Listen to the whole episode to learn more, and also get a sneak peak into Dr. Ja’s unpublished (oooh!) data. I also talk to a few other fly researchers to get a bigger picture about why people still use fruit flies, even when politicians call it a waste of government dollars.

To conduct this study, Dr. Ja and the members of his lab came up with a 3D printed system that allowed them to video tape lots of fruit flies at once and track their eating and movement. They called it the Activity Recording CAFE (ARC), building off of the previous development of the CApillary FEeder (CAFE) to measure fly “meals” alongside behavior like sleep. The lab made a series of tutorials and have shared the code on github to make your own system, if you are so inclined:

In the episode, Dr. Ja talks about the big dip in sleep during eating (since the flies can’t sleep-eat!), that then goes up afterwards, showing that they are sleepier after eating than before. Here’s one graph from their paper that shows it, where the y-axis is “probability that the fly is asleep.” Can you see that big dip at “Time=0” (while the fly is eating), and the bump up on the right (after eating) that is higher than the left side of the graph (before eating)?

We didn’t even get to the cool experiments Dr. Ja’s lab did, where they looked at the specific neurons involved in post-meal sleepiness! You can read more about it in the paper, where they found a set of cells in the brain called “Leucokinin receptor neurons” that were necessary for flies being sleepier specifically after eating protein! Here’s a slice of a fly brain where they used the power of fly genetics to label those neurons in green:

Thanks you all for reading! I loved making this episode of Frivolous Science for you, and aim to continue exploring these “seemingly silly” studies with you all - aiming for a more regular “monthly” pace in the future. :)

I actually interviewed a few fly scientists in the making of this episode, and will be coming out with a bonus episode that is my full interview with Dr. Christina May. I had only used a couple short clips of her in this episode, but we spent almost an hour talking about why flies are used by scientists, and about her previous work in fly feeding behavior (as well as why flies are adorable).

Subscribe to get notified when the next episode is out, here and on your favorite podcast app.

Subscribe now

What do you think about taxpayer dollars going to study food comas? Do fruit flies have a funny name in your language? Any ideas for future episodes? Please share your thoughts in the comments!

Leave a comment

Full Credits

Creator & Host: Houda Khaled

Link to the preprint: https://elifesciences.org/articles/19334

Music by Blue Dot Sessions (www.sessions.blue):

• Spark

• Pasta Frola

• GriddleKake

Additional music from https://www.classicals.de/:

• Brahms - Symphony No. 1 in C Minor, Op. 68 - II. Andante sostenuto

Soundbite Sources:

1. Rand Paul, Remarks at American Spectator Gala, 2/11/2015

2. Sarah Palin, 10/24/2008

3. Negin Farsad & Adam Burke, Wait Wait… Don’t Tell Me!, 6/7/2025

TRANSCRIPT

(ARTWORK DESCRIPTION: Text reading “Frivolous Science” with the “S” styled as “$” at the bottom. Blue background with a cartoon of a fruit fly under a pink striped blanket, with zzz’s coming from its head. A thought bubble above has cartoons of ice cream, french fries, and a burger, with text beside them reading “sugar…”, “salt…”, “or protein?”, respectively.)

HOUDA KHALED, HOST: If I were to ask your mom what you’re working on, what, what do you think she would say?

KAVIN NUÑEZ: Oh, God. I actually don’t know if she knows what I’m working on right now.

KHALED: What would she say?

NUÑEZ: Uh, she’s, “He’s working with mosquitoes.” “ And trying to understand the brain.” And that’s probably the extent she’d say.

KHALED: Dr. Kavin Nuñez is a researcher at NYU who studies fruit flies, not mosquitoes. As an undergraduate, he researched the effect of alcohol on brain health using human samples. As a PhD, he kept exploring alcohol in the brain, but switched from human samples to fruit flies.

NUÑEZ: They understood my undergrad work a little bit more. I started getting further into the flies, like, “Wait, you’re doing this all in mosquitoes, like little flies? Um...” And in th- this case, they were thinking like mosquitoes themselves. I’m like, “No, not doing any mosquitoes.” I, I don’t actually remember the name of how to say fruit flies in Spanish, but we had, like, little nicknames for them, like, “the things that go around the fruit when you leave them out.” I started losing them a little bit then. They were just a little confused.

KHALED: It’s not just Kavin’s parents who get tripped up, but politicians too. Here’s Senator Rand Paul talking about work into how biological sex influences aging.

SOUNDBITE OF RAND PAUL, AMERICAN SPECTATOR GALA: They spent a million dollars trying to determine whether male fruit flies like younger female fruit flies. I think we could have polled the audience and saved a million bucks.

KHALED: And former Alaska governor and vice presidential candidate Sarah Palin talking about research into the genes that cause autism.

SOUNDBITE OF SARAH PALIN: Sometimes these dollars, they go to projects having little or nothing to do with the public good. Things like fruit fly research in Paris, France. I kid you not.

KHALED: You can find jokes about research in fruit flies everywhere, especially if you combine that with drugs, like researchers did when they looked into genes that might make us more susceptible to cocaine addiction.

(SOUNDBITE OF WAIT WAIT… DON’T TELL ME!:

NEGIN FARSAD: Wait, so but why did these scientists do that?

ADAM BURKE: ‘Cause they were also doing cocaine. Exactly.

PETER SEGAL: They were like, “You know, you know what’d be really cool? All right, you know what’d be really cool? How about...”)

KHALED: But there is a certain segment of the population that loves fruit flies.

CHRISTINA MAY: Flies are cute. If you ever look at a zoomed in picture of a fly head, they’re super sweet. They got cute little antennae with little feather on them. And when they clean themselves, they just look adorable. So I, yeah, here’s, here’s some fly propaganda. They’re adorable. Go Google them. I love them.

KHALED: Fruit fly researchers revere fruit flies, and not just ‘cause they’re cute. The first thing a researcher will bring up if you show any doubt about the importance of fruit fly research is-

NUÑEZ: The Nobel Prize was for the work in fruit flies.

WILLIAM JA: I think there are six or seven Nobels now, given out for work that was primarily from the fly model.

KHALED: Six Nobel Prizes awarded for work done in fruit flies. That’s included research into the basics of how traits are passed down in families, how the body fights off infections, and the internal clock that controls our sleep and wake cycles. But why fruit flies? I’ll leave that to Folke Henschen, who gave this speech when awarding the first fruit fly-related Nobel Prize to Thomas Morgan back in 1934. And by that, I mean my friend Josh reading the speech in an old-timey voice.

JOSH WILKERSON (READING FROM FOLKE HENSCHEN, 1934 NOBEL BANQUET SPEECH): Another cause for Morgan’s success is no doubt to be found in an ingenious choice of object for his experiments. From the beginning, Morgan chose the so-called banana fly drosophila melanogaster, which has proved superior to all other genetic objects known so far.

This animal can easily be kept alive in laboratories, can well endure experiments that must be made. It propagates all the year round without intervals. Thus, a new generation can be had about every 12th day or at least 30 generations a year. The female lays about a thousand eggs. Males and females can easily be distinguished from each other, and the number of chromosomes in this animal is only four.

This fortunate choice made it possible to Morgan to overtake other prominent genetical scientists who begun earlier, but employed plants or less suitable animals as experimental objects.

KHALED: So lots of reasons to love flies. You can make tons of them super fast, which is convenient if you’re trying to understand how genes are passed on between generations. And what scientists have first discovered in flies has often been found to be true in humans.

Nowadays, fruit flies are very popular among neuroscientists. They have brains, but they’re much smaller and simpler than ours, giving scientists a more manageable way to study the fundamentals of brain circuits.

MAY: If you come at it from a neuroscience perspective, it really opens up the kinds of questions that you can ask. You can really dig down into how precisely certain work in the brain, not just sort of broad circuits of many neurons, but like little loops of like three neurons that talk to each other in some sort of feedback loop or something like that. And you can really probe how the brain works on a more microscopic level, I would say.

And even more than that, how pared down can you get and still do really cool things? Because the other thing that’s really cool about flies is that they are insects, and insects are an ancient form of life. And I, I love thinking about how much time evolution has had to really trim down and make as efficient as possible the fly brain versus the human brain, which has greatly expanded in size and neuron number from, you know, the, the great ape primates that we evolved from.

Now obviously humans have way more behavioral niches. I’m not expecting my flies to write a symphony or something like this um, or even write anything at all. I think of the fly brain as like- the Lamborghini. It’s really sleek and it’s really good at what it does, uh, which is get a fly around so that it can find food and mates and that’s really all it has to do and it does an excellent job of that.

KHALED: Wow, Lamborghinis, Nobel Prizes, all sounds pretty great, right? But this podcast is called Frivolous Science not for nothing. We are not talking about groundbreaking, disease-curing research. Nope. Today, we are talking about the science behind food comas. Yes, that drowsy feeling after a nice big meal. Does it happen in fruit flies? Why would anyone bother studying it?

