Training a Gundog Using Food Rewards

January 14, 2016

Some Observations

Ruby and Fern (Sue Brundrett)

Ruby and Fern (Sue Brundrett)

Many gundog trainers resist using food rewards when training their gundogs. Until your retriever has realised the value of expressing her innate desire to retrieve and you, as a trainer, have taught her self control around the things that she wants, food is the most desirable and the most convenient reinforcer available in the early days. But simply using food to reward desired responses is only a small part of the picture…

Using food, toys and praise is all about building a bond association with a puppy. As the training progresses the dog works with you, not against you. You become a team working together, using motivation not force. This relationship makes training easy in terms of the dog understanding what you want and trying hard to offer you behaviour you like. Positive training is about the big picture, relationship, motivation and association – not treats. So we manage the emotional experience while training, making sure the dog enjoys the behaviour with no frustration or confusion so that the dog chooses that behaviour rather than something else. If the learning is negative, then the conditioned emotional response associated with it will impact the choice, especially if there is opportunity to avoid it.

This is why, so often, dogs resist returning to the handler, go out to the retrieve slowly, lag when heeling, delay picking up the retrieve, to name a few. If this is happening to your dog try and see what he or she is avoiding. Check what it was in your training that created this particular conditioned emotional response.

Another important point with the use of food is that it should never be used to manage behaviour such as a dog running in on a retrieve, or producing food to tempt the dog to come to you. Positive trainers use food to create a “trained response” to a cue. Just like traditional training, the dog needs to learn to respond to cues in increasingly distracting situations and this is often where trainers fail to take their training to the next level where the reliability of the “trained” or “conditioned” response to the cue is proofed. Once the response is on cue (or under stimulus control) producing food should no longer be necessary.

The “take home” message I’d like to convey here is that ultimately the value of the reinforcer whether it is sausage, liver bread, tug toys, balls or the opportunity to display a hard-wired motor pattern, is decided by the dog!

 


BACK TO BASICS – RETRIEVER TRAINING

November 11, 2015

If your dog is not doing well in training or competition at the moment and nothing you seem to do works, then the safest route is to go back to basic foundation work.

Craig and Arrow SEcond series _First leg

Always remember that any retrieve from single marked retrieves to multiple memory retrieves to blind retrieves or a combination of all these can be broken down to a few very important basic skills. These are, in a nutshell:

  • Accurate marking and memory.
  • Steadiness.
  • Heelwork.
  • Focus.
  • Recall.
  • Delivery to hand.
  • Whistle sit at a distance.
  • Basic handling – taking a line, overs and backs.

All these basic skills, taken individually, can be worked on either in your home, garden or dog park for the odd 15 or 20 minute session during the week to become fluent. Unless they are executed flawlessly in a non-distracting environment expecting your dog to achieve the same results in an exciting environment is an exercise in futility. Spending hours on a Saturday or Sunday travelling to a venue with suitable terrain for our retrieving games and going over and over things that your dog is at best not achieving in a reasonably distracting environment is not a teaching exercise.

Ensure that your dog really, really understands the basic behaviours and concepts you have taught her in a non distracting environment before testing in a distracting environment. If this does not happen, there’s a very real danger that as the behaviours become more difficult you might get into REAL trouble because your dog didn’t, in fact, understand what you were doing in the first place. One way to be sure that your dog has understood what you require is to only make the behaviour more difficult (adding distractions or duration) when you can successfully get 9 out of 10 correct responses.

It’s worth thinking about this… One incorrect response is a mistake, two mistakes in a row, or close to each other, is a problem. Don’t fool around with it any more. Don’t keep letting your dog practise getting it wrong. Go back a step or two and build proficiency from there.


Honouring in Retriever Training

August 25, 2014

Honour: A dog that waits while another retrieves is said to have honoured.

Depending on the temperament and drive of your dog you will either sail through the honouring part of your training, or will go into line when you have to honour in a state of dread and anxiety, knowing that there is no way your dog will remain steady while another dog is running out to retrieve that oh-so-desirable dummy or bird.

