Canine Behaviour Foundation Course
Module 1 - Section 3 - Learning Theory
Module 1 - Section 3 - Learning Theory
1. How Dogs Learn
2. Other Training Methods
3. Lure and Reward Training
4. Using Lure /Rewards
5. Fading the Reward
6. Cue's, Signals, Commands
2. Other Training Methods
3. Lure and Reward Training
4. Using Lure /Rewards
5. Fading the Reward
6. Cue's, Signals, Commands
After you have completed the Module 1 Puppy section, we will go into more depth with learning theory and discuss various aspects such as reinforcement rewards, classical and operant conditioning, consequences etc, but for now, you will gain a basic understanding of how dogs learn, lure and reward training (positive reinforcement), negative punishment, and Fading and Cue’s and Signals, which is what you will need for the Puppy section.
"Increasing evidence supports the notion that humans and dogs share some social skills, with dogs' social-cognitive functioning resembling that of a 6-month to 2-year-old child in many respects," said József Topál of the Hungarian Academy of Sciences.
(Journal current Biology)
1. How Dogs Learn
Simply put, dogs will try different things and if the behaviour they offer has a positive consequence/end result (such as the dog figures out that sitting when requested results in a treat, or being called by the owner results in attention), then the behaviour is more likely to be repeated, simply because the behaviour, in the dogs eyes, worked.
However, you may now ask yourself, why then does a dog keep on jumping up when we may push it away or shout at it? While we may not think of the above scenario as being rewarding, the dog is gaining attention, and to a dog that is desperate for attention, it simply boils down to attention to that particular dog (especially in dogs that have a lack of human interaction) is attention, whether negative or positive!
In order for a dog to associate the consequence/end result, of its behaviour, the positive consequence/reward must be immediate i.e. as the behaviour is actually happening or only up to a second later. If the consequence/reward is not presented immediately, the dog is not likely to associate the behaviour it exhibited as being the one that resulted in the consequence/reward.
They also learn by the association of events that occur closely together. Such an obvious example of this is when the owner is going to take the dog for a walk, the owners tend to perform the same behaviours (we are after all creatures of habit!) i.e. changing into walking shoes, going to the drawer, cupboard or hook where the lead and collar hang, perhaps even going for a walk at the same time every day. The dog will then make an immediate assumption that a walk is about to be forthcoming and will get excited.
In the behaviour section, you will learn that one of the first things we would do to change over-excitement from a dog about to go for a walk is to change the owner’s habits and locations where the equipment is kept. Just doing a simple thing such as this will change the dogs association and make it much easier for both dog and owner to bring about the desired changes.
What we need to realize is that dogs are learning 24/7 and the learning process is not just restricted to training sessions, so your daily interactions with the dog will have a huge impact on behaviour as well. There are also no magic wands – what is needed is consistency and practice and this is something we always tell clients – we can give them the tools to change behaviour, but it is up to them to do the actual work.
A dog will keep on working at what gets it the reward – when the reward stops occurring, then the behaviour too will stop. A prime example of this is a Rotti I worked with many years ago. I was called in for serious destructive behaviour which was that the dog had literally gouged a hole in the fridge. On asking how this occurred I discovered that when the dog was a pup, it had learnt that pawing the fridge door resulted in the door being opened and a treat forthcoming (the owners thought the dog very smart and the behaviour cute). However the dog was now a full grown, large sized dog, and if the door was not opened on request, it just kept on scratching at it harder and harder – much like a child throwing a tantrum. The modifications? Ignore the behaviour and leave the room! Within two days the dog had stopped completely and the owner was able to buy a new fridge. If we keep on rewarding the behaviour, the behaviour is continually reinforced.
Simply put, dogs will try different things and if the behaviour they offer has a positive consequence/end result (such as the dog figures out that sitting when requested results in a treat, or being called by the owner results in attention), then the behaviour is more likely to be repeated, simply because the behaviour, in the dogs eyes, worked.
However, you may now ask yourself, why then does a dog keep on jumping up when we may push it away or shout at it? While we may not think of the above scenario as being rewarding, the dog is gaining attention, and to a dog that is desperate for attention, it simply boils down to attention to that particular dog (especially in dogs that have a lack of human interaction) is attention, whether negative or positive!
In order for a dog to associate the consequence/end result, of its behaviour, the positive consequence/reward must be immediate i.e. as the behaviour is actually happening or only up to a second later. If the consequence/reward is not presented immediately, the dog is not likely to associate the behaviour it exhibited as being the one that resulted in the consequence/reward.
They also learn by the association of events that occur closely together. Such an obvious example of this is when the owner is going to take the dog for a walk, the owners tend to perform the same behaviours (we are after all creatures of habit!) i.e. changing into walking shoes, going to the drawer, cupboard or hook where the lead and collar hang, perhaps even going for a walk at the same time every day. The dog will then make an immediate assumption that a walk is about to be forthcoming and will get excited.
In the behaviour section, you will learn that one of the first things we would do to change over-excitement from a dog about to go for a walk is to change the owner’s habits and locations where the equipment is kept. Just doing a simple thing such as this will change the dogs association and make it much easier for both dog and owner to bring about the desired changes.
What we need to realize is that dogs are learning 24/7 and the learning process is not just restricted to training sessions, so your daily interactions with the dog will have a huge impact on behaviour as well. There are also no magic wands – what is needed is consistency and practice and this is something we always tell clients – we can give them the tools to change behaviour, but it is up to them to do the actual work.
