The Basenji University

201 How Dogs Learn

Kinte Cloth

How Do Dogs Learn
An Introduction to Principle-Based Dog Training

Dog Breeds

First we must start with what kind of dog you have.

All dogs come with the basic starter pack of genes – they instinctively know that food, company and a tickled tummy are nice. But added on to that are the things your type of dog was originally bred for. Dogs find any instinctive behaviour – the things they were bred for – fun to do.

Labradors revel in retrieving, Greyhounds get a kick out of chasing & Collies get high on herding. That is not to say you can’t train a Greyhound to retrieve, but expect it to be more difficult than training a Labrador. Anything your dog has been bred for will be easier to train because they already like to do it. The down side is that, for the same reasons, it can be very difficult to train a dog not to do something they have been bred for.

Mixed breeds can get a mixture of behaviour. This can actually be easier, for example a lab/collie cross may like chasing after things AND retrieving them back! What types of behaviour has your dog inherited?

What’s a Reward?

We are going to use rewards when training, but to do that we must know what is rewarding for your dog. Most dogs like food treats, stroking and praise, but breed preferences might mean that your dog would rather have a game of chase, or fetch, or a tug on a toy.

What does your dog like?

If you decide on food treats, make it something extra special, not just ordinary old dog nosh. After all, we’re expecting your dog to perform really well, so they should get a really good reward, too.

Right from Wrong

Dogs learn the right way to behave by experience. What they are rewarded for, they do again. But you can’t tell them, “This sausage is for not jumping up at me when I came in just now”.

They learn by associating what they are doing at the time, with the reward. This means the BEST learning takes place if we can reward the dog when it performs the behaviour we want. One second afterwards may be too late. So, if we can arrange for the dog to be sitting quietly by showing them a piece of sausage AND THEN GIVE THEM THE SAUSAGE they will understand that the reward is for sitting quietly – not for jumping up.

Fortunately, there is a way we can extend the timing of the reward. Add a word as you give the reward – I use “Okay” – and your dog will learn that the word means that it has done the right thing and the reward is coming.

Lure and Reward

We now have principle based dog training. To teach a dog any behaviour, we lure the dog into the behaviour we want and then reward them. It will work for any behaviour, provided we can think of a way of getting the dog into the position, or performing the behaviour, we want.

Wrong from Right

But what if they do the wrong thing? We take away not only the reward, but also the possibility of the reward. We do this by putting the reward away (back in our pockets maybe) and then also taking away our attention. Turn around, fold arms, look at the sky: your dog does not exist. Then turn back round and try again with the treat.

If you add a different word – I like “ah!” because I never use it in any other context – it becomes a signal that the dog has done the wrong thing and will not be rewarded. Eventually, the word alone will be enough.

New Behaviour

All dogs go through the same series of processes when learning a new behaviour.

The first stage is ACQUISITION. The dog learns that something new is happening. Raise a treat above your dog’s head and say, “sit”. When the dog sits, drop the treat into their mouth. Your dog learns that if they sit when you ask, a reward follows.

The second stage is FLUENCY. You practice until your dog sits immediately. These two stages are usually quite quick. The next stage takes the longest; maybe months. In GENERALISATION your dog must learn that “sit” means sit, regardless of anything else. Sit means sit when: you say it quietly, or quickly; when you are wearing a hat, or stand on one leg; when you are in the kitchen, or the vet’s; when you are near, or far away. Your dog learns that nothing else matters but the word “sit”. The final stage, MAINTENANCE, means that you have to practice every now and then, so your dog doesn’t forget.

Attention Span

Dogs have a short attention span and learn best with short, 5 minute, bouts of teaching, several times a day. Start teaching a new behaviour somewhere with no distractions. When your dog is FLUENT, introduce distractions gradually, so that GENERALISATION takes place. The quickest learning will take place when your dog can focus on you and the task in hand.

For example, start training in a familiar place such as the living room, where there is nothing else to interest your dog. When they are fluent, take the training into the garden, where it will regress. Become fluent again before introducing more distractions, like on a walk, and then take another step back to become fluent again. Finally, practice in as many different places as you can.

Punishment

There should be no need for smacking, hitting with a rolled up newspaper or any other barbaric practice. If your dog is misbehaving, ignoring it is the best thing you can do. If the behaviour cannot be ignored, disrupt it in some way and prevent it happening again. Like in our doorbell example, train an alternative, acceptable, behaviour in its place.

Progression

You can progress by —

• No longer having the reward visible.
• Withholding the reward for all but the BEST performances.
• Changing the reward: instead of a food treat give praise or game sometimes.
• Rewarding intermittently: once every three, or five times, like a slot machine does.
• Doing all these at once.

And Finally, In a Nutshell —

• Take into account your dogs genes.
• Reward good behaviour and ignore bad.
• Disrupt any bad behaviour you can’t ignore, and then prevent it happening again.
• Substitute an alternative, acceptable, behaviour for the bad.

You now have the knowledge to train any dog to do anything. Give it a try.


This essay is reproduced here with permission from David Ryan Dog Behaviour Consultancy. The Basenji Club of America does not endorse any particular trainers or products. The David Ryan website is at http://www.dog-secrets.co.uk/
.


 More Information

There a a few other places to check out if you are interested in this subject.

At ClickerTraining.com there is a good video showing a variety of animals learning.

Visit Karen Pryor's website about her new book "Reaching the Animal Mind" where there are chapters from the book.

On Facebook "Why Positive Reinforcement Training?" shares ideas about how to turn dog psychology into a dog that does what you want.


Copyright the Basenji Club of America 2012                                                CLICK to return to College of Basenji Activities

Contact the Basenji University with comments, questions, updates and Workbooks at BasenjiU@basenji.org
.

We are indebted to the Basset Hound Club of America for the use of their Basset Hound University templates.