Dogs Honouring

Some of the watch points would be –

  • The futility of “cold” honouring: What many handlers do when teaching the dog to honour is to do many “cold” honours. A “cold” honour is when you let your dog watch other dogs working over and over again. I don’t believe that has anything to do with teaching a dog to honour. What your dog needs to learn is the sequence; first you run the series, then you honour the next dog. (Every once in a while the judge will play some games and not have that kind of honour, but that usually only happens in the higher stakes).
  • Picking up the bird or dummy if your dog breaks: Don’t make the mistake of thinking that the dog is learning the futility of breaking if you have the thrower pick up the bird. It’s the break itself — the exhilaration of the launch and outrun — that reinforces breaking. In order to avoid this train your dog with a tab attached to his collar so that when you’re in line you can hold the tab inconspicuously. By doing this when your dog is inclined to run in he won’t be able to move, and so cannot be reinforced by the launch and outrun. The tab is not meant to punish in any way, it is only meant to stop your dog from going anywhere. This is why I prefer a short tab to a long leash where if the dog does launch itself and manages to get up some speed before hitting the end of the leash, could be severely injured. It is worth bearing in mind that it is better to stop behaviours becoming a habit than to try and fix behaviours that have been reinforced so many times that they have become conditioned. So, don’t be in a hurry to get rid of the tab…
  • Get into the habit of leaving the line after honouring under control. Practice heeling away from the line by doing a U turn (turn into the dog) and heeling off lead a few paces before putting on the leash.

The Sequence

The event style of honouring goes smoothest if it makes sense to the dog. If you practice the sequence he can see that the cues differ from those that predict he will be sent. Also, for a young inexperienced dog it is possible to design the session so that the stimuli are not so overwhelmingly exciting that he can’t think about what’s going on and what he has seen.

So, for the young dog with a lot of retrieve drive, begin by making things simpler for him than a “formal honour at the line”. Once he has delivered his retrieve, heel him away from the line – far enough away so that it really does not look like the “line” to him and so “dilute” his excitement level a little. You are taking away the environmental cue of being in the line means a retrieve. You will also let him watch the “working” dog come into line as this will also become a cue that the next retrieve will not be for him. The departure of that dog to retrieve will also be a cue to relax and expect something different than being sent.

The moment the working dog is sent and after your dog has seen it take off, get your dog’s attention and get him focussed on you. Don’t even worry about getting him to sit – this will come later – he can just be standing there on his leash, still excited because he has just retrieved and wondering what is happening next. He hears gunfire, sees another dog take off, and instead of thinking about running out to get the retrieve he immediately orients back to his handler.

Orienting to you can be rewarded by giving him a treat as soon as you see him notice the other dog’s departure. After a couple of sessions once he is offering you his focus when the other dog departs, you can start cueing him to sit when he resisters the working dog’s departure, followed by a treat. If he won’t accept treats in this circumstance, substitute some praise or petting. Then walk him quietly away on leash “under control” so it is good to not develop habits that are the opposite of control.

With repetition you will see your dog visibly relaxing the moment the other dog is sent and voluntarily (no cue from you needed) turning his attention on you since he expects that a departing working dog means focus on my handler now. At this stage you can occasionally dor a formal honour where he is a sit adjacent to the line.

If you have a dog that is still going crazy even after you have left the area of the line, then move him to an area where he does not actually see the throw, but can hear the gunshot and hear the working handler’s release cue.


Spare the Rod and Spoil the Dog? I Think Not!

August 13, 2014

I have just been re-reading some of the Retriever training books which I acquired years ago when I started out training retrievers for field work. I used these books to plan and guide me in my training. In those days, the seventies and eighties most retriever trainers had never heard of operant conditioning, clicker training, classical conditioning, and, even if they had they hadn’t thought of applying the science to their training methods or even questioning what they were being taught by their trainers. These methods were often passed on from mentor to student over and over again.

Why is that?

I think it was probably because, harsh though they were, these methods worked. Trainers using these methods were winning field trials and working tests. There is also the belief that in order to compete at a higher level you need to be able to “correct” the dog for his mistakes. Mistakes that very likely came about as a result of inefficient, incomplete and outdated training methods. All this just goes to show what amazing animals our dogs are to put up with this abuse. Abuse at the hands of their handler who insists that he loves his dog!