A dog will keep on working at what gets it the reward – when the reward stops occurring, then the behaviour too will stop. A prime example of this is a Rotti I worked with many years ago. I was called in for serious destructive behaviour which was that the dog had literally gouged a hole in the fridge. On asking how this occurred I discovered that when the dog was a pup, it had learnt that pawing the fridge door resulted in the door being opened and a treat forthcoming (the owners thought the dog very smart and the behaviour cute). However the dog was now a full grown, large sized dog, and if the door was not opened on request, it just kept on scratching at it harder and harder – much like a child throwing a tantrum. The modifications? Ignore the behaviour and leave the room! Within two days the dog had stopped completely and the owner was able to buy a new fridge. If we keep on rewarding the behaviour, the behaviour is continually reinforced.
2. Other Methods of Training
The methods that we are employing nowadays (well, a lot of us at least!), differs dramatically from the type of training that was done prior to lure and reward training gaining popularity. We have already discussed aspects of this in the Pack Theory section, but will over this again, as it is a very important point.
The methods that we are employing nowadays (well, a lot of us at least!), differs dramatically from the type of training that was done prior to lure and reward training gaining popularity. We have already discussed aspects of this in the Pack Theory section, but will over this again, as it is a very important point.
I have a personal philosophy in that I will not, under any circumstances discuss or give comment on anybody that is practicing and advocating adverse methods of training. I will rather ask client when they mention it the following question: ‘did it work?’ (Of course not or otherwise I would not be there). They will say no, or a bit, I will then say ‘well let’s try something else and see if this will be effective’, or something to tha effect. By all means, know and be able to explain why an alpha roll for example will damage the relationship with the dog, but do try and get into the habit of not mentioning actual people – my opinion on this is that it is unethical.
A good example of this was a book by a well respected expert of his time, H. V. Beamish, who wrote a book titled ‘Your Puppy” which was published in 1957. In this book on the section called First Lessons – Discipline and Training, Mr. Beamish says: - ‘Discipline doesn’t mean giving orders. It means seeing those orders are carried out, whether it’s pleasant for you or not”. On the following page he continues “A multitude of difference exists between firmness and cruelty – calmly administered punishment, and a wild fury. Quite often the dog should be punished, but you must do it calmly, even though you are giving him a thrashing, which should be reserved for only one or two crimes. There are plenty methods of punishment, and often a moral thrashing is quite as effective as a physical one. Punishment should only be given on certain occasions, but always when the dog has deliberately done wrong, knowing he has done so. It would be absurd to give him a thrashing for mere ignorance”.
Well thank heavens that we have now learnt so much more about dog behaviour! This gentleman did not believe he was doing anything wrong at all; he was only doing what he believed to be correct. We now know that a dog does not do anything on purpose, the animal is simply responding to our pheromones and body language. We also know that to shout or scream at a dog does not accomplish anything - dogs respect a leader far more that is quiet and confident and to a dog attention is attention, whether negative or positive. If a dog keeps on repeating behaviours it is because it is getting the attention it wants.Somewhere, somehow, the behaviour is being reinforced!
In the sixties, seventies and eighties, dog training was largely based on coercive methods which were believed to have originated from police dog training and the military and are often called K9 training or Compulsion methods. The standard equipment used was a choke chain and the training was punishment based. Unfortunately, over the last few years, many of the old fashioned methods have crept back in due to TV programs which offer ‘instant fixes’ rather than teaching the dog another behaviour.
A good example of this was a book by a well respected expert of his time, H. V. Beamish, who wrote a book titled ‘Your Puppy” which was published in 1957. In this book on the section called First Lessons – Discipline and Training, Mr. Beamish says: - ‘Discipline doesn’t mean giving orders. It means seeing those orders are carried out, whether it’s pleasant for you or not”. On the following page he continues “A multitude of difference exists between firmness and cruelty – calmly administered punishment, and a wild fury. Quite often the dog should be punished, but you must do it calmly, even though you are giving him a thrashing, which should be reserved for only one or two crimes. There are plenty methods of punishment, and often a moral thrashing is quite as effective as a physical one. Punishment should only be given on certain occasions, but always when the dog has deliberately done wrong, knowing he has done so. It would be absurd to give him a thrashing for mere ignorance”.
Well thank heavens that we have now learnt so much more about dog behaviour! This gentleman did not believe he was doing anything wrong at all; he was only doing what he believed to be correct. We now know that a dog does not do anything on purpose, the animal is simply responding to our pheromones and body language. We also know that to shout or scream at a dog does not accomplish anything - dogs respect a leader far more that is quiet and confident and to a dog attention is attention, whether negative or positive. If a dog keeps on repeating behaviours it is because it is getting the attention it wants.Somewhere, somehow, the behaviour is being reinforced!
In the sixties, seventies and eighties, dog training was largely based on coercive methods which were believed to have originated from police dog training and the military and are often called K9 training or Compulsion methods. The standard equipment used was a choke chain and the training was punishment based. Unfortunately, over the last few years, many of the old fashioned methods have crept back in due to TV programs which offer ‘instant fixes’ rather than teaching the dog another behaviour.
On the old way lead was pulled up and the hindquarters pushed into the Sit position at the same time
The K9 method was to pull/yank the head upwards by tension on the lead and collar and push the h/q down into a sit
|
Choke collar - the correct position is higher on the neck
|
These TV programs talk about the dog owner needing to make themselves the ‘pack leader’ in order to get the dog do what they want it to and the way to do this is by using force, an example is of a dog being held down and forced to submit to the person. Techniques such as these actually make a dog more fearful and likely to be reactive. You may say, ‘but the behaviour stopped’. Yes it did, but not out of wanting to, rather the dog has gone into ‘learned helplessness’ and has just given up. This is not a dog that is working with its owner or even learning another way of acting, or overcoming fear. If we compare this to a human situation. If I put you in a room full of spiders if you have a serious fear or dislike of spiders, and hold you down so that you can’t get away and can’t do anything, the end result will be that you eventually just give up – this is what the dogs does and as above is called ‘learned helplessness’. Another example where I have seen learned helplessness on TV shows as mentioned is when a dog that was fearful of a walker was forced to go on one and could not escape it – yes the dog was walking, but just look at both the body language and the look in the dog’s eye – it gave up. This is not the way we want our dogs to react – rather we work with them to help them change their behaviour, not through the use of old fashioned, arcade methods.
