This is the fifth in a series of posts based on my notes from the 2018 Art and Science of Animal Training Conference that was held in Irving, Texas on March 24-25, 2018.
While I try to take accurate notes, it is possible that there are errors or that some detail is lacking. If you post a comment or email me, I can try to clarify or provide some additional information. Many thanks to the speakers and organizers who allow me to share. To learn more about the conference, you can visit the conference website.
Barbara Heidenreich: Exotic solutions to Exotic animal problems
Barbara works extensively within the zoo community, both as a consultant and as the trainer for long-term projects. She shared some of her experiences working with zoo animals and described how the zoo community faces different challenges than trainers working with domesticated animals and pets. Her presentation had a lot of video, which I can’t share, but I’ve included descriptions that highlight the main points. If you want to see some training videos, she does have a large collection of videos on her YouTube channel.
What is it like to work with exotic animals?
There are many similarities but with variations depending upon the species and the environment.
- Same principles
- In many cases, similar challenges
- She needs to be able to work with many different species
- It’s important to take advantage of modal action patterns – what is the animal “hard-wired” to do?
- Neophobia (fear of new objects) is a common problem.
- There can be challenges with hand reared animals – some hand reared animals have inappropriate behaviors around people.
She can’t be an expert on every species that she is asked to work with, so she presents herself as someone knowledgeable about influencing behavior. Her job is to work closely with the people who know the species well.
Zoo trainers may have a different kind of relationship with their animals -compared to trainers working with domestic animals
- Sometimes they do have trusting relationships.
- Sometimes it is irrelevant to the species or the behavior (sharks and rays, cobra).
- Sometimes it is discouraged because it can adversely influence the species or the behavior.
- Sometimes it is required that the trainer NOT interact (common with conservation projects).
Zoo trainers have to consider:
- How to manage groups of animals (many trainers? 1 trainer per animal? 1 trainer per group?)
- Teaching dominant animals to allow others to participate, shy ones to participate
- Might need to teach separation both for practical reasons and to facilitate training
- Multi species exhibits – sometimes the group dynamics involves other species that may or may not also need to be trained.
- Large exhibits where you might not see the animal (no easy opportunity to reinforce the animal)
- Control and choice- how much can you give? In zoos, may animals are not restricted within their own exhibit space
- Free flight – finding suitable locations, avoiding distractions, considering safety
- Facility design challenges – many are not designed with training in mind, not set up for protective contact, often use antecedent arrangement but the facility may make this more difficult
She shared a video of teaching pigs to separate for feeding. She had to consider how to mark and reinforce the behavior. The pigs were in an enclosure with hotwire, so she didn’t use a click as the sound was too similar. Pigs love tummy rubs and this can be used as a reinforcer. The training plan had to take into account species specific behaviors, like the fact that pigs like to bite…
She had a nice collection of videos that showed work with a lot of different species and with a wide variety of behavior problems. Most of the examples were consulting work where she was called in to solve a problem and only had a short time to resolve it. In many cases, she uses the same tools which are:
- training incompatible behaviors
- avoiding triggering the unwanted behavior
- creating positive associations with various objects
- changes in the environment or antecedents
I’ve listed a brief description of how the behavior and how she addressed it. While the videos made the presentation more interesting, I think it’s actually pretty interesting to read through the list and see how she had to make small changes for different species and different conditions. There are lots of great examples of how to set up the training so the trainers and animals could be more successful.
Monkey with undesired sexual behavior: This was a monkey that was showing inappropriate sexual behavior when his keepers were interacting with him. It only happened with them (not with visitors) and did not happen immediately, so they had time to train other incompatible behaviors in the beginning of each session. Over time the behavior decreased as he learned to offer other behaviors.
- Avoid reinforcing it
- Avoid triggering it
- Teach an incompatible behavior
- Immediately engage in incompatible behaviors before he has to time to do the unwanted behavior
- Build repertoire of acceptable behaviors
Hoof curl with Giraffe: This example showed teaching duration to a giraffe for hoof care. The giraffe had learned to rest her foot on a hay bale (the hoof curl behavior) but would not keep it in position. They extended the duration by adjusting the amount of reinforcement based on how long she held the position, and being consistent about not reinforcing for efforts that did not meet criteria.
