Are you bribing or rewarding your dog?

In a couple of my recent posts, I’ve tried to counter some of the arguments which some well intentioned although, misinformed dog people, use as arguments against some types of training methods. Here I’ll discuss some of the arguments more traditional, or force based, trainers use to illustrate why we shouldn’t use reward based training exclusively when training our dogs. Again, to reiterate, I am a 100% force free trainer and behaviour consultant.

The first thing to discuss is the scientific explanations of some aspects of learning theory which can cause some inexperienced reward based trainers to become unstuck. Firstly, a few terms and their explanations.

Fixed rate of reinforcement – this is when an dog receives a fixed number of rewards for a fixed number of behaviours. An example would be 1:1, where one behaviour gets one reward. A vending machine pays out on a fixed rate, money goes in, can of Coke comes out. Every time. Other examples could be 2:1 or 3:1, where the animal is reinforced after two and three behaviours respectively.

Variable rate of reinforcement – this is when the reinforcement occurs randomly. It is usually done on average. So, you could have a random variable rate of 3, which means the animal is reinforced after the 1st, then 5th, then 2nd, then 4th behaviour. A slot machine/fruit machine pays out on a variable rate of reinforcement. We keep playing for the chance of a pay out.

So now that the science is out the way, here is how we can apply (or misapply!) it.

When teaching a new behaviour, we want to reward our dog every time he does the behaviour. So when teaching a dog to sit, every time he sits, he is rewarded. Now the reason we do this is to teach him that it is worth his while to do it. When your dog reliably sits on request, it is time to move to a variable rate. The reason we do this is because if we don’t do it quickly enough, the dog will no longer work unless it is being reinforced every time. To use the vending machine example, if the Coke machine swallows your money, you may put in another coin. If it eats that amount, no one puts in more money in the hope that it will pay out this time. You assume the machine is broken and chalk is down to a bad experience. So, back to dog training. If we use a fixed rate of 1:1 for too long, when you try to stop doing it, your dog thinks you are “broken” and stops working.

Now, to add insult to injury, this isn’t as bad as it gets. Say we are using food to train. If we don’t get the food out of our hands quickly enough, the dog then doesn’t work unless he sees the food. At this point we are bribing the dog, not rewarding him. The sequence goes like this:

Bribing; food in hand, dog sees food, dog does behaviour requested, dog gets food (dog only does behaviour if he sees food)

Reward: food is hidden( e.g. in pocket) dog does behaviour requested, dog is rewarded with food for good work (dog is willingly working for the chance of reward)

To make the dog willingly work for the chance of reward, we need to put him on a variable rate of reinforcement as soon as we can. To do this, we begin with low averages, which means we reward the dog, on average, every second or third behaviour. When your dog is showing progress at this level, we can then increase the average to every fourth or fifth behaviour and so on. However high you want to set the bar is up to you, what you want to achieve and what your dog is capable of. Some dogs are willing to keep working and working (border collies are a good example, although this isn’t cast in stone). Other dogs reach a point of diminishing returns where they decide that the level of work they are offering isn’t worth the payout ( my own mastiff breeds for example). Each dog is different, as is the skill level of each owner/trainer.

Not applying the science of learning theory properly, can lead to more traditional trainers calling us “treat dispensers”. Proper application takes a good understanding of the science and capablitly to do it, which usually comes with experience.

To wrap up, a bit of practical advice if you have been bribing or using a fixed rate of 1:1 for too long. If you have been bribing your dog, start by not giving the dog every time. So you might have your treat in your hand and only give the dog the treat 8 times out of 10 (this is a rate of 10:8, ten behaviours for eight reinforcements) and reduce this number. Then try putting the treat away but increase your rate of reward.

If you have fallen into the trap of rewarding your dog every time, do likewise. Offer rewards 8 out of 10 times (10:8), then 7 out of 10 (10:7), then 6 out of 10 (10:6 or (5:3). When you get to about five out of ten (10:5 or 2:1) start to gradually introduce the random rate. This might take a few weeks, but if we reduce the reinforcement rate slowly enough,we should be able to rectify our mistakes. Push your dog enough that you make progress, but not too hard that she stops working, If she is finding it too difficult, go back a step, or half a step and try again.

Advertisements

What kind of trainer are you?

Positive reinforcement, traditional, balanced, non punitive. These are all terms dog trainers and behaviour consultants use to describe their styles. I use the terms non-aversive, progressive or force free to define how I train dogs and help people to train dogs.