If you are going to use physical punishment, then you have to have both the timing (the exact second the behaviour is occurring) together with the correct intensity applied – the very first time you punish - too little will not work and too much can severely damage the canine/human bond. All the dog learns from physical punishment it that is has too respond. This is an important factor for you to know and be able to explain to your clients.
Years ago while I was doing my Instructors course with SADTC there was a demo/competition I watched between SADTC, who uses positive reinforcement and the visiting team who used the compulsion method – this was enough for me to make up my mind that positive reinforcement was the way forward for me. The SADTC team raced through the course (which was difficult and challenging) with their dogs right next to them, looking up at handler expectantly and happy. Tongues – were lolling out, the faces were bright and interested and handler and dog were having a ball. The compulsion team on the other hand, did do well (and I really can’t remember who won –I believe I did!), but the dogs did not have the same look on their faces and looked more like a bunch of military people being forced to do something they were not really enjoying – there was simply no enthusiasm involved.
In the Behaviour Problems section you do have notes on what damage a choke chain can do in order that you have the information at hand for owners still using them.
Following this method a form of modelling was used by trainers. This is where the dog was assisted into position by the owner pressing on his hindquarters to put him into the sit position or sweeping the legs forward to get the dog into the down position. Very often choke chains were used in conjunction with this type of training and the owner used this to pull/yank the dog into the desired position while giving the command for the behaviour at the same time. To get a dog to recall the dog was ‘reeled in’ (yes a bit like fishing!).
The video below is an experiment in which people instead of dogs are in training sessions using two different methods: the positive reinforcement (R+) and punishment.
In both phases of the experiment the desired behavior is for the person to stand behind a chair.
In the first part the "trainer" uses a clicker (the clicker is a device that emits a click and that sound has been previously associated with a reward in food: serves to make the action exactly the moment that is taking place, as if It was a "very good" but more effective). The absence of the click indicates to the subject (dog or person) that they have to look for alternative actions to get it right and get the click (prize).
With the use of punishment, on the other hand, the trainer "punished " the subject with a mild electric shock for every "wrong" action that kept the person away from the desired behavior, while ignoring any other, even though the subject gets closer to the target.
Which of the two "dogs" will learn more easily?
In both phases of the experiment the desired behavior is for the person to stand behind a chair.
In the first part the "trainer" uses a clicker (the clicker is a device that emits a click and that sound has been previously associated with a reward in food: serves to make the action exactly the moment that is taking place, as if It was a "very good" but more effective). The absence of the click indicates to the subject (dog or person) that they have to look for alternative actions to get it right and get the click (prize).
With the use of punishment, on the other hand, the trainer "punished " the subject with a mild electric shock for every "wrong" action that kept the person away from the desired behavior, while ignoring any other, even though the subject gets closer to the target.
Which of the two "dogs" will learn more easily?
Other methods that were employed, especially in the case of a dog being reactive to another dog, was the handler literally lifting the dog off the floor with the use of a choke chain and ‘hanging’ it so to speak.
Then followed lure and reward training which is still very much in evidence today. Further research was done which showed that a dog will work harder if rewards are varied or given intermittently rather than all the time and different reinforcement schedules were brought in. This method of training is based on positive reinforcement, but the terminology is not entirely correct – one does not use only positive reinforcement, negative punishment is also used, so the trainer that says they ‘only’ employ positive reinforcements is not 100% correct!
This then led to clicker training is the best method to use, but unless a client has been properly taught the basics of how it works and the importance of timing, the majority of owners should only be taught the lure and reward method which is so very easy. Have you ever seen a class of first time dog handlers trying to handle their dogs, treats, give new commands and handle the clicker at the same time? If so you will know why we prefer to initially teach only the lure and reward method.
Then followed lure and reward training which is still very much in evidence today. Further research was done which showed that a dog will work harder if rewards are varied or given intermittently rather than all the time and different reinforcement schedules were brought in. This method of training is based on positive reinforcement, but the terminology is not entirely correct – one does not use only positive reinforcement, negative punishment is also used, so the trainer that says they ‘only’ employ positive reinforcements is not 100% correct!
This then led to clicker training is the best method to use, but unless a client has been properly taught the basics of how it works and the importance of timing, the majority of owners should only be taught the lure and reward method which is so very easy. Have you ever seen a class of first time dog handlers trying to handle their dogs, treats, give new commands and handle the clicker at the same time? If so you will know why we prefer to initially teach only the lure and reward method.
3. Lure and Reward Training
Lure and reward training has been very popular for approximately the last 12 – 15yrs years (if not longer) and is regarded as the most modern way, apart from Clicker training, to train a dog.We feel, however, that a form of reward training was probably in use many, many years ago, when the dogs ancestors started to encroach onto human settlements and this is how the relationship between man/dog/wolf started.
What is Lure and Reward Based Training? Lure and Reward training is exactly what it sounds like – luring the dog into doing something and then rewarding him for doing it. It is a very useful and simple technique that can be used for teaching a dog a huge variety of different behaviours. Dogs will work for rewards much in the same way that humans do. This is a modification of the system that Ian Dunbar uses:-
1. Lure dog into behaviour you want i.e. sitting
2. As dog performs the behaviour (i.e. bum touches floor), cue is given
3. Verbal Praise – essential as we don’t only want the dog to work for food.
4. Food Reward is given after the cue and praise is given.
This a simple as the ABC for clients to understand and grasp, and we use this together with NMR (Non-reward marker) and teach them about fading the rewards and in a very short time they have a complete training system that is easy to use.
