Transcript

Cold open [00:00:00]

Sébastien Moro: They wanted to know if the fish are recognising themselves by the face, or by the body, or a combination of both. We know, for example, that humans are using the face primarily.

So what they did is they took Photoshop, they took two pictures of fish — one of the focal animal and one of another one — and they cut the heads on the pictures, and put the head of the focal fish on the body of another one and the head of the other one on their body. And it appears that they’re attacking their body with the head of someone else, but they’re not attacking a picture of their face on someone else’s body. So it means that they can recognise themselves in pictures from the face, exactly as humans do.

This is the best mirror test ever. Actually, the animal that is succeeding at it the most convincingly, except for humans, is the cleaner wrasse.

Luisa’s intro [00:01:00]

Luisa Rodriguez: Hey listeners, this is Luisa Rodriguez, one of the hosts of The 80,000 Hours Podcast. I’m so excited to share today’s episode with you, which I had a lot of fun recording. I talk with Sébastien Moro about what it’s like to be a fish — from how they perceive the world and themselves to whether they can feel pain or emotions.

Fish are one of those animals that I have a hard time relating to — to be honest, part of me finds them kind of freaky and alien. But because Sébastien is super immersed in the evidence of what fish can do, and just generally excited to share how amazing fish are with the world, I came away from this interview with a wildly different feeling about fish and their capabilities and experiences than I had when I started it.

We talk about so much cool stuff, like:

• How some fish can beat primates at the mirror test and recognise individual human faces.
• How goldfish definitely don’t have a three-second memory — and an example of a fish that can do better on memory tasks than humans can.
• How some fish have sensory capabilities we can’t even really fathom — like “seeing” electrical fields and colours we can’t perceive.
• Whether fish can experience emotions, and how this is even studied.
• And of course, the ethical issues raised by evidence that fish may be conscious and experience suffering.

Without further ado, I bring you Sébastien Moro.

The interview begins [00:02:45]

Luisa Rodriguez: Today I’m speaking with Sébastien Moro. Sébastien is a science writer and a video blogger covering animal cognition. He focuses primarily on the animals that most of us find not so cute — so fishes, rats, pigeons, crows, et cetera. And that’s basically what we’re going to focus on today. Thanks so much for coming on the podcast, Sébastien.

Sébastien Moro: Thank you for asking me. I’m really happy to be here. And you’ll see that the “bad” animals are actually kind of cool.

Luisa Rodriguez: I’m genuinely really looking forward to it.

The wild diversity of fish [00:03:18]

Luisa Rodriguez: I actually mostly want to focus on fish today. I personally find fish hard to empathise with. They’re the group of animals I know the least about, and that I find the most intuitively hard to believe might actually be having some kind of conscious experience. But they’re also incredibly numerous. So if they’re suffering, that would be a huge deal. I think it’s even hard to roughly estimate the number of fish alive at any one time, but I think it’s somewhere in the trillions, and there are tens of thousands of known species.

So maybe just to start us off at a high level: I have a suspicion that I’m making a mistake when I think of fish as a homogeneous group. So how much diversity is there in the groups of vertebrates that are fishes?

Sébastien Moro: First, we should define what fishes are. And we don’t really have a clear scientific definition, because it doesn’t exist anywhere. “Fish” is more a common word for common people, but it doesn’t cover really well what all these animals are.

So if we want to follow the evolution, fish are coming from jawless animals which were like filtrating bags in the sea, something like that. And then you had a first split with the first animals with jaws so they could eat, and another one who didn’t have any — and lampreys are still belonging to this group. So today, modern fish are much farther from lamprey than they are from us.

And then obviously, when you have a jaw and you have big teeth, you start to win the race to evolution. So they started to kill pretty much everything, so lamprey and jawless fish pretty much disappeared. This is why lampreys are a very small group of animals.

And then you had a new separation, a new split between sharks, rays, skates — these kinds of animals who have cartilaginous bones and the bony fish. So today, it means that when we’re talking about sharks, we consider them as being fish — and they are, technically — but those fish are further away from the goldfish than the goldfish is from us in an evolutionary perspective, and so forth and so on. So many of the species of bony fish that we have today, they started to radiate, to explode, pretty much when we started to leave the trees. They are pretty recent animals.

We often think of fish as lower vertebrates, as if they were low on the tree of life and they didn’t evolve as much as us, which is pointless. They are as evolved as us. This is just that they followed another path; many of the actual species aren’t much older than us. So this is very important to keep in mind.

And the other thing is that when we’re talking about mammals or birds, altogether we’re talking about something like, I don’t know, 15,000 species. When talking about fish, you can double that, maybe even more. And they are living in a very wide range of habitats: they’re living in mangroves, they are living in oceans, in seas, in lakes, in rivers. So they have to face a lot of different challenges. And even in the ocean, they can be close to the surface or very deep. And these are not at all the same challenges they have to overcome; these are not the same problems they need to face.

So you have diversity in these animals that is probably bigger than the ones you have on land, in mammals. You have fish that live for something like one month; you have fish that can live… Actually, the Greenland shark, we have no idea how long they can live. We’re thinking they might live around 400 years, maybe even more, but this is not even sure. And you have fish with scales, others with no scales: catfish have no scales. You have fish who can breathe in the water, although they have to go back to the surface to breathe. I could continue for hours. The diversity is enormous.

Luisa Rodriguez: Yeah, you’re already pointing out some of my misconceptions when you’re pointing out that, in my head, fish live in water, and that’s a pretty uniform environment. But deep sea fish, fish in saltwater, fish in freshwater, that’s already going to be super different. Fish with access to lots of light, fish with access to lots of oxygen. So that’s already really helpful.

Can you give some concrete examples of the specific forms of diversity fish have evolved, given all these different environments?

Sébastien Moro: Sure. For example, one really easy thing to understand is the perception of colours. When we see colours, what is it? It’s wavelengths from the light from the sun. But the light in the water is not behaving the same way as in the air, because water is stopping different wavelengths. So the deeper you go, the less colours you have. For example, the longest wavelengths, which are red for us, stops at about one metre, two metres, something like that. Even five maybe. And the deeper you go, green is leaving, then blue is leaving, then there are no more colours.

And it means that for many fish — not all fish; you have counterexamples for everything — but many fish that live near the surface, like coral reef fishes, they are seeing four colours. We are seeing three colours, but many of them can see UV as well. So you have some fish who can recognise each other by the UV mask paint they have on their face. And it’s been shown that they only need this to recognise each other, they don’t even need something else — because scientists tried to train them with a black-and-white depiction of the UV spectrum, and the fish were continuing to properly recognise the fish they had to recognise. It helps them to precisely know who is from their species and who is not from their species, because they look alike a lot.

And we’ve shown, for example, that if you put a male between two cylinders and you have two other fish in the cylinders, if you have one of their species and one of the other species in the other cylinder, at first, if you just put a transparent cylinder, they will only attack the one of their species, because they know it’s kind of concurrency. But if the transparent cylinders are blocking UV, the fish don’t recognise each other and he’s just not attacking anyone. So we know this is how they do.

In the same idea, you have a deep sea fish where there is no more light, and especially you don’t have red anymore — which explains why, in some circumstances, many fish from the deep sea are red, because nobody can see red. But you have one fish, Malacosteus: they have a kind of small lamp just under the eye, and these fish are using red light to lighten the environment around them and to communicate. And at this depth, they are the only fish who can see red. So they can communicate and see other animals through a kind of secret channel that no one else can use. And this is exactly like IR security cameras. This is exactly the same thing. No one can see this light, but they can.

Luisa Rodriguez: That is pretty incredible. Are there any other high-level misconceptions people have about fish before I start basically asking you a bunch of questions about what it might be like to be a fish?

Sébastien Moro: Always. People only have misconceptions, they have nothing else. So I don’t really know what I could answer to that, because I could just spend all the rest of the podcast answering this question. There are many things.

The fish who can outperform chimps in cognition tests [00:12:24]

Sébastien Moro: I’ll just give a hint now, and probably we’ll talk about it later, but people have misconceptions about fish being less intelligent than land vertebrates. There are studies where fish are smashing chimpanzees.

Luisa Rodriguez: In what?

Sébastien Moro: And orangutans and capuchin monkeys. So…

Luisa Rodriguez: I actually just want to dive right into that. That’s too fascinating to not pick up on. What are the fish that smash chimpanzees?

Sébastien Moro: So we’re talking about a very small fish from the Pacific Ocean, which is named Labroides dimidiatus. The common name is cleaner wrasse. So these fish, as the name implies, are cleaning other fishes. The job basically is to eat the parasites and the dead skin off the other fish.

But the thing is, nobody likes to eat parasites or dead skin, so they prefer to eat the mucus off the other fish — which is their protective layer on them, a kind of gelatinous protective layer, which is more nutritious for them. So to do that, they have to bite their clients. And the clients hate that, so they consider it cheating.

OK, now we’re getting into a very complicated market system. These fish are opening cleaning stations. So they pick a spot, and then they go around to find different animals, and they start to massage their back with their fins. And we know that this rubbing with the pectoral fins is actually alleviating stress in the animals. It’s pleasant for them. They like having a massage. We have lab studies that show that clearly.