When I read that someone is a “balanced” trainer, I would suggest that it is usually a euphemism for trainers who use physically punitive methods when training, as well as reward based training. This is a clever marketing strategy as some of these trainers would have us believe that a “balanced” approach of both harshness and reward based training is necessary. There seems something quite appealing about a balanced approach in theory and it is only when we look at it more closely that we discover what many (not all) of these trainers actually mean by the term. This idea is further promoted by Cesar Milan when he says we need to “balance” our dogs, which could in turn suggest a more “balanced” approach.

Traditional trainers are ones who mainly employ compulsion based training methods, such as ear pinches and choke chains. The get results, they just don’t have a very good relationship with their dogs in many cases as, unless they use some other type of more reward based training, the dog behaves in the desired way in order to avoid painful or unpleasant consequences.

Now, before I move on to positive reinforcement (PR), I’ll define a few terms. Operant learning was first documented by B.F Skinner in the 1930s as part of his Masters degree and Doctorate. There are four elements to this, positive reinforcement, positive punishment, negative reinforcement and negative punishment. The term “positive” does not mean good, it means to add something to the environment. Conversely, negative means to subtract something from the environment. Think in terms of the mathematical + and – signs. Reinforcement means the preceding behaviour is likely to increase in frequency, duration and/or intensity. Punishment is the opposite of reinforcement, and leads to a decrease in the preceding behaviour.

Now the reason understanding of these terms is important is that a misunderstanding of them leads to misinformation being promoted as fact. Trainers who say they only use PR and do not use punishment are misinformed and can, as a result, mislead others. They may use PR primarily

Training loose leash walking using force free methods

but an element of punishment is always inevitable. But punishment does not have to be harsh/physically punitive/psychologically damaging/aversive. If we don’t want our dogs to jump on us and our dogs learn that by jumping, they receive no attention and stops doing it, then jumping is punished. If we play tug and the game ends if our dog misses the toy and catches our skin, then we have used punishment to stop the “teeth against skin” behaviour. If our dog body slams other dogs and we stop play and this reduces the slamming, then we have punished the slamming. As you can see in these three examples, there is nothing harsh or painful about what we are doing but we are imposing rules on ours dogs.

The term positive seems to have been partially hijacked and as a result, it has been misrepresented. Positive isn’t necessarily a good thing when applied to learning theory. I’m all for calling your dog business “Positive” for marketing purposes, but we need to know the science behind it to back it up and not misrepresent it.

I regularly use punishment in dog training but I don’t use harsh or physically punitive techniques. I find them unnecessary and in many cases, counter productive. It just depends where we draw the line about what we are willing to do to our canine friends in the name of training.

John McGuigan

http://www.glasgowdogtrainer.co.uk

Canine Social Superstars – part 2

Happy Hooper

In part one, I discussed how the way we breed and rear our dogs has effected their behaviour over the last 30 years.

On leash – dogs are now walked on a leash as they are no longer allowed to roam around the streets. Here in the UK, it isn’t required that they are on a leash in the street but since many pet owners don’t have the necessary bond with their owners or the required level of training, their dogs are rarely, if ever, off leash. The fallout of this is that their dogs never learn proper interaction with other dogs as the leash restricts their movement, body language and as a result their inter-dog communication.

Because they don’t learn this as much as they need to, many dogs greet each other inappropriately which can lead to fights. The dog who started the fight is often not the one who looks like the aggressor, so your social, civil and well adjusted dog gets a hard time from the other owner for rightfully telling another dog not to be so pushy, but it can look different to the uneducated. As a result of this, canine interactions can diminish in frequency and social skills reduce. They don’t learn the off leash communication they used to learn by having freedom to roam around the neighbourhood which they used to enjoy.

Our dogs now need not only to be friendly to other dogs but also to be able to tolerate, without reacting, other dogs which are rude, anti-social or aggressive. This is a lot to ask.

Central heating – Central heating became common place in the UK in the late 70s. Prior to that, coal or wood fires heated single rooms, mainly livingrooms, bedrooms and kitchens.  Internal doors were closed to keep heat in. As a result of this, dogs had less access to the whole house. They would either be in the livingroom along with the rest of the family where they could be monitored more closely, or they were sleeping alone in the kitchen.

In these times, internal doors in the house are open as the house is fully heated. This gives our dogs relatively unrestricted access to the house, which means the kitchen counter, the front door, windows looking onto the street, your shoes, wash basket, etc are more readily accessible and with that massive opportunity trouble.

