The information on this page was written by my mentor, Donna Conroy. Before her passing she was kind enough to give me copies of all her instruction outlines so that I could continue her teaching. Thanks Donna...you were a great friend and I miss you with all of my heart!

Positive and negative reinforcement are today's catch-words in dog training. Yet, this underlying principal of behavioural training, is probably the least understood by many trainers. Often considered as interchangeable with reward and punishment, reinforement is quite different.

REWARD is something given after an action to signify a job well done. E.g., a child has picked up his toys, is praised and given a treat; a dog is told to sit, he complies and is praised and given a biscuit...these are rewards. Certainly, rewards do play a role in training, but only once the child or dog has been taught what to do.


PUNISHMENT is also something given after an act takes place to express displeasure with that particular action. If a child uses a bad word and the mother washes his mouth with soap, he is being punished. The dog that chews his owner's shoe and is swatted after the fact is also being punished for the mis-deed.


POSITIVE REINFORCEMENT takes place during the act to increase the probability of the act occurring again. The child who is being praised as he makes his bed and the dog who is praised and given a food treat as he heels in the correct position, are being positively reinforced. During the early stages of teaching any exercise to a dog, positive reinforcement is what is used by the thoughtful trainer.


NEGATIVE REINFORCEMENT occurs during an act, and can be avoided by changing the behaviour. The irritating buzzer you hear if you do not fasten your seat belt is one example. Your own action, i.e., doing up the buckle puts an end to the noise. Similarly, the dog that is straining on his leash can 'turn off' the discomfort by resumming the heel position.

The major point to remember is that reinforcement can take place only when a behaviour is actually happening. Once the action has already taken place, nothing can 'un-do' that particular act.

How does one go about teaching the dog to perform a specific action? Let's use 'sit' as an example...traditionally, the collar was pulled up and the rump pushed down; the dog was then praised. Today's trainer will more likely also give a food treat once the dog is sitting. This of course constitutes a reward, and not reinforcement. If however, the treat is positioned in such a manner that the dog must sit in order to receive it, he would be being reinforced to sit. The biggest advantage to the latter method is that the trainer is keeping his or her hands of the dog. By performing the action completely by itself, the dog will learn more quickly. Once the dog understands what is expected, the food treat will be offered after the fact, and will become the reward.

Negative reinforcement is another tool that is not used until the later stages of teaching an exercise. Even then, it should be used sparingly. If teaching at earlier levels is properly taught, then negative reinforcing will seldom be necessary.

Punishment is NEVER a valid teaching tool and as a deterrent, it is virtually useless when used against dogs. Their canine brains are simply unequipped to understand why a punishment is being dealt out. It has been said that punishment may teach what not to do, but it can never teach what to do.

Another point about reinforcement is scheduling. In the early learning phase, the reinforcer is given every time, on a constant schedule. Once the concept is mastered, the trainer should resort to a random or variable schedule. This random style of reinforcement is extremely powerful; it is what keeps gamblers putting their loonies into slot machines...they don't win every time, yet never quite knowing when they will hit the jackpot keeps them coming back again and again. The same can be said of the dog being positively and randomly reinforced; he never knows just when he'll get lucky, so will continue to perform in anticipation of that piece of biscuit.

One exception to the variable or random schedule of reinforcement is when the task involves any type of puzzle or problem solving. An example of this is the scent discrimination test at the 'utility level' of obedience. The dog must be praised, at least verbally, every time he correctly selects if he is to know what to do the next time.

This article is but a brief introduction to motivational training. To better understand the principles of motivational training, it is suggested that books explaining behavioural learning be read. Karen Pryor's 'Don't Shoot the Dog' is an excellent starting point. To learn the applications of these principles in dog training, books by such authors as Ian Dunbar and William Campbell are highly recommended.