Luisa Rodriguez: We know that it reduces some stress hormone or something?

Sébastien Moro: Cortisol.

Luisa Rodriguez: Oh, it is cortisol. Wow.

Sébastien Moro: And then they start to have clients. The clients are coming to the cleaning station, and the cleaner wrasses are working in a harem way — so it’s a male and different female. You have to know that they can switch sex. So the male is the biggest fish of the group, but if another fish gets bigger, [that fish] becomes a male. So the male has to be sure the biggest female doesn’t eat too much to not be another male that will be a challenger.

Luisa Rodriguez: Fascinating.

Sébastien Moro: So the male is doing the policing there. And what happens is you have clients coming. They can have more than 2,300 interactions with different fish every day. They have between one and 200 different clients, all species included.

We know that these fish can remember each of the clients and their last encounter, which means if they cheated with one of their clients before and the interaction ended badly — I will explain in a second what that means — then they will make a really, really top-level cleaning next time.

So the clients are coming. Now, the cleaner wrasses know that there are different kinds of clients. For example, there are carnivorous fish, predators: you do not cheat with predators. Predators will always have a top-level cleaning.

Then you have different kinds of non-predatory fish. You have those who have very big territories, and in very big territories, they have different cleaning stations — so if they wait too much or if they see the cleaner cheating with the client, they will just go somewhere else. They have the choice. The cleaner wrasse knows this client, and she knows that if this animal is waiting for too long or is seeing a cheating interaction, it will leave. So these fish are always taken in priority, and they will never cheat if someone is watching.

And if you are a fish who is not a predator and who has a very small territory, then you have only one cleaning station, so the cleaner wrasse will cheat on you like hell. Then they’re just chasing the cleaner wrasses to punish them — and we know that it works, because in the next interaction, the cleaner wrasse will behave more cooperatively.

And we know that baby cleaner wrasses can recognise clients who are easily accepting to be cheated, and those who are really aggressive, and when they are a baby, they will follow what they learned from the adults.

And we know that the females are trying to cheat when they are out of sight of the male — which is the only study, from what I know, that is pointing toward the theory of mind. The theory of mind is being able to put yourself in the mind of another animal. It’s one of the highest levels of cognitive abilities, which is usually considered very linked to consciousness and higher order consciousness.

Luisa Rodriguez: Yeah. Actually just already, my alarm bells for a bunch of different cognitive abilities are going off. There’s memory of the last interaction; there’s theory of mind, as you’ve just said; there’s kind of making tradeoffs, and this ability to think about different options and choose between the best one in different contexts; there’s social learning, so learning from another fish, even as babies, which seems like that says something about how sophisticated the babies are.

I feel like there were others, but I’m losing track of them all. So I’m pretty struck by this already, that those are a lot of the things that I’ve learned to try to pay attention to when thinking about the potential sentience of a nonhuman animal. But sorry, I’ve interrupted. Go on.

Sébastien Moro: No problem. You’re actually right. They have a very complex social system, which makes them more able to answer questions that are more coherent for us humans, who are very social species. The questions we tend to ask are more adapted to these animals.

And last thing — because I could continue for hours on them — but last thing is that they can even understand the time intervals between when a fish is leaving and when the fish is coming back. The same clients. Why? Because the clients enjoy being rubbed, being cleaned — but sometimes they’re just getting cleaned, go away, turn around and get back in the queue. Because there are queues. And the cleaner wrasse knows that this client can’t get new parasites in the delay it’s been away, so it will not be taken at all. It will not be cleaned, because the wrasse knows that: “No, you can’t have any parasites. You just want massages. That’s it.”

So now that we know that, we start to understand why these animals could, on some tasks, be better than chimpanzees. And on many tasks. We’ll talk maybe about the episodic memory, and these fish are amazing for that. And all that I explained already is a pretty strong argument for episodic memory.

Luisa Rodriguez: Yeah, absolutely.

Sébastien Moro: So the task on which they have been put against chimpanzees is the following. You give them two plates. To make it easier, we’ll give them colours that are clear for us. You have a red plate and a green plate. You have the same food in each. You have to start with one plate. You have two, but you can’t eat in both at the same time — the capuchin monkey tried; they had to change the design because they were taking two plates at the same time. For the fish it’s not possible, so they have to choose either going to the red one first or the blue one first.

If you go to the red one first, the experimenter removed the blue plate immediately. So you only have 50% of the food. If the fish first starts with the blue plate, it can eat at the blue plate, then the blue plate is removed and the red plate is still here. So now the fish can have the second plate. So it’s pretty easy: you have to understand that to get most of the reward, you have to start with the blue plate and then the red plate.

They understood this task extremely fast. It was wild-caught fish, and they understood it in a very short amount of trials. While chimpanzees just had a freaking hard time. And orangutans, same thing, and same for capuchin monkeys.

But it doesn’t stop here, because already you could wonder, why are they doing better? There could be many answers, but one answer could be that in their world, they have clients who can leave, and they have to understand which clients will leave and which won’t to start with the one who will leave. It’s very coherent with their lifestyle. For chimpanzees, it’s not really something that they encounter in their everyday lives, so it’s not that important for them to do that.

And this is one of the most important things to understand: animals are developing cognitive abilities toward things they need in their life. It’s not, “You’re smarter than this one.” Making ladders of intelligence is pointless, because animals will develop abilities in domains that are useful for them. And this is not useful for the chimpanzees and capuchin monkeys, but it is for cleaner wrasse.

The interesting point is they did what we call a task reversal, which means that suddenly, without telling anyone — any fish, any chimp, any monkey — the blue plate becomes the fixed plate, and the red plate becomes the temporary plate. Once again, the fish were the first to understand the rule change — which cannot really be explained by their way of life, actually, but they were the most flexible about it. Which is amazing, because it needs to accept that what you’ve learned is now wrong, and you have to try something else. It’s the opposite from what I’ve learned. And they’ve done that much better than chimpanzees, orangutans, and capuchin monkeys — and to be honest, than the daughter of the guy who did this study.

Luisa Rodriguez: Cool. We should say now that you are not in particular a consciousness expert; you’re more of an expert in cognition for these animals. But I do feel like I’ve at least learned some — and obviously, you know a fair bit about things that point to consciousness in particular — and for me, a big one is this flexibility thing. Because when I imagine these fish learning some rule that they can apply both to their client fish and also to these plates, I can imagine them doing this kind of robotically. But as soon as they start to do things flexibly, and also to do other things that are quite flexible, like making tradeoffs, it really feels hard to start imagining them as purely following an algorithm that has been kind of evolutionarily selected for.

So I find this just extremely compelling. It’s also just quite compelling to me that I’d bet my house that chimps and orangutans have something it is like to be them — that they’re feeling things. So to, at least in some domains, have more flexible learning in these fish feels pretty remarkable.

The mirror test and problems with these tests [00:24:47]

Luisa Rodriguez: Are there other tests where these fish beat primates?

Sébastien Moro: Well, mirror test! First I will explain what we call the mirror test or the mark test. It’s a test that was developed in the ’70s by Gordon Gallup, which was first aimed at young humans. The idea is you put a subject in front of a mirror, and then if the individual can recognise himself or herself, he will go through different stages. The first stage is you don’t know it’s you, so you consider that it’s someone else. But you should start to see some social behaviour.

And after a little while, the subject will start to understand that there’s something going wrong there, something’s not right. So you should start to see weird behaviours that aren’t used in the social context, like moving your hand to see if the other one is doing the same thing.

Then the next stage is the subject understands that it’s his reflection, so you should start to see self-checking behaviour. For chimpanzees, who are very close to humans, they’re trying to check their ass, their bottoms. They’re just turning around and checking the inside of their mouth and their bottom.

And the last part to officially validate the success of the mark test, you put the subject to sleep and draw a mark where they cannot see. Usually it will be on the forehead or the side of the head, or on the chest, this kind of stuff. And then when the subject wakes up and looks in the mirror, either it doesn’t care, or it’s starting to try to remove the mark on him — so it means that the subject understands that it’s his reflection and he can orient the movement toward himself or herself.

Then there is also the option that the subject doesn’t react at all — which doesn’t mean it’s a failure; it could mean that many other things are important for this animal, and vision is not one of them. This is exactly the problem we have with pigs, for example. I don’t think pigs could succeed at the mirror test. To my knowledge, they haven’t been tested properly. But their vision is really bad, and they recognise each other by smell. So I don’t think that a mirror test would work.

So if we come back to the animals who have been tested: chimpanzees are passing the mark test with results that are not that crazy. Not all chimps succeeded. I don’t have the exact number in mind, but something like 30%. Not that much.

For elephants, the results were pretty bad actually. Again, to my knowledge, there have been three subjects, three Asian elephants, and only one in three succeeded. It’s not amazing.

Then you had tests on dolphins. and now we are having a problem: how do you remove a mark when you don’t have hands, when you don’t have anything to remove the mark? So for dolphins they’ve considered that as soon as the dolphin is trying to look at the mark in the mirror — the mark is on the chest — for a long time, then we can consider that the dolphin understands that the mark is on him or her. And that’s OK; we can say it’s a success. It’s funny because officially it’s not.

