Working_Setters

Sounds interesting, which book? It depends on how I think about it - in my limited understand of clicker training, I always used one click, one reward, and kept them close together for fear of unconditioning the click. ie if you've clicked a couple of times, but no treat, I'd have thought the dog is going to stop believing that a click = treat. OTOH there are times when I'll say "good dog" as a dog proceeds through a complex exercise, I might use "good dog" to mark a number of success points in the one exercise, with the dog needing to complete the exercise to get further reward. In this case there are a number of CRs and only one PR at the end of the exercise.

I think the point is that if you pick the right litter, it doesn't much matter which pup you get, so long as the pup is somewhere in the "normal range", it's what you do with the pup that matters. I was told the exact same thing by the most successful breeder of working setters in the south pacific, and they were being serious.

I agree, it's just that even the most experienced people don't agree on what those "select few" will look like as a pup. Obviously many, many, people are in the same position as you, they want to choose the "pick of the litter" for whatever task they have in mind, there have been all sorts of test devised, clapping you hands to see which ones have steady nerves, seeing which one shows the most interest in a quail wing, walking away from the litter and seeing which one will follow etc. There just doesn't seem to be many really experienced people that think this testing makes much/any difference. Here we are in HIGH agreement. Winning trials isn't easy, you have to beat every other dog there on the day, and there are some bloody good dogs/handlers out there, don't make it any harder for yourself by getting a dog that doesn’t have the instinct and desire to do the work. There was a thread on this topic recently on Australian Working Retriever Central, HERE, basically the consensus was to get a lab pup from the right litter. If I wanted to win RTs I'd follow that advice. I've seen some working FCs in OZ, if I were after a working FC, I guess I'd get one from these folk.

I'm not experienced enough to say, IMO you'd need to have picked dozens of pups and trained them through to adulthood to be able to draw any statistically significant conclusions. In the opinion of most gundog trainers that have that sort of experience, pick your breed, your breeder and your litter very carefully. Once that's done, close your eyes and grab a pup, else choose the one you whose colouration you like. Pups change sooooo much b/t 6-8 weeks and adulthood, the active one today is the sleepy one tomorrow. If you've choosen the right litter, then the genetics for the work will be there, after that it's a matter of environment and training.

Awhile back in Oz there was a second working gundog club formed that in many ways is in opposition to ANKC (they offer trials etc for working gundogs). My understanding, although it didn't happen to me personally, is that the ANKC (or prehaps it was only the VCA) threatened members/judges joining rival organisation would be banned from ANKC/VCA events. It went to court, ANKC/VCA lost. Now there are many people running their dogs in both organisations. Which is why I figured, they might not like it, but unless its illegal they can't stop you.

On the subject of conditioned reinforcers (CRs), how long is too long, b/t the CR and the Primary? Also it was suggested to me the other day that it's OK to use multiple CRs (in this case I was talking to a clicker trainer, so it would be click, click, click) then Primary much latter, say up to 5 min later. It was suggested that the dog was "counting up" the CRs and so long as the Primary (food) was appropriate to the number of CRs (mores CRs = more/better food), then this was OK. Any thoughts? Anybody train this way? I'm told that there are remote clicker collars, you push a button on a electronic box and the dog's collar makes a clicking noise, anybody seen/used them? I guess to be successful as a training tool, the dog must learn that CR now means Primary latter - which I' OK with, but if you're giving multiple CRs and not giving the primary until 5 min latter, I'd have thought there was a danger the CR would become unconditioned (ie multiple clicks now, reward 5 min later, there is a danger that in the dog's mind click nolonger = reward). Anybody got any experience with this one way or the other?

Hi Jeff - Am I correct in thinking that although the ANKC don't support it, they are not able to prevent their members from doing it? Nor are they able to discriminate in any way, against those who choose to join these (or any other) clubs.

If a dog has learnt a behaviour (stop to whistle) and has learnt that when they obey there will be praise and when they don't there will be punishment, then I think the dog anticipates prasie everytime it obeys and anticipates punishment everytime it doesn't obey. With regard to the timing, that's fine if the dog is on lead or on e-collar, but if dog is 50 meters away and no e-collar, timing is going to be a problem. That's why gundog trainers take their dog back to "the scene of the crime", repeat and enforce the command (generally a sit whistle the dog blew off), to help the dog pair the behaviour and the punishment. This can all be done much more efficiently these days with an e-collar, but when you're not using one, then during the time I'm covering that 50 meters my dogs know if praise or punishment is coming. I don't agree with the extent of this generalisation. I think most/all working dogs need an "off switch", a command that is completely handler dependent. For my dogs it's the stop whistle, I don't care what else is going on, I don't want my dog to think about it, I just want their butt on the ground when they hear that whistle. But in many other aspects my dogs are independent workers, they have to be as there is nothing a human can teach a dog about scent and scenting. Same with the police dog video (Rudy, I think) that was posted here recently. Dog was sent on perp, 2/3 of the way there the dog is recalled by handler, dog's response needs to be reflexive, it is up to the handler, not the dog, as to who gets bitten. Same with herding, sniffer and guide dogs, all needs to have a command the over-rides everything else, to allow the handler to stop/"turn off" the dog.