Changes in societal attitudes – without sounding too cynical, attitudes in society have also changed. My feeling is that there is more “me” and less “we” in society which in turn means that increasingly more and more people aren’t willing to train their dogs for the community’s benefit, accept responsibility when their dog does something wrong or anti-social, and are more likely to blame others for their own mistakes and their dog’s mistakes.

In conclusion, due to all of these factors, we now need to change the way we, as dog owners, educate our own dogs to cope with living in a modern world. Obviously there are no absolutes here but hopefully you get the idea. Proper exercise, both physical and mental is called for. Tonnes of classical conditioning during puppy hood, adolescence and into adulthood is called for so to make up for the reduced natural learning which was more commonplace in the recent past and so that our dogs have a wealth of positive experiences to fall back on should somethings stressful happen. Classical conditioning to loads of stimuli in the environment will hopefully offset deficiencies in temperament which can occur due to modern breeding practices. This isn’t to say that there aren’t a plethora of well educated, responsible, caring owners. There are, I come into contact with them on a weekly basis. But there are also loads who aren’t. This article is an effort to explain why we get some of the issues with our dogs in society and what we can do to combat it. Giving your dog something do do when you are out, such as eating from a frozen Kong, means he is less likely to chew your banister and bark at everyone walking past your front door, for example. There are loads of things we can do to offset these challenges which have emerged over the last couple of decades. We owe it to or own dogs and our communities to do it.

Wishing you responsible and happy dog ownership,

John

Glasgow Dog Trainer and Behaviour Consultant

http://www.glasgowdogtrainer.co.uk

Canine Social Superstars – part 1

Our lifestyles with ours dogs have changed over the last 40 years and subsequently, the way we breed, rear, train and live with our dogs has changed also. Much of what I’m posting here I’ve learned from the work of John Rogerson, Ian Dunbar and Stanley Coren as well as my own experience and interpretation.

Breeding – I grew up in a large new town outside of Glasgow. Living conditions were good and although we had housing estates they were not the same as the ones in the inner cities of Glasgow, London or Birmingham. As a child, I remember peoples’ pet dogs roaming around the streets. They were generally let out in the morning, would potter about and come home at dinner time to be fed. Of the dog’s I remember from my childhood, they were mainly collie-type mixed breeds. I do remember a few pedigree dogs, mostly terriers, a Dalamtian called Nick (who we knew better than to try to pet) and the Italian guy who owned the local chippie (fish and chip shop to non UK readers) who had a huge wolf sable German Shepherd called Lupo. The mixed breed dogs bred naturally as the bitches selected which dogs they wanted to mate with and, although I didn’t know it at the time, choose the big strapping lads with good bone structure and good temperament. Ian Dunbar’s PhD work investigated this among other things. He observed that a bitch would happily pal around with one or two dogs throughout the year but when she came into season, she wanted nothing to do with her pals and would seek the attention of the stronger, more stable dogs. She chose which dog to mate with. After her season ended, she would happily make up with her friends.

Breeding is now done by the breeders. Pedigree dogs are bred, not always for profit as the main goal but often so and professional show breeders often breed for looks which conform to a breed standard and not primarily for temperament. My opinion is that this is the wrong way round, as the majority of these dogs end up as pets and not show dogs. As a result, temperament suffers and we no longer the same amount of puppies which are suitable as family pets. The spaying and neutering programme which has been hugely encouraged has not stopped the flow of unwanted dogs into rescue centres and shelters. It’s just the dogs in there now are mainly pedigree dogs and have more training and behaviour issues due to poor temperament and owner lifestyle than they did before.

Rearing – 40 years ago couples got a dog after they were married and had a child or two. Dad would work, Mum stayed at home and looked after the kids. She would put the kids out to school (we used to walk to school then, in all weathers) and Mum would then walk the dog (which they had been given by the neighbour whose bitch had a litter) down to the local shopping centre where the dog would be left outside (not always tied up) while she did the shopping. As a result of this, the dog got used to being ignored in the house as other things were going on, got used to being outside walking, got used to being left on its own both inside and outside and got plenty of exercise and got used to greeting people (or not) off leash. Mum got back to the house and the dog was let out into the street where is romped around with other dogs so it learned to play with other dogs properly and with natural body language and communication. Dog fights were rare ( I never saw one) and bites were almost unheard of (I can vaguely remember Nick the Dalamation snapping at someone before we could tell them not to touch him).

