• Andy Davis

A new study evaluates learning and memory in wild and captive raised monarchs!

Hello everyone,

How cool is that title? Yes, you read it correctly, a new study was just published on cognition in monarch butterflies! Cognition is a fancy term that means assessing how smart they are. I just read this new paper myself, and so today's post will be my summary of the study, and I will also include some of the more interesting pictures and info. I would encourage folks to download the paper and read it directly to get the full picture, since this summary will not be extensive.

So, let's start off with the link to the paper, which is free to download. It is in the entomological journal, Insects, and it is titled, "Exploring the Role of Cognition in the Annual Fall Migration of the Monarch Butterfly (Danaus plexippus)". It was written by a single author, Robert J. Gegear, who is a researcher at the University of Massachusetts. I don't know him personally, but I do know of his work, which is focused on various pollinators (including bees), and how they sense floral nectar resources. He also has done some work looking into the neurobiology of monarch migration and navigation! So from what I can tell, he is a verified expert in the research around how insect brains function!

From my read of this new paper, it looks like the overarching question he was tackling here was to understand how much monarchs can learn when they are migrating. As he put it, monarchs travel thousands of miles during the fall migration, encountering varying landscapes and habitats along the way. How do they keep track of which plants they should visit to get nectar from, especially considering that the landscape changes as they fly south? Do they simply figure it out through trial and error every single day (which would be tedious), or can they actually learn which plants have the best nectar, and then most importantly, can they remember what they have learned so as to avoid that trial and error? Wow, what a cool question, right?

It looks like he conducted a series of lab experiments to find this out, and using two groups of monarchs. The first group were wild-caught migrants, which he caught one fall in Massachusetts. He also states that he obtained some captive-reared monarchs from Chip Taylor at MonarchWatch. There were about 30 of each group.

There were two experiments, and they each involved a behavioral test, which I'll explain here. The goal of the first experiment was to figure out if monarchs could learn something new. To do this, Robert positioned each monarch in a tiny plastic "harness" so that it was upright and its wings were folded. It could move its head and antennae, and also its two front legs - this is important because butterflies taste with their feet! Finally, the monarch could unroll its proboscis when it wanted to - this is very important too, because their proboscis indicated their "desire" for something. A picture of this setup is pasted below. Sorry for the low quality figure - this is what was in the paper.

Once the monarch was in place, he set about "training" it to recognize two different colors, blue and orange, and he made it so that the orange was a "valuable" color and blue was not. He first showed the monarch a square piece of styrofoam that was orange and at the same time he touched the monarch's feet with a qtip soaked in sugar water. Once they tasted the sugar they unrolled their proboscis and drank. This then made the monarch associate orange with reward! He then showed the monarch a blue color, and touched its feet with a qtip soaked in plain water. This made the monarch associate blue with no reward. He did this over and over for one day. He indicated that pretty soon (actually within minutes) most of the monarchs figured out that whenever they saw an orange square, they got a tasty snack. So by the end of the training period, the monarchs were all trained to unroll their tongue whenever they saw orange, and not when they saw blue. Crazy!

It is important to point out here that before the training started, he first ensured that the wild monarchs did not already have a preference for the color orange. See the paper for more details on that.

Below is the rest of the figure describing this training. Note the monarch proboscis in the lower two panels, which apparently are after the completion of the "color training". The monarch is sticking out its tongue when it sees the orange square and not when it sees the blue square.

I guess you could say that this first experiment was all about "how to train your monarch"!

There was another component to this first experiment that I won't get into much, but it sounds like he also trained monarchs to sense two different odors, using this same proboscis-unrolling setup. He trained the monarchs to prefer one odor over the other. I think this was very clever.

For the second experiment, Robert set about testing how much the monarchs could remember, from this training. He used the same harness setup and procedure here, and essentially, it sounds like he exposed each monarch to the same color cards for 7 consecutive days after the training period. But importantly, this was with no reward (no sugar water). He wanted to know if the monarchs continued to unroll their proboscis to the correct color, based only on the memory of their initial training. And, he wanted to know for how long they retain this memory. Thus, he scored each monarch on how many days they correctly reached for the right color with their proboscis.

Once again, he also tested their memory-retention of the odor training too, using the same design.

Before I show the results of these experiments, keep in mind that he tested learning and memory of two groups of monarchs - wild caught, and captive-reared. While he did not put much emphasis on this element in the paper, I am sure that this is something that a lot of people would find interesting, especially given the growing body of research around captive-rearing of monarchs.

Below are the results of the first experiment on learning ability. It looks like Robert created a number for each monarch that represented it's "learning ability", and the averages of all groups are shown in this graph.

Based on these numbers, there were no obvious differences between reared monarchs and wild migrants in their ability to learn something new. And, there were no meaningful differences between male and female monarchs in learning ability either. Very interesting.

Now for the memory tests, here is a similar graph, showing the average memory scores for reared and wild, males and females. Think of these scores as the average number of days the monarchs retained their training.

Here, there was a pretty obvious difference between reared and wild monarchs - reared female monarchs had very low memory scores compared to wild females, but this was only for the visual (color) tests. But for some reason, reared males had pretty good memory in each test. Wild males and females scored very well in all tests.

Collectively, these results indicate that 1) monarch butterflies are capable of learning something new when they need to, 2) wild monarchs at least are capable of remembering what they learned for almost a week, and 3) there is some evidence that captive-reared monarchs have lower cognitive ability than wild monarchs.

I read through the discussion section of the paper, and I noted that despite the lengthy summary and discussion of these results, it sounds like he was surprised to see the poorer cognition of the captive-reared monarchs. Either that, or, he chalked this up to the fact that they were apparently reared under conditions resembling summer. He may or may not be right here. Really, the rearing issue here is confounded with the seasonal issue, which means there could be a few reasons why the reared females performed so poorly in the memory tests.

Robert indicated that this study is really just a first-pass on this issue, and that there is a lot more to learn. He's right there. I think there is a lot more that could be done to study monarch cognition, like evaluating how anthropogenic effects like road noise, or agrochemicals affects cognition. Or how about testing the effects of parasites like OE on cognition? I can think of many scenarios and questions that a good scientist could ask, especially using this nifty test setup he established here.

I also got to thinking that memory and learning are probably important for a lot more than just finding nectar during migration. In fact there are many other facets of the migration where cognition and/or memory, would be just as important. How about finding appropriate roost sites for resting each night? Or, how about remembering which types of roost trees are best for roosting? How about learning to avoid crossing large, dangerous water bodies? Or maybe even learning to avoid cars? Wow, when you think about it, these monarchs face a barrage of new things on a daily basis during the two-month migration, and, their ability to learn and remember (the good and the bad things), must somehow factor into their success or failure.

Let me end here with a final thought about the cognition of captive-reared monarchs, since captive-rearing is on everyone's mind these days. Despite warnings from experts, there are still a lot of people releasing captive-reared monarchs each fall and sending them on their way, thinking that they will be successful. The results from tagging already tell us that captive-reared monarchs have a much lower chance of migration success than do wild monarchs. After reading this paper, I'm now wondering if this (poor cognition) is at least one of the reasons why reared monarchs don't do as well! Who'd a thunk?

That's all for now.


Direct link to this blog post:




The science of monarch butterflies

A blog about monarchs, written by a monarch scientist, for people who love monarchs