These are my notes from the 2012 Art and Science of Animal Training Conference.
Click here for more notes from this conference.
We got a treat at the Art and Science of Animal Training Conference this year. Ken Ramirez, who is head trainer at the Shedd aquarium in Chicago, spoke once again. As well, this year he was joined by Steve Aibel and Mike Pool, two trainers from SeaWorld San Antonio. These three trainers have years of experience working with dolphins and other marine mammals. They packed a ton of excellent information into their hour long presentation. Here are notes from two of my favorite sections of their presentation.
Dealing with “Oops” Moments: The Least Reinforcing Scenario
One thing that Ken Ramirez talked about during his lecture was the LRS, or the Least Reinforcing Scenario. One dilemma many trainers have is what to do when the animal makes a “mistake.” In these situations, traditional trainers often are quick to use to punishment. However, positive reinforcement trainers are often hesitant (for good reason) to resort to punishment. It is hard to force a dolphin to do anything or to try to punish a dolphin’s behavior, when the animal can just swim away. As well, punishment can have unwanted side effects. In the early days of positive reinforcement training, marine mammal trainers, out of necessity, had to find ways to work with animals without using punishment. According to Ken, the LRS technique can be particularly useful when working with large animals or when we don’t really want to tell an animal “no.”
So, what is an LRS? An LRS is the Least Reinforcing Scenario. What this means is that when the animal does a behavior the trainer doesn’t want, the trainer gives a brief and immediate pause, where no reinforcers are provided and the animal does not have a chance to do anything to earn reinforcement. Immediately after this, the trainer always provides an easy opportunity for the animal to earn reinforcement, to get the training game going again.
What does this look like in practice? Ken showed several videos of him using an LRS while training a sea lion. He was practicing a handful of different cues for behaviors that had been previously trained. If the sea lion made a “mistake,” he would pause for two seconds. Then, immediately after the pause, he would give a cue for an easy behavior that he knew the sea lion would be able to do successfully. The easy behavior was then reinforced.
To use a LRS successfully, it needs to be brief and immediate. Ken said trainers often are tempted to extend the length of time—he cautioned against doing this. This is often emotional responding on the part of the trainer and won’t make the procedure more effective. As well, an LRS doesn’t mean that you ignore the animal or turn away from the animal. During the LRS the trainer just pauses–any movement or change in the environment could accidentally reinforce the behavior. As well, the trainer continues to monitor the animal during the pause.
A Least Reinforcing Scenario is effective for trainers like Ken because they are training at a very high rate of reinforcement. By providing a short pause, the LRS briefly interrupts the rhythm or flow of the training session. If you do not have a high rate of reinforcement during your training session or if the animal is often making “mistakes,” you should not be using an LRS. Instead, you need to reevaluate the training situation or your shaping program to determine what changes you can make to help the animal be more successful.
An LRS also works for trainers like Ken Ramirez because they have a long history of positive reinforcement with the animal and because, when using an LRS, they immediately offer an opportunity for reinforcement after the LRS. The LRS gives a structured way to ignore unwanted behavior with an animal who understands positive reinforcement training. An LRS shouldn’t be used when training a new behavior or when you first start training with a new animal. As well, providing an immediate opportunity for reinforcement immediately following the LRS teaches the animal to continue to focus on you during the LRS.
The Least Reinforcing Scenario can also be used as an evaluation tool for the trainer. If you are using an LRS, you should be taking data during your training sessions on how often you use the LRS and which behaviors you use it for. This is important information that will help you structure the next training session. If you had to use an LRS multiple times in a session or multiple times for a particular behavior, don’t blame the animal! Instead, take this as an indicator that the animal either didn’t understand what you were asking or didn’t feel comfortable performing the behavior. In either case, you probably need to change your training plan during the next session, rather than blaming the animal or continuing to use an LRS.
Using HELPRS to Plan Your Training Program
Steve Aibel and Mike Pool talked at length about what animals need. Animals have three main types of needs, physical needs, social needs, and mental needs. We can meet many of our animal’s mental needs by well designed training programs and well designed environments that keep animals mentally stimulated.
It is really important for trainers to remember and consider all three types of needs, physical, social and mental. Sometimes, it’s easy to get caught up in what you want the animal to do or in what behavior you need to train for a particular situation or task. However, trainers also need to stop and evaluate if they are just training something that they need, or whether the training program also meets the animal’s needs.
Steve and Mike shared an acronym that they use when designing long term training plans for their dolphins and other marine mammals. The acronym is HELPRS. Here’s what each letter is for:
For each animal, they design training plans and goals for each of these six areas. For many of their animals, they want to train performance behaviors for the shows at SeaWorld. However, they also make sure they spend just as much time designing training plans and then scheduling training sessions to work on behaviors related to husbandry, exercise, learning, play, and relationships.
The HELPRS acronym is a great guide to consider when writing training goals and at the beginning of each week, day and training session. Have a set plan of what area (or areas) you are going to target in each training session. This is a useful model because it can help you balance your training program and make sure you are meeting all of your animal’s physical, social and mental needs. Steve and Mike said to remember that “Your animal’s quality of life depends on you!” Trainers have a responsibility to make sure they are providing a high quality of life for their animals.
I talked recently about some of Bob Bailey’s suggestions for taking data in my post about “Advanced Dog Training.” If you need ideas for taking data, I suggest reading some of the great comments that were left on that post by some of my readers. I’d love to hear how you plan what to work on during a training session. Do you have a long term plan or long term goals for your animal? How do you keep track of your progress on your long term goals? Which of the areas of HELPRS do you focus on the most? Are there any of the areas of HELPRS that you think you should be focusing on more?