Over the past few years we’ve trained many different behaviours, with a variety of animals together with a variety of teams. These have been trained behaviours for different situations and purposes. But what makes a behaviour completely finished?
The animal does the behaviour for a variety of reinforcers.
When we start to teach an animal a new behaviour we use the best reinforcer. Most of the time an unconditioned reinforcer called food because it’s an easy reinforcer. You make quick progress and feed the animal a lot everytime it has a new approximation. Behaviour finished! What happens when the animal only performs the behaviour with the pellets you’ve used every training session? No other reinforcer works. It’s this moment we become stuck. A behaviour that is finished should be maintained successfully with a variety of reinforcers, including conditioned reinforcers.
Environment doesn’t matter.
The behaviour is conditioned at one particular spot. The animal understands the behaviour very well. You then ask for the behaviour somewhere else and the animal doesn’t do the behaviour anymore, not even the first steps. Your behaviour became position dependent. A behaviour should be able to be asked anywhere to be able to say it is completely finished.
In this case we have to be careful to take into consideration all the changes applied in the environment, whilst still keeping the trained behaviour.
You train an animal for a blood sample. The animal knows it everywhere and the behaviour is very stable. You consider the behaviour finished but you haven’t thought about introducing other trainers. The animal begins to show some health implications, but you are on vacation for 4 weeks and you just left. If the animal will not perform the behaviour for other trainers, we now have a challenge on our hands. It is therefore very important to introduce a new learned behaviour to other trainers and we can do this in a variety of ways.
- Trainer who trained the behaviour performs the behaviour with his or her colleague beside him.
- The trainer who trained the behaviour does it first and then the new trainer asks the behaviour. Most reinforcement comes from the new person.
- The new trainer performs the behaviour with the trainer who trained it beside him.
- Behaviour introduced.
Maintenance of the behaviour.
At Kolmarden one of the veterinarians once asked us to train one of the giraffes for a radiography of the head for a dental check. This behaviour was finished within about 2 months. The behaviour was stable and had a good history, however after we did the procedure the behaviour was never asked again. Why?
It is important for the animal and the trainers that a behaviour is maintained, working towards being pro-active for future challenges. A medical behaviour such as this have to be maintained well because they are challenging to perform. We should aim to hold a strong reinforcement history connected to the behaviour. In the case of the giraffe, we are still able to use the chin rest and desense work that went on. We just have to practice those approximations again.
We practice different approximations to hold the behaviour high in maintenance. The history of reinforcement is kept up and the behaviour stays strong as an outcome. A finished behaviour is maintained well and can be asked at any moment in time.
Generalisation of the signal.
Teaching an animal a new behaviour is great fun. We get going quickly and want to finish the behaviour as quick as we can to be able to show others. Every session you only work the behaviour you are training and after a couple of weeks the behaviour is finished and you get to introduce others. The other trainers in your team start to ask the behaviour after they have asked another behaviour. The animal responds incorrectly to the other signals but does give the right response to the new the signal.
Before we introduce a newly trained behaviour to another trainer we have to discriminate the signal with signals the animal already knows. The easiest way accomplish this is asking small and simple behaviours that are incompatible to the new signal and behaviour asked. Slowly you can build this up.
A tip I always say is to ask the new trained behaviour first and then ask another behaviour the animal knows well. If you do this the other way around there is a higher chance of failure by the animal. After this step you add a simple behaviour before, the newly trained behaviour and another behaviour after. Slowly you introduce new signals. Remember it is about the signal discrimination not the actual behaviour.
We call this generalisation of the signal.
So you’ve trained a new behaviour as quick as you can. It’s done in just a few days and you think the behaviour stands strong. You introduce the behaviour to somebody else and directly the behaviour breaks down. You get it back quickly but the same happens when you introduce another person.
We can’t stress it enough, history is more important than the final goal. When your history is strong your behaviour will be too. This is important in any case but especially so with medical procedures where it needs to be even higher. Before we consider a behaviour finished it is advisable to repeat approximations and build them stronger for a better overall history. Focusing on how you train to reach the goal is more important than focusing on the goal itself. Take your time and reinforce the steps well along the way. Have you had a couple of good repetitions on a particular approximation? Move on, build it up and get that strong history together.
You most likely want to routinely take a blood sample and not just once.
Any questions about a finished behaviour and what it means? Send us an email or join our Facebook group. Zoospensefull is an international animal training and behaviour consultancy, for more information or to book Zoospensefull, please contact us at firstname.lastname@example.org or visit our website zoospensefull.com