Don’t nod off yet. We’re about to get into it.

(KHALED: In Frivolous Science, we talk about whether research should get funded or not, but at this moment, a lot of it isn’t.)

Yeah. How’s your lab doing?

JA: So my lab is always on the, the brink of death. If we’re not fully funded, the lab closes down. And so anytime one of our ongoing grants is near the end about to expire, we’re always nervous about the next one. Because we cannot really be in between grants, we have to have the next one to keep the lab going or the lab shuts down.

And usually I’m not so nervous. Usually something picks up that saves us and keeps us going for another few years. But this time the market or the overall environment is, uh, much more challenging with the defunding of NIH and other biomedical research institutions, uh, it’s just much more difficult overall.

(KHALED: This is Dr. William Ja, a professor at Scripps Research Institute, Florida Campus)

JA: Yes, except our institute has actually changed. And so the institute name is now a mouthful, but it is, um, hold on a second. Lemme get this right. We are the Herbert Wertheim, UF Scripps Institute for Biomedical Innovation and Technology. But I usually call us UF Scripps for short.

KHALED: Okay. UF Scripps. Okay.

Before we got to his study, I wanted to know more about his motivation for studying flies in the first place.

So I’d love to hear before we get into the details of the food coma study, I’d love to hear your motivation for working in flies, because flies are pretty distant from humans, right? We already get a lot of people talking about why do we study mice? Why not study humans, and flies feel even further away. So can you comment on that? Why would you want to use flies as a model for biological processes?

JA: Sure. Each of the popular model organisms we use in science have different advantages and, and disadvantages. With flies, there are some very clear advantages for why we use them. Powerful genetic tools allow us to manipulate genes, insert disease genes, make mutations, and test them in a living model. They are cheap to maintain, they’re easy to grow up hundreds or thousands of animals to use for our experiments, enabling high throughput studies.

For me personally, my first projects in the fly were on aging, and so we were looking at ways of identifying or understanding genetic mechanisms and other environmental manipulations like diet or nutrition that affect aging, that extend longevity or extend fly lifespan. And for aging research in particular, using mammalian models is very difficult because of the long lifespans involved, right?

KHALED: Sure,yeah.

JA: Mice live several years, so if you wanna test the idea that this diet or drug extends mouse lifespan, it’s a multi-year experiment. Whereas in flies, they live about one to two months, 30 to 60 days. And so it’s much easier to test hypotheses or test diets or drugs or whatever, to see if you’ve extended fly life.

And then to do those genetic screens to identify the mechanisms that are involved for that longevity phenotype. And then you hope that what you learn provides some models or ideas that you can then push forward into either mammalian models or, you know, even preclinical or clinical studies.

KHALED: Yeah, that’s so interesting. I mean, it makes sense from a practical standpoint. If you’re working at the timeline of a scientist conducting a study, if you wanna get answers that you would try and pick an animal, an organism with a shorter lifespan.

JA: Right. And that’s specific for aging research. But, in our other work that segues into our, you know, neurobiology or behavioral work, we’ve always been really interested in, in trying to understand how a brain works.

KHALED: Mhmm.

JA: But a brain, especially a, a human brain or mammalian brain is very complex. And so one of the ways you try to figure out or dissect how our brain works is to use a simpler model system. And the fly is a fantastic balance between simple and complex.

The fly does have a brain. It’s much smaller than ours, obviously. It has much fewer neurons and connections. Yet this simple brain enables a bunch of reasonably complicated behaviors. They can make choices. They eat, they have to make feeding decisions. They sleep, they have circadian rhythms. Um, they fight, they show aggressive behavior. They have social interactions. And so this rich repertoire of behaviors in a simpler brain allows us to try to dissect how a simple brain regulates these behaviors. And hopefully that teaches us some of the strategies that a more complex brain will use to regulate similar behaviors.

(KHALED: But we’re not just here to talk about how great flies are. You’ve already heard about all of those Nobel Prizes. Let’s get to the part about food comas, or as Dr. Ja calls it, postprandial sleep, which literally just means post-meal sleep.)

So then getting to the paper that first drew me to your work where you identified this change in sleep behavior of flies after they ate. I’m really curious about how you even got to look at that.

JA: Well, I guess that’s a good story. It was not born out of any hypothesis driven research, and I know that’s a, having a hypothesis and then designing experiments to test that hypothesis is a very important part of science. But a lot of the stuff we do in the lab is exploratory and, um, well fun.

And so that project came about because we were interested in the regulation of feeding behavior, understanding how animals decide when and what to eat and how much, uh, and so forth. And to, to do those kinds of experiments in the fly, we had to develop, uh, new tools or new techniques to quantify feeding behavior. Basically we used 3D printers to make arrays of enclosures to keep individual flies in. And we put a little food in there and flies can eat the food, and we have video cameras tracking the flies in each of these chambers. Each fly is individually housed in a, vertically oriented chamber that doesn’t have too much depth, but there’s enough depth for them to walk around.

And the way, the way we measure feeding behavior is by putting liquid food. So these are food solutions, uh, usually yeast and sugar water, that we put into a glass micro capillary. So it’s basically a tiny glass straw. We have a video camera or webcam just recording the straws and, uh, watching the food move, and we can calculate the amount each fly ate in each chamber.

KHALED: Then so you’re using the capillaries, watching that to track how much they ate.

JA: Correct.

KHALED: And at the same time, you’re tracking their movement, right?

JA: Right. And, uh, from locomotor activity, at least in flies, you can also derive sleep. It was shown previously by Paul Shaw and, and, and other fly sleep researchers that periods of immobility where the fly hasn’t moved for five minutes or longer are highly correlated with sleep, that it’s highly likely that the fly is sleeping if it hasn’t moved for five minutes. And so we had all of these behavioral metrics from a single recording, from a single video. And we just wanted to know what is the coolest thing we can analyze from this giant, you know, pot of data.

Um, And more, maybe more specifically, we had this tool that no one else had. And so another question I I, I gave to, to the grad student, Keith Murphy, who pioneered this work was what is something that our device can measure that no one else can, you know, at that time. And so we came up with the idea of trying to do this feeding-sleep correlation. At that time there was a lot of evidence that diet and nutrition and feeding behavior is associated with sleep, but at least in flies, it was very challenging if not impossible at the time for researchers to capture feeding and sleep data from the same individual fly.

KHALED: Okay, but this setup was perfectly built to be able to do that.

JA: Right, right. So we have all this feeding and food intake data, and sleep and locomotor activity and all these metrics for individual flies. And we, we asked what is the, you know, coolest thing we can analyze out of it. And so, um, postprandial sleep or food coma was an easy idea that popped into our heads. We just said, do flies show food coma? Do they sleep after eating?

And it wasn’t entirely clear how we analyze for that. If you could imagine flies are feeding from the straw. They suck from the straw, they run away, they come back, they suck from the straw again, how do you turn all these feeding events into, you know, something we could call a meal. And do flies eat meals, do or are they constantly grazing? Um, are there discreet enough feeding events that we can try to see if there is some sort of sleep associated with each of these feeding events? And so we eventually did a bunch of things to, to show that, yes, flies do eat, uh, what we are calling discreet meals. They eat five to 12 times a day, either bigger meals or smaller snacks, but they are just, you know, discreet feeding events. And so, by recording 60 flies over several days, you get hundreds or thousands of meal data points. And we had no real idea of how to analyze for postprandial sleep. And so I feel like what we did was very naive, but it seemed reasonable. We basically took these thousand meals that we had and we lined them all up at at T equals zero, and we simply asked, what does sleep look like before and after each of these meals?

KHALED: Okay, so you just, for all of the flies altogether, every time it ate, you just counted that as time equals zero. And then you measured 20 minutes before that and 20 minutes after that, right?

JA: Right. Roughly 20 or 30 minutes. Another way to say is we, we basically took all the feeding events, the meals, and we lined them up together at the same time. It doesn’t matter if they ate in the morning or at night, or in the middle of the day, we just lined up all those thousand or 2000 meals. Line them all up at the same time and ask what does the average sleep look like for the 20 or 30 minutes before that meal, and the 20 to 30 minutes after that meal.

Does that make sense?

Okay. So at T minus 30 minutes to the meal, there’s some average baseline of sleep. And as you reach the countdown to T equals zero, that sleep plummets to almost nothing. And that makes sense to us because flies cannot sleep and eat at the same time, right?

KHALED: Yeah, we’re not, sleep-walking, eating.

JA: Right, so sleep plummets. And then immediately after T equals zero, after that, that meal, uh, you see that average sleep start to come back up, start to rebound.

KHALED: Yeah.

