Posted elsewhere but appropriate to this... I will post some more on Paul Hutton, from personal experience, in a while
There are several reasons why i do not like treat training, some of them already outlined. The flippant, but accurate comment "Try throwing a biscuit at your dog when it's chasing a cat" kind of sums most of it up for me. However, there are some things to consider without becoming too much of a smart arse. (Not always easy for me).
The point you make about dropping a treat is valid ****** however the consequence of this handling cock up is greater than in most other techniques because the dog is focused on the treat, not the handler. In other techniques the dog would still not lose focus to the handler. You will see in this video that it happens to this handler as well and when it does she has completely "lost" the dog until all the treats have been recovered and she can again get some focus from the dog.
1.53 it becomes a joke! Whenever that treat is not in the dogs face this dog couldn't give a flying fox terrier about this handler and its patently obvious.
Here's the thing. There is one situation where treat training can work for the dog handler and that is for situations where the dog will be used competitively in competition and the more "stylish" the heel is perceived the greater the advantage to win the dog and handler have. I maintain the same style of heel is very achievable without a treat, the problem is the treat way is the easiest, though weakest from a control point of view, the method that gets the same result but has true control with it is unknown to most or in the "too hard" basket!. Now, no heel up has ever won a field trial, however if you have two dogs which are hard to separate on performance but one has a better heel than the other then it might give you an advantage, especially if the judge is impressed by such superficial things. The reality is that the heel that may appear not as polished is very often more ingrained and likely to continue in real world situations where as the treat heel is more correctly just a trick the dog will do in some specific situations but not when there is high distraction! Agility dogs, competitive obedience are all ingrained in treat training, are judged by people who treat train, that is their world. That's why those dogs get titles and ribbons, treat trainers rewarding other treat trainers with treats for doing it their way. Take those dogs out of the "Situation" (Because that training is "situation pattern obedience".) and the handlers have little or no control. Watch them with their dogs lose on the beach with beach balls, kids, other dogs and see they have CONTROL whatsoever. When I was working with Paul Hutton, and I'll elaborate a lot on that later, in his early days he had a huge client base with people who had "titled obedience champions" who were coming to him to try and get some control of their dogs when not in the ring. What does that tell you about the value of that treat? In my world if there was an obedience competition, at a certain point of the heel drill someone would let a sackful of cats loose so we could actually determine who in that ring has any control whatsoever! Why? because there are two stages to having control.
The first stage, which treat training can work in is teaching the command. Dogs learn anything well when there is something in it for them. It takes seconds to teach a dog to sit in order to get a treat. The same methodology is used on Dolphins to get the to perform tricks, it doesn;t give anyone control of the dolphin... take the dolphin out of the tank, put it into it's real world and you can wave a pilchard all you like. It won;t give a flying fish!
You can get a four year old child to make a Rottweiler sit if they use a treat, no one would be naive enough to believe the child just gained control of the dog. They can, however, get it to perform a "trick" using treats!
So using treats you have taught the dog the meaning of a command, it doesn't mean it will do it reliably when under high distraction. You have taught a meaning, you have not gained control or reliable compliance under high distraction. If a distraction represents, to the dog, a higher value than the treat your request to obey a command will be ignored unless your training regime has instilled in the dog respect for you and what you request... more than just a treat, otherwise we will constantly be balancing the environment to the value of the treat!. Let's face it, any dog that finds a treat more of a distraction then game when out hunting is probably not a dog most of us would want to hunt with.
We have high prey drive dogs, genetically predisposed to focus on game... that's what we spend big money on and hope to hell the breeder is producing.
The only factor that gives the handler compliance to commands is if he or she has the dogs respect. That cannot be established with a treat! Using treats a dog will perform the taught tricks any time there is nothing better on offer and may lull the handler or even observers into believing the dog is under control... it most likely isn't. It has a set of behaviors it has been bribed into performing in specific circumstances and in the absence anything better is prepared to perform those tricks and look like a wonderfully trained dog. Bring a fleeing cat, bolting rabbit, flushing pheasant, roaring stag into the mix and see what the dog values!
A dog trained to respect it's handler, that the handler should be their focus (Not what is in their hand) will quickly value the handlers praise and approval more highly than the treat trained dog appreciates the treat because it is balanced with respect for consequences to non compliance. In saying that, consequences do not have to be harsh physical treatment. Smart modern dog handling doesn't require the rough treatment of the past.
