Training a Tortoise to Help with Blood Work

This is an old study and it is not about dogs. What caught my attention though was the age of the subjects and the species it involved: Giant Aldabra tortoises.

Weiss E., Wilson S.(2003) The Use of Classical and Operant Conditioning in Training Aldabra Tortoises (Geochelone gigantea) for Venipuncture and Other Husbandry Issues.Journal of Applied Animal Welfare Science, 6(1), 33–38.

Introducing the 4 subjects

  • Male 1—May 1932, 252 kg
  • Male 2—July 1967, 142 kg
  • Female 1—December 1950, 102 kg
  • Female 2—July 1959, 123 kg

We don’t normally think of tortoises (land dwelling Testudines) as intelligent, motivated or all that trainable.  The four subjects of this study, two male-female pairs prove that is a poorly held prejudice. When you take the age of the subjects into consideration, ranging from over 36 to over 70 years, the trainability of the subjects becomes even more impressive.

I say we, because the authors also seemed surprised by how well their subjects responded, not unreasonable given that “all animals were naive to training procedures.

The tortoises responded better than expected to the training procedure. Because the training program is restricted to positive reinforcement techniques, we were restrained by the strength of our appetitive stimuli. One often does not think of tortoises as being easily motivated; in fact, they are quite motivated. They responded well to food reinforcers of carrot, papaya, and cantaloupe, but more interesting was the fact that they responded equally well to tactile reinforcers such as neck and leg strokes.

This is a lesson for people who like to say “My dog isn’t motivated”, and use that as an excuse for the use of force. If they would assume responsibility, then the statement becomes, “I am not motivating my dog.” By not blaming the dog and top blaming the dog and it’s something that can be worked on.

So how do you train a tortoise? Exactly the same way you would train a dog. The procedure should be very familiar to anyone familiar with clicker/progressive reinforcement training.

Phase 1: Training. The tortoises first were classically conditioned to associate the sound of a clicker with food. A training session consisted of 10 to 15click–food combinations. All four tortoises learned the association by the seventh session. It was determined that the tortoises made the association when they made a food search after the clicker presentation.

Phase 2: Operant conditioning. Once the tortoises had made the association between the clicker and the food, operant conditioning began. The first goal was to have the tortoises approach a target for a food reward. Targets can be any object, such as a stick, a paddle, a ball, or any object that the animal can sense and approach. Because Aldabra tortoises seek out bright-colored fruit in their natural environment, it was decided that the tortoises would be more likely to approach a brightly colored target than one that was dull. The target chosen was a red plastic ball attached to the end of a dowel.

Training began by presenting the target 1 to 2 in. in front of the tortoise’s face. If the tortoise made any movement toward the target, the clicker was sounded and the tortoise was given a food reward. Once the tortoises were reliably touching the target, they were placed on a variable reinforcement schedule. (This schedule of reinforcement allows a trainer to shape and modify a behavior with a lower risk of the behavior becoming extinguished because of nonreinforcement.)

The authors also note the importance of dedication even if things don’t go as well or as fast as you would want.

Male 1, refused training for 2 months other than slightly extending his neck to touch the target. However, this animal is now our most eager participant in the training.

This is not an uncommon experience when training animals.

Phase 3 continued with reinforcement and added some desensitization

The second trainer began by just squatting by the front leg of the tortoise, then touching the inner leg, and, finally, desensitizing the tortoise to a small-gauge needle. The tortoises always were on target while the second trainer manipulated the leg. The tortoises reacted to the needle insertion with a slight neck retraction or a grunt but then immediately returned to holding on target while the blood collection procedure continued. The tortoises are required to focus and hold on target during the procedure.

And that’s it. That’s how 4 old tortoises were trained to assist and participate in their own care. The authors report that training continued and the tortoises are now trained to climb onto a scale to be weighed without any of the stress inducing lifting that it once required.

So if your dog or cat acts up during visits to the vet, just imagine what a little training could do.

If a 71 year old reptilian herbivore can be trained to take a needle…

….you can figure out the rest.


8 thoughts on “Training a Tortoise to Help with Blood Work

  1. I have trained lots of animals that I owned, and horses I did not. Turtles and tortoises would seem untrainable, but I had a turtle that I found wild in my back yard, and after a year, would come when called. I had him for five years. I know this sounds weird, but for an ultimate test, I took him to a large open pool near the falls and let him go. He swam out about twenty feet. I whistled and made the motion of having food for him and he came swimming back to me. I finally did give him his freedom shortly after that. It always amazed me that a turtle could be trained. And a tortoise seems to me to be a ‘smarter’ creature. Not sure why either.

  2. Pingback: Gopher Tortoises and My Biologist Son « Old Things R New

  3. Pingback: Click for Health | Science of Dogs

  4. So great to see this happening in other zoos! Our vet team and I were the first to accomplish this in captivity with our Aldabras at the Audubon Zoo. This makes me so incredibly happy to hear about it happening elsewhere. Every species deserves the chance to be a willing participant in all areas of their life in captivity.

    • Self-regulation is very important to all species and I consider it – aside from the physical aspects – an essential aspect of animal (and human) welfare.

Comments are closed.