Thread: Dog training problems and answers

Originally Posted by el borracho

No difference Dougie in any dog .Clear precise objectives carried out in a fun consistent manner.Whilst I recognize yours methods are vastly different to Ruffs model if yours is working its working .I dont do treats , if I had them I would fight the dog for them and end up eating them myself

There is no single method in training just some that may work better than others and one needs to objectively sample those methods to make an informed decision.

Huge differences in any dog EB.... the dicipline sets the standard,... don;t worry I'll elaborate... at length.

I think you'd like my treats, cheese and possyum haha.

The difference between luring and I guess you'd call it positioning, is that by using a lure (or target in a hand-target trained dog) is that the dog puts himself in the position, using his own muscles. The dog's nose goes up and back for the target, as a result the bum goes down. Over repetition the lure is faded and becomes a hand signal (the hand being raised up, or finger pointing up in my case). Then when the handler can bet a million bucks that the dog will perrform the movement when the visual cue is given (or even just his body language suggests he will sit in a moment) then the verbal cue is given. The result is (with repetition) that both the verbal and visual cues are very strong as the dog has chosen to perform the behaviour himself, with the use of his own muscles.

Paired with positive reinforcement (positive means something added, in my case a treat or short game of tug; reinforcement meaning something that increases the likelyhood of the behaviour being repeted) and negative punishment (negative meaning something taken away, punishment meaning decreasing the likelyhood of behaviour happening again) I have found this an amazing method to teach dogs both behaviours and encourage them to learn (offer new behaviours, 'figure stuff out').

EG:

Sitting means getting a treat. - positive reinforcement
Jumping up and barking gets a turned back, no attention - negative punshiment

That's how my training works, sort of in it's most basic form not that anyone asked.....

*Note: again this is for companion and trick behaviours and agility...I have no idea how this would work with a hunting dog but I can tell you now that tracking dogs like police and SAR are all trained using positive reinforcement/negative punishment. In NZ anyway, to my knowledge.

Originally Posted by Ruff

Huge differences in any dog EB.... the dicipline sets the standard,... don;t worry I'll elaborate... at length.

That would be great Ruff -this is what this post is about discussing the problems or methods or answers to those challenges in training .
I would be surprised if training varys between dogs in the basic training of an animal of course depending on its use and natural abilities but am more than willing to take some learnin up !!! if so

Originally Posted by el borracho

That would be great Ruff -this is what this post is about discussing the problems or methods or answers to those challenges in training .
I would be surprised if training varys between dogs in the basic training of an animal of course depending on its use and natural abilities but am more than willing to take some learnin up !!! if so

It depends on if you prefer old fashioned methods that don;t work or new fangled methods that don;t work.

Essentially, most methods out there don;t work, they work to a level people accept, but nearly every handler I have ever met wanted more (Other than a handful of elite that discovered some of what I'll share soon)

This is a short video of exactly the antics my pup gets up to, mannerisms, intelligence and movements/sounds to a tee, an English Smooth Fox Terrier.

Great fun and extremely extremely challenging to train, and is a female, so definately a bitch - catches blackbirds in flight!

Dinner - YouTube

The real deal stolen from youtube, ie not my pup

Smooth Fox Terrier vs. Ducks - YouTube

Denzel my smooth fox terrier puppy ( 6 months ) - YouTube

Hatfield the Smooth Fox Terrier (www.CanineGuardiansForLife.org) - YouTube

First to correct a few misnomers that have already come up, then I'll address how different training applications can be applied appropriately to different disciplines....

Police dogs are absolutely not trained using positive reinforcement methods. Many people in other disciplines will latch on to a dog receiving some "reward" as positive reinforcement training. It is not. In it's pure form, as outlined above, positive reinforcement training dictates we reward the behavior we want and ignore those behaviors we do not.

This will work for some situations of low level distraction, it will never work properly in the real world and doesn't. It has as much logic to it as the IRD deciding it will, from now on, send thank you letters tho those who pay their taxes and ignore those who don't... Whose filing a return next year?

To outline how unrealistic this is... let's take real world scenarios... you have a dog which has become dog aggressive. You take it to the park and it attacks another dog... we are now going to ignore that behavior but when it's finished and is no longer fighting we'll give it a piece of cheese. Let's use common sense and think over if that will work.

Service dogs have been mentioned but virtually none of them ever work off a lead. How easy would gun-dog training be if the tasks we have for our dogs and their work never required them being off a 2m lead? It'd be a walk in the park as is most all service dog work. It requires very little discipline on the dog or handlers part. Virtually nothing a service dog does, with exception of Police Dogs being sent to hold an offender, has much relevance to gun dog work or control. As I said, how easy would life be if our dogs only had to find game on a 2m lead....??? We could "train" a dog to this in an afternoon.