In this video the trainer is NOT teaching heel but simply getting mannered walking on a lead. None of it is treat trained, the focus of the dog is on the handler because that is the most comfortable thing for dog. To not comply brings a correction, when complying there is no pressure. The dog gains respect for the handler and finds compliance to the handler it's most comfortable place to be. Compliance doesn't run out when the treats run out like we see in the first video. It only takes some subtle tuning of this methodology to teach a reliable, controlled heel. This continues off the lead with a few modifications.
To be fair, both videos are only a tiny glimpse into the beginnings of training and there's no point pulling either of them apart to debunk too much... but it gives an opportunity for me to point out my views on both methods.
The change of direction methodology, when done properly and taken through it full course is the same technique used to over come aggression, cat chasing, over excited interest in other dogs and more. The treat method cannot be used effectively for those issues because those issues will always, to the dog, represent a greater reward than a piece of cheese in the handlers hand.
The second method is the only one that has any elements which bring respect, compliance and actual control to the heel and when done skillfully (All dog training needs to be learned and the ability to control a high prey drive is a skill to be practiced. It doesn't just happen.) The treats will give you an appearance of it but the wheels will virtually always fall off under very high distraction in the real world and away from situation pattern obedience situations.
Having a dogs respect is the first essential to having the dog hunt with you and for you as though the pair of you are a pack working together. It cannot be done without it. You cannot teach respect with treats.
Some would maintain that by teaching the meaning of commands with treats, while simultaneously teaching consequence and compliance speeds the process up. It doesn't because in the treat process respect is breaking down. This means you may get compliance to the trick taught in the situation specific environment (Competition obedience ring, pigeon trial specific format) but that is unlikely to carry forward to having proper control in real world situations.
As ******* points out above you can't steady a spaniel to rabbits with the promise of a treat. *******, correctly pointed out that the work would be done with treats prior to steadying. This reinforces my contention. Treats can be used to teach the meaning of a command they cannot be used to gain control. Even if you taught heel, sit, down, stay, come etc all with treats when the chips are down in the hunting field the only way you are going to get a stop on game flushed off the dogs nose is if the dog has respect for the handler. To be fair in the past this has been gained more by ingraining fear of consequence in dogs than respect for consequence but that cannot be used as an argument against it, merely an illustration of poor handling for respect resulting, instead, in fear of the handler. But the high distraction situations such as flushing game off the nose and remaining steady cannot be done with treats. The game will always be of more interest than the treat. However, respect for the handler and the resulting successful hunt and retrieve (Pack success) will have more meaning to every dog than the promise of a treat or the threat of a hiding from its handler.
Are there any situations when using food is training is OK?
Yes, but not for teaching commands or trying to gain control. I have used food with some dogs to overcome phobias. I recently worked with a Labrador that had gained a severe fright whilst on a lead to the point that whenever the lead was put on the dog it would lie flat to the ground and refuse to move. By using food as a bribe the dog would move forward to get the treat breaking out of its fear mode for a second. If it retreated back into it's "safe place" of lying doggo another treat was thrown. Given the Labradors penchant for tucker it quickly forgot it's fear of the lead to find the next treat. It took about two sausage rolls, broken up, to have the dog completely forget it used to be scared of the lead. Having broken out of that mind set we could set about teaching it to walk politely on a lead without treats and set about some control work. In this situation the food actually became the distraction from the behavior... quite ironic.
While a treat trained heel looks flashy in competition with the dog, head held high, fixated on the handlers hand or face anticipating the next treat , it's not where I want any dog's attention in the field. I want the dog to maintain constant focus on me while simultaneously be taking in it's surrounds for game or just what's going on in general. Yes i get that in a retriever trial many of the behaviors can be taught with treats, but face it, In heavy cover in the Kaweka's with a Jap Stag roaring its nuts off 20 meters away you better have a better control methodology for a GWP, Visla, GSP, Lab, whatever... than a pocket full of Purina!
teaching the treat heel is of little use to the deerstalker. The dog at heel should be winding, not looking to see if a treat is coming.
At it's guts, you can teach some pretty looking tricks with treats in my opinion. It will never get you anywhere close to control. If that's OK with you then there's nothing wrong with it. It's not what I want from my hounds.
Bookmarks