Written by David Ryan, chairman of the Association of Pet Behaviour Counsellors. He's the tops in dogs.

The question is not, “What can I do to get my dog to stop chasing (joggers, sheep, rabbits, etc)”, or even, “Why does my dog chase…?” but rather, “What does my dog get out of chasing…?”
The answer lies in internally reinforcing behaviour. Dogs inherit instinctive behaviour that is too complex to be learned by every generation.
Chase behaviour is part of the inherited predatory hunting sequence. The sequence is genetically “hard wired” and prepares the canid to learn how to catch prey in order to survive.
The desire to chase exists on a continuum both between and within breeds. Labradors inherit more than Maremmas and some Labradors inherit more than others.
Because it is internally reinforcing the dog does not require an external reward for performing it. It does it out of sheer pleasure. In brain chemistry terms it gets a buzz of dopamine every time it chases. (Dopamine is the neurotransmitter chemical released by Cocaine and Ecstasy drug users).

Put simply, they enjoy it. Hugely. They enjoy the “high” they get from it to such an extent that they close down other senses to concentrate upon it. All focus is on what they are chasing as the source of pleasure. This is the first reason that owners cannot recall their dogs when they are in full flight. The dogs simply don’t hear them.
They also need to perform it. They are driven to perform the behaviour to receive the boost it provides. They are constantly looking for outlets for their chase behaviour. Deprived of the opportunity they become miserable.
If an owner has a dog that has inherited chase drive towards the top end of the continuum, they often find it difficult to control. The problem is that it is impossible to counter internally reinforcing behaviour with operant conditioning. A dog will not stop chasing for the promise of a biscuit simply because a biscuit is not as valuable as the internal dopamine boost from the chase behaviour. In fact, nothing is more valuable than the chase. This is the second reason owners cannot control dogs in full flight – they haven’t got anything the dog wants more than what it is doing now.
Owners of dogs with lower chase drives wonder what the problem is. The less internal reinforcement the dog derives from chasing, the easier it is to persuade to stop. There are things of greater value with which to tempt it.
The way to control chase behaviour is to control the target. Dogs are usually allowed to choose their own targets for their chase behaviour – often cats, rabbits, cars, joggers… Remember, they are actively looking for outlets for it, because it is so nice to perform.
Once a target is chosen, it provides a positive feedback loop in the dog’s brain: It feels good to chase > rabbits run > it feels good to chase rabbits > what shall I chase? > rabbits > must chase > rabbits > must chase > rabbits…
In this case rabbits become the preferred chase target. There will probably be others, less preferred, that will do at a pinch if it can’t access rabbits, but the biggest neural connections are between the desire to chase and the mental representation of rabbits. Rabbits are their primary chase target.
Owners need to control their dog’s primary chase target to take control of chase behaviour. Controlling rabbits is not possible, so they need to change the dog’s primary target. This means building neural connections between the desire to chase and a new target, whilst avoiding reinforcing the old one.

In practical terms it is pure dog training.
1. Remove the possibility of chasing rabbits – so long as the dog has the opportunity, it is receiving reinforcement. Change your walking places. This is not an option – it is an absolute necessity.
2. Do not allow the dog to chase anything else – build the chase drive up to the point where the dog is desperate to indulge it.
3. Play chase with a new target – ball, Kong, Frisbee. Start off indoors, or tie it out on a long line in the garden.
4. Continue to allow the dog only to chase the ball. Do not allow it access to the ball at any other time. You are controlling access to the chase behaviour.
5. Place the opportunity to chase the ball on command. Use a brand new word and simply associate it with the ball. Calling “Tennis” predicts the arrival of the ball.
6. Eventually the neural connections between “chase” and “ball” outweigh those between “chase” and “rabbit”. The dog prefers to chase the ball rather than the rabbit. The time and number of repetitions vary with each dog according to the internal reinforcement and the previously reinforced target. A spaniel can take a month at an hour or so a day.
7. To test the preference, take the dog on a long line to a place with rabbits. Wait until the dog looks at a rabbit, and then call “Tennis”. When it looks at you, throw the ball. If it doesn’t look at you – back to the garden and play some more ball.
8. The owner eventually controls the dog’s primary chase target, because it is in their pocket. As the dog wants to chase so much, it will focus on the owner as the source of its primary chase target.
The owner now has the ultimate operant conditioner! The dog wants the ball more than anything else on earth, and it can be asked to perform any behaviour to earn it. Recalls, sits, downs, eye contact, whatever.
These are the dogs that you see paying rapt attention to their owners.