M-J, I agree with much of your post. However, I would say for my dogs it's not a threat of an adversive, it's a promise. My aim is to have the dogs understand that EVERY time they obey the stop whistle they will be rewarded and EVERY time they don't they will be punished. IMO this consistancy is the key to achieving a consistantly obediant dog.

Collars come off in field and retrieving trials. (In field trials the two dogs currently running do wear a red or a white identifier around their neck, but it's not a collar)

Good call - I need to remember the diverse nature and experience of people reading this board. I'd certainly hate to think my posts contributed to somebody stupidly yanking and jerking a poor dog's ear. As you correctly point out they can be damaged by such stupidity.

DD - I think it comes down to the difficulty of conveying intensity, or the amount of force, using the written word. I suspect to most people "tug" means to pull, but I believe there would be considerable variation in the force people would associated with that action. To me, young babies frequently grasp and tug at your finger for example. Apply "force" of that level to my dogs, and I doubt they'd notice it, certainly it wouldn't cause them any discomfort. ETA - Here is my thoughts from a previous post on negatives and their use in recall training and I stand by them. Clearly some people have a different opinion and they're certainly entitled to it. However IMO (and as has been borne out by this thread) the majority of people that have achieved a truly reliable recall, as demonstrated by "the rabbit test", have used some adversive to achieve it (WMR and her whippet being a notable exception). As a result of their high level stop/recall training, my dogs get to enjoy a great deal of free running exercise in a variety of highly stimulating environments, something that wouldn't be possible if they were not able to be stopped/recalled in the face of heavy distractions. IMO it's a case of "short term pain for long term gain". Yes we agree that the short term pain should be minimised, but I don't believe it can be dispensed with entirely, certainly not with the driven dogs I train and love.

deleted - cause it's more trouble than it's worth. I would say ...........nah I wont say that either.

But if your whippet catches the rabbit before he comes looking for you, then you're in a heap of trouble as the chase has been rewarded big time. I use the running away and hiding stratergy with pups, but my older dogs are so driven to find game, it could be sometime before they come looking for me.

But the thing is when you're a Whippet and you're chasing a rabbit that's two meters in front of you, then the best thing in the world is two meters in front of you, there's no treat/reward that I know of that's bigger/better to the whippet at that point in time. Therefore (and this is the problem I've always had with positive only training) there will come a time when you don't have a reward that the dog will value more highly than the rabbit. However I'm thinking now, from where's my rock's posts, that it's not a question of having the dog weigh up the two rewards and decide which it would prefer, rather it's a process of very carefully training recall (and every other command) in low distraction enviroments, then gradually adding distractions. The dogs response needs to be a reflex response for success in high distraction settings. Which is also the same the way I train (traditional methods), it's just that I have a bigger margin for error, b/c if the dog ignores stop/recall, I can correct (negative) and still teach the dog something, where as positive only don't have that option.

Calling a whippet off a rabbit is quite a feat, sighthounds are not noted for their responsiveness, particularly during a chase - I'm impressed :D During the first year, when you are working on building that focus, what type of exercises are you doing, are they food, praise, play based, or something else? Do you have a core of general exercises you do with all your dogs, or is it very much dog specific? In a way our training isn’t all that different, neither of us EVER lets the dog get away with not obeying a recall, it’s just that you have the patience to work on a long line for a year, where as I cut them loose earlier, but expect to have to run them down (e-collar correct) a couple of times, to enforce recall.

So the key to a positive trained reliable recall is to practice over and over until it's a reflex action? Then it's not a question of the reward out weighing the pleasure of not responding (continuing to chase the rabbit for example), since the dog is so well conditioned to the command that there won't be any thought of not obeying? If the whipet did ignore a recall, how would you correct it? Are there no corrections? Just ignore and go back to a lower level distraction environment and more practice until response is better ingrained?

YES - I agree that is a huge part of it, stop/recall can't always mean end of fun, dog must know that there is an excellent chance that recall will be a brief pat and then back to more fun, but not recalling means no more fun

deleted - cause it's more trouble than it's worth.

Hi Helen - If your dog was out free running, say 50 meters from you, and put up a rabbit, could you reliably stop/recall the dog mid-chase? What reward can you offer a Brittany (or other gundog) that is more desirable that chasing a rabbit?

I very definitely choose the breed, the breeder, the mating and the pup(s) with specific goals in mind. I'm talking about my current and recent dogs, however back when I got my first dogs, that was a different story, ad in the newspaper, $50, bingo, I had bought my first dog.

Interesting thread. As I guess is widely known after recent threads I train with both positive and negative. I've been to 3 different obedience schools/classes (1 in Oz, 2 in USA) and in every case my dogs have recalled better than the instructors. I had one class in the USA where the instructor couldn't let her dog off leash out of her yard as no recall I agree, I've never seen a reliable recall from a positive only trained dog. Now it could be me, or the positive only people I've met to date, but IMO clicker etc is great for pups and great for teaching/shaping all sorts of interesting behaviour, stuff that would be really hard to teach in other ways. However the down fall of these methods IMO is distractions, when the dogs would rather continue what they are doing than get the reward, what do you do? No enforcement means, (in my limited understanding of positive only training) that if a command is ignored and you can't find a more desirable reward, then you are SOL. Would be very interested in hearing from positive only trainers that have achieved reliable high distraction recalls. How did you do it? Do you think there are advantages over more “traditional” (use of some negatives) methods?