And then we have another fish. Not another, because the dolphin is not a fish, but I say “another fish” because we’re not yet at cleaner wrasse. In 2016, there was a scientist who did the first mirror test on manta rays. They’re very big fish. And she started from a logical point of view, which was that we accept that dolphins can’t remove the mark and the animal just watching for a long time is considered as mirror self-recognition. So they’ve done this with manta rays, and manta rays did succeed. It’s very interesting, because it’s not even a bony fish — so it’s a very distantly related animal.

And they passed through all the previous stages — like behaving socially, all the different stages — and everybody said, no, they didn’t try to put a mark on, to remove the stuff, so it’s not a success. I don’t know. Why are you accepting that for dolphins, and not for fish?

Luisa Rodriguez: Moving the goalpost.

Sébastien Moro: Yeah, exactly. Because we think that dolphins are intelligent, chimpanzees are intelligent, elephants are intelligent. Fish? Nah. And then in 2019, there was a team — a mostly Japanese one, but kind of international — who tried the first real mirror test on cleaner wrasse. In the name you have “cleaner” because they can clean other fish, but they can clean themselves too. When they have a parasite or something, they’re rubbing themselves on the floor. This is a behaviour we can use.

Luisa Rodriguez: Nice.

Sébastien Moro: And actually, I’ll make it simple, because it’s a bit more complicated than that. But all the fish apparently succeeded at it, which means a success rate bigger than all the other animals, humans excepted.

So that was a problem. So much so that when they released the final paper, it wasn’t called “A fish passed a mirror test,” it was, “If a fish passed the mirror test, what does it mean for the mirror test?”

Because when they released that, they knew people who’d wait for them at the corner. So they were very cautious, saying that maybe the mirror test isn’t the proper test. But then like two years after that, they were already working on other controls. So all the sceptical comments they had, they were already answering it.

So a second study has been released and showed that there is no question about it: they can succeed it easily, like really easily. The first one didn’t have 100% success; it was like 80% success — just because they thought that two of the animals already recognised themselves in their reflection of the tank, so they didn’t go through the whole stuff. It failed the test because of this.

But the most amazing thing is: when you’re putting a mark on a fish, you can’t put it on the mucus because it would just be removed immediately, so they were injecting some kind of gel under the skin. So everybody was like, “But if they are rubbing, it’s because the injection is scratching.” The thing is that it’s not, because when the injection is transparent or the same colour of the skin around, the cleaner wrasses aren’t trying to clean it. And you can clearly see them getting in front of the mirror, going on the floor, rubbing, then going back in front of the mirror to check.

So they released last year, very recently, the most amazing test ever, that hasn’t been done on any other animal. They’ve made a photographic version of the mirror test, so nobody could say it’s scratching. It can’t, because the fish has no mark.

First they made them pass the mirror test, all the stages and everything. But this was the training. So now this test is just a training for them. Then the test was that you take a picture of this very fish and you draw a mark on him or her, or you take a picture of another cleaner wrasse with a mark. The animal tries to rub himself or herself only when it’s a picture of him or her.

That’s not all. I have worse.

Luisa Rodriguez: Wait, wait. I just want to make sure I understand that. So you take a picture of the cleaner wrasse and you draw a mark on them. And if they try to rub the mark off, that’s already pointing at something like they didn’t get any injection, they’re not feeling anything, there’s just like nothing physically there that would make them, for physical reasons, want to rub it off — except for the fact that they think that they see it in a reflection of them.

But it’s not just that: they only do it for the pictures that are recognisably of them, which is incredible. I guess that means that when they were doing the original training for the mirror test, they saw themselves in the mirror, and they learned what they looked like, and they learned to differentiate that from what other cleaner wrasses looked like?

Sébastien Moro: Yes.

Luisa Rodriguez: That is an amazing result. Sorry, you were about to say, I think, that it gets more impressive.

Sébastien Moro: Yes, you’re understanding right. There is a second part to this test. They wanted to know if the fish are recognising themselves by the face (as we do in humans), or by the body, or a combination of both. We know, for example, that humans are using the face primarily.

So what they did is they took Photoshop, they took two pictures of fish — one of the focal animal and one of another one — and they cut the heads on the pictures, and put the head of the focal fish on the body of another one and the head of the other one on their body. And they’ve tried it again, and it appears that they’re attacking their body with the head of someone else, but they’re not attacking a picture of their face on someone else’s body. So it means that they can recognise themselves in pictures from the face, exactly as humans do.

Luisa Rodriguez: Wow.

Sébastien Moro: This is the best mirror test ever. Actually, the animal that is succeeding at it the most convincingly, except for humans, is the cleaner wrasse.

Luisa Rodriguez: That’s pretty incredible. I guess I do have this reaction though that I put a lot of weight on elephants and chimpanzees having sophisticated cognitive abilities, and probably being sentient, experiencing things. And the fact that they fail… It’s not even just the fact that they fail and the cleaner wrasses pass. It’s mostly just like, if they fail, is it even really picking up on something that significant? What is your take on how much weight we should put on the mirror test?

Sébastien Moro: So they’re not all failing it. But the thing is, the mirror test is a very anthropocentric test. It’s been designed first and foremost for humans, then it’s been extended to primates, and the further you go, the more complicated it is.

But we’re not really sure about what it does assess. We don’t really know. It could just assess the ability to understand that this is a reflection of the body and make a kinesthetic answer to this.

But even though this test, when it’s succeeded, is a very interesting cue — it’s not a definitive cue, but it’s a very interesting cue — about whether an animal is self-aware or not. Which is not sentient: sentience is not the same thing, so the mirror test is not assessing sentience, because sentience is affective. It’s about emotions, and this test is not about emotion at all. There has been a proposal to modify it to add emotions to it. And they actually took the cleaner wrasse test to propose a new version with parasites eating them instead of just a mark — but real bites so you have a negative effect.

But the thing is, I say it’s anthropocentric because it’s based on vision, and vision in mammals is rarely the biggest sense for social recognition. Many, many mammals are using olfaction for social recognition. You take rats, you take most of mammals actually, and they will probably fail a mirror test — which doesn’t mean they cannot recognise themselves. It means that the test is maybe not well adapted to them. A failure to this test doesn’t mean an animal is not self-aware; it just means the test might not be for them.

And this is why for a while they’ve been thinking about doing an olfactory version of it. They’ve been working on that, on dogs especially, and these kind of things — which is funny because in 2009 fish already did this. They showed that in some fish they can discriminate their own smell from one of their brothers. And we’ve shown recently in guppies I think, I’m not absolutely sure of the species, but they’ve shown that when they’re trying to find out how familiar another animal is, they’re comparing their smell to their own smell.

Luisa Rodriguez: They’re using the similarities and differences in their smell to another animal’s smell to be like, “How closely related are we?”? That’s incredible!

OK, I just want to make sure that I’m understanding the implications. I think everything you just said does make a lot of sense to me. One issue is just that nonhuman animals use a range of senses — and maybe in particular, use their sense of smell for recognition much more than humans do, and much more than sight.

Also, it’s just a very specific ability. Maybe it’s tied up with reflections, and a concept that there can be reflections and that you could be in a reflection. Maybe it’s tied up in concepts of having a concept of oneself. And that does seem pretty impressive, but not a requirement for sentience, probably, on many theories.

So I think I’m basically sold that we should be impressed when an animal passes the mirror test, and passes these different versions of a mirror test, but not dismissive when one doesn’t. And we also shouldn’t dismiss the mirror test entirely, just because the animals we expect to pass it don’t, and the ones we don’t expect to pass it do. How does that feel as a takeaway?

Sébastien Moro: That’s pretty good. And because when you’re talking about ethics, the main point is sentience — which I define here as being able to feel subjectively negative events and positive events — sentience has an affective space that is not assessed in most of the cognitive tests. It’s not because you’re really smart that you’re feeling more, or you’re dumb so you’re feeling less. That’s just unrelated.

Learning and tool use [00:40:48]

Luisa Rodriguez: OK, so moving on a bit again: what do we know about what kinds of things fish can learn?

Sébastien Moro: They can learn a lot — exactly as we can see in land vertebrates, once again. Fish can learn how to use their environment. And there is also the question of social learning.

So if I talk about learning and using the environment: many fish have very good spatial skills, like very good ones. But something you might not expect is that some fish can use tools.

You have rays that are able to use the water flow to pull food toward them or away from them. It’s a study from Gordon Burghardt. He made for freshwater rays a test where there is a plastic cylinder, PVC, and inside there is some food. And the rays, with a weird body shape, cannot access food. So they are waving at the entrance so they create a flow of water that pushes the food outside, and they can use the water to get food.

He tried putting a kind of wireframe on one side and the food next to it so it’s close to an exit, but you cannot make it exit by there, to see how intelligent and how flexible the rays can be. And they are very flexible. They can understand, “OK, it’s not coming from this side. I’m trying from the other side.” And they are changing technique with time. After a while, instead of waiting, they make a suction cup with their body and sucking the food out of the tube.