Lure and reward training is also much more efficient than the old fashioned punishment based methods. With the punishment based methods there are many ways that the dog can get it wrong, each of which results in punishment, whereas with Lure and Reward training there is only one way – when the dog gets it right it is rewarded! This also makes it much faster for owners which I find results in better communication between dog and owner, and a lack of frustration for both dog and owner.
If you would like to know the more professional description: - The lure and reward method of training is based on the theory of using a primary (or unconditioned) re-in forcer UCR (e.g. food, ball, toys etc) to elicit and positively reinforce the desired behaviour. As every behaviour has a consequence, it is highly probable that the behaviour will be repeated. For example, if sitting works for the dog when receiving something it wants, then the dog will be more likely to sit (desired behaviour) when signaled to do so, rather and lie down (undesired behaviour).
Lure and reward training has been very popular for approximately the last 12 – 15yrs years (if not longer) and is regarded as the most modern way, apart from Clicker training, to train a dog.We feel, however, that a form of reward training was probably in use many, many years ago, when the dogs ancestors started to encroach onto human settlements and this is how the relationship between man/dog/wolf started.
What is Lure and Reward Based Training? Lure and Reward training is exactly what it sounds like – luring the dog into doing something and then rewarding him for doing it. It is a very useful and simple technique that can be used for teaching a dog a huge variety of different behaviours. Dogs will work for rewards much in the same way that humans do. This is a modification of the system that Ian Dunbar uses:-
1. Lure dog into behaviour you want i.e. sitting
2. As dog performs the behaviour (i.e. bum touches floor), cue is given
3. Verbal Praise – essential as we don’t only want the dog to work for food.
4. Food Reward is given after the cue and praise is given.
This a simple as the ABC for clients to understand and grasp, and we use this together with NMR (Non-reward marker) and teach them about fading the rewards and in a very short time they have a complete training system that is easy to use.
Lure and reward training is also much more efficient than the old fashioned punishment based methods. With the punishment based methods there are many ways that the dog can get it wrong, each of which results in punishment, whereas with Lure and Reward training there is only one way – when the dog gets it right it is rewarded! This also makes it much faster for owners which I find results in better communication between dog and owner, and a lack of frustration for both dog and owner.
If you would like to know the more professional description: - The lure and reward method of training is based on the theory of using a primary (or unconditioned) re-in forcer UCR (e.g. food, ball, toys etc) to elicit and positively reinforce the desired behaviour. As every behaviour has a consequence, it is highly probable that the behaviour will be repeated. For example, if sitting works for the dog when receiving something it wants, then the dog will be more likely to sit (desired behaviour) when signaled to do so, rather and lie down (undesired behaviour).
The Lure is held close to the dog's nose and slowly raised upwards.
|
As the head goes up, the h/q goes down and as soon as the h/q meet the floor, the Sit cue is given
|
What is important with training is that the owner needs to be consistent and the next essential is repetition, but remembering not to overdo it or the pup will get bored. Very few dogs learn to perform behaviour after doing it only once, it normally takes a lot of repetitions until the dog associates the behaviour with the reward.
Another important factor in training is not just to have one practise session per day. Practise throughout the day for a few seconds at a time, and most importantly, practice in different locations. We cannot count the amount of times we have heard owners say when the pup failed to perform an exercise in class - ‘oh but he does it perfectly at home’! Each and every minute we spend with our dogs is a training session – when the dog sits of its own accord, sit, good boy, the same when the dog lies down, when you call the dog to you etc. We miss hundreds of training opportunities daily just because we are not aware of the possibilities that are open to us.
How do we, as trainers, go about getting a dog to repeat those behaviours we like and not repeat those behaviours we don’t like? Although we have covered this in How Dogs Learn, I will repeat myself here. It is actually very simply: people, dogs and most other organisms perform and repeat behaviours because these behaviours work! Contrary to common belief, dogs do not possess some mystical gene that wants to please us, and neither do humans or other animals. In some way or other, the behaviour is being reinforced (rewarded). Mary R Burch & Jon S Bailey (1999) states “Reinforcement occurs when behaviour, followed by a consequent stimulus, is strengthened, or becomes more likely to occur again”. That, in theory sounds straightforward enough and should make it relatively easy for us to work out what reinforces some behaviours, but we may not readily understand what reinforcement (reward) certain compulsive behaviours represent.
For example, what reinforcement (reward) does a dog get if it licks or chews itself until its skin is raw and bleeding?? This particular behaviour is classified as compulsive behaviour (Odendaal 2001) and is reinforced because it is thought to relieve the dog’s anxieties by balancing the serotonin levels in its brain.
Another point we need to be very aware of is the body language and signals that we use when working with our dogs (we will work on this at a later stage). A dog is likely to pay attention to our body language and signals rather than to our spoken word.
Misconceptions and what can go wrong with Lure and Reward Training.
- The dog will only work when a reward is forthcoming. Wrong, once the dog has learned the behaviour an immediate varied schedule should be undertaken and the reward faded – and the sooner the better.
- Dogs get fat due to this method of training. Wrong, only if the owner gives too many treats or the wrong type of treat. The golden rule is you give as little as possible and unless training a new behaviour, use some of the dog’s daily rations. An important point here – you and your client may have a totally different idea of what constitutes something ‘small’ – show them the size treat to use . If your dog is at all ball or toy driven, then this can be used as a reinforcer instead, or alternated with food rewards.
- Using the Lure and Reward method is bribing a dog. Incorrect, a reward is something that a dog will be given after it has performed the desired behaviour. A bribe is something that is done before the dog has undertaken the desired behaviour.
