I don’t have any kids. I would like them though because I think it’s great to see such little creatures growing up and to be part of their adventures. Raising kids is hard and I have to give a lot of credit to my mom because if I look at myself…. Wow I wasn’t an easy one. My mom did a great job to “fix” me, my brothers and sister. I look at it and it seems like my mom had a very strong ingredient raising us. I still don’t know what it is but we have a thing in common and that is that we don’t get mad or frustrated quickly. We are willing to try more often and become more successful afterwards. We stay motivated to try a failure again and we see the best in every scenario. I still don’t understand how my mom did that. Although I find my mom the strongest woman in the world what makes me understand the underlying idea more and more. You get shaped by the experiences you get over time. This obviously with positive or negative thinking. I mean I consider my family quite the optimists. We try and many times we come out with success instead of going the other way of being a pessimist when we have more failure then success. Could this be because we accept our failures more and believe in reaching the goals?
I started working Sea lions in 2004-2005ish in a Zoo in the Netherlands. We had a pretty good structure and our main focus was on the positive. We used a great variety of reinforcements when animals did the behaviours on high criteria. When animals were incorrect we used the LRS.
LRS Least Reinforcing Scenario: A specific extinction technique based on differential reinforcement of other behavior (DRO). The trainer pauses 2 to 3 seconds following an incorrect response at the same point where reinforcement would have been applied following a correct response. The trainer’s response also functions as a stimulus for the animal to behave calmly and attentively. – IMATA Glossery
I learned that we used the LRS mainly as refocus of the animal with teaching the animals “look as long as you are calm and want to participate there is a chance of reinforcement” This brought our animals to a calmer way of interacting with us because we thought them that failure is ok. We reinforced the LRS once in around 50 times to teach the animals this particular message. I mean who wants to have a sea lion hanging in your leg? Before you reach this goal, I believe the LRS should be a well-trained and understandable concept.
Some of you might even say isn’t it just a DRI or DRO depending how you put it, this is why, when working in Canada with Californian Sea Lions we had one animal named Cleveland who did not accept failure for 1 single second. When the animal was incorrect it would run to another trainer and tried to bite this person. This was as you can imagine far away from acceptable. How we took this challenge was focussing on 2 different techniques. One was the balance of reinforcement and the other one was “do nothing”. This animal wasn’t able to stay with us after an incorrect behaviour and “do nothing”.
My first steps were to see what I could do in a training session, I asked her out of her pen to start my session. Over time I wanted to go back to her pen and asked her to go inside. Here we go! I thought. She didn’t want to go , she stayed with me for a split second. I bridged her and reinforced just a little bit anyway, I mean at least she didn’t go after myself or another trainer! But didn’t I just reinforce her NOT going inside the pen? I got her attention right away after I gave her a bridge. She looked at me quite surprised. I reinforced her and could send her in her pen afterwards, I guess she went because she wasn’t sure what would happen next. She went in and right away I gave her a higher reinforcement value then I had given for not going. We extended this time of response to my incorrect behaviour for 5-10 sec. This decreased the behaviour of running to other trainers completely. Not to mention that we thought her a very strong DRI as lie down as well what would help us if she would take the choice to go. Even then she would not bite at first and respond well to the lie down signal. LRS was in place and the animal was calmer. She didn’t bite anymore and we started to have her in the water with us and even being in the shows. SUCCESS!
Did you know that the LRS can be perceived differently as well?
Over time I had the privilege to work with Killer Whales in both parks in Europe. I discovered that the LRS could be different, not in a way that the message is different compared to the sea lion but in a way where there is a reason to reinforce your LRS yes or no. This is how it works, when we ask a behaviour to an animal, this behaviour mostly consist of more than just 1 criteria. If we talk about a specific task on a particular position there are multiple criteria’s present the animal has to reach to be able to be correct. 
The animal has to respond to the signal in an X amount of time, it has to be at that particular position and preform the next task to be able to come back quickly after. As you see there are quite some criteria’s to reach in just 1 behaviour.
Lets say the animal reaches 6 out of 6 criteria’s in a behaviour, we would consider this perfect so reinforcement will be present afterwards right away. What if the animal reaches 5 out of 6? Or 4 out of 6? We would consider this an incorrect behaviour and there for LRS. Should we reinforce the LRS? Yes, because the animal reached higher then 50% of the criteria’s in the behaviour asked although it’s still incorrect. When it is neutral, for example 3 out of 6, we have a choice to make, how well did the animal perform the 3? Was the animal with a dominant animal? Etc. Overall the behaviour was still incorrect. When the behaviour is performed lower then 50% reinforcement is not given after an LRS.
What happened after such a strategy of reinforcing your LRS is that the animal starts to think more about getting it right the next time you ask. The behaviour obviously has to be trained in a way that the animal does understand all criteria’s in the behaviour. When we asked the killer whales for a particular continues jump and they discovered that the first one wasn’t high enough they would already terminate the behaviour by themselves even though the whole behaviour was 3 jumps. The animals would know that they were incorrect and didn’t put any more energy into the behaviour. The training system has to be very black and white to make this work. It was very cool to see how the animals were thinking with us. After the given LRS we could ask again and mostly the animals would raise the criteria where they were incorrect previously. Like if they could read our minds.
Peter Giljam
Thinking Outside The Zoo
2 Comments