Training treats considered as detrimental

[mum24dog]

I disagree with this completely. Praise can be placed anywhere you want it if you understand how it works; most people don't BTW. I wonder if the dog doesn't pull out of the poles more often because he smells the bag of treats running in the other direction.

 

 

It works both ways. Those who insist on not using treats often don't understand how to use them properly, or how to fade them out.

 

People who use treats or toys don't find it difficult to accept that they may come across dogs that don't need them. Unfortunately those who have so far only had dogs for which they have not had to use them seem to dismiss the idea that one day they will meet the dog that will not fit into their previous mould. Maybe if they only work with dogs bred to be biddable they will continue to get away with it.

 

You say that you teaching agility. To what level? Once a dog has got to the stage of not having to be nursed through the weaves it won't follow a treat bag. If it is ever focussed on the treats more than the job in hand the handler is doing it wrong and it's the instructor's responsibility to help them get it right. Job first, reward follows.

 

But treats are not the first choice for most agility trainers, especially for training weaves once the initial introduction has been made. Toys for enthusiasm and forward motion, treats for those dogs that aren't toy orientated, for a really driven dog the chance to continue over more obstacles. Sometimes praise alone will suffice.

 

Rewards tailored for the individual dog and task, not for the handlers preconception of what the dog ought to be satisfied with. I hate it when some instructors assume that all dogs are interested in toys (or food for that matter) - they aren't.

[mum24dog]

What I find interesting is how many "treat-trainers" are walking around with treat pouches on their belts and their 9 year old dog in tow. Usually their dog is looking for treats from anyone they happen to pass, and the dog ignores them unless the squeak "cookie".

 

 

The people you meet need serious help then.

 

I do not allow other people to give treats to my dogs therefore they do not mug anyone else for treats. Simple as that. I don't give treats to other people's dogs without permission either. There may be occasions when they want me to for a dog or pup in training but I never do "just because".

 

I always carry treats in case there is something I want to reward in a particular way. I come home with most of what I take out with me. I have a 13 year old obsessive rabbiter with little interest in people that is going a bit deaf and blind. Her recall used to be OK but now she can't hear me so well so we have gone back to what we used to do in early training with lots of rewards to keep her within range, that way she still gets freedom to run. Having had an old deaf dog before, I started the retraining before it was needed and it's worked very well.

 

If you live in an area of constant high distraction, as I do, you need as many options as you think of.

[Shetlander]

 

 

I also question the "it works" rationale. My training philosophy is based on an honest realationship with my dog. You can pay your child every time you want them to do something, and they probably will, but does that really work in the context of life? I see treating in the same light. It maybe "works" to solve an immediate issue, but does it create an honest and well rounded dog that can be trusted in any situation? Not in my experience.

 

Not in your experience but it works for lots of us very well. Year's ago, I attended a wonderful obedience seminar by Connie Cleveland who uses a wide variety of techniques to train dogs. While I would not use some of her methods myself, I was extremely impressed with her. At the start of the seminar, she commented that dogs obey for three reasons:

-They want to get a reward

-They want to avoid a punishment

-They do so out of habit

 

Of these three reasons, Connie believed that habit was the most important in the long run. I totally agree. So if we train our dogs to obey out of long standing habit (that immediate response we all love to see), and we don't abuse our dogs in the process, isn't that a good thing? I have seen both fabulous and abysmal trainers across the spectrum of approaches. We should all do what works for our dogs and us.

[JohnLloydJones]

At the start of the seminar, she commented that dogs obey for three reasons:

-They want to get a reward

-They want to avoid a punishment

-They do so out of habit

 

Er, no, She missed the most important reason. Dogs are descended from highly social animals (wolves) and they have inherited a trait that is key to the survival of groups. They want have a deep seated need to be useful. Dogs inherently want to contribute to their pack (group, family, whatever). They respond to our commands, because they want to. Training simply reinforces their choices that we like.

[Sekah]

Er, no, She missed the most important reason. Dogs are descended from highly social animals (wolves) and they have inherited a trait that is key to the survival of groups. They want have a deep seated need to be useful. Dogs inherently want to contribute to their pack (group, family, whatever). They respond to our commands, because they want to. Training simply reinforces their choices that we like.

 

"Feeling useful" = reward.

[JohnLloydJones]

"Feeling useful" = reward.

OK, but fundamentally different from a proffered treat. Intrinsic, rather than extrinsic.