- Timing. This is something that an inexperienced dog owner often does incorrectly. An example of this is giving the reward before the full behaviour has been exhibited. A good example of this is an owner that awards the Sit after the dog has sat and then goes into the begging position. I had a lady that had a lovely little JR and she had rewarded as above, the end result was that every time the owner asked for a sit the dog went into the beg position!
In closing I believe that not only does lure and reward training work, it deepens our bonds with our dogs and the more things we teach our dogs the more we supply them with the necessary mental stimulation they desire. What is crucial however, is that Fading is started as soon as possible.
4. Using Lure / Rewards
The very first thing you or your client need to do is to find out what actually motivates the pup and then use this for training to get the behaviours they want. Many people believe that food only should be used, but this is not necessarily correct as many pups/dogs are not at all food driven or may view the food as a lesser reward. The client may find that the dog is ball driven, may love the Frisbee or perhaps prefer a soft or squeaky toy. By recognizing what motivates the pup you can take advantage of this to train good behaviour and any exercises you want to teach the dog.
Another aspect to consider for providing motivation, is what group the dog falls into? Is it a hound, a terrier or a gundog? By using a dog’s natural, instinctive behaviour, we can get it to perform the behaviours we want much more easily. Look at ways you could bring in what the dog does naturally as a reward.
Analyzing behaviours in the group categories will also help you as well in that you will have a basic understanding of the different breeds and will learn that working with a dog in the herding group (Collie for example), will take a different approach than when you work with a dog in the hound category (Beagle for example)
What owners do need to be aware of is to give as little as they can get away with where food is concerned. This often needs to be shown to owners; their perception of a little and yours may not be the same! It is all too easy to end up with an overweight dog and the risk of overfeeding is a very real one.
We prefer to use a softer type treat otherwise the dog tends to take time to chew/munch on the treat and the reason for the reward will be forgotten. Another aspect to consider here is that if a dog is full it will not be as willing to work, so work with the dog when it is hungry.
We use a variety of different treats when teaching a pup or even an adult dog new behaviours as this way it will never know what is coming and special treats are used as a jackpot for a difficult exercise, or when the pup/dog has eventually mastered the behaviour requested. Our most training treat of all time is Liver Bread and we have yet to come across a dog that does not agree!
Jackpots
Jackpots are what we use when a puppy does something exceptionally well or if he has been battling with an exercise, the down for example, and then he suddenly performs it, you can reward with a jackpot, which is a special treat, or several treats in a row, really making a fuss about how well the dog has done. It can be compared to a child receiving a gold star at school. A combination of food and toys can be used as well. My own dog learns new behaviours with standard kibble and to him the best jackpot in the world is to have his ball tossed in the air to catch – he would much rather have this than a food jackpot.
The very first thing you or your client need to do is to find out what actually motivates the pup and then use this for training to get the behaviours they want. Many people believe that food only should be used, but this is not necessarily correct as many pups/dogs are not at all food driven or may view the food as a lesser reward. The client may find that the dog is ball driven, may love the Frisbee or perhaps prefer a soft or squeaky toy. By recognizing what motivates the pup you can take advantage of this to train good behaviour and any exercises you want to teach the dog.
Another aspect to consider for providing motivation, is what group the dog falls into? Is it a hound, a terrier or a gundog? By using a dog’s natural, instinctive behaviour, we can get it to perform the behaviours we want much more easily. Look at ways you could bring in what the dog does naturally as a reward.
Analyzing behaviours in the group categories will also help you as well in that you will have a basic understanding of the different breeds and will learn that working with a dog in the herding group (Collie for example), will take a different approach than when you work with a dog in the hound category (Beagle for example)
What owners do need to be aware of is to give as little as they can get away with where food is concerned. This often needs to be shown to owners; their perception of a little and yours may not be the same! It is all too easy to end up with an overweight dog and the risk of overfeeding is a very real one.
We prefer to use a softer type treat otherwise the dog tends to take time to chew/munch on the treat and the reason for the reward will be forgotten. Another aspect to consider here is that if a dog is full it will not be as willing to work, so work with the dog when it is hungry.
We use a variety of different treats when teaching a pup or even an adult dog new behaviours as this way it will never know what is coming and special treats are used as a jackpot for a difficult exercise, or when the pup/dog has eventually mastered the behaviour requested. Our most training treat of all time is Liver Bread and we have yet to come across a dog that does not agree!
Jackpots
Jackpots are what we use when a puppy does something exceptionally well or if he has been battling with an exercise, the down for example, and then he suddenly performs it, you can reward with a jackpot, which is a special treat, or several treats in a row, really making a fuss about how well the dog has done. It can be compared to a child receiving a gold star at school. A combination of food and toys can be used as well. My own dog learns new behaviours with standard kibble and to him the best jackpot in the world is to have his ball tossed in the air to catch – he would much rather have this than a food jackpot.
5. Fading the Reward
One of the most common statement clients will make to me is “he only does it for food!” Now the last thing any of us want is for a dog to do what we ask only if we have food . How is this solved? Not at all difficult as you will read below.
When you are teaching a dog a new behaviour you reward 100% of the time until the dog ‘get’s it’ or using the correct terminology, until the dog is ‘proofed’ in the exercise. You then immediately start to stagger the reward. When a reward is staggered (in behaviour terms it would be an intermittent reinforcement schedule we will be working with here) a dog will actually work harder for the reward, simply because it never knows when it will be forthcoming! If we relate this to human terms, why do we keep on putting money into one armed bandits at a casino? Simple, we get rewarded for doing so – or at least some of us do! This motivates us to play again.