JA: And so it looks higher on the post meal side than the pre-meal side. And we did a couple different things to analyze that further. We summed up all that sleep post-meal and compared it to the same amount of time of sleep summed up pre-meal and we showed that the post-meal sleep was more than the pre-meal sleep. So that’s, that’s one way to do it. Another way you can do it is just to plot it. And you can see it temporally that as you get away from the meal, that sleep rebound from zero to 15 or 20 minutes after the meal is higher than the pre-meal sleep.

Does that make sense?

KHALED: Yeah, for sure. So then on average, these flies tend to be more likely to be sleeping after a meal than before a meal.

JA: Right. Right.

KHALED: How much variability did you see from fly to fly?

Do you,

JA: A lot. And from meal to meal. And so that’s one of the things that was kind of exciting by this work, when we dug into the literature. Because we all anecdotally feel like food coma is a real thing, right?

Especially after a big meal or a buffet, I get sleepy. How have people not studied this for years and years? But it turns out anytime, there are only a few papers on this, but anytime people try to study food coma or postprandial sleep in humans, they get a very variable result. Sometimes they see the effect, sometimes they don’t. Sometimes the subjects get sleepy after eating and sometimes they don’t. So it’s just a tough, tough thing to, capture or record. And if you believe our fly data, right, if you believe that what we’re studying is something similar, it might explain why it’s been so hard. So for any individual meal, after that meal, the effect on sleep is really, is very a, a very broad range of phenotypes, ranging from slightly arousing. Right, so that the fly didn’t even, didn’t sleep more, it was slightly aroused afterwards, all the way to deeply sleep inducing. And so the only reason we see this average increase in postprandial sleep is because we have one to 2000 meals as data points to work with. And that’s something that’s very tough to do in, in mammalian models to get that much data.

KHALED: Yeah, so this is one of the really big advantages of flies in general, where you can analyze so many and also your specific system where you’re assaying 20 or so at a time for a long time in an automated fashion. So there’s not someone who’s checking a box every time a fly is eating. So it’s something that you can scale up.

JA: Right, exactly. And this initial analysis, this idea of increased sleep after eating these meals wasn’t enough to convince us that we were seeing real postprandial sleep either. There was a number of experiments to follow that up that finally convinced us that what we were seeing was real sleep and it was really happening after meals.

KHALED: So this is one thing that you mentioned was that there’s previous data showing that inactivity means that a fly is asleep. Inactivity for five minutes is highly correlated with sleep. But then you went on to check, is that really the case in our data?

JA: Right, right. So that’s one of the things we did was to try to do another test of whether they’re sleeping or not. The not moving for five minutes is a pretty good measure and it’s really easy to make. But if we’re trying to come up with this brand new phenotype, this postprandial sleep phenotype, it would be good to double check that that was real. And so we did that by quantifying something called arousal threshold. If you see someone resting or, or not moving, how do you know that they’re really sleeping or just not moving because they’re resting, right? And the way you test that is because, uh, if they’re really sleeping, they have a higher arousal threshold, meaning when you poke them, it takes a harder poke to get them to move if they’re sleeping. Whereas if they’re just resting, it takes a, a lighter poke.

And so we did an arousal threshold test with the flies after meals by basically gluing, uh, cell phone vibrator coins to the back of the chambers and increasing the amount of force of those vibration, the G-Force of those vibrations after meals and showing that after meals, the arousal threshold , did indeed increase, suggesting that it’s really sleep.

KHALED: Mm. And did you actually get those from cell phones?

JA: You can buy them from electronic shops or Aliexpress or something like that.

KHALED: So you just, gently vibrated the flies at first and then increased it, and you saw which level of vibration led to

JA: That’s exactly right. That’s exactly right. You start at a low level vibration. You increase the force of that vibration, I forget what the program was, every five sec or 10 or whatever seconds. And then the, the level at which they start moving is the threshold that you score for that fly.

So we saw that there was an increase in arousal threshold after meals, uh, suggesting that increase in post-meal sleepiness is really sleep and not just post meal rest, restfulness or, or, maybe rest is the wrong word, too. It, it wasn’t just post-meal immobility. Right, you can imagine if you eat a lot, you just don’t feel like moving, but they might not be sleeping.

KHALED: Yeah. No, that makes sense. So you’re making this distinction between stillness and sleep.

JA: Right, exactly.

KHALED: Cool, yeah. So you found out this phenomenon, you’re able to confirm rigorously that there’s this postprandial sleep, post-meal sleep in flies, and then the rest of your paper is really looking more deeply into the mechanisms regulating that phenomenon.

JA: Yes. Yeah.

KHALED: So one of the things you looked at was how much they ate, right.

JA: Uh, yeah, just another way to break down that data. Uh, when I first described how we analyzed the data set, we took those one or 2000 meals and we lined them all up. But, by feeding the flies different diets that have more protein or salt or, or sugar, we can take those thousand meals and bin them in different ways and analyze the data again.

So the first thing we did was to take those meals and separate them into big meals, uh, medium sized meals and, and small meals. And you can see that the bigger the meal, the more postprandial sleep they show. And at least anecdotally that makes sense, right? It’s when we eat the big meals at Thanksgiving or at the buffet that we get our bellies really distended and we, I think anyway, that we, we feel the most sleepy after eating.

So the meal volume thing made a lot of sense. Uh, or we liked it anyway. But the other way we analyzed the data was to take again those 2000 meals. And then, uh, we first threw away the big meals because we already know the big meals, meal volume, drive sleep. So we just take the small meals, right. Now from those small meals, we can, uh, bin them separately into meals that had a lot of protein, a medium amount of protein, or a small amount of protein. And we can do the same bins for carbohydrates or sugars, high sugar, low sugar, and we can also do that for salt. So high salt, medium salt, low salt ingestion with these small meals. And that analysis revealed a specific meal components or ingredients that also drive this sleepiness after eating. And I think we were a little surprised because the drivers of that postprandial sleep turned out to be protein and salt, but not sugar.

KHALED: Yeah. I wonder what you made of that because it sounded like there was already data showing that sugar modifies sleep in flies in some way, right?

JA: Yes. But if I remember, those are usually if you change the diet completely and leave them on that for a while, how does, uh, sleep homeostasis or sleep balance change? This is sort of a more of a long-term effect, whereas postprandial sleep is really an acute or short-term effect that says when I eat that meal right there, if there is a lot of sugar in that meal, or very little sugar or a lot of protein in that meal, or little protein, what happens in the next 15 to 25 minutes afterwards to how sleepy I feel. So it is a, a different question, when we’re looking at it from this post meal perspective. And you’re right, we did find it kind of surprising because we thought, again, everything, all of these ideas come anecdotally, right? But we were thinking kids and, and Halloween and sugar, eating too much candy and the sugar crash so that these flies would also crash and get really sleepy, but they didn’t.

And in hindsight, maybe that makes sense. I think we had read or seen somewhere that this sugar crash temporally comes a bit later. Like after you ingest sugar, there’s some brief hyperactivity and if there is a, a real crash, if that, if that’s real, uh, it comes a bit later. So it’s just a, it may, the timing may be different for that than what we were seeing immediately after meal, after meals.

KHALED: Right. Did you ever go back and look that to see, because you already have that data, but you only really looked at 20 to 30 minutes before and after.

JA: Uh, yes, but I don’t think you can get anything out of that. It’s really difficult because keep in mind, we’re taking these thousand meals from all different times of day, right? Morning, night, uh, and whatever, and, and lining them up at the same time and looking at the sleep before and after. If you start looking farther away from the meal, like one hour, two hour, three hours, the time of day that that meal occurred really starts to matter, because your circadian sleep, the regular time that you sleep, starts to make a difference. Flies, like us, uh, sleep more at night than during the day, and that regular sleep pattern starts to make it challenging to see any longer term effects from meals.

KHALED: Right, so whenever you’re doing these additional manipulations, you’re trying to make this comparison of some baseline to, let me just change this one small thing, which is part of the benefit of working with these fly models. But no matter what, you still have so many other aspects. Food isn’t the only thing that’s affecting sleep. Like you said, it’s circadian rhythm, individual variability, and probably some other factors.

JA: We could do something. We, we didn’t do this, but we could do something like you suggest, uh, we could go back and take only meals that the flies ate between 11:00 AM and 12 noon. And then ask what does sleep look like between two and 6:00 PM, so a few hours later to look at a long distance effect. And that would sort of isolate the circadian component, right? Try to narrow it down and only select meals from a certain time of day. And then look at sleep only a few hours later, at another very specific time of day. Not during the night when they’re normally sleeping. And maybe that would show an effect, a longer term effect of sugar in the meal. But we never did that type of analysis.

KHALED: So what did you make then? So there’s a surprise about sugar not having an effect in the time that you’re looking at, but what do you think proteins are doing then? And you also saw an effect with salt right? So, what do you think? Sodium, yes. So what do you think both of those are doing that lead to this increase in sleep?

