dog instead of a mutually beneficial partnership. Trainers who espouse alpha theory often encourage the use of alpha rolls as well, even for young puppies. It's frequently viewed as a building block for the future relationship - teach the pup that it must submit from the day it comes home.

Alpha -or dominance- training does not always use harsh corrections. It depends on the dog. The dog gets rewards in the form of praise AND ONLY PRAISE. Using treats takes the dog's focus off the handler and onto that little piece of food. Using a method like Don Sullivan's method, at least, gives the dog praise for a reward, corrections if the dog doesn't obey, and no alpha rolls. This kind of training will work for the OP.

 

[Chickerdoodle13]

I know we've had this conversation before so I will not rehash it over and over. I'm still not sure you understand the psychology behind using positive reinforcement (treats, toy, etc).

Anyway, not all dogs will work for ONLY verbal praise. Some dogs will shut down if their only options are verbal praise or punishment.

As for the OP, I do think there are still training options available but she may never truly be trustworthy around cats. However, I wanted to chime in and say that even if you made the decision to euthanize her, it would not be wrong. Difficult, but certainly humane in her situation.

 

[DutchBunny03]

Treats should almost never be used for training. Not even in the very first training stages.If you give in to your dog and give him treats because he will "not work for verbal praise',all you are teaching your dog is that you can be manipulated. Training requires consistency.

 

[appps]

Haven't read the whole thread so sorry if duplicated but to protect your other animals I would be putting a muzzle on her like you need to with greyhounds to protect the other animals till you can find a solution.

 

[dainerra]

Treats should almost never be used for training. Not even in the very first training stages.If you give in to your dog and give him treats because he will "not work for verbal praise',all you are teaching your dog is that you can be manipulated. Training requires consistency.

but that isn't how it works.

you fade out the treats, a step that many owners skip. they try to go from over-feeding treats to cold turkey. Dogs learn to work for verbal praise. Sorry, but no creature works for nothing without force. Some dogs happily work for praise from day 1. Some dogs are never able to work for praise. They will happily wait for hours to get a reward but they know that it will come eventually.

Police dogs don't work for praise. Most work for toy rewards. They also work for the joy of the work, which is still followed by a reward.

 

[EggSighted4Life]

Haven't read the whole thread so sorry if duplicated but to protect your other animals I would be putting a muzzle on her like you need to with greyhounds to protect the other animals till you can find a solution.

I'm not sure if that had been mentioned, but it's a great suggestion.

I wouldn't leave it on though, only during training as a tool, when needed

 

[dainerra]

I'm not sure if that had been mentioned, but it's a great suggestion.

I wouldn't leave it on though, only during training as a tool, when needed

it will help a bit. but a muzzled dog can still easily kill a cat.

 

[Chickerdoodle13]

Treats should almost never be used for training. Not even in the very first training stages.If you give in to your dog and give him treats because he will "not work for verbal praise',all you are teaching your dog is that you can be manipulated. Training requires consistency.

This is the same point we got to the last time and it frustrates me to hear this sort of thinking because it is really quite erroneous. Dogs don't equate treats with "manipulation". Working dogs are still rewarded. Police dogs, military dogs, bomb dogs, search dogs are all rewarded even at the highest level. I can tell you these dogs would not perform as enthusiastically if they were not rewarded. Their ENTIRE job in life is all about the reward.

I've tried training my dog with praise only. It's not that she wouldn't do what I asked her to do but she did them with far less enthusiasm and lost interest in training sessions much quicker. When I'm training my dog, I want them to be happy and excited to work. I don't ALWAYS have to have treats with me to get her to do what I want. Many things are so ingrained in her head that she does them automatically and all she needs now is a "good girl" (things like sitting at curbs, sitting and waiting before running out the door, looking at me instead of barking at strangers approaching) You are not locked into treats forever if you use them, but I can guarantee that treat motivated dogs will learn more enthusiastically and be much happier if this is used as part of a training program.

I know I am much less enthusiastic about a job if there is no reward in it for me!