- Want her to put her foot on the bale of hay but she won’t keep it there
- Used different reinforcement value for different levels of response, 1, 3, 5 biscuits – contingent on what she does
Leopard that won’t go to a laser target: Laser targets are often used with the cats to move them from one location to another. They wanted to use a laser dot to move the cat from one enclosure to another, but this cat was ignoring it. It required a more creative shaping plan, one that didn’t require multiple repetitions, but just built the behavior one step at a time over two weeks. It’s a great example of how you can train, even if you can’t do long sessions. Even one repetition a day was enough to change this cat’s behavior.
- Creative shaping plan
- Put a large chalk circle on the wall, added some scent
- Leopard would go in and check it out, click and treat
- Did one repetition a day for 2 weeks
- Move the circle to a new spot every day
- Once the leopard had learned to go to the circle, they reduced the size
- replace circle with laser dot
Gibbons that won’t shift with the trainer in the holding area: These gibbons had become suspicious about going into the holding area when a trainer was present – because this usually indicated that they would be locked up. She had to work with their current environment and slowly shape the behavior of going into the building. The set-up is an island that is connected to the holding area by a log.
- They do have a remote feeder on the island
- Sound of feeder became the marker
- Have to cross a log to get to the holding area
- Gradually shape them to come closer to the holding area by using the remote feeder to provide reinforcement
Elephant who bites and spits out pills: This elephant requires medication delivered orally. She will take it but if she bites into it and tastes the medication, then she spits it out. They had to change some aspects of the pill delivery as well as teach her to swallow the pill instead of biting it.
- They had tried various things, but if she bit the pills and tasted the medication, she would spit it out
- Give fruit juice first – as lubricant
- Then freeze the pills and coat in coconut oil – so they slide
- Had to teach swallowing with a smaller object – they actually used a quarter of a peanut
- Once she could swallow the peanut, and maybe something larger?, they gave her the pill and cued her to swallow
Elephant that eats everything: They wanted to use the elephant in a commercial where she was waving an object around. But, this elephant liked to eat anything she could pick up, so they had to teach her to hold something without eating it.
- Find something she won’t eat – difficult (she eats everything!) – but they discovered that she doesn’t like eucalyptus
- Reinforce her for giving objects back – Offer eucalyptus – reinforce her for giving it back
- Then offer something with the eucalpyptus, – then reinforce her for giving it back
- Eventually she will take other objects without the eucalyptus and give them back
- They got their video of her waving with the object
Head squeezing surrogate mom: I think this was an orangutan (or some other larger ape – can’t quite remember). The baby was not being raised by the mom, but was allowed to spend some time with an “aunt.” Unfortunately, the aunt was accidentally reinforced for squeezing the baby’s head. They did make some progress with resolving this issue, but then the baby was able to go back with his mom.
- Identify when it is likely to happen (when baby is hungry)
- Build repertoire of more appropriate behaviors
- High ROR (rate of reinforcement), then thin it out
- Have some “bail out” behaviors – that they can ask the aunt or baby to do if it happens (they wanted a safe way of interrupting the behavior)
- End sessions at any of the pre-cursors
Giraffe that won’t stay in the chute: This giraffe would not stay in the chute for handling. First they had to reinforce her for being in the chute and build up a positive association with that location. Then they had to teach it to accept touch.
- Teach approximations to following a target
- High ROR
- Build up to R+ for staying in the chute
- Door open when they start to touch – so she can leave if he wants
- The word “touch” is used to tell the giraffe when she will be touched
Lemur that jumps on keeper’s shoulders: The lemurs are in an indoor enclosure with branches to climb on. One of the lemurs likes to jump on the keeper’s shoulders. She had them teach an alternative behavior.
- Teach target and station on a branch
- High ROR for targeting and stationing
- Make it harder to do the undesired behavior, easier to do the desired one
- Reduce the reinforcement value of the undesired behavior
- Practice by letting him wander a little and then come back – so he learns to return to the station. Watching where he goes when he leaves the station is useful information about what is reinforcing.
Dive Bombing conures: These birds were in a school and were dive bombing people who entered their enclosure. They did come up with a plan to teach alternative behaviors, but she presented it as an example of being realistic about whether a training or a management solution was going to work better.