What has prompted me to write this post is my re-reading of a book which is the handbook for many retriever trainers even now in the present day (the book was first published in the 1960’s). The author, to his credit, recognized that the work of Scott and Fuller about the Critical Period of Development in very young puppies was relevant to training any dog be it a family pet, guide dog for the blind, or a field trialling or hunting prospect. So, the first few chapters dealing with the training and socialization of young puppies is pretty nice. However, as the pup gets older and more is expected of it, “force” and “compulsion” rears its ugly head. You read such statements as “When he becomes downright ornery and stubborn, there’s only one way to straighten him out. Heaven protect me from the SPCA, but I’m going to say it. Thrash the dog. Do it with fervor, but with intelligence. I clip the dog with a folded leash until he cries out once”.  This statement horrifies me for many reasons. Do dogs actually know what stubborn is? Do they know exactly what they are being punished for? Has the trainer made absolutely sure that the dog’s “stubborn” behaviour is not confusion or a lack of understanding of what is required? Has the trainer actually sat down and thought about what part he has played in creating his dog’s “stubbornness”. I think not…

Child with labs

The science of animal behaviour has come a long way since those days and we are learning that physical, corporal punishment is a very rocky road on the way to learning. As recently as 10 to 15 years ago, most dog training was done with compulsion and force. These days, however, more and more trainers are using positive methods as they have been found to be more successful and less traumatic for the dogs and puppies. We try to reinforce correct responses rather than punish incorrect ones. We raise criteria gradually. We split behaviour into tiny slices instead of trying to teach large complex lumps. We learn to observe what our dogs are doing and how they are responding. We keep our rate of reinforcement high in the early stages of learning and we look at our timing. It is this consistency and clarity that will create a more confident dog who has good manners around people and other dogs, trusts his handler and looks to his handler for all his needs. Bad behaviour only happens because it has, in some way been reinforced, either inadvertently by you or members of your family, or by something in your pup’s environment.

Katelyn and Derby compressed

I’m not saying that we should be permissive dog owners – on the contrary, we need to teach dogs that there are rules to follow – these rules mostly require the dog to exercise self-control, which is always a good thing. The rules involve highly predictable behaviours for interacting and connecting with their handler.  And these rules can easily be taught without any force or compulsion. It is simply a contract that you enter into with your dog; “You can get what YOU want if you give me what I want”.


Retriever Training – Steadiness using Premack

March 19, 2014

In retriever tests and trials it is a requirement that your dog is steady in the line until sent by the handler to retrieve.  An unsteady dog will fail to mark the fall accurately, particularly if the fall is in cover of some sort, so steadying a dog is one of the skills it is important to master if your dog is to “make the grade”.

Punishment as Motivation – not Recommended

Traditional trainers through the years have used punishment to teach the dog to remain steady.  Some dogs are brought into line wearing a lead which will then bring the dog up short as he runs out and hits the end of the leash.  Some trainers clip their dog with a heeling stick or whip as they run out.  Some trainers “correct” their dogs at the slightest sign of breaking with a helper armed with a BB gun or slingshot shooting your dog in the rump.  As soon as the dog has been hit, you repeat the command “Sit”, smack it on the rump and command “Sit” again.  This is known as indirect pressure.  Later on, when the dog has been conditioned to e-collar corrections, the “Sit” smack “Sit” would be replaced with “Sit” nick “Sit”.  The nick being a short, relatively mild, burst of electricity.

Sadly, these methods are being promoted in some of the most popular gundog training books on the market, even though most dog sports are moving away from punishment as motivation.

Methods such as these obviously work which is the reason for their maintenance within the gundog community – but there is a more humane way.  Even the high drive dog with huge desire for the chase and grab predatory motor pattern can learn to be steady in the presence of gunshot and birds with a bit of patience and a progressive sequence of training goals.