But we could say using water is not really tool use. And then we’ll have to talk about archerfish. And here we have a massive problem, because archerfish are fish that are doing things they shouldn’t be able to do.

So to explain briefly, what are archerfish? They’re called Toxotes, scientific name, and they are hunting fish. They eat mostly insects, and when they’re young, they eat prey in the water, as every fish. But as they grow, they start to spit water spray out of the water. They spit, literally, to hit insects which are on the leaves of the plants above the water.

But this is not something that is instinctive. They have to learn it. At first they’re bad, and then they’re better and more impressive: they slowly learn to hit moving prey. And they can hit moving prey, they can learn that just by watching other fish doing it, without even trying it themselves: once you release them, they’re able to do it.

Now, this seems impressive, but it’s not as much as it is really — because we have to remember that these fish, they’re not territorial; they move all the time, which means that they don’t know the place, so they really use their sight to hunt. Then when you are in the water, when you’re comparing to the air, there is a refraction index that changes. So where they see the insect is not where the insect is — exactly how when you put your arm in the water, it looks like your arm is broken, but it’s not, it’s refraction indexed.

Luisa Rodriguez: Yeah, yeah.

Sébastien Moro: And they can just correct for the refraction index, knowing that this refraction index will change according to the pressure, to many things. And they can adapt it to the height of the prey, the speed of the prey, if there is wind or no wind — which they can’t feel, because they are in the water — and they can shoot a prey and calculate where the prey will fall to be exactly at the landing spot of the prey, to be the first one to eat.

Luisa Rodriguez: Incredible.

Sébastien Moro: Their visual cognition is so good that they can even recognise human faces — which is totally insane, because in their life they shouldn’t be recognising them.

Luisa Rodriguez: They never encounter human faces!

Sébastien Moro: Exactly. And even more impressive, they’ve shown that they can recognise faces of humans with rotations. So they learn to recognise the face of someone, and when you put this face next to 44 different faces, they can still recognise this one. But if you learn one face, and then I show you a profile of this person, it’s much harder to recognise this person. Monkeys, primates have a hard time doing it. The archerfish can do it. It’s insane.

Luisa Rodriguez: That’s so amazing.

Sébastien Moro: Their vision and their ability to understand visual stuff is crazy. Like, really crazy. So here we cannot say that water is not tool use. They are using water as a tool. And they even watched in detail how the jet is made, and they found out that they change the shape of their mouth when they’re spitting — so the end of the spit is bigger or smaller, the track behind is longer or not, just to be precise on the distance. So they’re really very precise on how to spit. It’s ballistic. It’s absolutely ballistic.

Luisa Rodriguez: Oh my god.

Sébastien Moro: And I’m waiting, so you can calm down… and then I have something worse.

Luisa Rodriguez: Yeah, no, my jaw is just on the floor. I mean, part of me is trying to be like, “How could one do this without being conscious?” But I don’t even care. That kind of visual math, the kinds of calculations… like, I can’t even. A related skill that’s coming to mind for me is archery.

Sébastien Moro: Archer-fish.

Luisa Rodriguez: Oh, my gosh. They’re archerfish. That just clicked. I just had not even considered that that’s why they were called that. And I’m getting the impression… I mean, human archers are very good, but I honestly feel like this is more impressive.

Sébastien Moro: I think it is.

Luisa Rodriguez: Because it’s through water, you’ve got the refraction, you’re using water sprays that you create with your mouth. It’s just actually unbelievable. But you said you’re about to blow my mind even more, so take it away.

Sébastien Moro: For people interested in digging a bit more, there is a very good review on archerfish that has been made by Stefan Schuster that is called “Hunting in archerfish – an ecological perspective on a remarkable combination of skills.” This is a very easy-to-read paper, and it’s a very good one. And otherwise, all the work of Cait Newport is really good. Schuster and her have been working together a lot.

So now, yes: if there was one study that blew my mind, it’s the one I’m going to explain now. And we’ve talked about the mirror test in cleaner wrasses, but this one is uncanny, really. It’s a paper on Atlantic cod. At first this study was about self-feeders for aquaculture. So in aquaculture it happens that the fish have self-feeders: some kind of devices they activate themselves to get food. If I remember correctly, they were trying to find food preferences in these fish.

So first, they need to recognise each fish. So they put a tag on their back with a bead on it, a coloured bead. So you have the fish with the blue bead, the one with the red bead. And then inside the tank you have a self-feeder. How does it work? Let’s imagine again a square tank seen from top. On the top left, you have the device. The device is a pull string with another kind of bead at the bottom. The idea is that the fish comes, takes the bead in its mouth, pulls, and then there is a light that switches on top of the device and the self-feeder is releasing food on the top right — which is kind of problematic, because it means that the fish that pulls is not the first one where the food is.

Luisa Rodriguez: Right. Because it’s on the other side of the tank.

Sébastien Moro: Exactly. So they were trying that, and it was working pretty good.

And then, it happened with three fish, but I will concentrate on one. There is one fish who got trapped in the bead of the pull string with his own bead, the one in his back. There is a video of it. So you can see the animal getting locked in it. It starts to get scared, tries to move, but can’t. But the device is working: the fish is pulling, so the light goes on, the food is given and the other fish are going to eat. And finally, this fish managed to unlock himself and move.

Then the weird thing is scientists started to see this fish trying on purpose to get stuck in the bead again. And they were like, “What the fuck is this animal doing?” And the fish started to be really good at it. And at the end of the experiments, the numbers are something like the animal used his or her mouth 40 times, and the bead 522 times. Why? Because when you pull with a bead in your back, you can have the head where the food is coming.

Luisa Rodriguez: I see.

Sébastien Moro: And what they observed is that each of the three fish who did that learned a different technique, and they improved it. When you watch the 10 first tries, they’re just trying in kind of an easy way to get locked in it and then free themselves to go to the food. But the 10 last tries, it’s amazing, really: the animal is going down, grabbing the pull string with their own bead, pulling while turning, which allows them to free themselves. And when the device is activated they are already free and they head at the food. Which means they are using an artificial limb that is in their back that they cannot see to overcome a difficulty.

It’s something that was not expected by the scientists. At no point did they think they would see that. They’ve been publishing a paper just on this, aside from the one they were preparing, because it was insane. And it’s Doctor Octopus. Like they have one more arm. One arm.

Luisa Rodriguez: Yeah. So they basically just made up a new way to make sure that they’re quicker to get to the food as it comes out. And then they experimented, improved upon it, and now it’s like the Olympic games of using their extra tool appendage bead on their back to be quick to the food. That’s absolutely incredible. It’s tool use, but it’s also so flexible. I mean, I can’t think of anything in the environment that would make you think that they have any kind of related skills, or inherited behavioural techniques.

Sébastien Moro: Yeah, this is really, really something. Honestly, the first time I read that I was like, okay, I’m going to read it again. I misunderstood and I didn’t. And nobody ever talks about this paper. Every time you can see stuff about how fish are smart, this one is never in it and it’s an amazing paper.

The paper is “Innovative behaviour in fish: Atlantic cod can learn to use an external tag to manipulate a self-feeder” by Sandie Millot, published in 2013. And I think the videos are probably coming with the paper, so they might be online.

Luisa Rodriguez: Great. Are there other impressive things or types of reasoning that fish seem to do? Planning, tradeoffs, other problem-solving stuff?

Sébastien Moro: A lot. I’ve said that some fish can communicate through chemical signals, and some researchers have discovered that some fish can learn the alarm signals from other species. It’s fathead minnows who can learn the alarm signal of sticklebacks. If they’ve never lived with a stickleback and you release in their water the chemical alarm signal, they’re not reacting. If you put them in a pond with the stickleback and try them later, they will react with anti-predator behaviour.

And if you let them live in the same pond for six years, they wanted to know if it was something that was transmitted to the babies, so they took eggs from this pond and hatched them — but in a tank, outside of any influence — and they weren’t reacting. So it’s something they learn.

Luisa Rodriguez: It really is learned. Wow. That’s great. That’s a good one.

Sébastien Moro: I have many things, like categorisation, that don’t test to show that. For example, for bamboo sharks, a snail is a snail: whatever image you show of a snail — like if it’s a photograph of a snail, or if it’s a drawing in black and white of a snail, if it’s a cartoon of snail, if it’s a painting of a snail with different shapes and everything — the shark will know it’s a snail.

We know that because this kind of research is following the protocol of the Y maze. It’s a maze in the shape of the letter Y: you have a corridor, and then on one side on the right, one side on the left it’s split.

So the sharks were having a reward if they were going to the picture representing a fish. One side they’ve learned with photographs of fish, and on the other photographs of snails, and it was always the same pictures all the time. And once they understood, the test began, and they started to show new images they’ve never seen before, and totally changing the style. For example, you have a photograph of a snail on the right side, and on the left side you have a really basic drawing of a pufferfish, and pufferfish can get very round, snail-looking-like, and the sharks were always able to recognise them. It’s been tested on different species, and many species are very good at categorisation.

I could go this way for hours. As I’ve said, a lot of times already, but it’s important: they have challenges that are as complicated as land animals have, so they can learn very complicated stuff. They can learn basic math, they can count. Some fish are able to use colours to know if they have to add one or subtract one.