JA: Well, with protein, we thought it was cute because the urban legend has always been that trytophan from Thanksgiving turkey, uh, causes that effect. So, so protein seemed okay, seemed reasonable. Salt was a little bit of surprise, but we do put a lot of salt in our foods. At the time, we really didn’t know. The data was what it was. And then we, we went on to try to identify neuronal or genetic mechanisms regulating the behavior. But we now have unpublished data that points to some possibilities.

KHALED: Ooh.

JA: And so, uh, after that work, one of our unpublished pieces of data was to try to figure out what the purpose of postprandial sleep was physiologically. If you believe that we’re studying post-meal sleepiness and that this is a conserved behavior that you see in, in mammals, then it must be something important, right? Something evolutionarily conserved across species should be important and do something.

KHALED: And why? Why does it have to be important?

JA: well, I think the idea is that, if it doesn’t do anything, it, it would be lost. Right, if it’s a behavior or a gene, uh, this works at, at many levels, right? Physiology, metabolism, genetics, behavior. If it’s something that is not useful to the organism, it will be lost in future generations. Especially something that seems like it’s so costly to maintain, right? Sleep is usually, I mean, sleep is necessary, that’s why animals do it. But sleep is, uh, especially in the wild for wild animals, is, is typically a pretty dangerous thing to do. You’re immobile, you’re less sensitive to environment, uh, environment, right? You’re, your arousal threshold is higher. You’re more prone to predation, so it’s, uh, potentially a very costly thing to do. So if postprandial sleep is maintained in animals, you’d like to think it, it serves some important purpose.

And so one of the hypotheses, one of the ideas of what that purpose might be, uh, was something to do with digestion. Uh, that, that was the most obvious, right? Um, you, you eat, you eat a big meal, you get sleepy. It must, well, we thought it must help you digest or do something for digestion. And so it wasn’t easy to figure out how we would test that idea. But we developed a pretty cool experimental approach that’s really hard to, to do. But, we basically fed flies single meals that were spiked with radio-labeled sugar or protein. Okay? And then we would feed them these single meals and then ask several hours later how efficiently they absorbed that sugar or protein.

KHALED: Okay, how do you measure that?

JA: Basically by seeing how much of that radio label, that radioactive tracer, accumulated in the fly body. If you see a lot of radiation, uh, radioactive accumulation in the fly body, it means it effectively or efficiently absorbed that sugar or protein.

KHALED: versus releasing it?

JA: Right, right. We also had to collect the, the crap, right?

The, the ex, the, the excrement. The excreta, um, and compare that. And if you see a lot of it in the excreta, then they, well then they excreted most of it and they did not efficiently digest it or absorb it. It’s a real tedious experiment to do, working with individual flies with individual meals. But the data is very quantitative and very, very cool to look at.

So in the control experiments, the first experiments we showed that no matter how much sugar, sucrose, is in the fly meal ranging from, I forget what it was, one to 20 micrograms of sugar in, in that meal, the flies absorb all that sugar, like 98, 99% plus. So flies are very good at digesting, processing, and absorbing sugar.

KHALED: Okay.

JA: With protein, however, ranging from a similar range, uh, in micrograms of protein per meal, uh, you get this decrease in the percent efficiency of absorption. And so with very, very low protein ingestion with low protein amounts in the meal, they absorb most of it 95, 98% plus.

But as the amount of protein in that meal increases, that percent efficiency of absorption goes down to 85, and then 75, I think down to at the highest level in the meal it was 60 or 70%.

KHALED: Interesting.

JA: So flies are very good at absorbing and digesting sugar, not so efficient at processing and digesting protein. So we were super excited by this because, this matches the idea that, oh, postprandial sleep must be important for helping with that digestion, digestive process, right.

KHALED: Yeah, because it was already, if it was already very highly efficient, then you wouldn’t need some additional mechanism to help with processing it.

JA: Right, right. And so, we repeated that experiment with the protein absorption radio tracer, but this time after feeding the meal, we took half of the flies and we put them on a vortexer or, and basically shook them for, for a few hours.

Um, not the best experiment, but they, uh, they couldn’t sleep right for those few hours. And the other half of the flies, we just let them sleep after eating as they would normally do.

KHALED: Mm. a gentle vortexer, right. They’re

JA: Yeah. But I mean, we could see that they were moving around so they weren’t sleeping.

And so when we did this, and then we asked what was the percent efficiency, uh, we were surprised to see that when you don’t let them sleep, protein absorption efficiency, uh, very clearly went up,

KHALED: Mm. When you don’t let them sleep, their efficiency goes up?

JA: Right, they absorbed more protein, they were better absorbing protein.

KHALED: Oh, that’s surprising.

JA: So if that’s true, that would suggest that the normal purpose of postprandial sleep is to actually reduce the amount of protein that’s digested and absorbed.

KHALED: Huh?

JA: So then we had to go back and think, could this be right? Why would this be right? Uh, and, and we, we came up with two things. So first turns out that flies are actually, compared to, to mammals pretty bad at getting rid of nitrogenous waste. Proteins are the major, you know, nitrogen source from food. And because of that, high amounts of protein are toxic or, uh, detrimental to fly lifespan.

KHALED: Hmm.

JA: And so is salt or sodium flies being so small, are very sensitive to dehydration stress. And so salt or, or high amounts of sodium are also very toxic for them. And so both of these compounds, sodium and protein, induce this postprandial sleep effect. They’re both bad for the fly. So maybe postprandial sleep is indeed, the purpose of it is to, you know, get rid of these components or not absorb as much of it, when you eat too much of it in a big meal.

KHALED: Interesting. And this probably, this fits in with the behavior of what they want to eat, right, what they go for. Do they usually go for food that are more sugary rather than proteinaceous or salty?

JA: That’s hard to answer. So first it depends on sex. So the males and females have different protein carbohydrate targets.

KHALED: Mm.

JA: But then at a second level, when you let the flies choose between how much protein and carb they eat, they do generally choose, depending on sex, but they do generally choose the specific ratio of carb to protein. So it’s hard for me to say whether that’s a high amount of carb or a high amount of protein.

KHALED: Okay.

JA: Like you’re asking me whether the flies usually go for carb or protein, and I, and it’s neither they go after a balance. So if you let them choose, they go after a target. I can’t really tell you if that target feels high protein or, or high sucrose, unless I’m comparing it to something.

KHALED: Sure, sure. Okay. No, that makes sense. I guess what you’re telling me is your theory about post meal sleep and it’s decreasing the efficiency of protein absorption so I was getting a little speculative about other things that a fly might do to help it have less of that waste to begin with by consuming less of it when it’s controlling its own diet.

JA: Ah, yes. It does regulate its feeding to choose the appropriate level of carbohydrate and protein. Um, but it could be that, it, it’s major source of protein in the wild will be, uh, live yeasts, you know, fungus, uh, and bacteria growing on rotting fruits. And so, it may not be able to say, I’m, I’m only gonna have a little bit of protein. It starts eating these yeast on top of the rotting fruits. And if that’s a highly proteinaceous meal, it may balance that out, by sleeping more after that meal and excreting more of it.

KHALED: Mm. Yeah, that makes a lot of sense.

Dr. Ja also discussed some human data that might support the idea that post-meal sleep is involved in regulating how nutrients are processed by the body.

JA: The only other thing we have that made me feel, comfortable with, with this result, even though we haven’t, uh, published it yet, is there’s one human study. From, I wanna say it’s like 30 years ago now, where they took college students and they gave them a free meal and then they let some of them sleep and some of them not, right? Some of them, they, they didn’t let sleep. And they measured diet induced thermogenesis, which is a measure of how many calories you were able to process from that meal. And the result they saw was in the same direction as what we saw with our sleep manipulating experiments in the fly. There was more diet induced thermogenesis when you don’t let the people sleep after eating suggesting they were able to extract, you know, more calories from the food, not less. So that was kind of exciting when we found that one paper, um, because it was consistent with what we, in the direct, same direction that we saw and, and might suggest again that the phenotype we’re studying is something conserved across species, and that gives us more motivation to dissect it further.

KHALED: Okay. Interesting. So if you were, I guess you’re working on flies, but there’s this idea of it connecting to humans. If you were to design a human experiment that kind of extended what you were doing, do you have any ideas of what you would love to see?

JA: That’s a great question. I guess I’d love to see some of the stuff that was already done in humans repeated with bigger numbers. It’s a little bit, I don’t wanna say disappointing, but disheartening because the old literature, it was hard to see a consistent effect. And one of the leaders in that field, or, or the guy who was doing the most work published the most papers on this. It was his papers that would sometimes show a postprandial sleep effect in his human experiments and sometimes not. And so he wrote a opinion paper, a perspective paper, uh, years later, lamenting, I guess, why is it so hard to measure this phenotype in humans that we all sort of anecdotally feel or, you know, feel like it is real and exists?

