A Fun Way to Better Heeling

When we think of heeling we usually picture the exercise we need to do for tests and competitions. Test heeling actually involves many parts; the dog walking with you, the dog looking at you, sitting, left turns, right turns, left about turns, right about turns, drops, stands, fast pace, slow pace and the higher the class the more you can add to this list. To get really good attentive heeling it helps to break the teaching of heeling into its simplest parts and the best place to start is with walking along with you.

You Get What You Reinforce

A history of reinforcement increases a behaviour. To get really pretty heeling you need to start with a picture in your head of what you want the heeling to look like and reinforce it. This means that when the dog gets into the position you want reinforce it. It is important you are aware of exactly what you are reinforcing; if your timing is slow or you are not attentive you may be reinforcing something other than what you think you are.

Reinforce Don't Bribe

There is a significant difference between reinforcing a behaviour and bribing a dog to get a behaviour. If you continue to rely on the dog following the food in your hand (or the pretend food in your hand) after the initial teaching you are bribing the dog and setting yourself up for problems. Bribed behaviours are only sustained for a short period, until the dog thinks it is not really going to get the food any time soon. Bribery does have its place, it can be used at the very start to get the behaviour you want to reinforce but it should not be relied on in the long term. When you are bribing you are saying to the dog “do this behaviour and I might give you some food”. When you bribe you are not using the dogs intelligence and so the exercise becomes boring. I am not making an ethical argument against bribery I am just saying that it doesn’t work in the long term.

When you reinforce you change the emphasis, instead of you trying to get the dog to do the behaviour the dog offers the behaviour, the dog is now saying “look at me, I’m next to you”. The dog is now using its brain, it is trying to make you reinforce. This is a massive change in emphasis. This offered behaviour is much stronger than any bribed behaviour. You and the dog become more of a team.

The trainer’s responsibility is to teach the dog. The responsibility for the learning is actually on the dog. What we want to produce is a dog who wants to learn, one who is really trying to perform reinforcable behaviours. A dog who thinks they are giving you cues to feed them.

Train One Thing At a Time

Dogs can be trained to do a number of things in a single training session but if you want to have your training move along quickly it is often better to think of training only one thing in a particular training session. In a class we may cover a number of exercises but on your own it is better to concentrate on one thing at a time. Lets take the heeling as an example. A typical heeling training session might include;

start walking then sit the dog,
walk a little more,
do a right about turn,
drop the dog (reinforce),
walk a little more,
do a left turn,
stand the dog,
walk a little more,
sit the dog (reinforce),
walk a little more,
do a right turn,
sit (reinforce),
finish.

The dog has received 3 reinforcements but it would be hard for the dog to work out exactly what it did to get the food. I have not made it easy for the dog to start offering me behaviours to get reinforced for because I have not made it clear exactly what I am reinforcing.

To help the dog only train one thing at a time (we can easily add a number of behaviours together later). If we want the dog to walk along looking at us that is what we reinforce.

Find Heel Game

Often it is better to go back to the beginning when you want to retrain a behaviour that is not exactly what you want. A good place to start with retraining heeling is with walking around the backyard, walk as you do when you want the dog to heel, don’t ask the dog to heel but let them know you have food and there is a chance they may get it, they have to work out how to get you to give it to them. It is important you don’t walk around with the food as a bribe, rather let the dog know you have food before you go outside. Walk in a big clockwise circle, you are not training turns here so it is best to not include any. Let the dog try to get into the correct position then reinforce it. Reward the attempts generously. When you reinforce you are telling the dog the behaviour is completed so you stop as you reinforce, don’t reinforce and continue moving. Give the dog a moment to eat then start moving again, don’t ask the dog to move off with you rather let the dog try to get into position to get reinforced again.

Take 10 pieces of food with you. At this point you need to decide what you are willing to reinforce. If your dog will not walk along looking at you then you need to reinforce the dog for coming close to you. In the next session you reinforce for being even closer, in the next session you reinforce for being close to you and looking at you.

So when do you reinforce the dog? You reinforce at the moment the dog is doing what you want in that session. Not after a step or two (that comes later), the moment the dog is where you want it to be reinforce it. Timing is everything, if you reinforce late you may be reinforcing the dog for looking away from you. The point here is to make it very clear to the dog what they are being reinforced for. Why only 10 pieces of food? You want to finish while the dog still wants to play the find heel game, you want the dog to really want to keep playing rather than keep going until the dog tells you it has had enough.

When the dog will find heel position by itself you are ready for Week One Training. Week One Training involves reinforcing the dog every time it gets into heeling position.

Reinforce (and stop moving) the moment the dog gets into heel position;
Start walking again when the dog finds heel reinforce (and stop);
Start walking again when the dog finds heel reinforce (and stop);
Start walking again when the dog finds heel reinforce (and stop);
Start walking again when the dog finds heel reinforce (and stop);
Start walking again when the dog finds heel reinforce (and stop);
Start walking again when the dog finds heel reinforce (and stop);
Start walking again when the dog finds heel reinforce (and stop);
Start walking again when the dog finds heel take 0 steps and reinforce (and stop);
Start walking again when the dog finds heel reinforce (and stop).

Each time you start walking don’t ask for heel, in fact it is better if you lose the dog for a moment, we want them to learn to try to get into heel position themselves.

INCREASING THE CRITERION

In week two we reinforce on a two step average. Take the dog outside and start walking;

Reinforce (and stop moving) the moment the dog gets into heel position (that is reinforce on 0 steps);

Start walking again when the dog finds heel continue for 2 steps and reinforce (and stop);
Start walking again when the dog finds heel continue for 1 step and reinforce (and stop);
Start walking again when the dog finds heel take 3 steps and reinforce (and stop);
Start walking again when the dog finds heel take 1 step and reinforce (and stop);
Start walking again when the dog finds heel take 2 steps and reinforce (and stop);
Start walking again when the dog finds heel take 0 steps and reinforce (and stop);
Start walking again when the dog finds heel take 2 steps and reinforce (and stop);
Start walking again when the dog finds heel take 3 steps and reinforce (and stop);
Start walking again when the dog finds heel take 0 steps and reinforce (and stop).

We are now using another very important training principle – USE A VARIABLE SCHEDULE OF REINFORCEMENT. Note what we are being variable about, the dog doesn’t know when the food is coming but it does know what it is being reinforced for.

In week three we move to a 3 step average. For example we could reinforce on steps;

1, 3, 2, 5, 1, 4, 5, 2, 3, 1.

In week four we move to a 5 step average. For example we could reinforce on steps;

2, 3, 7, 5, 1, 8, 2, 5, 7, 1.

In week five we move to a 7 step average. For example we could reinforce on steps;

2, 4, 2, 8, 3, 12, 3, 5, 10, 2.

In week six we move to a 10 step average. For example we could reinforce on steps;

1, 6, 3, 13, 3, 11, 14, 9, 1, 12.

In week seven we move to a 12 step average. For example we could reinforce on steps;

8, 2, 13, 17, 5, 11, 14, 4, 15, 2.

In week eight we move to a 15 step average. For example we could reinforce on steps;

4, 12, 20, 7, 14, 9, 12, 6, 20, 2.

In subsequent weeks you can move to 20, 25 or 30 step averages.

If you train in the backyard and then go to a park you will need to think about reducing your average in the new environment until the standard is back to where you want it. For example you may be at a 10 step average in the backyard but use a 5 step average in the park. To the dog walking in heel in the backyard can appear to be a different exercise to walking in heel at the park which can be different to walking in heel at the park with dogs around. When you change the environment you need to consider your reinforcement schedule.

Sometimes when you are on a 10 step average you may like to throw in a 3 step session, you don’t want your dog to think things keep getting harder.

You might decide to do this training 3 times in a week (each session should be less than 5 minutes). You can train other exercises on other days or after you have done your heel work but don’t do any heeling while working these other exercises.

The idea of this program is to help the dog to really understand what it is being reinforced for and what heeling really is. Avoid the temptation of racing through heeling training. Training your dog is not a race, it is best to build strong behaviours through the use of high reinforcement histories. In this way you won’t find you have to retrain again later.

When you get the 7 step average you can introduce the word “heel”. Note we haven’t used the word till now because the dog was not actually heeling before, now we link the behaviour (walking next to us attentively) with the word “heel”.

If the dog moves to offering the heel rather than you trying to bribe the behaviour you will find that lagging will be a thing of the past. You won’t need to worry that your dog doesn’t “like” heeling because heeling will be a game the dog wants to play.

This article was printed by kind permission of Greg Barker