Luisa Rodriguez: What?!

Sébastien Moro: I’m going to make your jaw fall again. This paper has been taken from a famous paper on bees. Bees are able to add and subtract. So basically, how does it work? It’s based on the match-to-sample protocol. So match-to-sample is where you show an animal a stimulus — for example, a white paper with two red squares on it — and then it goes inside a Y maze, and you have the two corridors going on each side, and on one side you have exactly the same pattern, and on the other side a different pattern or a different number of red squares. And in a match-to-sample, they go to the same signal they saw at the entrance. But it doesn’t mean they can count; it just means they can memorise this, and that gets you the answer.

Luisa Rodriguez: Yeah, pattern match.

Sébastien Moro: But what is interesting is they taught the fish. So it was cichlids and rays — which is interesting because bony fish and cartilaginous fish — and they showed that they were able to understand that when the entrance stimulus is yellow, the good answer is the number of stimulus at the entrance minus one. If blue, it’s plus one. So if you have three blue and you get inside, you have to go on four symbols. If three are on yellow, you have to go on two.

Luisa Rodriguez: Wow. That is counting. And it’s not just counting. This is fascinating, context-dependent counting.

Sébastien Moro: Yeah. And as I said, it’s coming from bee studies. Bees can do it. And it’s just one: it’s adding one, subtracting one. Chickens can do much better, but I won’t go there. This is not the topic.

But yes, we could go this way for a long time. But the fish that have been tested — which are, I’d say, five or six species — all of them have basic math abilities, which is logical. Almost every animal works the same with numerical abilities, like the way humans are working around that is the same as fish do, bees do, dogs do. Every animal tested has the same kind of system.

And we enhanced it through language, but we have two basic systems: understanding small numbers and assessing big numbers. And we have a limit around four: we can count precisely, at first sight, up to four or five and no more. This is exactly how every animal is working. And in humans’ populations which don’t have a strong language about math, usually they’re counting, “1, 2, 3, 4, many.” There is nothing above four.

Luisa Rodriguez: So we’re just the same on this?

Sébastien Moro: Yeah. Exactly the same. It’s amazing how much we’re the same.

Luisa Rodriguez: Wow.

Consciousness and sentience [01:00:07]

Luisa Rodriguez: This is fascinating and mind-blowing, purely from the perspective of “fish are incredibly cool.” But do these feel like they tell you anything about whether fish are experiencing these things in some conscious way or affective way? Or do you feel like, at least for the kinds of things we’ve been talking about just now, that they’re incredible, we can’t fathom them, but they don’t really tell us anything about conscious experience in fish?

Sébastien Moro: As I said earlier, we have to split consciousness and sentience, which are not the same thing. It’s very hard actually talking about real consciousness, like high-order consciousness, like humans: we don’t know how to assess it correctly, even in humans. We don’t have proper assessing tools.

So today we’re trying to build new ways to assess consciousness and sentience and split them properly. I understood that you’ve interviewed Jonathan Birch about that. He’s a pioneer in it. He’s a very important person on it.

There is a very good book, especially on fish, a study which is named, “What is it like to be a bass? Red herrings, fish pain, and the study of animal sentience.” It’s a publication from 2022. It’s really interesting because it’s coming back on all of this, and the famous study that was talking about modification of the mirror test we were talking about. Many of the studies that were done that we’ve talked about, they aren’t made to assess sentience.

Luisa Rodriguez: From your perspective, you know so much of what there is to know about fish that we have studied so far, so you’ve got all of this wealth of knowledge: what is it that feels most compelling to you, that makes you feel like you’ve got really high confidence that fish are experiencing things?

Sébastien Moro: Most of the papers we have are going in this way, and very few — very, very few — are going the other way.

So what makes me so certain? I’m really talking about my personal opinion here. I tend to think that emotions… What are emotions? What are emotions used for? They are putting a gloss on what is around us: “This stuff has a positive gloss; I need more of this. This one has a negative gloss.” And we know that emotions are very, very closely linked to learning. So emotions are something that attract or repel. And it seems pretty obvious that it must have appeared very, very early in evolution, because this is how we work.

This is maybe one of the biggest differences with algorithms. Algorithms are following closed loops. And this comes up when animals are more driven by emotions, by value of things, which is made by a kind of limbic system that says, “This is good, this is bad. You want more of this, you want less of that.” And I don’t understand why other animals couldn’t have had that.

And another thing is, I’m reading a lot about bees, so I know very well the corpus of knowledge on bees at the moment. And we’re starting to have the same results in bees. So I’m not 100% certain now that bees could be sentient, but the biggest leader of bees research today, Lars Chittka, has said on Twitter that bees are sentient for him. And the results we have are going in this way. So they have a one-million-neuron brain, and the brains of fish are much, much bigger. And when we split from insects, brains were not existing either. So it’s just a convergent evolution.

Fish have complicated lives; they have social lives, very social lives. I guess we’ll talk a bit more about this later, but I already introduced this with cleaner wrasses: their lives are very complicated. They have challenges that they have to overcome that are as complex as what we find in mammals and birds, maybe more sometimes. So them having no sentience, when we recognise sentience in birds and mammals? It’s either you refuse it for everyone, or you accept it for everyone at the moment. Not for everyone, because animals with a brain or central nervous system, today we think at least there should be a kind of global network; everything should be put in common to make a unified vision of you and this kind of thing.

But consciousness probably has many degrees; sentience has pretty much many degrees — but not degrees on a ladder, degrees more on a circle. But that would sound just weird actually, that they wouldn’t be sentient.

Luisa Rodriguez: Yeah, it would just be really surprising to you.

Nociception and pain [01:05:34]

Luisa Rodriguez: If there are any studies that you know of that do look at things like fish perception of pain, or experience of pain in particular, I’d be pretty interested in that. To start, do most fish have nociceptors, the type of nerve cell that senses noxious stimuli?

Sébastien Moro: As I said earlier, we don’t know much about most of the fish. So the experiments to find out if fish have nociceptors, there aren’t that many studies.

So first, what are nociceptors? Nociceptors are the nerve endings that are related to noticing if something is damaging the skin. So it can be pressure, it can be temperature, acid, it can be pretty much anything. But you have different kinds of nociceptors for each of these, and you have multimodal nociceptors that can do many things at once.

As far as we know, the very first study that has been made about this topic was, if I’m recalling correctly, Braithwaite, Sneddon, and Gentle in 2003 — which is called, conveniently, “Do fishes have nociceptors?”

Luisa Rodriguez: Incredible.

Sébastien Moro: It’s a study on rainbow trout. They actually found nociceptors — mostly on the face, around the fins and the tail, but also along the body. But it seems that in these animals, most of the nociceptors are located in zones that would be, we could imagine, very sensitive. A few other fish, I think we found nociceptors of this kind, but it hasn’t been tried this much.

And there is an important thing. You have two kinds of nerves there. Oh god, we’re going very much in detail, but it could be important to understand that. You have two kinds of nerves. You have type C fibres and type A-delta.

The type A-delta are very fast, and they’re not linked in humans to conscious pain; they’re more linked to very fast reactions. For example, when you put your hand on a hot stove, you will remove it immediately. At no point did you make a conscious decision. It’s a reflex arc; it’s something that you did not process at all. So type A-delta fibres are participating in this kind of stuff. So the simple fact that a fish reacts, that it has nociceptors, doesn’t mean that they can feel pain in a subjective way.

In humans, the type C fibres are those linked to pain mostly. It’s why when you put your hand on a hot stove, at first you move the hand, but you don’t feel anything. And a few milliseconds later, you start to feel pain — because these fibres are slower.

This is one of the first weird things they found out: the rainbow trout, they have type C fibres and type A-delta. But the type A-delta are in bigger numbers and type C are in small numbers — which is the opposite in humans. But the further you go from humans, the more you have chances that it’s going to be different. So it could be that type A-delta also acts like type C. We’re not really sure about that.

But that’s not all, because you need to have a pain pathway. Are these fibres going through the spine? Yes, they do. Are they going to the brain? Yes, they do. In humans, we know that it will be the hippocampus and amygdala that will treat this kind of information. Did we find in the brains of fish something like that? Yes, we did. We found the full pathway, and we found homologous parts in the brains to the amygdala. And we know we can modify the reaction through antidepressant drugs or anxiolytic drugs. Actually, zebrafish are one of the biggest animal models to test antidepressant drugs. For many mental disorders, zebrafish are one of the main animal models.

Sébastien Moro: And this is one of the problems: when you have to try to assess subjective pain, you have to hurt animals. So what do we do about that? That’s a very complicated question, because for a long time we thought that fish could not feel pain. It was a consensus somehow until 2000, something like that. It’s really recent. The debate is really, really recent. And most of the studies have been released between 2003 and 2012, something like that. It’s really been a rush of studies.

So yeah, if you have lots of questions, I have lots of answers, because many, many studies have been made. So we now have a pretty clear view — at least, not going too much in detail, but today we have as many clues that fish can feel pain as we have for birds or even mammals. Pretty much.