And he just said, well, in the lab environment, it’s just tough, right? First, sleep experiments in humans is tough. You have to do it in a, you know, a sleep lab and people just don’t like sleeping when they’re being watched. I mean, it’s already a, you’ve already really perturbed them by putting them into this, this environment to do the experiment. And then circadian variables, just a lot of variables, that go into postprandial sleep, including, uh, when they last ate, how well they slept the night before, time of day. And all of those many variables that are harder to control or, or expensive to control.

And, uh, again, we were just fortunate that that was something we could do in flies by having these hundreds or thousands of meal events that we could dissect and study from, and then split further into meal components.

KHALED: Hmm. Yeah, you wish that someone would take this on, even though it’s pretty challenging to do because of all of those variables that you mentioned in people.

JA: Yeah, I would like to see some of the key experiments connected to meal processing or caloric processing or uptake from meals. And how they affect both sleepiness and the actual change in physiology or metabolism.

(KHALED: So even if post-meal sleep is something that really can be observed in humans and it does have some impact on metabolism, what then? For now, the answer is still we don’t know.)

JA: It’s hard to say right now, just because we don’t know quite enough yet about the biological relevance of the phenotype. It’s just attractive right now because the phenotype, the behavior looks like it exists in multiple organisms. And so, you know, we’re still conjecturing, we’re still speculating and hoping that that means it’s a relevant and important behavior for further study.

But on the other hand, I mean the, the human body, your body, you know, tells you what it wants. And so if you get sleepy after eating, maybe you should listen to your body and, and take a nap. And we don’t quite know what that does yet, but it might be important.

(KHALED: And this research lives on not just in the form of their continued work on post-meal sleep, but in the simple setup they designed that’s been recreated by other labs to study their own questions about individual fly behavior. That’s included works on the effect of social isolation on sleep, brain circuits controlling circadian rhythms, and age-related changes in metabolism.)

JA: Most of the stuff we do, especially for this project, are fairly inexpensive. It used to be 30 years ago when you made custom behavioral chambers, you’d have to have a machine shop make them for you out of plastic or acrylic, and they’d, they’d, they’d cut them for you and, and assemble them for you and be thousands of dollars. Now they have 3D printers that cost, you know, a hundred or $200. you can 3D print and test all sorts of custom chambers until you get the one that that works best. The cameras can range in price, but, we found that we didn’t need high resolution cameras to do what we wanted. And so I think we were using Xbox cameras in our first version of the experiments that were, you know, 50 or a hundred dollars each. So I really did enjoy this project and, and developing the, those methods, in that everything we’d done was pretty, uh, inexpensive . And that other labs, uh, a few other labs, picked up. So that’s been really, uh, great to see, when you develop method and, and a few others, you know, use it and are happy with it.

KHALED: Thanks so much for listening. You may have noticed the large gap between my first episode and this second one, but PhD life got the best of me. I can see the light at the end of the tunnel though, as I’ll be defending in September. And I’ve got plenty more to come before that. As a small gift to thank you for your patience, I’ll be putting out a bonus episode with my interview with Dr. Christina May. You might remember her from the beginning of the episode.

MAY: This is what I want you to tell the world. Flies are cute.

KHALED: We actually talked for almost an hour, and she had lots more interesting things to say about flies in general and her own research, so keep an eye out for that.

You can find the entire transcript and a link to Dr. Ja’s paper on our website, frivolousscience.substack.com. I’ll also be uploading some videos of Dr. Ja’s 3D-printed fly array that was used in the study. Frivolous Science is made by me, Houda Khaled, with music by Blue Dot Sessions. Thank you to Josh Wilkerson for his fantastic voice acting, to all the fly scientists I talked to, including Dr. Kavin Nuñez, Dr. Christina May, and of course, an extra special thanks to our guest, Dr. William Ja.

I don’t know how we got onto the subject, but he started complaining to me about his friend who never finishes his iced coffee because the bottom gets too watery.

JA: They suck from the straw, they run away, they come back, they suck from the straw again.

KHALED: Huh, couldn’t be me. Well, I’m Houda Khaled, and this was Frivolous Science.

So excited to share this first episode, where I talk to Dr. Ralph Emilio Peterson about his work following rats around New York City with a mic and a camera. We talk about how he first got interested in tracking rats around the wild, what sorts of differences he’s noticed already in wild rats that hasn’t been observed in lab rats before, and what his hopes are for future studies! Spoiler: They involve a “research dumpster” and rat “hearing” tests, among other things.

Take a listen to learn more, and also hear some recordings of rats “talking” that we can’t perceive with our own ears! You can also hear a clip here, overlaid over my thermal camera imaging of NYC rats in a community garden:

I, as many other New York City residents, see rats every day. But I was surprised how much fun it was to quietly look for them in the dark, and watch them hop around on camera! However fun, these unsteady videos with finger pointing popping up every once in awhile (thanks Anat :D) are definitely not ideal for the types of sophisticated analyses Dr. Peterson et al. developed in their study. He kindly sent over a sample of the videos they actually used, which you can see here:

The difference is obvious, right? If you read the full preprint, you’ll be able to see some more details about how they analyzed the videos. This is the graph that Dr. Peterson and I discussed that measured rat sizes from these thermal imaging video, including the “rogue squirrel”:

While the thermal camera used in the study is on the pricier end, the microphone is surprisingly accessible, which is what makes it potentially viable as a way of tracking rats throughout the city as a pest control method. It’s been used by scientists to record from birds, bats and.— after adding a protective case — even fish! Here’s a photo of it from the designer’s website:

Thank you so much for reading, I had a great time making this first episode of Frivolous Science, and beginning this exploration of what sorts of studies are (or aren’t) worth funding. If you’d like to get notified when the next episode comes out, you can subscribe on your podcast app, as well as on this substack:

Subscribe now

I’d also love to hear your thoughts after listening! Do you think this type of research should receive public funding? What do you think these rats are talking about? Any theories about whether wild rats need hearing aids or not? Share your thoughts in the comments:

Leave a comment

Full Credits

Creator & Host: Houda Khaled

Link to the preprint: https://www.biorxiv.org/content/10.1101/2025.07.21.665423v1.full

Music by Blue Dot Sessions (www.sessions.blue):

• Spark

• Jadie Grange

• Highway 94

• GriddleKake

Soundbite Sources:

1. Rand Paul, Fox and Friends, 5/15/2025

2. Joyelle Nicole Johnson, Wait Wait… Don’t Tell Me!, 11/22/2025

TRANSCRIPT

(ARTWORK DESCRIPTION: Text saying “Frivolous Science” with “S” as “$”. Purple and orange gradient background with simple cartoon rat in white with sound waves coming from its mouth. In the background is a city skyline.)

(AMBIENT LEAF CRUNCH SOUNDS)

HOUDA KHALED, HOST: Do you wanna tell me where we are?

SHWETA PATWARDHAN: Yes. We are in the M’finda Kalunga Garden on the Lower East Side of Manhattan, walking through the, over on the pathways with crunchy leaves, and so far no rat signs.

KHALED: On a cold night in November, community garden volunteer Shweta Patwardhan took a few of us on a tour, to look for rats.

PATWARDHAN: If you see pink flags, that’s where we found a rat burrow. Oh, so they might be close to where the burrows are actually. Then we could use your camera.

NATALIE OPPENHEIMER: Oh, yeah?

KHALED: Yeah.

ANAT MANO: I think there is a rat squeak.

KHALED: As in many other parts of the city, M’Finda Kalunga Garden has historically had a big problem with rats.

PATWARDHAN: We do try to avoid piles of garbage in the garden.

KHALED: I hope so!

In cooperation with the city, garden. volunteers have been making efforts to reduce rat presence. Not just to stop them from eating garden vegetables, but because rats can act as disease vectors to humans, damage infrastructure, and even cut wires to start fires.

PATWARDHAN: Sorry, Houda. Not that many rats, I guess.

KHALED: Yeah. It seems like your garden’s doing a great job. That’s awesome.

And while it took some hunting at first...

PATWARDHAN: Yeah, overall, I’m happy about this development.

KHALED: Eventually we found quite a few!

PATWARDHAN: Oh, there’s definitely something there. Oh, wait, wait. There it is! It’s running.

OPPENHEIMER: Oh my God, it’s running.

PATWARDHAN: Wait, is there two, are there- there are two rat!

KHALED: There’s two rats. Yeah. Oh, what are they doing?

KHALED: Using a thermal camera - in my case, a cheap one from Home Depot - we could see the rats pretty well, even in the dark or in their burrows.

PATWARDHAN: Is that…Huh, oh wait, do you think that’s it? Oh my gosh. I think we can see it in the burrow. Look, can you see it? How it’s moving around? Yeah. Whoa, that’s a lot deeper than I would’ve guessed.

KHALED: It’s coming back out. Oh, it just walked out.

PATWARDHAN: Oh my God.

KHALED: What do you think about your community garden pest management now?

PATWARDHAN: Well, technically this is outside of our formal jurisdiction, so I’m still feeling pretty good.