[mum24dog]

Er, no, She missed the most important reason. Dogs are descended from highly social animals (wolves) and they have inherited a trait that is key to the survival of groups. They want have a deep seated need to be useful. Dogs inherently want to contribute to their pack (group, family, whatever). They respond to our commands, because they want to. Training simply reinforces their choices that we like.

 

 

You've been around BCs too long. (My bold.) You might think differently if you'd met Hazel or Charlie. Rabbits far more important to both of them than people or other dogs.

 

Dogs do what works for them. Whether human control figures highly for them depends on the dog. Whether interacting with other dogs figures highly depends on the dog.

 

Dogs are not wolves. Dogs do not have the pack structure of wolves.

 

If they want to respond to our commands it's because it is worthwhile to the dog to do so and what is worthwhile will also depend on the dog. I'm sure that survival comes into it, but the individual survival of that dog.

[Maralynn]

 

I understand that most people who are treaters can't understand how to operate without them, but it's certainly possible, and, imo, far more efficient and effective.

I know of many people who used to train without treats but now use them since they found that they liked the result much better when using treats. Treats can add intrinsic value to the work. Value that stays after the treat has been faded out.

[alligande]

I feel the need to respond to this, as an amateur trainer who uses treats to train basic things like sit, as well as silly tricks and skills needed for agility, but does not walk around with pockets of treats (I dont even own treat pouch) and only uses them to train new things, to reinforce those behaviours I do use the premack principle extensively. Walk next to me nicely on an off leash walk, the dog gets to go back to do his own thing, give me down at a distance, dog is released to carry on his own interests, I practise all sorts of skills on walks with not a treat in sight, yet the basic behaviours where taught using them.

 

Recently I finally got around to putting a command on stretch, took me a couple of evenings to reliably get the stretch, I was using a verbal marker and treats as well as praise, using treats does not prevent the use of praise, within a week stretch has been added to the list of commands and a treat is not needed.

 

It would be a pretty useless trainer who was still using treats, even months after first training something let alone years latter.

 

I am not in any way militantly a positive trainer, I strongly feel if a dog can understand a marker for correct it can understand a marker for unwanted behavior. If you put your paws on my counters there are consequences, show any interest in food on the coffee table and we will use a verbal negative (depends on the human, i think it works on tone not word) but when learning a new behavior there are only rewards and of course praise, as the lack of understanding is more likely to be my ineptitude as a trainer, rather than the dog being slow.

 

What I find interesting is how many "treat-trainers" are walking around with treat pouches on their belts and their 9 year old dog in tow. Usually their dog is looking for treats from anyone they happen to pass, and the dog ignores them unless the squeak "cookie".

 

The most common rationale I hear for the use of treats is that it's easier to mold behavior by using them. What people seem to disregard is how long it takes to decondition the dog from needing a treat to perform the behavior reliably. No matter how much you "randomize the treat reward, if you create a behavior using treats, your dog will always expect a treat for performing that behavior. That creates two negatives, imo, 1) a dependence on always having food, and 2) a WTF moment for the dog when he doesn't get the reward. I don't understand the "random" reward model in general. I want my dog to know every time that its done the right thing.

 

I also question the "it works" rationale. My training philosophy is based on an honest realationship with my dog. You can pay your child every time you want them to do something, and they probably will, but does that really work in the context of life? I see treating in the same light. It maybe "works" to solve an immediate issue, but does it create an honest and well rounded dog that can be trusted in any situation? Not in my experience.

[Root Beer]

Like alligande, I also don't walk around with a treat pouch. The idea that handlers of treat trained dogs must have treats on them at all times is going to strike those of us who successfully train with food/toys/etc. (meaning - learning happens and in the end the dog knows what is expected and generally complies) as ludicrous.

 

For example, I regularly hike in the woods with my treat trained dogs off leash, and I don't take treats into the woods with us. They stay close and come when they are called. They are trustworthy and reliable - and yes, there is wildlife out there, and they are very much aware of that fact. I have never once resorted to vainly begging, "cookie, cookie, cookie" as the false characterization of food based training often portrays us.

 

I would say that 95% of the time that I am with and interacting with my dogs, I do not have treats on me. And they do all of the things that I have trained them to do. That in spite of the fact that almost all of my training is food based in some regard. (At the introductory stage, to build precision and to aid in comprehension of complex tasks, to facilitate generalization and fluency, etc)

 

I would categorize those who have a treat pouch at all times, everywhere they go, right along with those who always have a prong or choke chain on their dogs, or those who would never dream of allowing a dog off leash in public without a shock collar on for "just in case" and consider that "trained". This, of course, presumes a dog of normal temperament without any true issues that ultimately hamper learning no matter how training is approached.

