Science History: Do elephants ever forget?

It’s time to go back into the annals, to a time when applied behavior analysis, a budding biological and philosophical branch of psychology and medicine, was in its relative youth. A time when the exploration of behavior was catching on at a fever pitch, and the whole scope of life on earth could be put under the microscope of a growing analysis that wanted to peel away all excess explanatory fiction and get to the truth of it all for one burning question: Do elephants ever forget? That’s right. It was 1975 and even now, we’ve all heard the reputation about elephants. Intelligent creatures. Great memory. Potentially endless ability to recall. Was that true? Could it be possible to test it? With three elephants, and a few boxes of sugar cubes, it was a great day for science. It was the day we will all remember as the day we became sure that elephants are very, very good at memorization.

Published in the third annual issue of the Journal of Applied Behavior Analysis in 1975, four researchers took on the age old question and became heroes of empiricism and elephants alike. We had Hal Markowitz, Michael Schmidt, Leonie Nadal, and Leslie Squier. In the far and distant land of the Portland Zoo, they had an idea. The idea was that they could investigate the memory of elephants by constructing a test based around a plywood, plexiglass, and slate operant panel, and this apparatus would create a light-dark simultaneous discrimination task. What’s an operant panel? Think about the word “operant” here as something which modifies behavior by reinforcing or inhibiting certain effects and consequences. Operant be used to describe spontaneous behaviors that come under the control of other stimuli, which is also accurate for what the researchers were trying to get the elephants to engage in. Basically, it’s a thing where a stimulus is delivered (light or dark) an elephant can reach its trunk, press a disk, and if the light or dark disk matches the stimulus they were given, it records it as correct. This was a miracle of technology at the time, mind you. It had a galvanized steel feeder. It had microswitches. It was incredible. Built specifically for the elephants to be able to comfortably reach in with their trunks, press the disk of their choosing, and then get a treat when they were right. They learned to be right very often, and as you will see, retained that knowledge over a long period of time when they were set up to do it again.

Our subjects were three Indian elephants in the Portland Zoo, two originally from Thailand, and one from Saigon. Their names were Rosy, Tuy Hoa, and Belle. They were very willing participants in this experiment, uncoerced, and not even deprived of food to help their motivation. Why? Because all it took was sugar cubes. They loved sugar cubes. Once they got a taste of the sweet treat, wildly different from their usual diet, they were willing and able to learn this contraption and do the best operant chamber work an elephant had ever done before. By using the positive reinforcement of these sugar cubes for correct responses, learning came quickly. It would also turn out that with a great reinforcement system, and a willing elephant, they absolutely remember the best ways to get sugar cubes.

This was also the start of a great behavioral research program at the zoo. This experiment was unfortunately stopped within a decade due to administrative policies, but for a stretch in excess of 8 years, we learned that elephants truly had a knack for memory. The raw data, unfortunately, was also destroyed in a fire at Reed College later on, but the published work remains for us to enjoy.

Markowitz, Schmidt, Nadal, and Squier had a long term plan. When you break it down, this experiment was actually two experiments. First, to see if the elephants could master the original task of discriminating light with dark on corresponding plexiglass and slate circles. We learned that within six minutes of reinforcing correct responses, Tuy Hoa was amazing. She was a fast learner and had a hunger for sugar cubes. The other two also did well with some modifications (we’ll talk about that in a second). The second experiment happened a little over eight years later where a (mostly) salvaged duplicate of the apparatus was created, no extra training was given, and these three elephants tried again with stunningly accurate results. Tuy Hoa herself was able to reproduce mastery with only two mistakes. She still enjoyed sugar cubes eight years later too. Rosy and Belle were a little slower to catch up, but as it turned out with some further medical investigating between experiments, medical consultants on vision were able to confirm retinal damage through vascular deficiency in these two. With Belle, they attempted to use different colored lights to rule out the damage, but at the time the literature on elephant color vision was fragmentary and they were unsure if elephants in general could see color. They couldn’t be absolutely certain, but with the colored lights, the researchers did see much better responding from these two pachyderm ladies with blue and green lamps.

This study also had the benefit of shining a light (get it?) on the limits of medical research and parameters of elephants, which were common in captivity at the time, but had no matching depth of research. The dearth of specific biological information on what elephants could perceive took this simple study into their capability of memory, and raised questions on all other forms of visual acuity and behavioral research potential. As people would learn later, elephants actually have dichromatic vision, and they can see reds and greens fairly well during the day. It also turns out they can see blues and violets better at night. They can distinguish color, just not as well as humans do. That is about as far as my understanding in elephant vision goes, but I imagine that if our researchers knew these details back then, they might have had an interesting restructuring of their experiment based on light/dark using colored lamps. We can only speculate.

What we can take away, is that due to some fascinating research in 1975, we can be confident that elephants have the capability to remember a task they’d completed 8 years prior, and with enough sugar cubes, reproduce their results without further training. It was in their repertoire, and studies since have only confirmed (and exceeded) this original eight year span of memory.

Do elephants ever forget? Maybe.

Do they forget something that got them sugar cubes eight years ago? No way.

Comments? Questions? Leave them below.


Markowitz, H., Schmidt, M., Nadal, L., & Squier, L. (1975). Do elephants ever forget? Journal of Applied Behavior Analysis, 8(3), 333–335.

Image Credit: 

Stock photo:

Edits by: Christian Sawyer, M.Ed, BCBA

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s