How dogs learn

CONCEPTS

 

How does my dog get to know what I want?

 

Dogs usually learn best by positive reinforcement i.e getting a reward - because this is a happier experience for them  which they will want to repeat.   Positive reward can be you looking at them, touching them or speaking to them, giving them a treat or playing with them  or retrieving the dummy which is a game for the dog but one which we control.

 

Rewards have to be suitable for occasion, the individual dog, the level of excitement he is at, the difficulty of the action he has just completed and how often the dog has successfully carried out the action.  For example, if the dog is very excitable, a calm vocal good boy may be sufficient or you will promote over excitement.  A dull or bored dog will need winding up with a much more over the top reward.  A dog carrying out an action successfully for the first time will need and deserve much more praise than one which is familiar and normally responsive to the relationship between a command and action.  REWARDS MUST BE GIVEN WITHIN HALF A SECOND OF THE DOG PERFORMING THE DESIRED ACTION WITH THE COMMAND.  NOT AFTER YOU HAVE, FOR EXAMPLE, FUMBLED IN YOUR POCKET TO FIND THE TIT BIT.

 

A word of caution it is easy to inappropriately reward your dog for the wrong behaviour e.g. touching it or talking to it if it starts to get over excited or whines.  (For a better course of action see below).   To see a short clicker training video with a ten week old puppy

 

How does my dog get to know what he should not do?

 

There is much confusion about negative reinforcement and  punishment (see explanations below)

 

Negative reinforcement is providing a pressure which is released when the dog performs the action required.   For example, we press on the dog's bottom and say "sit" and release the pressure when he does.  Therefore, he learns he should not stay standing when we say sit.

 

Pressures can be physical or mental  

 

When using negative reinforcement  it must be balanced by at least as much positive reinforcement when the action is carried out.  This helps keep a happy motivated dog.  Without motivation the dog will not work as well for you.

 

Punishment

The are two types of punishment direct punishment and indirect punishment.  

 

Direct punishment is something which when carried out reduces the likelihood of the dog performing a given action  For example a slap on the nose reduces the likelihood of the dog stealing the biscuit.

 

To carry out effective direct punishment it must be performed within 1/2 second of the undesired action, repeated every time the undesired action occurs and in all environments the undesired action occurs.  The punishment needs to be at a level that is salutary to the dog but not at such a level that he is physically damaged.  Direct punishment therefore is often difficult for us to get right in respect of timing, consistency and intensity.  Gradual raising of intensity over time can also lead to punishment toleration in the dog rendering the punishment ineffective.   Also incorrect timing or application can sometimes result in the dog associating the punishment with a different action to the one you were trying to build an association with.  This can lead to phobias, fear and reduced or wrong learning.  Using the example above, the dog might bite its owner  to avoid the slap or get the biscuit anyway.  Therefore, we often need to use our brains to think of alternative approaches.

 

A frightened dog reacts mainly on instinct and does not learn as well as a relaxed dog.

 

Indirect punishment is something which when taken away reduces the likelihood of the dog performing a given action.  For example, If the dog refuses to sit for his dinner we just stand there calmly, with dinner out of his reach and refusing to give it to him until he sits.  Normally, the dog will try out several behaviours i.e. circling or barking before he hits on the right one which is rewarded instantly by dinner.  In gundog work indirect punishment is useful.   If a dog is over excited by other dogs retrieving.  It is important he is smartly walked away (within half a second of the first sign of excitement).  This greatly reduces the sight, sound and smell  of the exciting stimulus.  The dog then calms down and the reward for calmness means he can return to watch (usually for a short spell, to start with, before repeating the walk away)  (remember you may have to repeat this a lot especially when dogs are working at water retrieves).

 

Use of indirect punishment needs to be balanced by positive reinforcement for the correct behaviour i.e. The dog is allowed to watch or you can say good boy.  Indirect punishment does not seem to be associated with the same problems as direct punishment and even where our timing is not at first so good it accommodates owners needing time to improve their timing and therefore makes it a more effective training method.

 

A tool box of different strategies, selecting the most appropriate for  any given situation is what makes the most successful trainer.

 

Unfortunately, because we are working within the natural environment and there are many distractions for trainee gundogs, it is very easy for them to be self rewarded for inappropriate behaviour e.g. Chasing a rabbit.  Therefore, we have to work harder about making ourselves the most interesting thing in our dog's life when training,  Prompt action when the dog starts to think about going wrong is often our best defence.  However, reading the dog's mind is the art and skill of the game that requires practice and experience.  It is because of these difficulties that sadly the trainers of some gundogs resort, out of ignorance or laziness, to some rather unpleasant methods of training which we do not find palatable.  We hope these notes can give you and your dogs sufficient positive reinforcement during training for you both to wish to repeat the experience in the future.  Well trained dogs are happy dogs because they have more freedom.

 

Happy training    Joy and Chris

 

TIPS ON HOW WE TRAIN

 

We are not "one method" trainers.  If what you do works and is legal we will not be trying to change what you do.  We will however try to work with you to improve both you and your dog.

 

Even when not actually working with your dog you can learn by watching others with theirs and listening to what we say to them.

 

We train by conditioning dogs to respond to a sound or, more often a body signal. which is strongly associated with a given action.

 

To condition a strong association we must make sure the sound or signal is always given within 1/2 a second of the action DURING THE STAGE OF TRAINING. Therefore, we need to be in a position to ensure the action is carried out.

 

To condition a strong association we may have to repeat the above 50, 100 or 200 times depending upon the individual dog.

 

However, dogs can get bored if we repeat the same thing too many times especially in a short time span.

 

ONCE A DOG UNDERSTANDS A GIVEN COMMAND OR ACTION VARIOUS ACTIONS AND OTHER COMMANDS HE UNDERSTANDS CAN BE CHAINED TOGETHER TO PRODUCE AND END RESULT. E.g. a series of hand signals and stop whistle eventually finds the dummy.

 

If you are not in a position to enforce a command, or cannot be bothered to do so, do not give it or you will simply teach the dog to disobey you.

 

We try to catch the dog in the act of doing it right.  Some tines we have to save the dog from its own instincts.

 

Nothing is cheating if it works.    Commands should be short and clear.

 

A sit without using the stop whistle is a wasted sit

 

Get your recalls right and you will get the retrieve right.  The retrieve is just a recall with the dummy being carried

 

Many dogs are head shy.  They have to learn it is nice to have the head handled and caressed in a certain way, particularly when holding a retrieve, so that you can teach them a good delivery

 

A good recall or delivery means the dog is in front of you, in close and looking up at you.

 

Training is a lot of black and white but it gets greyer the better you and the dog get.

 

Try not to chase your dog. Let it chase you.    It takes two to pull.

 

Your whistle should always be in your mouth when sending for a retrieve

 

Training is not a perfect continuum whe often have to go back a step or two to go forward.

 

 

 

 

 

 

DOWNLOAD

You can download this article here along with a sheet to enter your commands as you train to ensure consistency.

click here