There’s more than one way to train a dog. A walk through the “Pets” section of a book store or a quick search of Google for “Dog Training” will give you a dazzling array of methods, techniques, and “systems” to get your dog trained just the way you always wanted. Basically, to my eye, they break down into three general camps: the “Do it or else!” methods that we sometimes called “Traditional” dog training, the “Do it for this!” methods that use positive reinforcement to reward desired behaviours, and the growing group of “Hybrid” methods which advocate selective use of both “Traditional” and positive reinforcement techniques in varying measure.
To say that dog training represents a sizable revenue stream is an understatement. International pet store chain PetSmart is estimated to derive over $4.5 million annually from their dog training services. According to a report by AMR Research, PetSmart reports that 1% of their overall sales come from dog training services. This says nothing about the other large dog training service chains and the literally thousands of independent dog training service providers in local markets.
So is there money at stake in defending a particular dog training methodology if you are a professional dog trainer? You bet there is!
Each camp has its experts, proponents, and supporters. “Traditional” training has been with us for decades, possibly even centuries. It has a long history. Training based on Positive Reinforcement and behavioural science is relatively new and has only got widespread attention and exposure in the last 15 or 20 years. The results of positive reinforcement training, specifically Clicker Training, have been nothing short of astounding. Behaviours and achievements in dog sports that took months or years to achieve using Traditional methods are now being accomplished by skilled positive trainers in a matter of days and weeks. Yet in the face of such amazing success, a number of misconceptions and half-truths have started to circulate about clicker training and positive reinforcement.
Myths About Positive Reinforcement and Positive Training
Positive Reinforcement training only works with small/soft/easy to train dogs — This is one of my favorites. My dogs are all trained using positive training techniques and behavioural science. My Belgian Shepherds (Groenendaels) are considered one of the difficult to train dog breeds. Our dogs do many dog activities and sports including dog agility (two have earned championships), dog obedience, tracking, herding, and others. In addition, they are wonderful companions in our homes and regularly socialize with friends and family — and they have lots of dog friends at the dog park. In fact, hundreds of species have been trained using positive reinforcement, including: horses, hyenas, rhinos, bears, dolphins, killer whales, cats, lizards, and many more. If you can’t train your Malamute with it, you’re not doing it right!
Using food in Positive Reinforcement training is just bribing the dog — By definition, a bribe is paid before you get what you wanted. When you give something after you get what you want, it is considered a reward or payment. Our human society is full of examples of payments and rewards beginning with the pay cheques most of us receive for our work each week. Payment or reward can also come in the form of a promotion, extra time off, or a trip paid for by the company. All of these things are offered as motivators to provide incentive for us to do a good job, to do good work. It just so happens that dogs value food a great deal and it makes a perfect substitute for money when it comes to paying them for their efforts. Where bribes represent payment for a promise of something later, food rewards used in Positiive Reinforcement training amount to payment for working with us. No more and no less.
Using a toy to reward your dog with tugging is dangerous/makes your dog aggressive — There is absolutely no scientific evidence to support the assertion that playing “tug” with your dog makes them more aggressive. One of the key lessons for the dog in Positive Reinforcement training is “Cause and Effect”: if you do X for me, you will be rewarded with Y. One of the first things we teach our dogs is to “drop it” and “take it” when playing with a toy. If our dogs don’t “drop” the toy, we will stop the tug game. So, quite the opposite of making our dogs aggressive or dangerous, teaching them to play “Tug” with us using OUR rules of “drop it” and “take it” teaches them to be controlled. We can get them to drop anything at any time because they are frequently rewarded for letting go of things.
Trainers who use Positive Reinforcement have unruly/unreliable dogs because they don’t say “NO!” to their dogs — This is probably the most misunderstood aspect of positive training. Positive Reinforcement can be used to teach behaviours but it can’t be used to stop behaviours. Every dog will do something undesirable and a good trainer will use something to stop or interrupt unwanted behaviours. Properly timed punishments or corrections or even the removal of a desired item or a “time out” can be used to communicate that a behaviour is unacceptable. Ideally, the owner can find an alternative behaviour they would prefer and teach that rather than simply saying “No!” but in some cases there is no alternative. Positive Reinforcement trainers tend to have a great sense of timing and so when they do need to correct or punish their dogs, they frequently need to do it less often and with less intensity than other trainers. The goal is to teach the dog what you WANT instead of what you DON’T WANT.
And There’s More!
You might think there would be a lot of debate about these myths on the internet already — and it’s certainly out there. Yet these myths persist. Even in the face of scientific research and empirical evidence supporting Positive Reinforcement and Clicker Training, its detractors persistently make false claims and misrepresent the facts about these forms of training. Positive Reinforcement and Clicker Training has produced highly-motivated, high-performance dogs that have achieved the highest awards in dog activities of all kinds.
It might seem to be a mystery but consider that change is hard for many people and new ideas can be even harder. Humans have had dogs in our lives for thousands of years and we’ve gotten along fine with them for the most part. Do we really need to look at our relationship with dogs and find better ways to train? In my opinion, YES! You can put screws into wood using a hammer but, as we now know, there is a better way (the screwdriver).