 

Incomplete training is incomplete training, no matter how the handler personally chooses to train. Food is not the problem in the case of the handler who must have a treat pouch in order to get the dog to do anything ever - the problem is that the training is nowhere near complete.

[JohnLloydJones]

Dogs are not wolves. Dogs do not have the pack structure of wolves.

Dogs are certainly not wolves, and, unsurprisingly, they social interactions are subtly different too.

You've been around BCs too long.

All my life.

 

I do have interaction with other breeds, and I agree that the border collie is special (which is, of course, why I work with them). The desire to live in a social group* and contribute to that group's commonwealth is deep and exhibits itself in all breeds I have met; albeit in varying degrees.

[terrecar]

I am one of those who trained without treats (back in the old days of obedience, it was frowned upon), but now use some operant conditioning with treats. I understood operant conditioning purely from a classroom perspective, but that wasn't my first introduction to it. I knew a psychologist who headed the training of military dogs at APG in MD in the 70's, and he used classical and operant conditioning [ETA: the former in research]. He gave me a tour of the facilities and a demonstration of his techniques, and at another time at someone's home, he demonstrated with my own dog how to teach sit. As a military person, he really wasn't the sweet talking kind. So, the caricature that I see here is just laughable to me. I am pretty sure his dogs performed once the treat was phased out. There were lives at stake. I also don't think his dogs felt it beneath their dignity or any other such anthropomorphic nonsense.

 

As much as people rail against Skinner, his contributions to learning theory were not without merit. They were, however, limited. I also see a disconnect between someone who wrote Walden II (utopia via controlled environment, basically), yet recorded the "aha!" concept--the latter which pretty much translates as "I have no idea how that rat did that thing but I sure can't explain it with my theories".

 

Of course it doesn't make sense to use those techniques with sheepdogs. One of the reasons I am in such awe of the dance (now I'm anthropomorphizing) between the shepherd, dog and sheep is the uniqueness of the communication between them. It really is much more complex (and moving) than anything you might find in Skinner's box.

[GentleLake]

Er, no, She missed the most important reason. Dogs are descended from highly social animals (wolves) and they have inherited a trait that is key to the survival of groups. They want have a deep seated need to be useful. Dogs inherently want to contribute to their pack (group, family, whatever). They respond to our commands, because they want to. Training simply reinforces their choices that we like.

 

HA! I don't think you'd be saying that if you'd had a couple of the dogs I've had (and I'm sure many others; I'm not suggesting that my dogs were exceptional or worse than others) in the past.

 

I've had 3 dogs now over a period of time who had little to no interest in responding to my cues, or even commands, because they wanted to. They wanted to do what they wanted to do, and if I wanted them to do something other than that I had to give them a damn good reason. I could have used fear of punishment as the motivator, or I could use more positive motivators. In their cases, praise alone wasn't a sufficient motivator, so I've used food, which was.

 

I've heard heard the there-are-no-bad-dogs and dogs-all-want-to-please arguments before, and it just ain't true for all dogs.

 

With all due respect (and I don't mean that sarcastically), John, I have to wonder how much experience you have working with spitz-type breeds, which are very independently minded and don't seem to have (in my admittedly limited experience) much, if any, desire to please their handlers. And AFAIK, they're genetically closer to wolves of any other group of related breeds.

 

And have you tried to get a reliable recall on a lurcher or sighthound anywhere other than your living room or small fenced back yard, especially without food?

 

I agree that there are other great ways to train dogs, and some no-so-great ways, as well. But not all methods work equally well with all dogs, even with all dogs within one breed. I prefer being knowledgeable on various methods and being able to use what works best in a given situation and/or with a particular dog. I do use food, but I phase it out and practice very hard with my dogs so that they don't become reliant on it.

[JohnLloydJones]

With all due respect (and I don't mean that sarcastically), John, I have to wonder how much experience you have working with spitz-type breeds, which are very independently minded and don't seem to have (in my admittedly limited experience) much, if any, desire to please their handlers. And AFAIK, they're genetically closer to wolves of any other group of related breeds.

I started out with a disclaimer that my experience is mostly with Border Collies. Note that I also mentioned in the post you quoted that the "desire to live in a social group and contribute to that group's commonwealth" exhibits itself to varying degrees (and that it doesn't always imply doing what *you* want). A Buhund (a spitz-type), for example, has a very strong need to alert when strangers approach that makes it an excellent choice for that purpose, but might not be a very good choice to train in agility (I've never seen ot heard of one, at least). Training recall, however is not that much of a deal.