KHALED: All right. We need Shweta on it. Oh my gosh, there’s so many here.

Hi, I’m Houda Khaled and this is Frivolous Science where I talk to scientists conducting seemingly silly science that kind you hear complaints about on Fox News...

SOUNDBITE OF RAND PAUL, FOX & FRIENDS: Well, I compliment Secretary Kennedy and the Trump administration for finally saying, enough’s enough. We’re not gonna keep fundingfrivolous and crazy research.

KHALED: ...and even NPR...

SOUNDBITE OF JOYELLE NICOLE JOHNSON, WAIT WAIT… DON’T TELL ME!: Those scientists know we need to cure cancer, right?

KHALED: …but that can have a big impact on our lives.

Today, I talk to scientist Ralph Peterson, whose passion for understanding rodent vocal communication might indirectly lead to new solutions for rat pest control, just by listening in on their conversations.​

KHALED: Ralph Emilio Peterson is a Ph.D in Neuroscience from New York University, and he’s also my former classmate. For his dissertation, he looked at the ways that rodents communicate vocally. He found some interesting things, like that gerbil families have their own unique dialects compared to other gerbil families. He’s now a Postdoctoral Fellow at Basis, where he’s continuing his work studying animal vocal behavior in increasingly naturalistic environments. Would you say that’s accurate?

RALPH PETERSON, PhD: Yes, very accurate. Nailed it.

KHALED: Yeah? Also, you’re a musician.

PETERSON: Yeah, I grew up, playing jazz trumpet. That was of my first loves in life and passions. and that was actually what inspired me to go to grad school and study, the auditory system. I was always fascinated by. How music could invoke emotion. There’s this really surreal state that you can enter when you’re playing music and specifically jazz, when you’re improvising and, it feels like you’re transported somewhere new. it’s this very kind of euphoric state. And so I was always curious how it could be that music, could. interface with your emotions, how they’re so intertwined.

When I went to grad school, I decided that I wanted to study sensory processing, auditory processing to figure out, how sound hits the ear and travels through the brain and may or may not interact with, with emotional systems. And I specifically focus on vocal communication because, it’s so central to everyday life and, every species. And Very much intertwined with emotion. For example, if you’re on the phone with, a family member and they’re not feeling well, they’re sad or something, you could tell in their voice that they’re not doing great. So the, voice in a sense is like a readout of, I don’t know, the soul, if you will. There’s a lot of information other than just the semantic content of the vocalization that you can read out from speech.

KHALED: Yeah, I think that’s pretty intuitive in people’s experiences and I feel like that’s a really good example.

So then, what made you veer towards animal research rather than, I don’t know, looking at humans and emotions? It’s hard to read out the emotion of a gerbil or a rat.

PETERSON: Yeah. it’s a good question. there is some great work going on in, in human, vocal communication, people like Eddie Chang or Michael Long are doing experiments in humans as they’re, as they’re freely, vocally communicating. But there are some caveats, these are people going in for procedures to treat strokes and whatnot. It’s a very stressful environment. I thought it’d be interesting to look at rodents. They’re a standard model organism in neuroscience. They’re highly social and they speak in this ultrasonic frequency that humans can’t hear, so there’s a bit of, mystery as to what they’re saying.

There had been, decades of research in gerbils, specifically studying their auditory system because it was very amenable to auditory experiments. And in a parallel but totally non-overlapping area of research, there’s been a lot of field work done in gerbils to study their social capabilities. They have this kind of intricate social structure. They live in these underground borough systems with multiple generations pups in there, of children in there.

KHALED: Versus other animals where the pups would leave the nest and form their own families.

PETERSON: Exactly, yeah. That’s like the human model where you grow up with your siblings and then you are forced to disperse when you’re 18. that’s not the case in gerbils. It’s like a big, Catholic family where they’re all helping take care of each other. But the unique and kind interesting thing about gerbils is that some of the older siblings will take on the role of caretaker. So they’ll undergo like a sexual suppression where they actually cannot biologically reproduce. So they assume this role of caretaker called alloparenting. There’s, a study showing that the mother will force dispersal of specific individuals in the family based off of criteria that we don’t really well understand. And there’s studies in other, rodent species that show that in their burrow systems, they have dedicated chambers for using the bathroom or caching food or just nesting.

So there is this idea of they’re living in a big group that’s not just a random big group. There’s structure to it. Yeah. And there’s hierarchy and, So I wanted to understand how they use vocalizations to coordinate their social hierarchy and basically understand how the brain processes vocalizations at a neural level.

KHALED: So what makes you think that’s the likely way that they’re communicating? Because there could be all sorts of methods of communication. They can touch each other, look at each other, and bite each other, or use pheremones.

PETERSON: Yeah, absolutely. I think that it’s likely that those other senses are involved as well. But, if you think about a day in the life of a gerbil or a rat, or most rodents living underground, there’s not many visual cues. They have this really acute sense of hearing, and vocalize a ton. If you record a family of gerbils 24/7 like I did on my PhD, they’re vocalizing hundreds of times per hour. They’re not vocalizing to no end, I assume, and in very low lit conditions, I assume they’re using it for something. And the act of talking is, one, metabolically expensive. Like you need to expend energy to actually talk. Two, it’s, evolutionarily potentially precarious because you’re exposing yourself to predation.

So there are species that can hear in the ultrasonic range, like cats for example, that could in theory hear them and run after them and attack them. So all that to say, I don’t think that they’re vocalizing just because they can. I think that there is a purpose and, it’s not clear in the field what that purpose is.

KHALED: Yeah, but then, the way you say it makes it sound like it’s actually pretty obvious. So then there’s the alternative question, like, why haven’t people been able to figure that out? Does it have to do with it being computationally difficult, since that’s a big focus of your thesis and preprint?

PETERSON: Yeah, it’s, difficult. it’s just technologically difficult generally. So yeah. On the computational side, quantifying behavior in animals is a difficult thing to do. As I mentioned, there were decades of field work that were done in gerbils that were done by humans that wrote down on pen and paper field observations. Over the last, 10, 15 years or so, there’s been this, emergence of, machine learning in the field of animal behavior and neuroscience.

Where instead of, looking at animals, behave and writing down with pen and paper, what they’re doing, you can actually use machine learning to extract their calls and, quantify their vocalizations in principled fashion.

KHALED: And what are the benefits of that? I feel like we’ve learned so much from pen and paper and just observations. But what great information do we get  by collecting such large amounts of data and quantifying it?

PETERSON: Yeah, so one of the things we found in our studies in graduate school, we found that the gerbils emitted new and different vocal types than had previously been observed. And we think that this was in part because we put them in a very kind of naturalistic situation. So a family altogether, and we recorded 24/7 for the period of three weeks. So we just eavesdropped on whatever they did.

If an ethologist, 30 years ago were to do the same study, they would have to listen back to this audio over three weeks and like individually annotate every single time there was a vocalization. it’s just not a very scalable approach if you want to record data at scale.

And I think what scale gets you is insight into new and potentially, rare events that occur. A lot of these field studies, if you read back to the methods, they like, observed behavior for two hours because, that’s all they could muster basically.

One of the things I began to appreciate and become increasingly inspired by in my PhD was this idea that animals evolved to thrive outside of the laboratory. Yet we bring them into the laboratory and kind of constrain them relatively reduced environments in order to gain experimental control to understand how the brain works.

So the idea that the lab most that we study has been born in a lab, was shipped to us in a FedEx box from one lab to another lab and lives in a small, very reduced cage. And then we go in and we do a bunch of neuroscience experiments and claim that we’ve discovered something about how the brain works feels a little unsatisfying to me. Because this mouse, evolved to live outdoors and, in, the natural world and with its family and with the stress of predation,

KHALED: And not unlimited food.

PETERSON: Yeah, exactly. So there’s this idea that everything that we understand about the brain comes from this limited set of behaviors that are studied in the lab and whether or not they generalize to how the brain works outside of the lab is still an open question.

KHALED: So you think that in auditory communication, there’s a limited repertoire? So that a lab mouse will only communicate in more limited ways because they’re not having all of the life experiences that a normal animal has, or that a human would have - to extrapolate from these animals to humans. So that if you look at an animal in a real environment that’s experiencing a full life, you’ll find out  new things about auditory communication as well.

PETERSON: Exactly. You nailed it. Yeah, that’s a hundred percent the thesis for, this type of work is let’s go into the wild in first. Just document and record what the different types of vocalizations are. Then let’s quantify them and see how they match with what we see in the laboratory.

(KHALED: Dr. Peterson’s fascination with studying animal behavior in natural environments began during graduate school with his research on gerbil families. However, he’s now taken it to a whole new level. Today, instead of working with animals in a lab, he ventures into the wild - exploring places like Central Park, subway stations, and rat control zones around the city, following rats where they go, with his microphone and camera in hand. He talked about how this project got started, how they were able to do it, and the interesting things they’ve discovered so far about rat behavior in the wild.)