Don’t always have the treat in your hand either, pop some food into your pocket or even keep some titbits throughout the house. This will stop the dog only working when it can see the food – be a magician! Start fading as soon as possible. A client I had a few years ago called me in for really bad inappropriate elimination problems. We worked on the situation and I told her to really go overboard with the reward when the dog did eliminate in the designated place. All went well and then a few weeks later I got a call, to say that she now takes the dog out to eliminate, but it just stands there looking at her and she does not know what is wrong. I popped over the following day and asked her to show me what she was doing – always much better to observe things for yourself. For the life of me I could not figure what she was doing wrong! I then started questioning her and asked her if anything at all had changed in any way from a few weeks ago – had she changed to a new area in the garden? Has she changed the treats ? I eventually discovered that major change had taken place – to the dog at least – the moon bag which she wore around her waist to put the treats in had broken and she had not used it the last week – bingo!! I asked her to go and get it and take the dog outside again – the result? – the dog eliminated immediately – the dog had associated the reward coming from the moon bag rather than the owner!
So how then do you stagger a reward? When the dog is performing the requested behaviour proficiently, then repeat the request and only offer a reward every second or third time. Mix it up, sometimes every fourth time, sometimes ever second time, until you gradually fade the reward almost completely and can use praise instead. I find that it is better to give the odd reward rather than cut it out all the time – think about it, would you work month in and month out for no pay??
Remember that instead of a food reward, you can also tell the dog ‘good boy’ or similar. Food is not always the be all and end all of rewards, as to some dogs praise is more important, to others a ball being thrown and to some a scratch is the best thing in the world.
A lovely exercise to do with both pups and older dogs to start fading is the ‘3fors’ i.e. the dog is taught to sit and rewarded, the dog is taught the down cue and rewarded, the dog is taught the stand cue and rewarded. When all 3 are in place start to ask for a sit and then a down and then reward. Mix these up being careful not to always use the same sequence and you really don’t have to stop at 3! Using these 3 simple exercises you can practice lots of different sequences and start fading at the same time.
If you come across a situation where in the dog knows the behaviour and suddenly becomes deaf and totally ignores you, stand silently for about 30 seconds to see if the dog will comply (if it does offer the treat immediately and even Jackpot with 2 or 3 pieces), if not, waggle the food treat in front of its nose and say ‘too bad’ and walk away. If there is another dog around, immediately ask that dog for behaviour and when it does comply, give it the treat – this has a great impact on dogs! If there is not another dog around, try the exercise about a minute later in a different location. If the dog complies, then do ‘jackpot’ it by giving two or three rewards very quickly. The dog will work out for itself what works and what doesn’t.
Don’t only have training sessions, start to bring the behaviours, such as sit, down etc into everyday life by asking for a sit before playing, sit and wait before going out a door, etc. Also remember to practice in different locations. You may have a dog that sits perfectly every time you ask in the house, but does that dog do the same in the garden?
One of the most common statement clients will make to me is “he only does it for food!” Now the last thing any of us want is for a dog to do what we ask only if we have food . How is this solved? Not at all difficult as you will read below.
When you are teaching a dog a new behaviour you reward 100% of the time until the dog ‘get’s it’ or using the correct terminology, until the dog is ‘proofed’ in the exercise. You then immediately start to stagger the reward. When a reward is staggered (in behaviour terms it would be an intermittent reinforcement schedule we will be working with here) a dog will actually work harder for the reward, simply because it never knows when it will be forthcoming! If we relate this to human terms, why do we keep on putting money into one armed bandits at a casino? Simple, we get rewarded for doing so – or at least some of us do! This motivates us to play again.
Don’t always have the treat in your hand either, pop some food into your pocket or even keep some titbits throughout the house. This will stop the dog only working when it can see the food – be a magician! Start fading as soon as possible. A client I had a few years ago called me in for really bad inappropriate elimination problems. We worked on the situation and I told her to really go overboard with the reward when the dog did eliminate in the designated place. All went well and then a few weeks later I got a call, to say that she now takes the dog out to eliminate, but it just stands there looking at her and she does not know what is wrong. I popped over the following day and asked her to show me what she was doing – always much better to observe things for yourself. For the life of me I could not figure what she was doing wrong! I then started questioning her and asked her if anything at all had changed in any way from a few weeks ago – had she changed to a new area in the garden? Has she changed the treats ? I eventually discovered that major change had taken place – to the dog at least – the moon bag which she wore around her waist to put the treats in had broken and she had not used it the last week – bingo!! I asked her to go and get it and take the dog outside again – the result? – the dog eliminated immediately – the dog had associated the reward coming from the moon bag rather than the owner!
So how then do you stagger a reward? When the dog is performing the requested behaviour proficiently, then repeat the request and only offer a reward every second or third time. Mix it up, sometimes every fourth time, sometimes ever second time, until you gradually fade the reward almost completely and can use praise instead. I find that it is better to give the odd reward rather than cut it out all the time – think about it, would you work month in and month out for no pay??
Remember that instead of a food reward, you can also tell the dog ‘good boy’ or similar. Food is not always the be all and end all of rewards, as to some dogs praise is more important, to others a ball being thrown and to some a scratch is the best thing in the world.
A lovely exercise to do with both pups and older dogs to start fading is the ‘3fors’ i.e. the dog is taught to sit and rewarded, the dog is taught the down cue and rewarded, the dog is taught the stand cue and rewarded. When all 3 are in place start to ask for a sit and then a down and then reward. Mix these up being careful not to always use the same sequence and you really don’t have to stop at 3! Using these 3 simple exercises you can practice lots of different sequences and start fading at the same time.