I believe the dogs stopped b/c they have been conditioned to stop. I have spent a lot of time training stop. I start in a distraction free environment and teach stop to whistle, then gradually increase distractions. I do it gradually, over a period of time, stop to whistle is the bedrock on which I build all my other training, so I want it solid, not rushed. It's nothing negative, dogs understand sitting to whistle will please me, they are happy and confident, knowing that stopping will earn my praise. My dogs are NEVER allowed to get away with not stopping, once I blow that whistle the dogs WILL stop, even if I have to run them down, drag them back to the spot where they ignored the whistle and physically push their butt to the ground. (E-collar can make that process much easier on both me and the dogs). I guess the whistle becomes "god-like" to the dogs, whenever they hear it they WILL sit, one way or the other. Usually, if training is correctly paced, dogs will only "try it on" (not stop to whistle) a couple of times, after that, they understand that stopping to whistle is unconditional. We practice it over and over, dogs are just out playing, running flat out, I hit the whistle - they hit the deck, I praise and release, they resume playing and running flat out. Everybody is happy. Over time (my current dogs are just achieving this) stop becomes a reflexive action, there is no thought of not stopping. Dogs hear the whistle, their training kicks in and their butt hits the deck, no matter what is going on around them. Dogs ALWAYS stopping instantly to whistle is of premium importance to me. However dogs that run HARD and hunt with great DESIRE are also of premium importance to me. I go to great pains to achieve both. I would be very interested in hearing your thoughts.

Hi Erny, The reason I insist that my dogs bring back the dumbell (training dummy) is that I'm preparing them to retrieve shot game in the field. It's unacceptable to me, as an ethical hunter, that dead/dying game be left in the field b/c the dog refused the retrieve. The dog must have the physical/mental ability and the training, to search for as long as it takes to recover game went ever possible. Whether or not I'd FF if I didn't hunt is an interesting question and I'd need to think it through. Off the top of my head I'm inclined to say I would, b/c to me a command, is a command, the dog doesn't have the ability to decide which commands are potentially life-saving and hence REALLY need to be obeyed, and which commands aren't and so only need to be obeyed if the dog feels like it. I guess I could throw the ball and not say anything, the dog could chase/play/retrieve it if it choose to or ignore the ball if it wanted to do something else. However if I said "FETCH", I'd need the dog to go get the ball, as IMO ignoring "FETCH" today will lead to ignoring "SIT" tomorrow, and "SIT" could well save the dog's life.

I hope these folk were in the process of engaging your services, as IMO they are clearly in need of assistance. Minimal experience is enough to teach that sort of physicality is going to kill the drive in all but the most hardened (wilful/hard-headed) dogs, and even with these dogs sustained physical abuse must eventually take it's toll. My setters are very responsive dogs and would be quickly destroyed by such a training approach. I think working gundogs are different to many other dogs. Any working gundog that I would consider worth feeding is going to be driven by a STRONG desire to find game (my love is Pointers and Setters). Many folks refer to the control I'm talking about as teaching the dog manners, and in a way that's correct. It's teaching the dog control in the face of MASSIVE distractions. So in answer to your question IMO, its a combination of teaching the dog self-control and ultimately of me controlling the dog. A good example of what I'm talking about happened on Friday, I let the dogs out for their morning run, they run around emptying out etc, when they smell a rabbit in a brush pile. They start working the brush pile enthusiastically, round and round, squeezing their way deeper and deeper into the pile. After about 10 mins the flush the rabbit from the depths of the pile, they see the rabbit flush and begin to give pursuit, they've taken about two steps in pursuit of the rabbit when I hit the stop whistle. They were in a state of high excitement, they badly wanted that rabbit, but the instant I blew the whistle, they sat, and their attention focused on me. The dogs weren't cowered into submission, I walked over to the dogs, praised them for a very nice stop and cast them off in a different direction, they took off full of drive in search of new game. The dogs are trained to the extent that they have the self-control, the presence/clarity of mind, to be able to take commands despite the high level of excitement and are trained to the extent that they will obey the sit command despite a strong desire to do otherwise. So who controls the dog, I do. But a large part of the way I achieve this control is by teaching the dog self-control. A dog has to hear a whistle command before it can obey it, and in many untrained gundogs, dogs lacking self control, they are driven to fever pitch by the excitement of the rabbit, their mind is focused on only one thing and I think they don’t have a chance of obeying the whistle, as they don’t even hear it, all their attention is focused on the rabbit. ETA My dogs are still young, and not yet fully trained. With a fully trained dog, the sight of a flushing rabbit (or the sight/sound of flushing quail), is by itself the stop command. A fully trained dog would not have required a stop whistle in the above example, as they would have “stopped to flush” and they would have done so, whether or not I was in sight.