PETERSON: The project came about when, I reached out to. Emily Mackevicius who is one of the co-founders at Basis. She had been recording birds in Central Park using thermal imaging, interested in how groups of birds forage. So we met in the park one day, she was showing me her equipment and, I was coming from an interest in communication, so I’d asked her, “Are you recording sound? Are you, have you seen any rodents? what’s going on?” She’s like, “We haven’t, focused on rodents,” but she’s interested generally in urban species. We basically decided we would go out together, she would bring her thermal camera and I would bring my ultrasonic microphone, and that we would attempt to just document some rat behavior, not really knowing what we’d expect.

It was that kind of fun, like pure type of science that you do just because it’s cool, not because you have a super precise hypothesis. so it was very discovery based to start. And yeah, we went out a few nights and just scouted some areas. We went up to Harlem. There’s this area that was marked as the rat mitigation zone where they had a high density of rats. So for anyone else, probably in the city, they wanna stay away from the neighborhood. But for us, we decided to go right in there, and it was quite amazing.

One of the advantages to Emily’s thermal imaging camera is that it detects heat. so you can image through, occlusion. So if the rat is running around a bush in the park, you can see through the bush because the heat propagates through the bush. Yeah. So in cases where a regular camera would be occluded, the thermal camera kind of opens up a whole different realm for you. And on the microphone side, I have this little handheld thing here. It’s like maybe half the size credit card in length and width, and about as thick as like my thumb. Basically, I went out with this thing and I would point it at groups of rats, just hold it in my hand about two meters away from them.

KHALED: Two meters. Ok, so you have some distance.

PETERSON: There was some distance. Yeah. and they weren’t, they’re, these are New York rats, they’re not scared of people. it was plenty of distance for me to record some really high quality vocalizations. And it was just amazing that first time coming back, loading the SD card on the computer and seeing them.

KHALED: Because you can’t hear it at all, since our ears can’t detect...

PETERSON: Yeah, so human ears can hear up to about 18 or 20 kilohertz. Rats vocalize anywhere between 20, around 20 and 50 kilohertz. So yeah, all the, next time you’re around in New York City at night and you see a bunch of rats, just think to yourself like they could be, having a full on conversation to themselves and, we would never know.

So there’s something about that really intrigued me. First of all, we were able to record them. We saw that there were these beautiful sequences of ultrasonic vocalizations. It wasn’t just like one or two, it was like a sequence of 20 or 30 in a row.

KHALED: Very Chatty

PETERSON: Very chatty. Presumably it was coming from one of the animals because there weren’t a lot of, overlapping calls. You can usually tell if two of them are talking at once because the calls will overlap with each other. So again, it like piqued my interest about this idea of kind of social coordination. It’s what is the relationship of these animals? What are they talking about? What are the features in the world that they care about? And, just why are they talking?

(KHALED: Before we keep going, I think it’s time for us to finally listen to what a rat actually sounds like. This is a recording from Dr. Peterson’s study, pitched up [***Correction: down***] so that we can hear it:

[Rat vocalization recording]

So cute, right? Like Dr. Peterson explained, these calls can come in a range of frequencies, that have been associated with different meanings, or social contexts, based on studies in the lab. For example, the one I just played was originally close to 20 kHz frequency, which is usually thought to be an alarm call, while calls closer to 50 kHz are thought to happen while rats are eating or looking for food. And these are both different from human-audible squeaks, which are mostly in negative situations, like if a rat gets hurt or in a fight. However, all of that data comes from watching rats in the lab, and what Dr. Peterson found in rats in the wild did not always match up.)

PETERSON: So that was one of the big takeaways from this pre-print is that, there’s a couple things. The first thing is that the context in which we see some of these 22 kilohertz alarm calls emitted, have not been observed before. as I mentioned, 22 kilohertz is an alarm type of call. And to my knowledge it hasn’t been shown that they use that call type during foraging context. One of the things that I saw was I saw a rat jump into a trash bag, and I pointed the microphone at the trash bag And, eventually came up with a piece of food and ran away. And it was emitting, very long sequences of this, 22 kilohertz alarm call. in this foraging context.

(KHALED: Here is a clip of that rat in the trash can:

[Rat vocalization recording from trash bag]

What a cutie!)

PETERSON: So it was a little bit paradoxical. It was like, shouldn’t it be, shouldn’t that be a positive, situation? But, maybe things work differently in the wild. Maybe, actually it is, the intent was to alarm people because it is up above ground and it’s scared that it might, run into something while it’s in the trash.

KHALED: So its just like, trying to tell other rats, “look out for me”?

PETERSON: Could be, yeah, could be. That’s one of my theories for what the function of these, calls are. Is essentially like an acoustic beacon. I don’t know if you’ve ever played Marco Polo.

KHALED: Sure.

PETERSON: Like that, where you say a vocalization, it reports your position in space. Another animal says its vocalization, reports its relative position in space. That way you’re creating this spatial map of where your conspecifics are, where your friends are.

Again, the difficult thing about doing these experiments in the wild is that - and, you know, this wasn’t necessarily the intent of the study - but it’s more difficult to do a manipulation to prove any one of these theories to be true. If we’re in the lab, we would have a, a rat with different types of food sources and a speaker and a microphone and all these other equipment to test this idea that maybe they’re using different calls depending on like the nutritional value of whatever they found, for example. In the wild, we’re just eavesdropping and like hoping that we record enough data to make these kind of, these correlations, I can’t really say with any certainty what it is they’re talking about. Yeah. Just because we don’t have enough data to really say. A kind of zany dream of mine is actually to have a research dumpster, equipped with microphones and cameras and the ability to actually manipulate the types of food that we’re putting in the bags. I think it would be really fun next direction here.

KHALED: Yeah, because it seems like the big issue, the limitation of environmental recordings, is that you don’t have a good handle on manipulating the environment as you would in the lab. But the limitation of the lab is that you’re limiting everything about what the rat is experiencing. So it seems like the research dumpster is a good in between.

PETERSON: Exactly. Yeah. And the more and more I’m working with this kind of unwieldy data, the more and more I’m feeling like some sort of compromise like that is, is necessary to get to these causal questions that you’re asking.

(KHALED: In addition to their findings on rat vocalizations, using thermal camera imaging, Dr. Peterson was able to draw some preliminary conclusions on rat behavior from their video recordings using thermal cameras. He describes how rats seem to coordinate their movement, moving all at the same time in what he calls a “burst,” kind of like what you might see when a group of deer suddenly all run into the forest. Or, more NYC relevant, pigeons taking to the sky all at once.)

PETERSON: So we tracked a big group of like 20 rats or so, and in looking through the video we noticed that there are these periods of time where they would burst and they would leave the frame, leave the field of view, and come back.

But it wasn’t every, it wasn’t all 20 animals. There was like, and these bursts happen every, couple minutes. I think something, in the environment scared them away or something. So we were like, okay, what is, what can we say about the animals that are bursting versus non-bursting? Is there any difference between them? So we estimated the size of each of the animals in the, of the 20 animals.

KHALED:  I loved that figure, by the way, because when you showed the estimations of the size, there’s like this nice bell curve with all of the rats. There’s some variability in their size. Just like some are bigger, some are smaller, maybe some are babies, and then there’s like a low tail, and then there’s this little bump.

 And then at that bump, there’s just an arrow with a picture of a squirrel.

PETERSON: Yeah, it’s our rogue squirrel. Yeah. That made it in, which is roughly two times the size of a rat. Yeah. It actually took, some pretty sophisticated like probabilistic modeling to, to get to the, the three dimensional estimate size of each of the animals. So one of our co-authors developed a procedure to estimate the size of the animals. And then using that information, we started to look at the bursting behavior to see whether or not there was anything interesting there.

And basically what we found is that larger rats seem to be more bursty. So the bigger rats are the ones that run out of the frame right away when there’s presumably, a an inversive event or something. And the smaller rats were the ones that kind of stayed put and seemed to be, oftentimes they would burst, but only after the other animals had already burst. So they’re like a step behind. One of the things we’re kinda speculating about here is that probably the smaller rats are somewhat younger. So there could be this type of, real time social learning that’s going on where you’re seeing, a real time example of something scary happening in the world, large rat running away and showing, the younger rats basically how to behave in a proper context. Again, very speculative.

KHALED: So, just speculating, like the big rats already know how to respond because they’ve seen maybe like this, sound or thing they’re seeing is something they should be afraid of. But the younger rats don’t know it yet. So they have to learn from the adults. But the idea is now you  have the tools to investigate that.

PETERSON: Exactly, yes.

(KHALED: We also talked about some future potential ideas Dr. Peterson has for the project, besides research dumpsters, that would help deepen our understanding of rodent social behavior in the wild.)

