The Secret (Smart) Life of Bees

IRA FLATOW: This is Science Friday. I’m Ira Flatow. A group of researchers from Queen Mary University of London have trained bees to push a ball around the ground. And then when they do that, they receive a sweet treat reward.

Researchers reporting in science noted those soccer-playing bees were displaying something called cognitive flexibility. And it got us wondering, are bees the smartest insect? Numerous studies show just how good they are at telling the difference between things. They’re great at navigation, even teaching stuff to other bees.

So what does this tell us about the bee brain, and what about bee cognition? Do they actually think? Helping us to dig deeply into what’s behind bee intelligence are my guests, Dr. Lars Chittka, professor of Sensory and Behavioral Ecology at Queen Mary University of London. Dr. Felicity Muth–

LARS CHITTKA: Good evening.

IRA FLATOW: Good evening to you– is a post-doctoral fellow at the University of Nevada in Reno. And if you have questions about the intelligence of bees, give us a call. Our number, 844-724-8255. You can also tweet us, @scifri. So Lars, are bees the smartest insect?

LARS CHITTKA: I don’t know.

IRA FLATOW: [LAUGHS] They’re pretty smart, though.

LARS CHITTKA: The reason why a lot is known about bee intelligence is because they come in large numbers, at least if you’re talking honeybees and bumblebees, and so it’s relatively easier to gain large quantities of data from them. But you find in other insects, very remarkable abilities. There are some species of wasps where all members of a colony recognized one another individually by their facial features. There are some solitary wasps that, as opposed to bees who have only one home, have multiple nests at a time and have to remember the locations of all of them. So there’s probably plenty of intelligence in other insects that has been poorly explored.

IRA FLATOW: But you explored the bee in the study. You did the bee study, the soccer bee study that came out last week. What did you learn about the bee behavior in that experiment? And were you surprised at how well the bees did?

LARS CHITTKA: We were very surprised, especially by one particular twist that I’ll come back to in a moment. Now it’s long been known that bees are able to associate rewards with certain aspects of flowers, such as colors and patterns and sense and so on. But in scenarios where the reward isn’t that immediate, it wasn’t that clear how well they would perform at that. And so in this particular case, the bees had to use a ball as a kind of tool or a token in the same way that you use, perhaps a token to put in a vending machine in order to obtain a sugary sweet bar or something like that. They had to learn to move that ball from a particular destination to a goal, a target destination, in order to obtain a reward.

And that alone, I think was remarkable in that so far, we knew that bees would be able to sort of push various things out of their way in order to gain access to a reward, which is sort of similar to what they do in a complex flower, like a snapdragon. But in this case, they really had to have the flexibility to move an object from A to B. But what was even more remarkable is the way in which uninformed individuals that had no prior exposure to this task learned it from their sisters. And the remarkable thing was, not only did they copy the right technique and copy the notion of where to place the ball, but these observers actually did something better than what they had seen the skilled bees do.

And this is how we did that. We played a little trick on the skilled bees, in that there were three balls available, each at a different distance from the goal. And obviously the smart thing to do would be to move the closest ball to its target. But we’d glued the closest ball and the second closest ball down so that the skilled bee only ever moved the furthest ball to the goal destination. And at the same time as these skilled bees performed that task, we let an uninformed individual that had no idea what to do with that ball into the arena to allow it to observe what that demonstrator, as we call it, did with the ball.

And once the ball was in the target destination, they always both got a sugary reward. And after doing that three times, we put the previously naive individual on the spot and let it alone into the experimental setup up without the presence of a skilled demonstrator. And now this observer had a choice between stupidly aping, so to speak, what they had seen the demonstrator do– that is, pick the furthest ball and move that to the target.

Alternatively, they could copy the technique and do the right thing with the ball and pick the closest one. And the latter is what they did. So rather than just imitating, if you wish, what these naive bees had seen the demonstrator do– that is, move the furthest ball, the previously naive individuals now did the clever thing. They did the right thing with the ball. And they picked the closest ball to the target and moved it there.

IRA FLATOW: Felicity, does this mean that bees are intelligent, do you think?

