By Emma Judson, Canine Consultant
We hear this a lot, and for a variety of reasons. Sometimes, it’s because a person doesn’t fully understand how positive reinforcement based training works; sometimes it’s because someone wants to justify using harsh methods. Whatever the reason for this statement, the fact is, it isn’t true. Keeping things simple – if an animal can perceive something as a reward, then positive reinforcement will work. This means that all dogs can be trained by positive reinforcement. In fact, all mammals can. Also birds, fish and even reptiles can. This however, doesn’t mean that you can use exactly the same specific technique on all animals, and get the same results. You may have a dog who will work his butt off just so you will throw a tennis ball. I can assure you without a shadow of a doubt that you will not ever find a snake who you could train using a tennis ball! So, there is a scrap of truth to the myth, which is of course how myths carry on.
Every animal is an individual, and you need to take that into account. A snake is only ever going to find a few things rewarding and of those things (food, escape from predators, sex) you are only likely to be able to manipulate one of them (food). A dog might have a list of potential rewards as long as your arm (ball, bone, rope toy, walks, food, high-value treats, attention from you, physical contact, verbal praise, sniffing smells, chasing, walks……) and you can use all of these things very easily.
So, the science works for all dogs. All dogs learn the same way. But how you APPLY the science; how you decide which of a variety of specific techniques that all fall within the ‘positive reinforcement based training’ banner, does very much depend on the dog!
Within force-free, positive reinforcement based training, there are a huge number of ‘variables’ in each technique (and there are lots of different techniques as well). Just thinking about clicker training. The basics ‘click the behaviour you like, reward follows swiftly’ apply to all dogs. But I might have a dog who has to have a behaviour broken down into lots of tiny steps, so I use a process called ‘back-chaining’, where I teach the final step of the trick first, and add in more until I have the whole behaviour I want. This may be a very sensitive dog who needs lots and lots of success and to learn in little bite-sized bits, or it maybe because I am teaching a really complicated routine.
On the other hand I could have a dog who generally makes quite big ‘leaps’ in progress. He likes to guess at, and try out, new behaviours, and isn’t worried if he makes a mistake. With that dog, I may teach in far fewer steps. We’d proceed quicker, though the process might be a bit more haphazard.
This is what is meant when trainers say they look at each dog as an individual. Both the dogs in these examples are learning by positive reinforcement, (and also operant conditioning!) but one dog is more sensitive and can’t handle ‘failure’. She needs to work slowly with a high rate of reward. The other dog is ‘bomb-proof’, likes to work quickly, and needs a lower rate of reward.
Another, possibly more obvious, difference between dogs, is what motivates them. I can teach Fido and Rover to sit, using a reward – but Fido doesn’t like petting and praise, he likes cheese. If I were to attempt to only reward Fido with petting and praise, I wouldn’t get very far, because to him, these are NOT rewards. Rover on the other hand isn’t motivated by cheese OR petting/praise – he adores tennis balls, he’d sell his grandma for a tennis ball! It would be foolish in the extreme to attempt to teach Rover using the things I think he SHOULD find rewarding but in fact doesn’t!
So the bottom line really is, the science works on all dogs – how you apply that science, how you twiddle the various factors and specifics, is down to you and your dog!