Force-Free Motivation

Initially you will be teaching your pup (with positive reinforcement) the basic obedience behaviours he will need in his career as a working retriever; things like self control, responding reliably to whistle cues, walking off lead at heel and the “Stay” and “Release” sequence.  You will also, in these early days, encourage his inherent desire to retrieve with appropriate games.  The chase and grab games will become intrinsically reinforcing which will trump any extrinsic reinforcement that you can offer and in the early days of training you will not want to kill or subdue this eagerness and enthusiasm for the game by dampening his retrieving spirit with punishment or aversive training.

Eventually, though, you will need to ask your dog to be steady at heel in the presence of the huge distraction of gunshot and either a bird or training dummy as well as somehow harness this self-reinforcing chase and grab behaviour.  This is where the Premack Principle comes in.  David Premack, a psychologist, proposed the rule that high probability behaviours can be used to reinforce low-probability behaviours.  We can use this principle when training dogs very successfully.

High-probability behaviours are activities which are performed voluntarily and which are enjoyed for themselves without the intervention of the handler or owner.  Some examples of this would be playing with other dogs (if your dog is a social butterfly), chasing monkeys or squirrels, barking at the neighbour’s dogs through the fence, sniffing interesting smells, or in the case of our working retrievers, running in on a retrieve without waiting for our signal to fetch.

Low-probability behaviours, on the other hand are behaviours which are often the learned or trained behaviours, such as the recall and walking at heel, or, in our case, the patient waiting at heel for the handler to signal a release so that we can complete the retrieve.

What we are trying to convey to the dog is that if he practices self-control and waits until we release him (low probability behaviour) then he can have the delirious pleasure of the outrun and the retrieve (high probability behaviour).

So, how to apply this in real life? 

First of all we need to proof our Sit/Stay and Release behaviour with distractions such as balls rolled around and food scattered close by, singly and in the presence of other dogs who might be competition for the valued resources of balls and food.  Many gundog trainers these days practise “denials” in an attempt to steady their dogs.  A “denial” is simply not allowing access to the retrieve by removing the retrieve article before the dog can get it.  Access to the retrieve article is also limited by the trainer picking up the article more times than the dog is sent.  This is a relatively force-free way of proofing distractions and it is pretty worthwhile in the early days.  However, there is one problem with this method later in the game that makes it ineffective.  The dog is unlikely to learn the futility of breaking if you have the thrower pick up the bird.  It’s the break itself — the exhilaration of the launch and outrun — that reinforces breaking as much as the retrieve itself.

Once the sit/stay has been proofed to your satisfaction, you can introduce the retrieve.  Have your helper standing a fair distance away (about 50 to 80 metres).  Because the sit/stay is likely to be unsuccessful the first time around, your dog needs to be set up to succeed at this early stage.  Mechanically prevent him from failing by holding on to his collar tab.  Use as little restraint as possible, partly because this can be counter-productive around a prey-type stimulus, and partly because you will want to monitor for any tiny bits (droplets) of self-imposed self control.  It helps the cause to keep the tossing of the dummy as low key as possible.

Put your dog in a sit/stay, lightly holding the collar tab.  You are asking for only a droplet of self control so as soon as the dummy is on the ground and provided there is no tension on the tab, release him with your “Fetch” cue.  Your dog might start to break as soon as he sees that the helper is preparing to throw in which case he must cease his “wind up” and you must remind your dog that he is to sit and stay.  Try again; and again, if necessary.  He will soon realise that the only way to get what he wants (the high probability event) the more he has to stay put and wait for the release (the low probability event).

Once this has happened you can gradually crank up the waiting for the release and with a few repetitions you will find that he is waiting with rapt attention for your magic release cue.


Lining to a Pile drill – hidden advantages

September 26, 2013

British Ret Champs 2007

Everyone who has ever trained a gundog to work on blind/unseen retrieves has done this drill at one time or another.  The “primary” role of the basic pile retrieve exercise is to teach confidence for the send away and part of the introduction to sight blinds.  One of the principal aims is to teach the quick pick-up and prompt return.  But, in addition, it also assists in the hold-heel-sit-deliver sequence becoming very automatic in the dog while he is concentrating on being sent for another retrieve.  In the field the dog has to remember how to swing into heel position and sit down focussed on the field (not up at your face) while holding onto the one he’s carrying and then delivering it properly.  He has to do all this while he is excited about watching the next throw, or going for the next bird of a multiple mark.  Having a pile of objects in front of him that he knows he will be sent to again provide the forward focus and of course keeps the subsequent retrieve aspect in the mix while he practices all the sub skills.