So, because of the above, dogs were well exercised, had daily stimulation and play and were well socialised with people and dogs. It is different now. In many cases these days, couples move in together and buy a pedigree dog, perhaps simply because they like the look of it, perhaps because they have heard about certain breeds of dog being more adaptive to a modern lifestyle, and perhaps because they have not yet contemplated children. The dog now spends loads of time in doors by itself as they are both working. They come home from a day’s work, tired. If they do take the dog out, it’s either round the block on the leash while they talk on their mobile phone or their pedigree dog (insert breed but think Husky for the sake of this discussion) gets driven to the park where it is let off to play freely without any human interaction or intervention or structured play and training and then driven back home again, sufficiently energised to wing off the walls and torment the daylights out of them for the remainder of the evening.
Not every owner or every breeder behaves like this b in general terms, these things have changed the way our dogs behave.

In part two I’ll discuss how keeping our dogs on leash, how we heat our house and changing general attitudes in society have influenced our dogs’ behaviour.

Use of physical punishment in training – why it works and the harm it can do (part 1)

Before I start this post, I want to preface it by saying I am now a 100% non aversive trainer and have been for some years now. I can’t remember the last time I shouted at a dog other than my own (which was a long time before I knew any better) and don’t even say “No” any more, rather I may repeat the cue to give the dog another opportunity to respond.

If any of you have read my first post, you will know that I came from a background of traditional dog training. I took my young Dogue de Bordeaux to a sports dog club, where the use of choke chains and prong collars was common place and shock collars were sometimes seen.

The reason I used metal collars was because

1. I wasn’t shown anything different and

2. I was getting results.

My dog Bosco was a terrible puller on the lead and I taught him, in the space of about 5 minutes, to walk to heal using a prong collar. I also taught a great down stay, sit stay and recall, all using a prong collar. Prong collars work, that’s why I used them and that’s why people continue to use them. Now I wasn’t a barbarian who enjoyed hurting my dog. I loved my boy and wanted what was best for him which was to mind his manners and do as I asked of him, I just went about it the wrong way. In my experience this is generally true of most dog trainers (always exceptions of course).

Prong collars work by, according to the operant model, positive punishment. The positive part is a plus sign(+) where something is added to punish (reduce) the undesired behaviour. So in the instance above with Bosco, pulling on the lead, getting up from a stay and not coming back were all punished/reduced by me adding a correction (i.e. painful experience) with the prong collar. The father of operant training, B.F. Skinner did this in a lab using rats. Rats were put in a box where, upon pressing one lever they obtained a food pellet and on pressing another one got an electric shock. It doesn’t take too long before the rat learns which button to press and which to avoid.

As progressive/non-aversive trainers, we need to understand why other “balanced” trainers do what they do. I recently had a discussion on a web training forum on this issue. The other person argued that using force in training doesn’t work. She argued that I hadn’t trained a down stay from Bosco, merely taught him not to get up, that I hadn’t trained heeling, merely punished him for pulling on the lead. My response to this is, really what is the difference? The picture looks that same, Bosco stays down and walks on a loose leash.

I think arguing against the effectiveness of forceful training is futile. What we should be doing is educating about the fallout or effect it has on our relationship with our dogs. I am currently making my way through Steve White’s DVD set, “How Police K9 techniques can transform your everyday training”. Steve is a progressive trainer who has trained police dogs forever. In the seminar, he states that when punishment is being used (he’s not condoning it), the dog should not associate the punishment with the handler. This is because it breaks down the bond of trust between handler and dog. The dog is less likely to trust someone who hurts it, which is hugely problematic when your dog is your police partner. The breakdown of handler/dog relationship is the best which happens. The worst that can happen is that you end up with a dog who is frightened and/or aggressive to the environment.

For instance, I taught Bosco to be super reactive to other dogs. I did this really well. He pulled because he wanted to say hello to other dogs and people he met. The harder he pulled, the more I corrected him. The more I corrected him the more he associated other dogs in the environment with pain and the more reactive he became as he learned that the pain stops when the dog goes away. What could have happened because he knew it was me who was hurting him, would have been to turn around and attack me. That would be among the worst things which could have happened.

By understanding why traditional trainers do what they do, knowing our own craft inside out so that we can produce better results and be able to demonstrate that our dogs are working for us because they want to and not because they fear not doing it, rather than dismissing these techniques as ineffective then hopefully we can continue to change things.

