Basic Training – Theory
Once you understand how dog’s learn, you can use this knowledge to become a training ninja. By getting inside a dog’s mind you begin to understand how to motivate them and how to avoid certain training mistakes.
2.1 How Dog’s Learn
2.2 How we can help dogs to learn
2.3 Reward-based training
2.1 How Dog’s Learn
Dogs share our homes. But for this to work we expect the dog to obey certain rules. High on the list is the dog doesn’t relieve himself in the house, doesn’t destroy the furniture, and people get to sit on the best seats. We also expect the dognot to show us up when we’re out, so that means sitting on command and coming first time when we call.
But since dogs aren’t born with a computer chip for a brain ready loaded with an obedience program, it’s up to us to teach them the rules. In Module 1 we touched on some of the traditional methods of dog training and how the tide has turned against them. Behavioral scientists now understand more about how puppies and dogs learn, so it makes sense to use this knowledge and work with the pet to help them learn, rather than fight against nature.
At what age to start training?
Puppies in particular are miniaturized learning machines. Up until the age of 18 weeks their brains are hugely plastic, absorbing what they see around them and logging down their reactions to it. It is no coincidence that this is known as a ‘socialization period’ since it’s when the puppy soaks up new experiences, such as meeting people and other animals, visiting placing and accepting them.
With regards to when to start training, the answer is right away!
There is a popular myth that you shouldn’t start training until a dog is 6 months old, but this is wrong as even a 2 day old puppy learns which teat he favors and what his mother smells like. True enough, up to 8 weeks of age a puppy is too young to respond to requests, but as soon as you take that puppy home he’s old enough to start potty training.
The Learning Process
A puppy needs to learn quickly about the big wide world if he’s to survive. Think about a wild dog evading predators and knowing what to eat and you begin to see why. A puppy’s learning is geared to keeping him safe, and he assigns life experiences into one of three boxes, labelled:
|Good outcome||Worth repeating|
|Bad outcome||Avoid in future|
|Not sure||Check again later|
To see how this works, let’s look at an example.
Let’s say a puppy pokes an object with his nose; the outcome will influence whether the puppy engages with that object again. If the object was a ball, and it went bouncing off down the hill this is a good outcome for the puppy that created a game of chase. Woof! Poking the ball was rewarded with a game. Next time the puppy sees a ball, he’s likely to associate it with fun and be happy to approach it again. Woof woof!
However, if the object he poked was a bumble bee and it immediately stung him on the nose, he’ll have bad memories of bees and avoid them in future.
If the object was a rock, he had neither a good or bad time. His mind subconsciously decides he doesn’t know enough about rocks yet (Are they all this dull or was this a one off?) and he’ll check other rocks out. However, if every stone he investigates is as boring as the first, eventually he leaves them alone.
Now this is all very well, but how does it help when it comes to training?
2.2 How we Can Help a Puppy Learn
What we learnt in Module 2.1 gives us tremendous power because we now know three golden pieces of information:
1) If a puppy feels good about something he will seek it out in future.
2) If that something is unpleasant, the puppy avoids it from now on.
3) If the experience was underwhelming, the puppy may test it again but then forget about it.
Can you see how this might help us with training?
Feeling Good about Something
We know that a puppy is likely to repeat actions which lead to a reward (Think of the puppy nudging a ball with his nose). If we can get the puppy to link an action with a reward, we have a strong training motivator.
For example, if the puppy knows every time he uses the toilet spot in the yard he gets a great reward, then he’ll want to save up his bodily functions and ‘spend’ themoutdoors in exchange for a treat. This is the theory on which potty training is based. Likewise, this principle can be adapted to teach a dog practically anything from sit, stay, and recall, to fancy dance routines. The trick is to have the dog link the action to the reward, which is where clicker training can be so useful. (See Module 3.2)
An Unpleasant Experience
Those golden rules can also stop us from teaching the dog to dislike us!
Let’s look at how a dog avoids unpleasant experiences and so that we learn a salient lesson about punishment. See if you can spot the mistake in the following scenario:
You were only gone ten minutes. How much trouble can a puppy get in to in that short a time? Quite a lot as it happens. You return to find the puppy has emptied his bladder and bowels all over a priceless oriental rug. In fact, he’s still at the scene of the crime, just finishing off. Thinking to teach him a lesson, you stride over, smack his firmly on the rump, and rub his nose in the leavings. For added value, (while you clean up), you send him to his bed.