[rushdoggie]

The people you meet need serious help then.

 

Agreed. Doghaven, I found your entire post to be head-scratchingly incorrect. If the only people you know who use treats by carrying them all the time and handing them out constantly, well I guess I can understand how you might feel that way.

 

With 20 years of training dogs: mine (and I have had quite a few of various breeds including Siberian Huskies, Dals and Papillons as well as BCs), foster dogs (also a bunch including all the breeds mentioned) who came to me with limited to zero training and in some cases serious behavior issues) and student dogs (I have taught pet classes, intermediate for competition bound dogs and agility) I have seen literally hundreds of dogs successfully transition from getting a treat for a behavior to getting occasional reinforcement (which includes treats but also praise, toys, people interaction and release for life reward).

[Laurelin]

I disagree with this completely. Praise can be placed anywhere you want it if you understand how it works; most people don't BTW. I wonder if the dog doesn't pull out of the poles more often because he smells the bag of treats running in the other direction.

 

When I train weaves the reward is always moving in the direction of the weave poles in the form of something thrown (toy or food toy/container).

 

Do you work agility to a competition level with all types of breeds? I am honestly not trying to be snarky but I just don't understand how you could go about it. I'd like to see the thing in action because I hear people say you can do it with praise as a reward and yet I never see these people out competing.

[Shetlander]

spitz-type breeds, which are very independently minded and don't seem to have (in my admittedly limited experience) much, if any, desire to please their handlers. And AFAIK, they're genetically closer to wolves of any other group of related breeds.

 

Lhasa Apsos, despite their absurd teddy bear appearance, have also been identified as one of the “primitive breeds,” most closely related to wolves. Having lived with two of the critters, I admit a piece of the puzzle fell into place when I first read that. Does my fuzzy little guy want to please me? Sure! He loves to please me! He just doesn’t always know how without some extra guidance and motivation. And he is such a tough minded fellow, that most correction (including corrections from other dogs) is virtually meaningless to him. He doesn’t like it but it doesn’t seem to compute. He either immediately shakes it off or is upset but no wiser. Both my Lhasas were intelligent dogs, though it is a different type of intelligence than I find in my herding breed dogs. Sort of monkey like, but without tools.

 

What both of them have clearly lacked compared to my Shelties and Border Collie is sensitivity to my moods and feelings, so they were at times utterly clueless of what would please me and what would send me over the edge. For instance, each Lhasa happily jumped onto my chest after I fell to the ground (the first time writhing in pain, the second time just lying there, surprised I hadn’t hurt myself). They both acted as if I was playing. On the second occasion, my Border Collie and Sheltie appeared rather mortified to see me go crashing down on a icy patch and didn’t for a second think I was goofing around. My joke about Chili’s insensitivity to the social cues of people and other animals (not to mention his relentlessness) is that he can’t help himself. Like Lady Gaga, he was born that way.

[Maralynn]

I posted a couple of shaping vidoes back on the first page. My dog quickly went from getting rewarded for every step in the right dirrection that she made to generalizing that I wanted her to pick up ojects and put them in a container (like small objects such as a pen and credit card in a small box instead a large crate and socks). I quickly went from using 30 treats in a session to using 4. Now she'll pick up and deposit multiple objects between rewards. This was done over the course of 3 or so weeks with only intermittent training sessions. Still trying to figure out how those treats slowed down the learning process or caused her to be dependent on them. I mean, if she was dependent on treats then why could I phase them out so quickly? Why is she excited when I bring out the container and objects to put in it even when I don't have treats on me?

[Shetlander]

Still trying to figure out how those treats slowed down the learning process or caused her to be dependent on them. I mean, if she was dependent on treats then why could I phase them out so quickly? Why is she excited when I bring out the container and objects to put in it even when I don't have treats on me?

 

Quinn is very helpful around the house (putting toys away, bringing me his empty meal container to wash, putting clothes down the laundry drop) and at the office (picking up a fallen pen or set of keys, taking empty pop cans to the recycle bin). He was also trained with treats and I sometimes still toss him a tiny cookie for completing such little tasks. But mainly, he just loves to be involved and a simple "Good boy." "Yay!" or even an absent-minded "Thank you" puts an extra bounce in his step.