Luisa Rodriguez: That really surprises me, partly because when I was trying to just learn a little bit about fish in preparation for this interview, I tried to learn a little bit about fish brains. And my understanding was that they do have some analogous components that seem relevant to pain, but they really lack major neural structures, even relative to birds. So I thought that was at least some evidence of less going on, and maybe less experience?

Sébastien Moro: OK, so we have to talk about brains there. For a long time, we thought that you needed big brains to be smart and to feel — and especially a neocortex, which is the most recent part of the brain in the evolution of mammals, mostly. But the more time passes, the more we’re discovering that that is not true. That is absolutely not true.

Today, we have results in insects that are very, very impressive, and they can do things that they shouldn’t be able to do. I’m really thinking about bees right now. Bees are doing things that they shouldn’t be able to do. They can count, they can use abstract concepts to find the direction in order in a maze. Bees are really amazing animals. And then they have a one-million-neuron brain. We have 86 billion.

So yes, brain size, brain complexity has its advantages, but not as much as we think. It seems that it makes it easier to memorise a lot of things and to treat information in parallel and these kinds of things, and maybe higher abstract concepts. Even fish and bees can understand abstract concepts — like “over there,” “under there,” “different from” — they could do this, they have concepts. Almost all animals have concepts and categorisation systems. But the more we progress, the more we find out that most of the abilities humans have have been useful in animals in every kind of environment for a long time. And many things are probably not as mentally costly as we thought.

So when you’re talking about birds… Actually, for a long time, birds had the spot that fish have today — birds don’t have a neocortex, they can’t feel pain, they can’t have higher-order cognition — until we’ve discovered that the brain of birds has nothing to do with a mammal brain.

For example, in 2016, there was a paper that found out that in parrots and some passerine birds with brains the same size and weights as a mammal’s brain, they have threefold the numbers of neurons. They have many, many more neurons per centimetre in the brain than we do. Actually, the brain of birds is brain 2.0: it’s a very small, efficient brain made to fly. We don’t have to fly, so we don’t have to be light, so we can have big brains. But birds can’t, so they found another way.

Luisa Rodriguez: Interesting.

Sébastien Moro: And fish seems to be a rudimentary brain — and yet: cleaner wrasse. A cleaner wrasse is 10 centimetres long. The brain is not even the size of my nail. It’s a really small brain. And yet they can outperform chimpanzees.

Luisa Rodriguez: Yeah. I basically do buy that we should definitely put some weight on exactly the kinds of brain structures, and certainly brain size not being a requirement for experiences in general. There’s certainly not a clear consensus here. For anyone interested in learning more about these debates, I can recommend our interview with Jonathan Birch, on his book, The Edge of Sentience.

My impression is that best practice is probably to look at lots of different pieces of evidence, including neuron counts and brain structures, but also other things like whether antidepressants that work in humans seem to have mood-boosting effects in other species, or whether animals seem to have clear preferences painkillers like analgesics.

Have studies like these been done in fish? Where basically they see if they can harm fish in some way or do something that would be, in theory, painful to them — and then offer them painkillers and see if they, one, seem to prefer the painkillers to not painkillers, and two, see if they avoid that negative stimuli less than they might otherwise if they didn’t have the painkiller? Is that something that’s been done?

Sébastien Moro: So first things first, I’m going to be the fun killer there: analgesia studies aren’t teaching us much about objective feelings for the simple reason that you can have the same results with plants.

Luisa Rodriguez: Fascinating. Can you say more about that? As in, there are certain kinds of analgesics you can give a plant that will make them less…?

Sébastien Moro: Yeah. You have plants that react to touch — you know, they’re closing their leaves. And I’m not a specialist of plants, so I know what I say is true, but I’m not sure about the species and the analgesia. But if you put some kind of analgesia on the leaves, then they will react much less, or even not react anymore.

Luisa Rodriguez: That is really, really interesting and something I had not heard before. I don’t know whether to be like, “Oh dear, plants might feel more than I thought,” or, “OK, so analgesic studies are weaker evidence.”

Sébastien Moro: Probably that it’s weaker evidence. We have two studies like that. Both have been made by Lynne Sneddon, if I’m not wrong, who is the most important person to talk about when we’re talking about pain in fish. She should have a Nobel Prize, probably.

She’s made a first one, where it’s on a zebrafish, and they have a tank, and this tank is split in two compartments. In one there is enrichment — so there are algae; other zebrafish who are doing very well, everything’s fine for them; and it’s pretty dark as well, because they like when it’s dark and they can hide. And on the other side, the other compartment is fully white, bright, no enrichment, no other fish.

And when you give the choice to the zebrafish of what to do, it will go with the other zebrafish. So now, what happens if you inject acid in the lips, or in the body? I’m not sure on this one: it’s often in the lips, but I’m not sure about this one. Anyway, you inject acid, then this fish will go with its mates and just lay on the bottom of the tank. So the behaviour has changed already, and it’s not moving much and it’s hiding. But if you spread, I think, lidocaine in the very bright compartment of the tank, then the fish will leave the security to go in this part of the tank. So it’s taking the risk to be in the open, to take the drug. So we have what we call a “motivational tradeoff.”

And we have different kinds of motivational tradeoffs that have been discovered in goldfish and rainbow trout that will accept to go in a zone where they receive electric shocks to be closer to a mate or to be closer to food. For example, a goldfish in an aquarium, in a tank, if it goes after the centre of the tank with some electrode, they will apply electric shock on its skin. And the fish will learn that as soon as it crosses the middle, it will receive electric shock, so the fish is not staying in the same place all the time. But if it’s food deprived, it’s still staying in this zone, and the more they’re food deprived, the more they would tend to go in the electric part to get food. So it’s not something totally automatic because they can have control over it.

Luisa Rodriguez: Right. Depending on how hungry they are. So it’s not just, “If hungry, do X; if not hungry, do Y.”

Sébastien Moro: Exactly.

Luisa Rodriguez: It’s more of a scalar.

Sébastien Moro: There is another really interesting paper that was published in 2005, and the topic was trying to find the pathway from the nerve fibres to the brain of the fish. It has been done on rainbow trout and goldfish.

I’ve said that in trout we have mostly A-delta fibre — so the very fast one — and fewer type C fibres, as we find usually in humans, in mammals. And actually, that’s the case for trout, but not for goldfish. Goldfish have a nervous system that is about nociception, so the ability to feel noxious stimuli. The ratio of type A-delta and type C fibres are much closer to what we find in mammals, and it might be because they’re living in different environments.

Just to give one example that can help you understand how different it can be. Trouts are often living in very cold water, so their thermal nociceptors — their nociceptors, fibres for temperature — are not responsive to low temperature. It means they can’t be hurt by cold as we are. It depends on a lot of things. And actually, their A-delta fibres aren’t behaving like ours, because A-delta is usually very short lived and they are responsible for every reflex movement, very acute and very fast and pain, while the type C fibres are longer, dull pain — and actually, their length of frequency at the time they are still active is longer than what we usually see. So this is why some scientists think they could behave a bit like type C fibres.

It’s a very complicated matter, because different animals live in different environments and there are things they need to react a lot to and others they don’t need to react to. So this is why some don’t tell the difference between just a small contact and a really powerful contact, like a spike or something. Trout are reacting very strongly to any contact, any physical contact, while goldfish much less. So it depends on how they live, how their life is made.

Luisa Rodriguez: Just out of curiosity, what do we know about the way they live that might explain why trouts are much more sensitive to any physical touch?

Sébastien Moro: I couldn’t answer properly. I have maybe some ideas, like adult trout are much less social than goldfish. So that might be a reason. There is much more aggression. They’re very territorial fish. But another thing that I actually just remembered: in this study, it’s juvenile rainbow trout and adult goldfish. So maybe the nervous system of juvenile trout is different than the one of adult rainbow trout, which wouldn’t be surprising. Because, to give another example, some fish can see in UV. Juvenile trout can see in UV, but when they are adults, they can’t anymore. They lose the UV ability. So maybe the nervous system is going through changes as well. I hope it’s trout and not salmon, but I think it’s both, actually.

Luisa Rodriguez: It sounds like you’re putting a lot of weight on just — and I’m kind of bought in to this approach — like, what are the lives of these species like? And what is advantageous for the species to either find at least reflexively bad, and maybe even consciously painful? So I feel like that in particular is a takeaway I’ll want to keep.

Sébastien Moro: And it seems pretty logical that animals are feeling pain, because it allows you, first, to move from something that could damage you, but also to remember that the thing is bad. So you need something long term. And, as we said, learning is very linked to emotions — so an explanation without any pain would almost be harder to get strongly than one that involves pain.

Boredom and depression [01:24:50]

Luisa Rodriguez: Yeah. Turning to feelings besides pain, do we have any indication of whether fish feel things like boredom?

Sébastien Moro: Not really. We have some hints from the welfare in aquaculture. There are many, many papers about the welfare of fish in aquaculture, which is pretty funny. Well, not that funny, actually. It’s just pretty weird, because fish have almost no welfare protection in aquaculture. So all the scientists are doing is totally unused at the moment.

