If you haven’t read “Don’t Shoot the Dog” from Karen Pryor you should! In her book on page 38 & 39 she explains the 10 laws of shaping. A great tool for the trainer to make progress with their learners. In this article I would like to add my thoughts on each point of view, reviewing them and their importance in your training.
1. Raise criteria in increments small enough so that the subject always has a realistic chance of reinforcement.
As trainers we want to reach our goals quickly, which is not always the right thing to do. When we train behaviours, our goal is not only to finish the behaviour but have a reliable behaviour we can ask for in future. This means that there is a strong history connected to the finished behaviour. To reach this, as this law says, it is absolutely best to take small approximations so the learner has a fair chance of reinforcement. I stand by the fact that we as trainers have to observe our animals enough to see which approximation will give them the success they need to receive a reinforcer.
2. Train one aspect of any particular behavior at a time. Don’t try to shape for two criteria simultaneously.
As a coach I’ve observed trainers adding too many criteria to reach their goal. They end up with a lot of problems along the way, because they do not think about this specific topic. Adding one criteria at a time helps the learner to be successful. This law can be difficult because we need to evaluate what we see as a criteria. You can basically look at it like this.
Scaling an animal in a group setting: One criteria is ‘be calm when you are with me’. Remember this individual is still in the group. Moving the animal away from the group is the next criteria. Adding a weight scale in the environment, changing the environment, is the next criteria. Asking the animal to step onto the scale is next next criteria and so on. We can even make smaller criteria in between. If you go for 2 criteria at the same time you would ask the animal to leave the group and accept a new object simultaneously. You won’t reach the success you want.
What I always say is “it’s not worth trying”. Trying is a 50-50 chance. Lets make sure they we get that 100% success rate instead.
3. During shaping, put the current level of response on a variable ratio schedule of reinforcement before adding or raising the criteria.
A variable ratio schedule basically means randomly changing the delivery of reinforcement. What we will be doing is making sure that the learner understand a variety of reinforcement, maintaining interest and success rates.
4. When introducing a new criterion, or aspect of the behavioral skill, temporarily relax the old ones.
Remember the criteria story. To relax one criteria for example, you could say let’s keep the individual in the group to then introduce the weigh scale. This way we relax one criteria and work our way up to a second where the individual goes away from the group to be at the scale. This way of thinking gives you more success in your training. This has helped me to progress a lot faster and really is a key ingredient.
You can do this continuously until you reach the goal behaviour you are looking for.
5. Stay ahead of your subject: Plan your shaping program completely so that if the subject makes sudden progress, you are aware of what to reinforce next.
This is a very interesting law which I see very often. Trainers not making a shaping plan because they think it takes too much time. The interesting part which regularly occurs is that they fall into problems training the behaviour, which could have been solved from the beginning if a plan was in place. Instead, it actually takes a lot longer than planned to train the behaviour, and the trainer has thrown away valuable time in the long term.
Staying ahead of your subject is not only important for the animal, but even more so for the trainer. The trainer is ready to reinforce the moment when the animal jumps forward a couple steps. Just because the trainer knows what the plan is and maybe what the trainer is looking for in future steps in deep detail. Click here to find out how to build a shaping plan!
6. Don’t change trainers in midstream. You can have several trainers per trainee, but stick to one shaper per behavior.
A huge difference in progress is when one trainer shapes a behaviour. I believe you can do this with 2 trainers as well but remember that communication will be more challenging with every trainer you add. It mentions in this law that you don’t want to change trainers midstream, this is a bit of a discussion point for me because I believe it can be done.
Once in a training session I tried to get a dolphin to do a beautiful front flip behaviour (they jump head out of the water make a 1,5 complete rotation to enter with their head back in the water). My plan worked well until I had to add some airtime in the the behaviour. I was struggling more and more eventually didn’t know what to do, so asked for help. I asked another trainer to take a session with her, because I realised that I had started to display tunnel vision with this behaviour. Not knowing how I should solve it, when in reality it looked easy on paper. I gave the behaviour away mid-flow and asked for help. The trainer showed me his ideas and I directly got new motivation, making quick progress from there.
