Every Dog Every Day

Training is a constructive process

Positive reinforcement has become the trendy, frequently used, marketing qualifier for books, classes and training systems. This is very laudable when considering the alternate fashions of punishment, domination and other such views in relation to our dogs.

Even though the marketing wraps the training in shiny, crinkly paper and calls it “positive reinforcement”, there is sometimes an underlying thread of de-constructing the dog as a dog. Regular, every day protocols recommend suppression of the dog-like behaviours we do not enjoy, re-packaging them into twenty first century companion cushions.

We should be able to celebrate a dog for being a dog, loving them and wanting to have them in your life because they are dogs. Certainly there are some habits that make our lives less than perfect with this co-existence but we must consider that any suppression can cause hidden stresses that are likely to lead to other irresolvable issues. Don’t be tempted to hide behind positive reinforcement and compress your dog into an unrecognisable package.

We cannot train any animal “how to behave”, they were born knowing how to behave. But to live in small communities, our family groups, certain behaviours need a little help and endorsement to enable the whole group to live together without conflict or distress.

We can select the behaviours and magnify them with successful outcomes of pleasure through reinforcement, but dogs are the ones that have to learn how to behave.
The closer we are to the learner-selection of the behaviour the stronger that selection will be. That leads to a greater chance of long term success and a more content dog.

I dare say some selective reinforcement, all positive, could have shaped me into a com- pletely different lifestyle and I may have appeared to enjoy it. But the contentment I receive from living a life of my own choices, not my parents, is beyond words. I was lucky with parenting that gave me an education of sufficient depth and quality to open up a broad range of choices.

About positive reinforcement

All of us would choose to be learners in a world of positive reinforcement.

That world is full of reinforcement for doing the right things, protection from traumatic error and a continuation of the learning desire for life.

That world does not actually exist. We are reared in a mixture of trial and error, learning how to change what we do to get nearer to success; from learning to walk to remembering to back up your valuable files on the computer. It often takes a serious error to force a change in behaviour. In the parenting role, for children or for animals, we strive to protect them from error that can cause trauma or harm and select the errors that can be considered just “learning information”.

Protection from learning by error should not be avoided but because our human learning history is filled with poisonous learning errors, associated with embarrassment, ridicule and shame, we are pressured into avoidance in our teaching as much as our own learning. The older sibling that is always faster, better and more successful – usually because they have more life under their belt, or the school peers that are reinforced by ensuring that someone is worse off than they are.

A desire to learn should be nurtured in all animals, it is the normal state of being, where the environment provides us with answers to our trials.

A learning error should be regarded as simple information without emotional context except perhaps frustration. When success, or reinforcement, does not happen it should not be painful, harmful or have punishing consequences.

Computers are quite good at “learning errors”. They do not seek to embarrass just do not perform the expected task. The answer is the same every repetition and we need to change our behaviour to achieve success. The skill is “change our behaviour”, but sadly a repetition of the same behaviour, over and over again is often the procedure.

The dog that begs at the table – if for the first 20 opportunities you do not fold and deliver the corner of your toast, then the behaviour will fade and only an occasional testing event may happen again. But if you fold on the twentieth occasion you have now set up success based repetition.

Consider that the dog was wired to try something 20 times before giving up, and the 20th occasion was successful. Now we have built an expectation that something will be successful on the 20th occurrence and that batches of 20’s need to be performed before a result. The dog could logically try 20 x 20 events and expect success on the 400th occurrence. What a pain that would be.
I do not think dogs consider anything in batches of 20, but that one occasion will certainly serve to push an increasing number of repetitions before another single successful event occurs. Far better to remove the dog from the opportunity of learning to beg if the table users have such poor self-control.

A dog sitting to greet visitors is not an innate behaviour and needs continual reinforcement to be maintained. Similarly returning to you when called (coming away from social interaction or rabbit chasing is not innate), jumping into the car, going to bed, releasing the socks from their mouth etc. all need a lifetime of reinforcement to maintain.

The world of positive reinforcement is the one we choose for our young animals of all species. We protect them from trauma and harm with boundaries, dog leads and electric socket covers. But they will occasionally run into the leg of the table, trip up when trying to climb a staircase and need to practice several hundred different types of motor skills and co-ordination. This is called play, and we can devise many games that encourage learning these life skills. Games and learning should be designed with errors for knowledge and on many occasions extra information when success occurs.

Need to Pee?

Puppies are very predictable when they need to empty out. When they first wake up, they may be sleepy, but as soon as they begin to move about their first need is to pee. If that is designed to happen in your garden on grass, you can provide extra reinforcement with a treat or social approval.

This is where the pup learns that the weirdest things make us happy.

But if you were not present when they woke up, and they pee’d on your carpet, then punishment after the event will only ensure the puppy is apprehensive when they first see you in the morning.

Fact: Puppies are predictable when they need to pee: when waking up, after eating and at least every hour when awake.

Your choices are:

Anticipate: move the puppy to the grass, or area suitable for a pee.

If you can ensure the First In The Morning events are in this location, then the location will have significance for the puppy and success is more likely. But if they have never pee’d in this location you may need to wait. For the events you cannot anticipate, when playing for instance, then ensure play is happening in a place you can clean up easily. In this case there is not social reinforcement, treats or punishment. It simply happens.

Punish the puppy for peeing indoors.

This event then becomes filled with anxiety. Is this young animal supposed to work out what the transgression is? Should they go behind the sofa to pee? Should they avoid you at all costs when needing a pee? If you are not good at delivering clear explanation for causing stress then simply do not do it. I have yet to meet any- one who could explain the reason for not peeing indoors to a puppy. Be awake, be a good student of your puppy’s behaviour and be the parent you should be.

The outcomes are:

Behaviour you desire: reinforce, be happy, give treats.

Behaviour you do not desire: get the cleaning products, plan how to avoid this happening in the future.

The key skill to successfully reinforcing the desired behaviour is your ability to be a student of your dog’s behaviour and understand the reasons that drive that behaviour.

All Things Gradual

Becoming more skilled at something is best achieved when the degree of difficulty increases so gradually we hardly notice the change. If we feel the change, then this is a fair indication that our skills were not ready for the increase in difficulty. We spend all our lives learning skills and in many cases those skills continue to develop.

When setting up a learning experience for puppies or adults we need to be able to break down the experiences or the increasing difficulty into small gradual steps.
For instance, the overnight “being nearby”, can begin with the crate at the side of our bed, then to the doorway of the bedroom, then to the landing, then to the foot of the stairs etc. Teaching your puppy to wait for their dinner can begin with 3 seconds and increase to 3 minutes in very gradual steps.

The changes should happen at such a rate that the puppy does not notice or experience stress. If the move from landing to foot of the stairs causes uncertain whimpering it was a step too far. If the waiting for dinner is broken by a jump and bark, you have held out for too long. Go back a step and maintain a temporary plateau until the behaviour you are looking for is established and then take a smaller step.

At the other extreme if the puppy sleeps by your bed for the first 6 months and then a changing household requires the puppy to now live in the utility room you can expect some strong, and reasonable, objections. You have caused a serious and impossible to understand rejection. If you jump from 10 seconds to 3 minutes for the dinner wait the dog is most likely to fail, and you have put pressure on yourself to scold, or punish, when the error was your own.

Too much too soon.