But we know that when we add enrichment to fish, you change a lot of things — like their abilities to learn better, to cope better with the situations. We are really discovering that each time you add something, it helps the fish in some way.

So they are working at maybe adding some current in the water so the fish can swim, can exercise. It’s something for salmon especially, because salmon are migratory species, and keeping migratory species inside the tank is a massive problem. In my opinion, salmonids should never be in aquaculture. Well, I don’t think any animals should be in aquaculture, to be honest. But salmonids are really the wrong species for that.

And, for example, for rainbow trout — because it’s one of the topics I know best — rainbow trout are animals that are having shelters. When you put them in barren tanks, there are no shelters. So you can have aggression or these kinds of things. They are not very social animals, especially when they’re getting to be adults. When they are young they can be social, but not really as adults. And it goes the same with salmon. So yeah, we can be pretty sure they can get bored.

But more than all, we found out that they might develop depression, or at least depressive-like behaviour. And we have a few very interesting studies about that, especially from Norway, I think, from Marco Vindas and his team. And they’ve discovered that 30% of the salmon were just stopping growth, to get bigger, and they were just dying. They stopped eating, they stopped doing anything, and they were just dying.

Luisa Rodriguez: Oh, no. Under what conditions?

Sébastien Moro: Classical aquaculture. They were in tanks, in big circle tanks. I don’t remember if it was in the sea or not. But anyway, that’s not very surprising. And when they tried to check physiological cues, there were a lot of alterations that were reminiscent of depressive-like results in mammals. So we can’t say with certainty that it’s depression, but it looks like that at least.

Luisa Rodriguez: Sure. Can you say specifically what was similar to what we found in mammals?

Sébastien Moro: So for people interested, the paper I’m talking about is “Brain serotonergic activation in growth-stunted farmed salmon: Adaption versus pathology,” from Marco Vindas and his colleagues. So basically it’s the serotonergic system that is just overflown, and everything is just going crazy inside.

Luisa Rodriguez: I mean, I’m just beyond a nonexpert, but it’s at least a little striking to me that those feel like all of the same words that we do use to describe and understand depressive states in even humans — though I’m sure there are reasons that we shouldn’t actually take it as really strong evidence. But cortisol, serotonin: those are just the things that come up when you talk about depression in us.

Sébastien Moro: And oxytocin is also working in fish in kind of the same way. It’s linked to social behaviour, as we see in mammals.

Luisa Rodriguez: Yeah, interesting. So how do we study emotions in animals, and fish specifically?

Sébastien Moro: When we’re trying to assess emotions in animals, emotions are usually split in three parts. You have the behavioural part, like I won the lottery, I’m jumping, I’m screaming, I’m laughing: behaviour. There is a neurophysiological answer, like I’m releasing dopamine. And these two things, you can see them from outside. Or a bit from the inside if I take an injection or something, and I check what you have.

But there’s a third part that is a subjective one, and we do not have access to that. So when we’re trying to assess emotions in animals, we can only see the behaviour and control how the body is reacting inside: the cortisol, dopamine, these kinds of things.

But scientists started a few years ago to develop protocols to try to find out what is happening in the minds of those animals. So we have one big model that is usually considered the main model: the dimensional model, which means that emotions are in a graph. So you have a horizontal line which is the balance. It goes from bad to good: every emotion is either good or bad. And you have the arousal, which will be very excited or not very excited. And emotion will be pretty much on this graph.

Luisa Rodriguez: I think I’m getting the picture. If you picture the graph, then you have kind of quadrants. And maybe on the side of very positive and very aroused, you have ecstasy: extreme joy and excitement about something. Maybe in the low arousal positive part, you have something like feeling at peace, very calm. Then on the negative side with high arousal, maybe it’s terror. And on the negative side with low arousal, maybe it’s some dull, deep sadness. I don’t know, I’m kind of making this up. But am I getting the broad picture right?

Sébastien Moro: Yeah, that’s it. It’s an old model that has been updated a lot since then, and it’s often used. Then one thing we’ve discovered is that you have emotions and moods. Moods are more long term, where emotions are more of a short reaction to a special event. Mood is not linked to an event.

And from that, there is a very important protocol that has been developed: judgement bias. The idea is to see if an animal can be optimist or pessimist. And why would an animal be optimistic or pessimist? Well, maybe because when you live something bad, you might think that more bad things will happen.

This is why an animal that lives in very harsh conditions or environments will always have a pessimist bias. It will always think, “No, I’m not going to do this, because it will fuck up again.” But an animal that always has good stuff — every time the animal is taking a risk, it pays — then it will take more risk. And this is exactly what we see in humans as well. The silver spoon theory, it’s just that: when you come from a very rich family, that is protecting you; you can take more risk, because you will not have a very bad comeback.

Let’s say that in front of you, you have five doors. The door to the right is green. Every time you go there someone gives you £10. On the left there is a red door. When you are going to the red door and open it, someone slaps your face: negative stimulus. Green: positive stimulus.

But I said there are five doors. What happens in the other doors? You might be curious and try to go there to see what happens. So if you are someone pretty pessimistic, you will probably go to the green one, the one maybe just after, and that’s it. You will not go much to the centre. Someone who is very optimistic will maybe go to the almost last one, the one just before. Because what can happen that’s bad? Maybe just a slap. It’s not that bad.

And the judgement bias protocol is exactly that: you’re teaching an animal that some stimulus is positive, another one is negative, and then you test how the animal reacts in an ambiguous situation when it’s between both.

So how did they do that with fish? They took fish that are pair bonding for life. The pair, the couple, is very bonded, they’re very close to each other. The first part, they take a female. For the people interested, they’re Amatitlania fish. There is a tank divided in three compartments. In the middle there is a female. And in front of her you can have a small cup. When it’s on the right side and there is a white lid, then the fish knows there is food. When it’s on the left side and there is a black lid, there is no food. Negative, positive. And sometimes they present a new cup in the middle with a grey lid. And we count how much time does the fish take to go there.

So first, they just count the times the female is going in the white one and the black one and the grey one. This is our base. Then we remove that, and we put in two male fish. We’re trying to find out which one she prefers. We know, because she’s staying close to this fish in her tank compartment. So the scientist can find out, “She likes this fish. And this one, she likes this one.” And now what they do is they take either the one she likes or the one she doesn’t like, and they put her with the other male in the same tank now. And they let them do their life, like mate, make babies and everything.

What we see is when they are with a male they like, they’re having a very good life: they’re mating very well, there is no fight, there is a lot of survivability of the babies, and they are helping each other a lot because they’re participating a lot. We often think that fish are not taking care of the babies, that they’re just spreading them around. And some species of fish are doing that, but many species of fish are protecting the babies and taking care of them. You even have species where other fish from the community are helping parents to raise the kids when they’re not reproducing themselves.

So I’m coming back: if they put the female with the male she doesn’t like, then if you watch what happens next, there are fights, they are not mating that much. When the man and she have babies, the survivability is not very good and the male is not helping much. The couple is not working.

And then they’re doing the judgement bias test again to see what happens when females are with their boyfriend and others are with the crappy male. And we find out that when the grey cup is put out, the female who has their boyfriend, the one they like, will be much faster to go to the grey one than the pessimist one, than the one who is not with the other one.

What does it mean? It means that having a positive situation or negative situation changed the way they perceived the situation from themselves before. They are their own control. And this is the only time, to my knowledge, that a judgement bias test has been made on fish. It’s a very common test on farm animals, like pigs and cows: it’s very used for farm animals, for welfare questions. And there has been a student in theses who did it. But the paper is just a thesis paper. It hasn’t been published. But this one is the only real, good one that has been made.

Luisa Rodriguez: Yeah, that is fascinating. And it does feel really compelling to me to know that welfare scientists use this test a lot on mammal farm animals.

There’s a clear, very anthropomorphised story, where the fish gets to be with her soulmate fish. And they live a happy life, and she’s a happy fish who then goes about life thinking that things are going to go well and has a more positive affect. And the fish who has to marry — or “marry” is really anthropomorphising things — but who has to partner with a fish who she doesn’t like very much gets depressed, they fight, and it’s not a happy situation. And she becomes more depressive in her interactions with the world, and thinks that the world is worse than it is.

Is there a story we can tell where this doesn’t have anything to do with affective states?

Sébastien Moro: I’m not sure, because the very test itself has been developed to assess affective states.

Luisa Rodriguez: Wow. Yeah.

Sébastien Moro: Which doesn’t mean necessarily that affective states are conscious, because you can have unconscious affective states, but this is still one of the best clues we have today that gives us an idea that these animals can have affective states, and probably in a conscious way — because it can influence their behaviour; because these affective states change the way they behave before and after and for a long time.

This is not the only thing. We have other clues from other studies, not especially on that. But as much as we know how to assess objective emotions, this belongs to some of the best clues we have. I’m not even sure that we will ever be able to know for sure. But the thing is, how certain are we? With the knowledge we have today, and we’re progressing, I think we’ll have more efficient clues or protocols in the upcoming five to 10 years. I’m pretty sure we’ll have some more accurate stuff, and I have pretty much no doubt that many fish will pass it. I’m really confident on it.