- DRI (differential reinforcement of incompatible behavior) – stationing
- High ROR – then thin it out
- But realistically, the easiest thing was to adjust the environment so that the birds were not in the same space with the people
- Be realistic about your options and resources
Fear of New Objects: A giraffe was afraid of new objects and wouldn’t touch a target. They were working on a high platform at head level (for the giraffe). They started by feeding the giraffe in the presence of the target and waiting until he chose to interact with it.
- Giraffe doesn’t like target
- Present target, feed as long as giraffe stays within reach
- Eventually the giraffe gets closer and closer to the target and touches it
- It’s about waiting for the animal to choose to interact with the object
- Once he is choosing to touch the target, then they can mark and reinforce for a targeting behavior (switching over to an operant response)
Fear Response to Touch: This giraffe was being taught to accept touch. They had him touching a target while standing behind a barrier (leaning over a doorway). One trainer was holding the target and reinforcing the giraffe. Another trainer was positioned so she could touch the giraffe on the neck as he reached forward to touch the target.
- Systematic desensitization (they started with little hand movement toward the giraffe and very light contact)
- Teach the animal to initiate contact – they wanted the giraffe to learn that he could choose when to start the next repetition and control when the person touched him
- mark and reinforce for accepting contact (they started by reinforcing targeting).
Undesired vocalizations that had been reinforced: This bear had been accidentally reinforced for making loud vocalizations. Over time it had gotten worse and worse because trainers had not waited out extinction bursts. Barbara had a video showing how extreme the extinction burst had become.
- Avoid reinforcing undesired behavior
- R+ for silence
- She was reinforcing for longer periods of quiet and pushed too far so the bear went into an extinction burst
- This was a somewhat distressing video – the bear sounded horrible – but it did show how extinction bursts work, and how bear did seem to be able to turn the vocalizing on and off.
Bear outside grabbing people: This was a short clip that showed a young bear who had been allowed to interact with people in ways that were no longer safe, now that the bear was getting bigger. Barbara had them teach the bear to station on a log instead of grabbing at or trying to climb people.
- Young bear used to interacting with people
- Teach stationing
Rough play with fox: This fox had been rescued and allowed to play rough with people. He had learned to bite at hands when on a person’s lap. The zoo wanted to be able to have people pat him, without the fox getting excited and biting at people. The first step was to teach alternative behaviors that were appropriate when he was handled by people, and then introduce patting.
- Avoid triggers
- Teach other desirable behaviors
- Gradually raise criteria
Spitting orangutan: If you thought it might be fun to work in a zoo, well…maybe not. This was an orangutan that had developed an annoying habit. I think Barbara said that she did this to avoid interacting with the keeper, but I’m not sure.
- Would slurp up urine and spit at keepers
- Avoid triggering it
Bouncing leopard: This leopard’s cage was at the end of a hallway and the keepers had to go down the hallway to access something on the wall opposite the leopard’s cage. When they did that, the leopard would start bouncing off the walls – even though they were not intending to interact with him. Ignoring the leopard made things worse so they had to teach the leopard what to do when someone was in the hallway.
- Would bounce off the walls when keepers were near
- Cue for appropriate behaviors
- Keep engaged
- Be quick with reinforcers and bridging
A lot of the examples did use the same basic strategies, but it was interesting to see how Barbara had to adjust based on the animal’s individual needs, the environment and whether she was addressing a problem behavior or teaching something new.
In some cases, the behavior would happen as soon as the trainer approached so it was a matter of focusing on antecedents. In others, an unwanted behavior had accidentally been reinforced, so it was a matter of teaching alternative behaviors and building a strong reinforcement history for them. In others, it was about focusing on teaching new behaviors. There was a strong emphasis on teaching the animal to participate and reinforcing for desirable behavior.
I think we can learn a lot from looking at how zoo animals are trained and while many of her “problem” behaviors are not the same ones we tend to encounter with horses, there are good ideas that we can take from all of them. Targeting and stationing are just as useful for horses as they are for zoo animals. One that sticks in my head is teaching the leopard to go to the laser dot by doing a little each day. I’ve done this with some horse behaviors and found incorporating a minute or two of training as part of routine handling can make a big difference.