There are situations where you can change trainers midstream. Is it ideal? In some cases yes and in some cases absolutely not. Choose wisely.
7. If one shaping procedure is not eliciting progress, find another. There are as many ways to get behavior as there are trainers to think them up.
Shaping a behaviour is not just going with the flow. For each animal there is a different way. Each individual learns differently and needs different shaping methods. If one doesn’t work, try another. Different ways to shape behaviours are:
– Modelling. Click here to find out more information about modelling.
– Capturing. Click here to find out more information about capturing.
– Targeting. Click here to find out more information about targeting.
– Mimicry. Click here to find out more information about mimicry.
Whenever you are training an animal you are using a combination of various methods. Observe and listen to the animal to find out which shaping strategy would work best.
8. Don’t interrupt a training session gratuitously; that constitutes a punishment.
The last thing you want to do is interrupt a training session. In the past I’ve seen many times trainers disturbing another trainer for non important reasons mid session. The interrupted trainer is now unfocussed and gives a different consequence to the animal than actually wanted. The first thing we always do when coaching teams – we do not interrupt each other while training. If you want to ask a question or have something to say we do this after the training session. This is only fair to the animal being trained.
There are still many occasions I have witnessed where the trainer that is mid-training session is asked something and leaves the animal promptly to do something else. This is unfair to the animal and the animal might receive this as punishment.
Stay focussed in your session!
9. If behavior deteriorates, “Go back to kindergarten.” Quickly review the whole shaping process with a series of easily earned reinforcers.
A huge topic to discuss. Because as mentioned at law number 5, a shaping plan is needed. Not only for you to stay ahead of the subject, but as well as for a behaviour which could deteriorate.
Many trainers that do not work with shaping plans think they take an approximation back, but in reality what they are actually doing is training something new to solve an existing problem. Now the animal actually becomes more confused. The trainer now asks somebody else for some help, who may do exactly the same. A shaping plan will help you to retrain a behaviour in only a day, sometimes 2 (depending on the challenge). Which leads me to an important point. Putting some extra time in making a training plan is not throwing time away. You have a higher chance to succeed with your animal, save time in future sessions and not even mentioning the confidence you have training the behaviour now! All because you are prepared.
10. End each session on a high note, if possible, but in any case quit while you’re ahead.
The first thing I want to say right away is “A high note” must be from the animals point of view. The reason I mention this is because there are plenty of trainers that will say “yes but he has his friends with him” or “the grass is so nice and green that should be enough” thinking this way only gives you problems. It MUST be about the animals perspective of that “high note”.
Starting a session with a call to the animals, having the session based on reinforcement techniques and then having a calm and successful end, even when you are leaving, is a huge change in the relationship you have with the animals. Everything they do becomes errorless.
Makes sure the animal knows it has done a fantastic job!
Want to learn more about this and other topics? Check out our website for more information and other webinars about common training subjects. If you have any questions about the topics in this article or about training in general, we now have the option of online coaching! Book your FREE 15 minute call right HERE.
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.
2 Comments
PeterGiljam · April 10, 2023 at 16:43
Anytime you bridge you already reinforced your animal. The main point here is that you want to reinforce on different moment but as well make sure the animal actually does it for a variety of different reinforcers. This is the important part. Not that when you do a blood sample you can only do this with Mango. The animal should preform the behaviour for all types of reinforces 🙂
Sarah van Herpt · January 12, 2023 at 04:57
I am wondering what you think about number 3. “During shaping, put the current level of response on a variable ratio schedule of reinforcement before adding or raising the criteria.”
In the book, it seems like Karen is saying to not reward the behaviour at all sometimes. Is it also a variable schedule if you vary the amount and type of reinforcer? I believe that if you have asked for a behaviour to be done, and the animal consents to do the behaviour and meets the criteria, then they should earn reinforcement for it. I am interested to hear what you think?