If you come across a situation where in the dog knows the behaviour and suddenly becomes deaf and totally ignores you, stand silently for about 30 seconds to see if the dog will comply (if it does offer the treat immediately and even Jackpot with 2 or 3 pieces), if not, waggle the food treat in front of its nose and say ‘too bad’ and walk away. If there is another dog around, immediately ask that dog for behaviour and when it does comply, give it the treat – this has a great impact on dogs! If there is not another dog around, try the exercise about a minute later in a different location. If the dog complies, then do ‘jackpot’ it by giving two or three rewards very quickly. The dog will work out for itself what works and what doesn’t.
Don’t only have training sessions, start to bring the behaviours, such as sit, down etc into everyday life by asking for a sit before playing, sit and wait before going out a door, etc. Also remember to practice in different locations. You may have a dog that sits perfectly every time you ask in the house, but does that dog do the same in the garden?
6. Cues, Signals, Commands
When you give a pup or dog an instruction, it can be termed any of the above. We prefer the word Cue, as this means giving the dog an indication or prompt for the behaviour required. We do not resonate with the word command, as it implies that forcing a dog to do something rather than having the dog work with/for us. This is all a matter of personal choice and for example,we prefer to make use of the word signal, for when using the body to indicate to the pup/dog what we want.
What is very important is that the cue you do give must always be the same. The importance of consistency cannot be overlooked. By giving each behaviour requested (e.g. sit, down, here) a different name, and sticking to this, we are giving the pup a clear indication of what we expect. We need to bear in mind that not only must the owner be consistent, but that every member of the family is as well.
The better an owner is at associating the same cue with the same behaviour, the quicker a dog will learn.
The main things to remember about Cues:-
Why Cues don’t always work
This normally starts happening when an owner will repeat the verbal cue over and over again. We will call our dog to come here, nothing happens and then we repeat the process and more often than not our voices will get shriller and shriller. We are now starting another problem; our dogs just don’t want to come back to an angry owner! This can happen several times a day and in no time at all we will find that we have conditioned our dogs to ignore us.
Another reason cues don’t always work is due to inconsistency on the owner’s part. The importance of consistency cannot be overlooked. By using the same cue for each different behaviour, we are giving the pup/dog a clear indication of what we expect. We need to bear in mind that not only must the owner be consistent, but that every member of the family is as well.
Two other aspects why cues don’t work is that the owner often forgets to get the dogs attention before giving a cue, and secondly, the owner may be training a new behaviour and omit to give a reward for same. I have had owners tell me that their dogs are disobedient but I would rather say that the owners have failed at their attempts to get their dogs to do what they do want, in other words, the training program has failed in some way – there was no communication.
Another aspect that often causes problems is due to poor timing when it comes to delivering the reward. The reward must be given at the exact moment the dog performs the behaviour correctly.
Tone of Voice
This is an important part of training a pup or dog and needs to be explained to owners. If I am leaning over, screaming “Fido come!” not only does my voice sound aggressive, my body language of bending over is also threatening to a dog.
By the same token, owners who plead with their dogs to perform a behaviour by saying ‘sit, Fido, please sit” etc, just become white noise to their dogs and they are seldom listened too.
Women have the most natural voice to use for dog training – it is often high pitched, softer, and could even be described as a falsetto. However, women tend to sometimes be a bit shrill, so, with an excited dog especially, it is better to lower the tone of the voice, but still keep it soft and happy.
Men, on the other hand have more problems with the tone and often tend to have the ‘bad dog’ tone of voice, which is great for when a reprimand is given, but not so great when calling the dog!
I find that with all dogs, we need to give our cues in a firm manner as well with the voice being slightly lower than usual. Dogs, as people, will listen to a much lower tone of voice and will be more likely to pay attention. Do you remember the teacher at school (we all had at least one!) who would speak in a quiet, low voice and you had to almost strain to hear her? Well guess what, it worked with you and it works with your dogs as well!
Another problem with cues from owners is that they have a tendency to voice the cue as if it was a question ‘waaaaiiiiiittt?’ Rather use a firmer voice, a hand signal at the same time and say ‘wait’. It is a good idea to play around with different tones of voice yourself and see for yourself which works well with your own dogs. Do the same thing with your client’s dogs, you will quickly see what manner is the best for the dog, demonstrate to your client and this will make the training so much easier, as once again, the owner is giving firm, easily understood cues.
Signals or Visual Cues
Although dogs can discriminate between different words and can learn many, we have to remember that dogs are more likely to watch our body gestures and hand signals. If we think about it, the majority of communication between dogs is based on visual cues, so this is an area we can take advantage of to teach our dogs. We can then give our dogs signals without using our voices at all. This makes more advanced training much easier and can even save your dog’s life in the event you can lift up your hand for the dog to wait, and even ‘down’, when in a dangerous situation.
When the owner starts training the pup, get them to give both vocal and visual cues every time. When training the sit cue, get the owner to practice bending their hand and arm upwards just before they give the cue for sit. This way the dog will associate the hand signal with the spoken cue. When teaching the pup to lie down, put your hand and arm, with the palm facing downwards and moving the arm towards the floor just before giving the cue for down. When we ask the dog to wait, we are once again making use of the hand signal (arm bent upwards in the stop position).
Context Cues
We may not be aware but we also give our dog’s context clues without even realizing it. Take the example of my own dog Brady. If I put on either one of my two pairs of walking shoes he knows immediately that a walk is forthcoming without me having said a single word or picked up any piece of equipment I normally take with me. When we are dealing with a dog that suffers from separation anxiety, the context cues we give assume a greater importance and our behaviour of simply picking up our handbag and car keys can signal to the dog that we are on the way out the door. Owners will often say ‘but the dog knows when I am leaving’. Of course it does, the owner just gave it all the context cues necessary! The clothes we wear for different activities will be noticed by our dogs as well as what we do at different times of the day.