FELICITY MUTH: Well, I think that one of the problems is that we use this word “intelligence.” But a word in itself, it doesn’t actually mean that much, because the word “intelligence” probably means something different to every person. And as humans, we tend to think of intelligence in very human terms. But of course, every animal’s cognition is just evolved to deal with the situation that the animal has to deal with. So how animals think reflects what they need to think about in their environments. So I think what this shows here is a kind of very specific form of cognition that bees use in their natural environment.

IRA FLATOW: And why, Felicity, do we know so much about bee behavior? Why do we study them so much over the other insects?

FELICITY MUTH: Yeah, that’s a good question. I think that we’ve had a kind of fascination with bees for a really long time. So the very first learning experiment with bees was actually done around 100 years ago. And the person who did that experiment, Karl von Frisch, won the Nobel Prize for his findings.

So I think that because of the relationship we have with bees– for example, honeybees– we’re kind of naturally drawn to them. But also they work really well in a lab environment. They’re kind of the best subjects you could have, because they’re endlessly motivated to partake in experiments. And they have all these interesting cognitive abilities that makes them really rewarding to work with.

IRA FLATOW: Mm-hm. Lars, tell me about the bee brain. Does the bee’s brain look impressive because it can do all these things?

LARS CHITTKA: Well, it looks impressive because it can do all these things, but it also looks impressive if you look inside. So we are impressed, of course, by the tremendous diversity of both innate as well as learned behavioral routines that a bee’s brain can produce. And I guess all too often, people will assume that because a bee’s brain is little– which it undoubtedly is, it’s no larger than perhaps a pinhead– that it might therefore be simple or not complex.

But of course, once you actually examine its structure– granted, it has only about a million nerve cells. But each individual one of them might be as complex in its structure as a fully grown oak tree. And of each of these million cells, a single one of these nerve cells might make contact with perhaps 10,000 or 100,000 other cells in that same brain. So it’s a long way from being a simple brain, but perhaps it’s simpler than obviously a human brain with its 85 billion neurons. And so therefore, we were hoping that we can use bees as a shortcut to understand integrative brain function and multitasking.

IRA FLATOW: It’s fascinating, the structure of the bee brain. Felicity, I know that you study animal cognition. And so I’m asking you, how do bees stack up against other animals? Can you measure the intelligence against, let’s say, against a dog or something like that?

FELICITY MUTH: Well, I think that what we try and do as animal cognition scientists is not necessarily look at cognition or intelligence overall, but rather pick out one particular cognitive ability. And what I really like about the recent study that Dr. Chittka had out, and a handful of studies he’s had out recently is that they showed these seemingly complex behaviors that bees have, but because of the ease of working with bees, we can actually get at the mechanism underlying these behaviors. And so this is useful if we want to compare it to behaviors in other animals– for example, dogs or chimpanzees or even humans– because they’re showing similar behavior. But actually it’s a system where we can get at what might be the mechanism underlying this behavior.

IRA FLATOW: Here’s a tweet in from Angela who asks, “Did you use a controlled color arrangement for the balls, like similar or dissimilar to flowers?” Lars?

LARS CHITTKA: That’s a good question. So the balls in the training were all yellow. And the background on which they are presented is blue, so in that sense these are colors that will often be found at natural flowers. But of course, the task at hand was very different from what bees do at natural flowers.

So in the sense that you can perhaps explore animal intelligence to the effect that the cognitive performances depart from what you see in natural behavior, bees are obviously very good at learning colors. In fact, much better, faster at doing that than pretty much any other animal, human infants included. And the reason is because they’re natural-born color choosers. Colors have natural meaning to bees, whereas they don’t to cats, obviously, for example. So in that sense, the task was perhaps facilitated by, as far as it is from a bee’s natural task, but it might have been facilitated by the colors we chose.

But it was interesting to observe that the bees actually didn’t stick narrowly to the colors that they had encountered or experienced during training because, if for example, after this demonstration phase by a skilled demonstrator, we replaced the yellow balls with a black one, which is really not what you typically find in flowers, certainly not a black movable item, the observer bees were still able to solve the task and even solve it when they were faced with this threeway choice of which ball to move. So they disentangled their strategy from the particular color that they had encountered during training.

IRA FLATOW: What kind of bees were these?

LARS CHITTKA: These are bumblebees of the European species Bombus terrestris.

IRA FLATOW: Would we then extrapolate to all bees the same kind of ability, or do you need to test each one?