OK, if you didn’t spot it, the error is in punishing the puppy.
But, you say, didn’t we just learn that punishment stops an action from being repeated.
Well, yes it does, BUT, (and it’s a big but), it rather depends what the puppy links the punishment to.
In this case, the puppy doesn’t associate the punishment with relieving himself, but to you. His puppy mind is liable to think you have a strong irrational dislike of his bodily functions (the Chinese rug doesn’t enter into the equation).
Unfortunately, the upshot is the puppy is now reluctant to relieve himself when you’re there, in case you ‘Lose it,’ again. See how much harder this makes house training? You take the dog into the yard, but your presence inhibits him. He hangs on, which means when you go back inside he’s even more desperate to toilet and liable to have an accident indoors.
In addition, he’s afraid of your reaction so he squirms behind the sofa or dives under the bed to relieve himself in secret. By driving the problem underground you just made it almost impossible to clean up after him. This means he now has lots of doggy signposts all over the house, marking out his toilets. Not a great outcome, is it?
The upshot is, avoid punishing a dog as the lesson he’s most likely to learn is to be fearful of you.
An Indifferent experience
OK, so what about that indifferent experience, surely we don’t have to worry too much about that?
Actually, yes, you do.
Let’s consider the puppy who hears the mail dropping through the letterbox onto the mat. It’s a strange noise and he’s not sure what it means so he tries a tentative “Woof”. What happens next could shape the sound of the future for the household. If you are that puppy’s owner, which do you think is the correct way to react? Decide between A and B.
|Scenario A:||The puppy barks and you ignore him.|
|Scenario B:||The puppy barks and you shout “Be quiet”.|
The correct answer is “A” – Do nothing. The reason for this is that by consistently not reacting, the puppy slowly learns it’s not necessary to bark at mail and that there’s nothing to be gained by doing so.
Contrast this with B. The puppy can interpret shouting in one of two ways. He may think you’re trying to bark and join in, which gives him feedback that he’s doing the right thing by barking. Or, by shouting you give him attention, which all puppies love. Thus he learns by woofing he gets your attention and so he’s likely to bark every time he hears the mail.
Long story short, some behaviors are best ignored.
2.3 Reward-Based Training
Imagine being a stranger in a country where you don’t speak the language. Someone tries to tell you something important but you have no clue what they’re saying or what they want you to do. Despite your best efforts, they get progressively more and more angry. You try harder, but instead of cross words, they punish you.
Not a pretty picture is it? And yet this is the position that far too many dogs find themselves in.
Dogs don’t speak English, and actually it’s amazing they obey us as much as they do. However reward-based training offers dog’s a way to understand when they did good. Once they understand which actions please us, they can offer those actions on cue (which is largely what language is about). In short, reward-based training gives us a way of communicating with dogs.
Some trainers use clickers (See Module 3.2) but, you don’t have to. It’s possible to train without a clicker, using a timely “Yes!” to mark good behavior. The disadvantage to not using a clicker is that some dogs tune out to the spoken word. After all, they hear you say “Yes”, when your partner asks if you’d like a cappuccino, so what’s to say whether the dog should be all ears to you or not. (That’s why clickers work so well, because it’s a unique noise that the dog knows means “training in progress, listen to Mum”.)
The crucial thing to understand about reward-based training is the reward must happen within 2 seconds of the action. So if your dog does a fancy new move on cue, but you reward him five minutes later, he learns nothing and it’s anyone’s guess if he’ll repeat it again. Reward that behavior the instant it happens and he’ll know what repeat.
And finally, the great thing about reward-based training is that the dog wants to work to get that reward. This means he’s focused and eager to please, which is something you can’t buy.
Oh yes, those rewards don’t just have to be food-based. Work out what really interests your dog, such as scents, playing tug, or chasing a ball, and then use these. For example, if your dog really enjoys tug, click and reward with a game of tug (First, teach him to drop the tug toy, so you can end the game in a timely manner.) Also, as a reward to an excellent recall, put his lead on but reward him with a walk round the park. Be open minded with your rewards and keep him interested.