One disadvantage of pile work is that most dogs will try to shop, at least a little so the pile retrieve exercise can be considered a form of proofing to NOT do it.  Shopping is likely to appear in a minor or major way with more than half the retrievers out there.

Troubleshooting the Shopping Habit

Some dogs just love to test each dummy in the pile before selecting the one that is “just right” and then returning to the handler.  There are some things you could do to stop shopping:

  • Make sure that the recall component of the retrieve is much more fun than shopping.  A solid, reliable recall is the best way to nip this behaviour in the bud from the very beginning.  A dog cannot be spinning around to answer a recall cue if he is trying to shop. It is an incompatible behaviour.
  • You could also limit options with a long line so that he cannot shop once something is in his mouth thus taking away the opportunity to self-reward with this undesirable behaviour.
  • What would also work is following him to the pile so that you are close to him when he picks up which would make you the biggest thing in his world right then, not the next dummy.  Then gradually lengthen the following distance.

With a beginner dog that is only just learning the lining exercise you will probably get a couple of attempted shops and if you do not allow the shopping to become a habit by following the options outlined above the shopping will be over so quickly you will wonder if you really “trained” the dog not to do it.  But you did.  Dogs do what works.  If shopping does not work for him the first time he tries it, compared to alternatives that work better (give him more pleasure), he is not likely to continue to do it.

Another way to increase the no-shopping proofing once your dog has the basic understanding is to use a pile of non-similar objects.  For example, large and small dummies, some canvas, some plastic, and a Dokken duck (if familiar to the dog). Do not expect the dog to grab the closest object, and it is okay for him to inspect the pile (quickly) for his preferred object. But whichever one he picks up first needs to stay in his mouth and be delivered, even if he wishes he’d grabbed a different one. He will get an opportunity to do that on the next send!

Louise’s Comment:  These insights come as a result of a discussion on the Positive Gundogs forum.  http://groups.yahoo.com/neo/groups/PositiveGunDogs/info


The Pathway to Excellent Blind Retrieves

May 16, 2013

British Ret Champs 2007What is a blind retrieve?  It depends on your point of view –

  • As a judge……..
  • As a handler/trainer………
  • As a dog………..

From a judge’s point of view it is two-fold.  The handler must be seen to have enough control over the dog to be able to direct him to an area in spite of any distractions on the way and then get him to establish a hunt in that area to come up with the bird.  The dog must be seen to be responsive to the handler’s whistle and hand gestures, go out with style and enthusiasm, and use his nose to guide him to the fall once in the area.

From a handler’s point of view it is the culmination of hours and hours spent on many repetitive lining and handling drills which started when the dog was a pup and will probably continue until the dog is in his 4th and 5th year.  It is, in the case of the team that has done all the preparation and proofing necessary for the level of competition they find themselves in, the incredible feeling of total focus and communication between you and your dog that makes a successfully executed blind retrieve the ultimate and most satisfying game you can play with your dog.

From the dog’s point of view it is knowing exactly what is required of him when the two of you walk into line for a blind.  He will be watching from signals from you to indicate where he must look and in which direction to go.  It is the only time when his natural ability to retrieve the bird take second place and he understands this.  It is his confidence and trust that you will guide and direct him to where these natural abilities can take over in his final hunt for the bird.  This attitude does not develop overnight but takes (dare I say it) years to build.  In a young dog, confident in his ability to mark a fall, establish a hunt and find the bird without any help from you, this realisation has not yet penetrated and I know how frustrating it is to send a young dog out to a blind only to have him wildly hunt an area the size of a rugby field, totally ignoring any whistle commands or hand signals you might be making from the line.  Take heart – the penny does eventually drop, and when it does all the seemingly mindless repetition of drills and exercises you have been doing in training all falls into place, and you have a good “blind” retriever.

Blind Summary Easy Beginnings

Lining Drills and Transition

Handling Drills and Transition

Transition


%d bloggers like this: