Modelling as a Preference to Shape Behaviour

There are 4 main ways we can shape behaviour: targeting, modelling, capturing and mimicking. Modelling is a common shaping strategy but have you ever thought about what it looks like and how it should work?

Modelling is a technique where you physically manipulate the animal into the specific position or behaviour you want to have. Often used in the domestic pet and horse world but it is also used in other areas, such as, marine mammal training and falconry. As with many training techniques, modelling comes with its own set of challenges.

For the animal to accept the use of this technique, we first have to condition the animal to being touched. This is a challenge on its own if you have animals who are not comfortable being touched. In the past, I’ve worked with animals that only accepted touching but did not see it as reinforcing. This makes the use of modelling a lot more difficult. To do it correctly the animal has to understand that we are going to touch them and they have to want to be touched or see it as a positive experience.

With modelling we use pressure on a body part to be able to move it. Often we have to hold it firm to do so, and this can be aversive if we as trainers progress too fast. We should go as fast as what the animal allows in order for the session to be as positive as possible. Taking small steps from touching to holding to firmly holding takes time and shouldn’t be under estimated. We have to be observant at all times and watch for the responses the animal gives us. The biggest challenge is actually not the animal, but the trainer using the technique correctly.

We as trainers want to see quick results and most often do not have the patience to teach behaviours another way. While many trainers choose different techniques and don’t mind how long it takes. In the end we should try and train our behaviours in whatever way is best for the animal.

The trainers that want to see quick results often expect quick results and therefore progress too fast into moving the body part. What usually happens next is that the animal pulls back, but because we want to be in charge you end up getting into a tug-of-war with the body part of the animal. There are plenty of trainers that will say “the animal should not be so stubborn or train me, that’s why I hold it while it kicks” making this way of training completely aversive to the animal. Modelling, like most training is built on trust, trust between the trainer and the animal.

So we should start building a strong trust account with our animals. If you as a trainer have not established this yet, you might not want to use modelling. When the relationship is established, then we can start modelling if we choose to do so. What we have to think about is the detail, when the animal kicks or pulls away, let go and start from the beginning again. If the animal still allows you to touch that particular body part, the animal shows it trusts you to continue and that is the type of behaviour you need, from there you can build your behaviours again. If the animal doesn’t even want to come to you anymore, you might not want to focus on the modelling part but more on the relationship you have and where you are in your bank account. If the animal doesn’t come back, then it is a clear indication that you went into the negative of your bank account.

The main mistake people make with modelling is that it can be over done or progressed too quickly, resulting in a damaged relationship. It’s a balance between force and acceptance, you want the animal to happily accept being touched then use a positive reinforcer. You want to force it and decide what is acceptable, then use a negative reinforcer like releasing pressure (we at Zoospensefull do not encourage you to do so). We believe that modelling is not build on choice and control but more on what we want to do. Even though this is a sensitive balance especially if the animal sees us modelling their feet as negative, then why wouldn’t it leave? Remembering the bank account, and constantly reflecting on our relationship and our balance.

Training behaviours and opting for modelling as your first option is not always the best way to go. It might be the fastest but not the best one choice for the animal. We at Zoospensefull are big fans of targeting, the reason is pretty simple, the animal has full control over the whole environment. With modelling this is not always the case.

To reach this goal we should reinforce the animal after a proper approximation, not by releasing pressure but by using positive reinforcement and that’s the difference.

Let’s say you train a behaviour and you need an extra person. We teach the animal to give their body part to us, trainer A reinforces and Trainer B uses modelling. Trainer B should always be the same person, in any case where you will be using modelling, to be effective, you need the trainer who does the modelling to be the same. It is impossible for us to describe how much pressure we use to lift a leg or to move a wing. To be black and white to the animal we should keep trainer B the same. Whenever the animal hears the marker, it means for the animal it did well and for trainer A to give it reinforcement.

Modelling is not for every animal or trainer, but it is important with this technique that the animal decides and that we listen to their signals. It is a quick way to get a behaviour but we have to make sure our bank account stays in the black. If modelling is done well your bank account will go up and as result your animal becomes more tactile opening it up to a variety of other more complex behaviours.

Have fun! Zoospensefull is an international animal training and behaviour consultancy, for more information or to book Zoospensefull, please contact us at info@zoospensefull.com or visit our website zoospensefull.com