There is a reference above to having a sit you "could bet a million dollars on". I defy anyone to get such a solid thing using purely positive reward or treat training on a high prey drive dog in a high prey drive situation. e.g: An 18 month German Short-hair Pointer is sent out on a retrieve for a pheasant which has been winged and has planed into cover 150 meters away. It is clearly hit and clearly retrievable. The dog has mis-marked the landing and is 100 meters out heading at full steam in the wrong direction which, unfortunately is also upwind of where the bird went. As this is a hunting area the area the dog is heading to is likely to contain a myriad of scent from both recent and not so recent game, it may even have live game in it. The dog, convinced it is doing the right thing is in full flight.

Ask it to stop. If it does, no problem... you can then give it a re-directional signal into the area of the fall and retrieve the game. If it doesn't you now have a pursuit (Hoping like hell this isn't on the other side of the river) you have to try and re-establish contact with your dog bring it back to the area of the fall and start the hunt for the runner... by this stage it has a 5 - 10 minute head start and may never be retrieved which is the point of the gun dogs existence to ensure a humane and prompt end to the birds life.

Treat training has its basis in Dolphin Training and it works on Dolphins in concrete tanks the same way it works on dogs in confined areas of low distraction with little else to do. Take the dolphin out of the tank, put it in the ocean and wave an anchovy and see if it wants to play ball or bugger off? Combined with "Situation Pattern Obedience" methods it can create a fine display of an agility dog, a dog that doesn't crap in the rug at home, it stops Paris Hilton's Chihuahua from biting her hairdresser (Maybe) it does nothing to train a high prey drive dog in high prey drive situations to have self discipline and to respect its handler enough to comply with requests when there other options on offer.

Food will only work as a reward when it is the best reward on offer. Of any decent gun-dog I know off, when confronted with hot scent I could be holding a side of beef and not be able to get its attention. Further you then have to deal what is essentially an issue of perception. Many of these methods are excellent for teaching tricks (Tricks which will usually only be performed in low distraction areas under a regime of having repeated the exact same pattern over and over again for the same result, something which can never be replicated in the hunting field. In the following video you will see someone purporting to teach sit to a "puppy" which has obviously had this "lesson" a thousand times... I prefer the videos I watch to be teaching a dog that has had no prior training, it adds credibility to the methodology. However, watch it and see if you came to the conclusion I did. I do not see a woman teaching a dog to sit. I see a dog teaching a woman to give it a piece of cheese. Notice he usually sits before the command, she rewards the behavior and says "He has sit, down"... no he has teaching you to feed it "down".


Unfortunately we live in a dog world of two major contrasts which, also unfortunately, fuel each other. Traditionally high prey drive dogs have been trained using a lot of physicality. Not using a reward method does not mean brutalizing a dog or requiring strong physical discipline although many people make that giant leap.

The method mentioned above by KaimaiCockher is, without question the fastest and most humane method for teaching a sit. I will elaborate in a bit, but in the interim I will point out why the food method, is much, much harder on the dog, more stressful due to the confusion it creates. First you have to accept you are dealing with a canine brain, not a human one.

In the video, if you also go to the next lesson of teaching a down you see the handler ask the dog to sit, give no reward, be essentially teased with the food which previously it has been taught it would get for sitting, it has complied but received no reward, the handler keeps holding the cheese lower and lower until the dog finds itself on the floor, she says "down" and gives the cheese. In a dogs mind the first lesson is now rendered meaningless and the new way it has learned to teach the woman to give it cheese is to lie down. It is going to take a long, long time, a lot of stress and confusion for the dog until it links all those words and actions together to decipher that when the right action is associated with the right sound it will get a piece of cheese. (I would also add she totally confuses the dog by just about running "down" and "good" together... the dog is not being given a specific identifiable command to associate with the behavior unless you were teaching the command "downgood")

Contrast this, keeping in mind how the canine brain works, with a simple push on the hindquarters which is combined in the very instant with the command. The meaning is learned virtually instantly. The need to comply is learned through repositioning the pup into the sit whenever it chooses to move so the combined lesson it receives here is "sit until told to move" In very young pups the time I expect is seconds and this gives me the opportunity to also teach the "OK" release command in the same lesson without any confusion. Among it's very first low level lessons is to respond, to comply until told otherwise. By teaching this lesson in a light way under low distraction you have essentially set your entire training path and the psychology with which the dog will understand the rest of its training.