[mum24dog]

Like alligande, I also don't walk around with a treat pouch.

 

 

I don't have a treat pouch, just a pocket of kibble, whether I have a dog with me or not. It has come in useful on occasion.

 

I don't carry food because I have to to get my dogs to perform behaviours they already know; I do it because I never know when I might have a mind to train something new or reward something unexpected, never know when I might spot something I can capture and shape. I rarely have defined training sessions nowadays.

 

It's just being prepared for any eventuality, not a crutch.

[mum24dog]

I am one of those who trained without treats (back in the old days of obedience, it was frowned upon

 

 

You surprise me. I trained my first dog in a competitive obedience environment 50 + years ago and we were encouraged to use treats.

 

When I got back into dog owning around 15 years ago after raising my children it was rather like a time warp. The class I went to still used choke chains and jerking (maybe not quite so much though), which were the norm first time around, and still used treats.

 

I didn't stay long and went to a class that had progressed over the decades instead.

[dogrsqr]

My own dogs are working sheepdogs. I train other people's dogs in just about any context there is; herding, agility, behavior issues, etc. The only time I have ever used treats is to get feral dogs close enough where I can get a loop on them or to lure them into a crate.

 

I understand that most people who are treaters can't understand how to operate without them, but it's certainly possible, and, imo, far more efficient and effective.

Fear is a great motivator. Here in Minnesota I see many hunting dogs that are sent off to a "trainer" where they promptly slap a shock collar on the dog. It works great to get the desired behavior, but it also creates unthinking dogs. Who wants to try something new when you don't know when or for what you might get zapped.

 

I compete in agility and flyball as well as herding. I've used rewards, treats and toys, in training agility and flyball. My dog does find doing these activities themselves to be rewarding so no need to ever have treats at any of these competitions. Rewards are used for training. Once a dog knows something you go to a VSR, expanding time and distance and eventually no reward is necessary.

 

I do think people get confused because with clicker training if you click you must reward. So they tend to think that this is what you do forever. Not so ... only when you're training a new behavior.

 

I believe you can train without rewards, especially if the activity is self-rewarding, but my experiences have taught me that training with rewards is faster, longer lasting and a much more pleasurable experience for all. You can train a dog (and I'm talking about those breeds that don't find it to be much fun) to retrieve using an ear pinch or a clicker and rewards. Guess which one the dog prefers.

 

What I don't understand is the people (usually men) who say the dog should do "fill in the blank" because I told them to! Really ...... is this what you expect from your wife and kids too? If so I'm guessing you're disappointed a lot.

[mum24dog]

A Buhund (a spitz-type), for example, has a very strong need to alert when strangers approach that makes it an excellent choice for that purpose, but might not be a very good choice to train in agility (I've never seen ot heard of one, at least).

 

 

We've had a Buhund in our agility classes. Didn't stay long but the owner didn't know how to motivate the dog so she could have had unreasonable expectations. I wouldn't generalise about the breed's ability to learn - I've known dimmer dogs. They aren't really built for flexibility and speed though.

[Root Beer]

You surprise me. I trained my first dog in a competitive obedience environment 50 + years ago and we were encouraged to use treats.

 

The acceptance of the use of food (and even +R in general) in competitive Obedience training in this country is still very much a work in progress. How much it is accepted, frowned upon, or even flat out condemned varies from region to region, training group to training group, and even handler to handler.

 

Progress in this regard is happening, but there is still quite a lot of backlash to be dealt with if one chooses to brave those waters. And let me be clear - I do not say this to insult the sport of competitive Obedience. It is simply an honest acknowledgment of the culture of that particular sport in this country.

[mum24dog]

The acceptance of the use of food (and even +R in general) in competitive Obedience training in this country is still very much a work in progress. How much it is accepted, frowned upon, or even flat out condemned varies from region to region, training group to training group, and even handler to handler.

 

Progress in this regard is happening, but there is still quite a lot of backlash to be dealt with if one chooses to brave those waters. And let me be clear - I do not say this to insult the sport of competitive Obedience. It is simply an honest acknowledgment of the culture of that particular sport in this country.

 

 

I'm always amazed that the US has produced so many innovative and compassionate thinkers in the field of training and yet the other extreme is so widely accepted. Maybe that's the reason - the bad side acted as a catalyst to come up with something better and more humane.

 

I have seen welcome progress at Obedience shows here over the last few years but even those who compete seriously will admit that their fellow competitors are resistant to change. The sport is seriously declining in popularity for that reason.