In Part B of this series, beginning tomorrow night, we’ll look at some common myths about Clicker Training. We’ll also look at why these half-truths and falsehoods remain so popular and what we can do to test them out for ourselves. Until then, keep reading, keep watching, keep thinking for yourself. We all want to do the best for our dogs and sometimes that means doing things a little differently even if it means we have to learn to accept some new ideas and information about our dogs and dog training.
Until next time, have fun with your dogs!
Photo & video Credits
Dog books – leerburg.com
Tugging – CarbonNYC 2005 from Flickr
Carolyn Scott & Rookie video – Musicaldogsport 2006 from YouTube
In the 60’s and 70’s we used to think we were hip. Then I learned to click and my dogs do a trick. So now I feel hip and click! It works, but timing can be a challenge and is critical. Stay hip and be more click – double bonus!!!
Rob
Eric you say that there is “scientific research and empirical evidence supporting Positive Reinforcement and Clicker Training” in your blog above, Would you kindly give me the source of individual empirical studies or a compilation of empirical studies that have specifically experimented and proven that positive reinforcement training is more effective than other paradigms of dog training. I find articles that extrapolate from chimps, monkeys, dolphins etc. regarding learning but I cannot find any scientific double-blind or even empirical studies with controls that directly prove that positive reinforcement and clicker training are more effective than other training methods. I know of the original work by Pavlov etc. that dogs associate one thing with another but that does not seem like genuine support of Positive Reinforcement as a training method. I have read the modern studies that dogs following a person hand pointing but they do not seem to be testing positive reinforcement training methodology. I would really like to read such studies that are directly on-point for dogs and positive reinforcement, all four paws, you might say but I do not find them easily. Please give citations to the scholarly works regarding dog training that you refer to. i want to understand more exactly.
Hi Bob –
I say there is scientific research and empirical evidence supporting Positive Reinforcement and clicker training for a simple reason – it exists. In addition to the many cited references in books like Culture Clash by Jean Donaldson, How Dogs Learn by Burch and Bailey, and others, a simple search of Google for “positive reinforcement canine studies” will yield a number of useful results. Similarly, you can use scholar.google.com to search for the actual research papers on the subject although some results may require a fee to download.
To get you started, I would suggest looking at the work of behaviourist Jesus Rosales-Ruiz of the University of North Texas. His team have done a number of studies with dogs that have been published covering various aspects of positive reinforcement training process and its effectiveness versus other kinds of training using specific criteria and methods. You can get an idea of Dr. Rosales-Ruiz’ credentials and publications at his UofNT faculty page here: https://faculty.unt.edu/editprofile.php?pid=2033
Also, please see this fine article by Pat Miller for additional references: http://www.thebark.com/content/positive-reinforcement-dog-training#.T_zXbXjcDnM
If you have specific questions on what I refer to in my articles, I try to include hyperlinks to source material whenever I can. If you are not finding something, please email me privately at eric@caninenation.ca and I will try to get you the reference you would like.
As far as proving that one training method is more “effective” than another, one would have to define the criteria by which that effectiveness is judged. For example, if you wanted a recommendation for how to permanently stop an unwanted behaviour in a dog by the fastest possible method, I might suggest using a handgun. The side effects, however, may not be to your liking. So I think it’s important that we define what “effective” means before we make any judgements.
You are correct, the work of Pavlov does not directly support positive reinforcement training. Pavlov researched Classical or Respondent Conditioning in animals. This is a passive form of association like a dog learning to run away when you use a certain tone of voice that reliably predicts that you will scold him. Positive reinforcement training is better supported by the work of B.F. Skinner, Edward Thorndike, Marian and Kellar Breland and others who researched Operant Conditioning – active conditioning wherein the animals learn that if they perform a particular behaviour, they can get a reliable response from their environment (or trainer).
Hope this all helps. Thanks for reading.
Eric
Very nice piece which I just shared.
Nan Arthur, CDBC, CPDT-KA, KPACTP
Author: Chill Out Fido! How to Calm your Dog.
Excellent post and so true. R+ techniques are still a more inferior way to train dogs to those who believe in ‘becoming dogs’ themselves in order to train. But the fact of the matter is, we will never be dogs, never be able to communicate using dog body language and signaling completely, and never be able to ‘correct’ dogs in a way that other dogs correct others in the group. With our fellow humans, we have spoken language, which allows us to carry out ‘training’ and building relationships.
But with dogs, we don’t. Thus, we need to realise that dogs see us as where many valuable resources come from and that the more we train dogs to understand that “For you to get Y you must do X”, the more we will be able to ensure a safe and enjoyable life for our dogs. When it comes to behaviour modification, we also need to use these resources that the dog wants in a variety of ways that helps the problem.
Anyway, I ramble.
Excellent Post, Eric!
Totally agree.
I think part of the problem is that training with Positive Reinforcement requires that the handler give up control over the dog. That is crazy talk to many people and throws the whole concept of dog training, as understood via Disney and Koehler, on it’s ear.
peace!