Training as the dog gets older embodied exactly the same principles but with the addition of higher and higher distraction. In the "old school" methodology alluded to though only negative consequences are introduced to ensure compliance. This is as flawed as the food training method because it establishes a basis of conflict in all but the softest of dogs. The dog still wants to do something else but is refraining for fear of physical correction. The problem with that mentality is that if the dog at any stage spots an opportunity to do as it wants and thinks it can dodge the correction it will.

Both methods fall apart because when the chips are down the wheels fall off. The "perfect agility" dog taken to the beach with kids, beach balls, other dogs, seagulls etc when let off the lead forgets every lesson it ever learned in its life. The gun-dog in the field in a patch of maize and a nose full of pheasant scent from three birds runs riot because the only alternative is to stop doing it and return to the handler, bad option for a good dog and it is prepared to take the irrelevant tantrum from its handler on return.

I can go on, but I have already gone on and one, but I think it's important that people understand this fully to "get it".

Several years ago, having been an advocate of old school training methods for many years, i realized that to a great degree it didn't work. As I had worked with Paul Hutton and also done a lot of remedial training in other dogs, not just gundogs, I was fully aware that food training and positive reward training were totally flawed except for some application to low distraction and competitive situation obedience sports for dogs. 99% of the problem dogs i encountered and had to "fix" had been trained using the later methods. Quite simply when the dog had something better to do it had very little respect for the handler or a need to comply, but importantly, a good reason to comply.

A food treat was only a reason to comply if it was the best reward on offer... chasing a rabbit was considered a far better reward than any food treat i had ever seen. Conversely, in most high prey drive dogs it was also high enough reward to accept the tantrum of it's handler once finished, so while the handler had given the dog a "sort out" for the chase it rarely, if ever, stopped the chasing... the handler just felt like they'd done something. Up until this point the handlers only alternative method was to stand and hold a piece of cheese in the futile belief the dog would want to stop chasing and return for the food reward. They never do.

I got to thinking about how young dogs are taught to behave in a pack and how to behave on a hunt. The pack 100% teaches the specific roles each dog has in order to ensure the success of the entire pack. The greatest reward a dog can receive for behaviors is the success of the entire pack, not just itself. The issue with this thought process is that most people believe indiscretions within the pack are handled with physical punishments dealt out by Alpha dogs within the pack. That isn't true either. The role of the "leader" within the pack if the leader is of any use, is not even up for discussion. Displeasing the pack leader does not usually involve the perpetrator being given a hiding, all that is usually required is a warning, in a dogs case a look and the dog pulls itself into line. non wanted behavior while hunting will usually just entail a warning look and the pack moving off to start the next hunt. When the correct behaviors are enacted the successful hunt ensues. The dog learns its role within the hunt and pack through compliance bringing success for all.

The relevance this has to training a gun-dog are this. Most handlers never think of themselves and their dog as a pack. but that is the only way the dog sees it. This is a small book already and I'm not going into all the hows where's why's right here. I had to, using these principles, devise my training method to incorporate this mentality... but it boils down to this.

The dog learns to do certain behaviors on the hunt, to understand the leader is dictating the hunt and to be successful through the joint efforts of all involved. It is not steady because someone will give it a piece of cheese... the chase would be too much fun, it is not steady because it will get it's arse kicked if it doesn't because sometimes I can get away with it. However, if steadiness is intelligently taught to be an integral part of how your hunt is successful the dog complies because that is simply what it does, not through threat, not through bribe.

What the vast majority miss about this is that this is how dogs are pre-wired for us. It's why they are useful. They learn all these lessons by themselves in a pack without being brutalised or being fed tidbits along the way. It is simply how they exist and how they work to success collectively.

Most handlers fall down because they are focusing on how to make the dog do something whereas the successful method is one of bringing a young dog up to know it's role in pack. From the gun dog handlers point of view this leads to a totally different mindset when hunting. A dog quarters knowing it's job is to find game and finding game holds a reward. By flushing or pointing the game and remaining steady the hunt is successful (You can start to join the dots... the behavior on how to act when it screws this up will require a whole other post) the dog's reward is returning to the pack with the bounty.

By careful and controlled handling the dog learns it gets reward for the distant stop and redirection in the success of the pack.

This entire mentality crosses over into just about any canine training application you want to apply it to. But it requires several things most handlers are devoid of and this is what takes the time to learn. How to act like a leader in such a way it is absolutely not up for question... you won't get it with biccies or hidings.