We can use these context cues to further train our pups in routine situations. A good example of this is you do not put on the lead until the pup sits. The pup quickly learns that the quicker it sits, the quicker you are to put on the lead and then the good times begin – a walk is forthcoming!
When you give a pup or dog an instruction, it can be termed any of the above. We prefer the word Cue, as this means giving the dog an indication or prompt for the behaviour required. We do not resonate with the word command, as it implies that forcing a dog to do something rather than having the dog work with/for us. This is all a matter of personal choice and for example,we prefer to make use of the word signal, for when using the body to indicate to the pup/dog what we want.
What is very important is that the cue you do give must always be the same. The importance of consistency cannot be overlooked. By giving each behaviour requested (e.g. sit, down, here) a different name, and sticking to this, we are giving the pup a clear indication of what we expect. We need to bear in mind that not only must the owner be consistent, but that every member of the family is as well.
The better an owner is at associating the same cue with the same behaviour, the quicker a dog will learn.
The main things to remember about Cues:-
- Use clear signals.
- Always be consistent and use the same cue every single time in every single situation.
- Use simple cues that are both easy for you to remember and also for the dog to understand. An example of this is that the owner requests a ‘sit’ from the pup. They do not say ‘will you sit down now please’. This sounds so simplistic, but you cannot believe the amount of people who do this!
- Tone of voice
- The above applies to all member of the family.
Why Cues don’t always work
This normally starts happening when an owner will repeat the verbal cue over and over again. We will call our dog to come here, nothing happens and then we repeat the process and more often than not our voices will get shriller and shriller. We are now starting another problem; our dogs just don’t want to come back to an angry owner! This can happen several times a day and in no time at all we will find that we have conditioned our dogs to ignore us.
Another reason cues don’t always work is due to inconsistency on the owner’s part. The importance of consistency cannot be overlooked. By using the same cue for each different behaviour, we are giving the pup/dog a clear indication of what we expect. We need to bear in mind that not only must the owner be consistent, but that every member of the family is as well.
Two other aspects why cues don’t work is that the owner often forgets to get the dogs attention before giving a cue, and secondly, the owner may be training a new behaviour and omit to give a reward for same. I have had owners tell me that their dogs are disobedient but I would rather say that the owners have failed at their attempts to get their dogs to do what they do want, in other words, the training program has failed in some way – there was no communication.
Another aspect that often causes problems is due to poor timing when it comes to delivering the reward. The reward must be given at the exact moment the dog performs the behaviour correctly.
Tone of Voice
This is an important part of training a pup or dog and needs to be explained to owners. If I am leaning over, screaming “Fido come!” not only does my voice sound aggressive, my body language of bending over is also threatening to a dog.
By the same token, owners who plead with their dogs to perform a behaviour by saying ‘sit, Fido, please sit” etc, just become white noise to their dogs and they are seldom listened too.
Women have the most natural voice to use for dog training – it is often high pitched, softer, and could even be described as a falsetto. However, women tend to sometimes be a bit shrill, so, with an excited dog especially, it is better to lower the tone of the voice, but still keep it soft and happy.
Men, on the other hand have more problems with the tone and often tend to have the ‘bad dog’ tone of voice, which is great for when a reprimand is given, but not so great when calling the dog!
I find that with all dogs, we need to give our cues in a firm manner as well with the voice being slightly lower than usual. Dogs, as people, will listen to a much lower tone of voice and will be more likely to pay attention. Do you remember the teacher at school (we all had at least one!) who would speak in a quiet, low voice and you had to almost strain to hear her? Well guess what, it worked with you and it works with your dogs as well!
Another problem with cues from owners is that they have a tendency to voice the cue as if it was a question ‘waaaaiiiiiittt?’ Rather use a firmer voice, a hand signal at the same time and say ‘wait’. It is a good idea to play around with different tones of voice yourself and see for yourself which works well with your own dogs. Do the same thing with your client’s dogs, you will quickly see what manner is the best for the dog, demonstrate to your client and this will make the training so much easier, as once again, the owner is giving firm, easily understood cues.
Signals or Visual Cues
Although dogs can discriminate between different words and can learn many, we have to remember that dogs are more likely to watch our body gestures and hand signals. If we think about it, the majority of communication between dogs is based on visual cues, so this is an area we can take advantage of to teach our dogs. We can then give our dogs signals without using our voices at all. This makes more advanced training much easier and can even save your dog’s life in the event you can lift up your hand for the dog to wait, and even ‘down’, when in a dangerous situation.
When the owner starts training the pup, get them to give both vocal and visual cues every time. When training the sit cue, get the owner to practice bending their hand and arm upwards just before they give the cue for sit. This way the dog will associate the hand signal with the spoken cue. When teaching the pup to lie down, put your hand and arm, with the palm facing downwards and moving the arm towards the floor just before giving the cue for down. When we ask the dog to wait, we are once again making use of the hand signal (arm bent upwards in the stop position).
Context Cues
We may not be aware but we also give our dog’s context clues without even realizing it. Take the example of my own dog Brady. If I put on either one of my two pairs of walking shoes he knows immediately that a walk is forthcoming without me having said a single word or picked up any piece of equipment I normally take with me. When we are dealing with a dog that suffers from separation anxiety, the context cues we give assume a greater importance and our behaviour of simply picking up our handbag and car keys can signal to the dog that we are on the way out the door. Owners will often say ‘but the dog knows when I am leaving’. Of course it does, the owner just gave it all the context cues necessary! The clothes we wear for different activities will be noticed by our dogs as well as what we do at different times of the day.
We can use these context cues to further train our pups in routine situations. A good example of this is you do not put on the lead until the pup sits. The pup quickly learns that the quicker it sits, the quicker you are to put on the lead and then the good times begin – a walk is forthcoming!