PETERSON: Eventually down the line we’d like to do some perturbations or even potentially neural recordings and say, how is the auditory system different in the real world, what are the hearing capabilities of these animals in the real world and how do they match up to laboratory based animals?

KHALED: Oh, hearing capabilities, too?

PETERSON: Yeah, we didn’t touch on this in our study, but if we’re to like, speculate a little bit about what might be going on is you and I ride the subway in New York, daily I imagine. And it’s extremely noisy down there and. subway is home to many, rats that seem unperturbed by the subway system.

So I would strongly predict that they have some degree of hearing loss like going on. I would imagine, but no one’s ever done a, like a hearing test on a wild rat to my knowledge. So this would be like a base level type of question is measure ,you could do, it’s called the auditory brain stem response, ABR, to measure, hearing sensitivity at different frequencies. And I bet if you were to compare, subway rat’s ABR versus a lab rat’s ABR it would look significantly different.

KHALED: Wow. I was thinking about it in the other direction. Coming from my background, I think about like learning and memory, and enriched environments, and how much that changes animal cognition. I would think maybe that they would be more sensitive to a wider range of things because a lab mouse is experiencing a limited number of sounds versus a natural rat in the world.

PETERSON: Yeah. I could see that happening too, but maybe at a different level. So I would imagine that they could, that their auditory system is really good at categorizing sounds, more so than lab rats, for example. There’s this idea of like categorical coding of different auditory objects that people sometimes talk about in the field. And, like you’re explaining, living in an enriched acoustic environment with many more acoustic objects that your brain has to like experience. I would imagine that the brain of a wild rat would be tuned to, maybe differentiating between more acoustic objects. But I think at the, before getting to there, that might be like, at the level of that might be like a central nervous system processing question. I think before you even get there, like at the, at the periphery, like at the level of the ear, I think that there’s just gonna be some probably cochlear damage.

KHALED: So just at the level of being able to sense sounds versus interpreting them.

PETERSON: Yeah, exactly. And this is, being a bit speculative. but yeah, I think it would be, nevertheless, I think it would be cool to, one, at least do some of these, auditory brainstem responses from wild rats. And two, to your point, it would be great to do some recordings in like the auditory system and do some psychophysics. So whether or not they can discriminate between two different sounds And, wild rats versus lab rats.

(KHALED: Now, let’s revisit the big question: as taxpayers, is this research worth funding? Dr. Peterson’s study is relatively low cost. The priciest equipment was a $10,000 thermal camera, while many tools, like a computer vision software called YOLO - short for “You Only Look Once” - were free. Still, research like this has relied on government funding over the years, covering costs like scientists’ salaries and equipment to study rodent communication in the lab and in the wild.

So what do we gain from studies like this, besides that we should consider distributing earplugs to rats to help protect their little ears from the subway?

I think it’s important to remember that vocal communication is an integral part of most people’s lives. I was most reminded by this after my interview with Dr. Peterson, where he was showing me videos of his adorable baby.

[Baby giggles recording]

Do you hear those giggles? Can’t you sense her happiness talking with her mom? Vocal communication is something that is so important to us, its starts immediately after we’re born. And it carries so much meaning way before we learn to speak, with frustration in cries and joy in those adorable laughs.

And for those who struggle with vocal communication, research in the animal auditory system has already led to life-changing advances. People with hearing loss who want to regain their hearing can receive cochlear implants. But these technologies aren’t perfect: it isn’t an option for all types of hearing loss, and many recipients of cochlear implants still struggle to interpret sounds as language. Because of our limited understanding of how we interpret sounds in the brain, we don’t always know why this happens or how to fix it. A deeper understanding of social vocal communication could lead to better hearing solutions, and even help other people with vocal communication challenges, like individuals with Autism Spectrum Disorder or after a stroke.

In a totally different realm, Dr. Peterson’s team is working with New York City to improve rat control. With an estimated 3 million rats roaming the city, and millions of dollars poured into mitigation strategies that aren’t solving the issue, this is a growing problem. And it’s not just in New York - other cities aren’t faring much better.

Their research suggests that ultrasonic microphones could track rat hotspots in real time and monitor how rats respond to control efforts.  By understanding rat vocalizations, cities could refine their approaches - for example, using contraceptives in areas with many mating calls, or employing other strategies based on different rat behaviors. At the same time, researchers could gather more data to even better map vocalizations onto specific behaviors and improve pest control efforts.

Dr. Peterson emphasizes that this is just the beginning, with much more left to learn. But this research could not only help cities combat rat infestations smarter and faster, it may also lead to breakthroughs that improve the lives of those with hearing and communication challenges.)

So what’s your final verdict?

KHALED: Thank you so much for listening. This is my first episode, but look forward to another one in the new year, and if you enjoyed it, please share with your friends.

You can find the entire transcript and a link to Dr. Peterson’s preprint on our website, frivolousscience.substack.com, along with thermal imaging videos from my trip to the community garden, and much better ones shared with us from the original study.

Frivolous Science is made by me, Houda Khaled, with music by Blue Dot Sessions. Thanks to our rat garden party, Shweta Patwardhan, Anat Mano, and Natalie Oppenheimer, and to the many friends and family who have been incredibly supportive.

An extra special thanks to our guest, Ralph Emilio Peterson. You know, we were talking after, and I asked him about how he met his wife. He said it happened at a party where he walked right up to her and won her over with one line:

(PETERSON: So I’d asked her, are you recording sound? Are you, have you seen any rodents? What’s going on?)

KHALED: Huh, I’m surprised that worked. Well, I’m Houda Khaled and this was Frivolous Science. See you next year!

Are your taxes being wasted on scientist pet projects? What do headlines get wrong when they lambast studies on "fruit flies in Paris" and "shrimp on treadmills"? Welcome to Frivolous Science, where we meet with scientists behind some of the silliest sounding studies to get to the bottom of what their research is really about. The first episode premieres December 1st - tell all your friends!

Creator & Host: Houda Khaled

Music: Borough (Mole Rider) by Blue Dot Sessions (www.sessions.blue)

Soundbite Sources:

1. Rand Paul, Senate Health Committee Hearing, 3/5/2025

2. Fox News, 4/2016

3. Fox News, 5/10/2016

4. Sarah Palin, 10/24/2008

5. Ralph Peterson (Episode 1 sneak peek!)

6. Bankrupting America, 8/15/2011

TRANSCRIPT

(ARTWORK DESCRIPTION: Text saying “Frivolous Science” with “S” as “$”. Light blue background with simple cartoon depiction of a shrimp running on a treadmill.)

SOUNDBITE OF ARCHIVED RECORDING, RAND PAUL: “Well one way to have more precious resources directed towards good science is to get rid of all the frivolous studies.”

HOUDA KHALED, HOST: “Politicians will tell you that millions of taxpayer dollars — your hard-earned money — are being funneled into ridiculous scientist pet projects.

SOUNDBITE OF ARCHIVED RECORDING, FOX NEWS: “Next, more than $82,000, that’s how much money is being spent on creating stuttering mice.”

SOUNDBITE OF ARCHIVED RECORDING, FOX NEWS: “Alright I’m standing in front of our wheel of waste here, so let’s spin the wheel, and I want you to explain some of the waste that our government is spending”

SOUNDBITE OF ARCHIVED RECORDING, SARAH PALIN: “Some of these pet projects, they really don’t make a whole lot of sense, and sometimes these dollars, they go to projects having little or nothing to do with the public good. Things like fruit fly research in Paris, France. I kid you not.”

KHALED: “And while that last one might sound like a fun sequel to Emily in Paris, is it really worth the cost? In Frivolous Science, I sit down with researchers behind some of the silliest sounding studies.”

RALPH PETERSON, Ph.D.: “A kind of zany dream of mine is actually to have like a research dumpster.”

KHALED: “We get to the bottom of what headlines get wrong, what these studies are really about, and why it actually matters. Because while picturing shrimp on tiny treadmills might make you laugh…”

SOUNDBITE OF ARCHIVED RECORDING, BANKRUPTING AMERICA: “Wow, as in sea shrimp, as in, shrimp and scallop shrimp? …Exactly.”

KHALED: “…some of the biggest scientific breakthroughs often start with questions that sound absurd. You’ll hear directly from the scientists behind those unconventional questions, the surprising discoveries they lead to, and the joy of curiosity-driven science. Frivolous Science is a new podcast and the first episode comes out December 1st. Subscribe now to get notified when it drops, and tell all your friends.”

Politicians love to mock “wasteful” government spending on research — from shrimp on treadmills to the sex lives of screwworm flies. But what’s the real story behind these headline-grabbing studies? Discover what your tax dollars are actually funding, and experience the unexpected joy of scientific discovery that happens when curiosity leads the way.

Subscribe to get notified once our first episode comes out, and feel free to reach out with any ideas you have for future episodes!

A teaser for our first episode: Why should we care about what rats are saying while they dig through our trash?

Subscribe now