Most handlers think they become a leader by being a bully. you don;t , you do it by being the leader... always. The start of grasping this is when you are out with your dog stop thinking of this as two entities... it is one... a pack. As such it must have a leader... that's you, as leader common sense will allow you to teach a dog how to behave in your pack and all the rules that go with it. It won;t brutalize or annoy other dogs (With violence or ridiculous and unwanted play and affection), it will be compliant, it will be effective and it will think the sun genuinely shines out of your arse... YOU will be its universe and it's access to everything in universe it values...

Sound unlikely? Come and see my dogs... or ask someone on here who has seen my dogs. They don;t get treats, they don;t get hidings, they get to hunt with me every day and they love life and I love showing them off...

and for the record... very little of this thinking or methodology is available to the public and it's why so many struggle. While not totally my own methodology what I have developed over the past few years most certainly is and it is better than any other method I have used or seen for communicating with a dog because the ultimate thing you learn doing it is that most all problems with dogs break down when the communication does. And if you go back to the video at the top and wtach it could be a lesson in how to confuse a dog. Just as equally as a video for someone giving a dog a hiding could be construed the same way.

There are better ways, the dogs won't get any smarter, but the handlers should.

Whew, enjoy.... the discussion from this should legendary.

Dead Right Ruff, I nearly lept out of my chair and applauded.

So A little cyber hug instead {{{Ruff}}}

Ok, but not so close

I stopped reading half way through. I understand I will never change your mind Ruff and I am fine with that. I however think you have missed some very important parts about positive reinforcement (and sorry Ruff as much as you hate it, it's happening all around you right now, to you and your dog and every animal - many behaviours are self-rewarding such as the border collie hearding his sheep or you sitting at your computer enjoying the forum, it's been a positive experience so you are likely to repete it)

The key with +R training in my opinion is that we as handlers are in charge of setting up for success. Yes you are dead on correct that the reward (in your example food) must be the highest value reward (higher than chasing the rabbit across the field). If we set our dogs up to fail, like practicing only in the livingroom then asking for a down-stay in the middle of the dog park with ten other dogs running free, they WILL fail. Taking a dog aggressive dog nose to nose to another dog is OBVIOUSLY going to fail and this is a failure of the human who was so stupid and inconsiderate to put their companion in that situation.

With reactive dogs and +R, it's about providing an acceptable behaviour as another option. Eye contact is a great one. Rather than lashing out, a more low stress calming behaviour is to look the handler in the eye/face. Again setting up the dog to succeed is the key. Building up to those situations where a higher threashold is required, not just jumping right from easy to difficult.

+R means putting a great deal of thought and time into setting up successful situations. The video you put up isn't the best example of clear training and I agree with your comments on it - it is not clear and I do not quite use the same methods as this lady. Also, Expert Village sucks lol. I make fading the lure a very quick process - lure two or three times with food. Lure with hand as if a treat is there, but it's not, three times. Then open hand targeting. The reward is still there at the end but the food lure is gone.

As I have said before, the training in a high distraction environment and with something I am not familiar with myself - hunting - is really interesting to me. I am really looking forward to the tracking course Jet and I are doing come January. Again, different skill set, this is on a tracking harness.

Ruff do you label your methods? Sorry I haven't read the entire post up there but what would you call it, and do you think other people use the same techniques?

Train yourself to read the whole post Dougie.

Okay I read the second half and I liked it you and I aren't so different and if you have read my previous posts you'd understand why....stop getting hooked up on the food reward thing, start thinking REWARD reward thing. Your dogs want the success of the 'pack' as you call it, and that in itself is a reward. Behaviours that have a positive outcome are more likely to be repeted....you just proved my point. My dog loves to tug on a towel and you bet that is a shitload more rewarding than lots of things (including chasing rabbits or cats!) and is used accordingly...it's my highest value reward so is used during highest critera situations.

Hidings are stupid, I am glad you aggree with that. Also that video is stupid haha. I couldn't bring myself to watch all of it to be honest.

Many behaviours are self rewarding (whether we like the behaviour or not, eg some dogs bordom barking). For a hunting dog, hunting is rewarding! Just like chasing the rabbit.

I am not going to get into this as a debate. The answers are very clear and unobtainable on an internet forum. The real word does, however, make them very clear. We are not even close... we are eons apart! You don;t know it, but I have the advantage... I know the methods you are advocating, you have no idea of mine or quite literally the control level obtainable... tracking in a harness... why? Running on natural instinct with no control? We don;t have that luxury!