LARS CHITTKA: I think we may have to test other ones on a case to case basis. But my guess is that if you were test it on any other common bumblebee species, you will find similar affordances.

IRA FLATOW: Mhm. Let’s go to the phones. Let’s go to Daniel in Statesboro, Georgia. Hi, Daniel.

DANIEL: How are you doing, Ira?

IRA FLATOW: Hi there. Go ahead.

DANIEL: I have a question. At my property, we have a large bee that comes back at the same identical location year after year. And if we go out there and get close to it, it doesn’t attack us. It just kind of looks at us. It runs off every other flying thing in the air and just comes back year after year.

I’m just wondering, how old do these bees get? I just can’t imagine it’s the same bee. It’s probably been eight to 10 years. And do they pass their knowledge on, to be able to keep doing this behavior?

IRA FLATOW: Interesting question. Let me get an answer for you. Thanks for calling. Lars, what do you think?

LARS CHITTKA: I suspect that in that case, it’s not a social bee, but a solitary bee. And it’s likely that what you’re seeing from one year to the next are sequential generations of the same species of bee. But that’s very hard to tell without actually seeing what kind of bee it is. But it’s quite possible.

IRA FLATOW: I’m sorry, do I understand though, that bees can recognize faces, right? Would they know? If you had the same bee coming back, would you be able to know it was you?

LARS CHITTKA: So we have trained bees to recognize black and white images of human faces. No, that doesn’t mean that their face recognition is as elaborate as ours– that is, that they could necessarily recognize a human face under natural conditions. So when beekeepers ask, will my bees recognize me? Maybe, maybe not. So I think the task of recognizing a moving face that has to be recognized from all kinds of angles and in different lighting conditions is probably more complex.

IRA FLATOW: I’m Ira Flatow. This is Science Friday on PRI, and Public Radio International. The delay is a killer. [LAUGHS] I have a few minutes left. Lars, if you were to do more research, what more do you need to know? What would you like to know? If I gave you a blank check to study bees, what would you use it for?

LARS CHITTKA: One of the things that we are extremely curious about, which you’d already mentioned, is how can such a little brain juggle so many tasks and learn so many things and store so much information? And for that, we’re working with a number of very skilled modelers to build as realistic as possible models of the bee brain to explore how, with minimal circuitry, you can get so much done. And that’s something that we are trying to explore over the next few years.

IRA FLATOW: Felicity, does studying bee intelligence tell us anything about the causes of colony collapse, or what we might do about that?

FELICITY MUTH: Yeah, I think studying bee intelligence or cognition is really useful for informing the decline that we’ve had in bees. Understanding cognition is kind of what we would call basic research. It doesn’t have an obvious application. But if we want to understand how things like pesticides and habitat loss are affecting bees, then we need to have a baseline of how they’re behaving under natural conditions without those environmental perturbations. So I’d say that it’s really important, informing how we think about the declines that we’ve had in bees.

IRA FLATOW: Mm-hm. Is there any reason why you’d want to train bees, Lars, to do any other things, anything useful? And by training a few, teach the rest of the colony?

LARS CHITTKA: It depends on what you have in mind by anything useful. So I think one reason why we study bees, and we actually have quite a bit of funding from an engineering research council at the moment, because of the obvious question of how can you navigate reliably over long distance with essentially 0% errors, by using such a tiny microcomputer as a bee has? So I imagine you could steer a jumbo jet from start to finish with a computer as large as a bee brain.

And yet, the tasks in essence are by and large the same. You need to get an object to fly through three dimensional space and get it reliably onto the ground and off the ground, and so on. And so of course, there is potentially quite a bit of applied value in studying how they process information.

IRA FLATOW: Dr. Lars Chittka, professor of Sensory and Behavioral Ecology at Queen Mary University in London. Doctor Felicity Muth post-doctoral fellow University of Nevada in Reno. Thank you both for taking time to be with us today.

FELICITY MUTH: Thanks, Ira.

IRA FLATOW: You’re welcome.

Copyright © 2017 Science Friday Initiative. All rights reserved. Science Friday transcripts are produced on a tight deadline by 3Play Media. Fidelity to the original aired/published audio or video file might vary, and text might be updated or amended in the future. For the authoritative record of ScienceFriday’s programming, please visit the original aired/published recording. For terms of use and more information, visit our policies pages at http://www.sciencefriday.com/about/policies/