Part 2 – https://glasgowdogtrainer.wordpress.com/2013/01/05/use-of-punishment2/

Bosco without his prong collar, working for the ball and much happier for it

BATting with Reggie

I attended Grisha Stewart’s Behaviour Adjustment Training (BAT) seminar last weekend. I have been using BAT both on it’s own and in conjunction with other behaviour modification methods since I read Pat Miller’s article on the subject in the Whole Dog Journal in 2010. Since I was first made aware of BAT, I’ve also used elements of it in working with clients and there dogs. The weekend was excellent, loads of new information and I got to catch up with some people I hadn’t seen in a while and meet some new contacts too.

The basic premise of BAT is that it uses functional rewards. These are things which the environment offers rather than we offer such as the feeling of relief a scared dog experiences when a dog it is frightened of goes further away or the chance to play with another dog or sniff the grass. From Grisha’s website http://www.functionalrewards.com, Grisha defines BAT as

“Behavior Adjustment Training, or BAT, rehabilitates dog reactivity by looking at why the dog is reactive and helping him or her meet his needs in other ways. In a nutshell, BAT is a dog-friendly application of ‘functional analysis’ that gives the dogs a chance to learn to control their own comfort level through peaceful means. It’s very empowering to your dog, in a good way.”

Eager to put my new found knowledge into practice, I had a session with Reggie, the adolescent Boxer, the following week. Reggie is a great dog, very typical of a Boxer at his age. He is eager to please, super friendly with people but can lack manners in greeting other dogs. He is very keen to meet other dogs but pulls on the leash to get to them (and this isn’t always possible) so much so that he has ripped his pads a few times against the ground and is now starting to show signs of frustration when he doesn’t get to other dogs quick enough or he doesn’t get to greet them at all.

We met at a great park where I know loads of the owners and their dogs, so I was able to choose appropriate dogs for us to work with. We made some really good progress and it took us about 20 minutes until Reggie was able to approach the fenced in area where there was another dog playing (here in the UK we are extremely fortunate that we can let our dogs off leash in public parks and aren’t limited to using dog parks, but some parks do have fully enclosed dog parks also).

We were about 40 minutes into our session when we met Evan who is about a year old. On speaking to Evan’s owner, he told me that he had adopted Evan from the Dog’s Trust a couple of months earlier. Even is a mixed breed about the size of a Labrador. Evan was also a canine social superstar. His communication skills were among the best I’ve ever seen from a dog. Evan gave Reggie enough signals to convey to him that he needed to approach with a bit less energy than Reggie was accustomed to. It also helped that we had been “BATting” Reggie for about 40 minutes at this point.

Reggie greeted Evan from a side angle, they did their “butt sniffing” dance and then Evan invited Reggie to play by bowing and then skipping away, asking for Reggie to follow. Reggie duly obliged but at about 100mph and with as much restraint as a child in a sweet shop. Evan immediately told Reggie that this level of energy wasn’t acceptable by stopping running and sniffing the ground, which is a calming signal for both dogs.  You could almost hear Reggie thinking

“What happened there, I thought we were playing?”

Reggis started sniffing the ground too. As soon as he did this, Evan bowed again and skipped off, telling Reggie the game was back on. Reggie responded but went straight to 30,000ft and Evan responded by ending the game as before. Reggie calmed down after a few seconds and you could almost hear Evan thinking

“Will we try this again and see if you understand the rules this time?”

Reggie then played but with much less intensity as before. When dogs play chases, they usually take turns chasing and being chased. Some dogs like being chased but if they do, you can usually tell by relaxed body language and not a panicked look in it’s eyes. If you seen this from your dog when it is being chased or when your dog is chasing another, you need to intervene and stop the chase as it is no longer fun for the “chasee”.

At this point, Reggie stopped chasing Evan, bowed very slightly and then hopped off with Evan chasing him. Real progress. They then played for several minutes and I asked Bryan, Reggie’s owner to call him a couple of times during this time and then release him to go and play with Evan again, so Reggie was still paying attention to him. The whole exchange from initially seeing Evan and Reggie bowing to Evan took less than three minutes. It was a joy to watch, and I wish I had my video camera with me so I could have captured it.

It took us about 40 minutes for Reggie to begin to understand what the rules were when we were trying to teach him. Evan did it in about 3 minutes. We don’t always have superstars like Evan available, but we can use them when they appear. Other than that, the 40 minutes it took Reggie the first time will become 30 minutes before long and when he really understands the rules of what we are trying to teach him, will quickly take less than 10 minutes, then 5 and before we know it and with enough practice, will become his normal behaviour. It’s worth it for Reggie’s sake, for the sake of our own joints so we aren’t getting our shoulders and backs hurt and for other dogs and owners because we owe it to them to raise and teach a sociable dog.

Reggie enjoying his Kong after training