I am not in the slightest bit surprised you ceased reading half way through. That is a very typical reaction to facts in that way. In the 25 plus years I have been involved in commercially training dogs total denial to the point of not wanting to hear or accept is standard for most people who have been largely indoctrinated into the dog world following one dogma (lol). It is nearly impossible to get anyone to objectively listen to anything that may have contrary conclusions to what they have been taught. I would imagine getting them to read it is even harder... of course curiosity always wins out, which is why you finished it.

However where you and i diverge is that over the years I have witnessed, researched and most often used every method out there and evaluated it for effectiveness... the bottom line is, very few of the modern training methods really work. They will produce tricks, situation pattern obedience and polish to competitive canine pursuits, but they offer no real world control and rarely create a positive bond between dog and handler, though many do misinterpret their dogs stress signals as positive bonding. However the dog training standard is, across the board, so low that often times the person who has a dog that sits or shakes hands is revered among their peers as a wonder dog trainer. This makes it even more difficult to have them let go of their own little world, despite really knowing it’s bollacks, and admit there are better ways to do things, they just don’t know them.

I have said nothing which proves your point in any way shape or form, I was hoping you might have offered a little more of that yourself rather than taking pieces of my post to try and make your own case. Nothing I have covered in my post is in any related to “Positive reinforcement training” at all.

That there are rewards in my methodology is plain, as there is in every training method. That there are consequences to incorrect actions is undeniable, which is absent from many modern methods (absurdly) I would add.

You could make a case about positive reinforcement being all around us... I would respond by asking if you stop at STOP signs because the nice policeman is going to give you a thank you note or because he is going to give you a ticket if you don’t? If there was no consequence, how many would stop when there was nothing in it for them? That’s where your contentions turn to custard. Literally, they actually make no sense and as I appreciate you have an awful lot invested in them I won’t bother you further.

But as is my custom, I am always available anytime to provide dogs to work alongside other dogs so people can show up the very real failings in what i say.

Ultimately I am not interested in a training competition. I appreciate you know nothing of gundogs (by your own admission and as evidenced by your rabbit/reward scenario... chasing rabbits bad... result self reward) I understand you are not training to the level most will expect or want their gundog to be at... but all the more reasons why, if you don’t know what the discipline is, that you should perhaps not offer too much advice re food/reward training because there are some real gundog trainers here that will help newbies... and getting them off on the wrong foot doesn’t serve them or ultimately you.

If you cannot differentiate between being a pack, leading a successful pack and shaping the pack behaviour to one of a highly effective and bonded unit and think being a mobile canteen (It has been reported some dogs will become aggressive to protect their food supply) will get you a bond with your dog you are just misreading and most likely anthropomorphising your dog. You must remain at all times acutely aware of what your dogs intellectual and moral capabilities are... dogs will never equate reward with “love or bond”. They are too selfish and just don’t think that way. Proof... My 12 year old dog, his late seven year old brother... kenneled next to each other... traveled the country in the same crate, hunt together in the same pack, appear to enjoy each other... want to see the love???... try one kennel, two dogs, one bone and this from non-aggressive working springers. Ultimately the bond a dog will form with someone will be dependent at how good THEY are at their role in the pack, and if that is handing out cheese the respect level will be relevant to the importance of that position.

and I'll leave my contribution to this thread there. I have no desire to debate it... I have seen the results from just about every method out there, have consulted with several government agencies (and your claims in how they condition their dogs to scent (Not training) is totally wrong with, perhaps, the exception of airport beagles) have trained and worked with thousands of dogs I put no store in theory, trickery or bribery... I simply get up each day and train dogs, and make better dogs for people and that is my goal, cause when i see my methods turn a dog destined for euthanasia turn into a fully workable, obediant, well mannered mate for someone I walk on air.

Good luck to you....

I love your stance on the love thing, with your two male dogs so true.

Thoughts appreciated Ruff, good luck to you too, I hope I have made it clear to people that I am in no way a gun dog trainer and this thread encouraged 'all types' right at the very start. I'm not sure if you've hooked any clients here. I can assure you the latter has never been my expectation, I just love dogs and training with them.

I knew this would end up in some sort of debate as I am sure so did El B, he's just popped out to get some snacks after setting the whole thing up. It interests me how others train their dogs and hence why I asked how the conventional sit or down was taught by every day Joe blogs.

Anyway unfortunately I think we have both scared off anyone who would have even thought of posting here. Sorry to those people. I might stay away from the dog threads for a while. Just to let you all know though, I am not 'giving up', I am just tired of the politics that surrounds dog training unfortunately. In an ideal world I would like to see man and dog out enjoying each other.

I'll be out the back with Scribe if anyone is looking for us

Oh and how do I label my methods????


EFFECTIVE!

Haha, love it