And we were talking about the different studies that have been made to try to get into the subjective aspect of feelings, of emotions. Actually, I’m using “emotions” and “feelings” as if it were the same word, but in this field of science, “emotions” are more like the physiological behavioural reaction, but they aren’t necessarily conscious. Usually when they use “emotion,” consciousness is not implied in an obligatory way, but “feelings” is.

So I talked about the judgement bias study, but we have another one that is really interesting — because there is another model which is not something that is one or the other: two models are working together. So we talked about the dimensional model, and you have the appraisal model — which postulate that every animal, before feeling something, will pass it through a lot of different appraisals or evaluations. So, is it predictable, is it unpredictable? Is it positive, is it negative? And this is what will define the emotion they will feel. And there are many different things that will be evaluated. Some are conscious, some are not.

It’s been massively studied in sheep. A French study, actually; we have a French team specialised on it.

And we have one paper on fish, on sea bream, of exactly the same kind. So how did it work? They decided, exactly as in the dimensional theory, they follow the valence: so we have positive stimuli, which is food, and negative stimuli, which is taking the fish out of the water. And now, for the other part of the graph, it’s not arousal as it was before, but it’s predictable or unpredictable.

So the fish will have different situations. Either the food is warned by your light switching on, so it’s predictable and positive; or the lights and the food have no link — so sometimes the light is going on and food is not here, sometimes food is here, so it’s positive and unpredictable. Then we have negative predictable and negative unpredictable exactly the same way.

And when they’re observing what the fish are doing, and how the body is reacting, we notice that, for example, when negative is predictable, they’re trying to escape before even the scientists catch them. But when it’s unpredictable, they’re not, or way less. When it’s positive and predictable, you see some more mobility, some more stuff like that; when it’s unpredictable, much less.

And it goes the same for cortisol. Cortisol is linked to stress, usually. And what we see is that when it’s negative and predictable, the stress level is much lower than when it’s negative and unpredictable. Why? Because you can get prepared. So we have many different kinds of stuff like that. What we see in the end is that in each part of the test, the body is reacting another way and the fish is behaving in a different way — which means that it’s not the stimulus that makes the emotion; it’s the way the fish perceive the situation that makes the emotion or the feeling.

Luisa Rodriguez: That’s fascinating.

Sébastien Moro: It really is, because it gives us the idea that the fish himself, it’s his view of the situation that has an influence. And this is what we see in mammals and in humans. Sometimes the same stimulus will be positive and another day it will be negative. Sometimes you will be happy to go to work and sometimes you won’t, because the situation is different — and it’s not going to work that makes the feeling; it’s how you are going to work. That’s exactly what we see in sea bream.

Luisa Rodriguez: Yeah, yeah. So you’ve already made this connection, but it just really does feel to me like this is mapping onto a thing that describes the human experience, at least in some cases, very well. When things feel like they are predictably bad, maybe I can try to work on it to fix it. If it just feels like a bunch of things are going unpredictably wrong, that feels much more likely to cause something more like depression in humans.

Sébastien Moro: It’s really interesting to see how lots of human behaviours that we think are really just human, well, they’re not, and it’s pretty normal. Everybody needs control in their life. And it works for animals.

There is a behaviour that is called “contrafreeloading.” Contrafreeloading is a word to explain that animals would rather work for food than have free food. So if you give the choice between just a cup with food inside and a toy to get the food inside, usually the animal will prefer the toy.

And we’ve found that in cows. We know that cows are more happy when they learn how to open a gate than to just wait at an automatic opening. There’s been a very fascinating study on goats, where we’ve discovered that some goats would rather play a video game to get water than to just press a button to get water — because everybody wants to have control of the situation.

And it’s exactly where the difference between frustration and anger will appear. It’s very important for animal welfare, because why do we see that stereotypy? Because the animals are in some situations where they can’t do anything to help it, so they’re becoming helpless. So if you can control a situation, you get angry — because you can do something, you can fight — but when you can’t do anything, you get totally depressed and helpless. And this is exactly what we see in pigs, for example.

And we don’t have that much work on fish about this, but I’m pretty sure that contrafreeloading is something probably very present in these animals.

Luisa Rodriguez: Cool. OK, so we don’t know much about fish, whether fish get frustrated or aggressive when they have a lack of control or lack of opportunity to work for the things that they want —

Sébastien Moro: The probability is very high. Like really high.

Luisa Rodriguez: Yeah, you’d bet on it.

Sébastien Moro: There might be some studies that I’m not remembering right now because we’re really talking about thousands.

Luisa Rodriguez: Because you’ve read hundreds of studies.

Sébastien Moro: Maybe there is. But right now I mostly remember farm animals.

Memory [01:48:22]

Luisa Rodriguez: Sure, sure. OK, let’s leave that there. Pushing on, I’m curious what fish memories are like. Do they remember things for a long time?

Sébastien Moro: We have animals that have a lot of challenges: they need to remember where they live, where the nest is, where the prey are, where the predators are, when the predators are here or not. They need a good memory, as land vertebrates do. So there is no surprise that we can find different kinds of memories in different kinds of fish.

So you have fish, especially migratory fish, who have tremendous memories. You have fish that have been studied for, I think the longest one is 12 years, and they were still finding an exact spot 12 years later for migration. But for the most well known, for goldfish, it hasn’t been that much studied, because most of the studies are hardly going above one month, for technical reasons I think.

But for example, when they were taking the cleaner wrasses I talked about, these cleaner wrasses most of the time are wild fish. So they’re catching them in their habitat, testing them, and then putting them back where they found them. And it’s really interesting, because they found out that a year after the catch and release, the population of fish where they caught fish was really hard to catch again, while the other populations around were not. So it meant that the fish remembered this catching. It was highly aversive, so they learned to avoid that.

And we found that in carps. We know that they learn hook avoidance for at least seven months, and you only have to catch them one time and they will remember. Not for every animal, but many will.

Even sometimes in aquaculture: they were thinking about conditioning fish to come back when you were playing a sound. So they played a sound, and seven months later, the salmon were coming back at the sound. So their very good memory has been used for this kind of stuff. It hasn’t been pursued. Nowadays I don’t think this is used at all, but this was something that has been tried.

There is a very interesting and funny story. It’s one of the first papers we have on fish. It was in the ’50s and ’70s, so it’s really old. And one guy noticed that gobies, these fish are first living at high tide, and they’re checking the topography of the floor of the sea. Then the tide is going away, and they’re choosing a pool to be their home pool. And when the tide is away, they’re living in a small pool.

And the guy working on that, Lester Aronson, discovered that these fish were sometimes jumping from one pool to the other. But the problem is, when you’re inside the pool, you don’t see the other pool. But the fish were insanely accurate. So the question was, can they somehow fly, so they jump to the spot where the next pool is and they’re flying there? Or do they remember?

So he first checked what they were doing, and it seemed they were scanning at high tide, memorising the full area, and then jumping from memory. And he proved it. How did he do it? First, he took the fish from the home pool and brought them somewhere else, with different pools that these fish had never seen before. The fish were not jumping anymore. And when they were jumping, they were falling on the stones around on the rocks. They were missing pools all the time. So that was the first interesting thing.

And then a few days or even weeks later, he brought them back, and they still remembered. So they had a topographical map in their mind that they saw from top, at first in high tide, and then from inside and subjective view. And they could still jump precisely.

And there have been many tests on these fish, even a recent one, that showed that they can remember the area for at least a month and probably more.

Luisa Rodriguez: That’s insane! So it’s something like there are hills and low points in the sand, or whatever the bottom of this area is, such that when the tide goes down, some of the low points still have water in them, and those are the pools. And while the water is still high, not only are they able to work out that the deep bits are going to be the pools, but they also create a map — and they create a map with enough accuracy to, while in one pool, know that if they go like 36 degrees to the left, they will be in another pool.

Sébastien Moro: And they can go back to the sea.

Luisa Rodriguez: Jumping pool to pool. I mean, I feel like I’m like, “But do they remember things for more than three seconds?” And you’re like, “They actually remember better topography than humans do.” There’s no chance I would do that. That’s amazing.

Sébastien Moro: I’ll talk about cleaner wrasses in a second, and you’ll see they can memorise stuff that I’m not sure I could. But just to pinpoint a last thing, because I wanted to start with that, and I said we have many different kinds of fish with different kinds of needs: we have memories for food, memories for social stuff, and they can be very different from one fish to the other.

But for example, one very interesting paper was to make a comparison between the memory window of fifteen-spined sticklebacks, which are marine sticklebacks, and two different forms of sticklebacks. So the three-spined sticklebacks have a freshwater form and an anadromous form — “anadromous” means they go from seawater to freshwater and back. And they wanted to see if there were differences between their memory windows, because they have very different ways of life.

So they taught them to manipulate different kinds of prey, especially [amphipods] — this kind of small [crustacean] with shells on them, an exoskeleton. They taught them that, and then they stopped for a while bringing them this kind of prey. And after a delay, they brought back the prey. And what they found out is that the seawater sticklebacks forgot how to handle these prey after about one week; the anadromous after about 10 days; and the freshwater, after 25 days, there was no loss. So we don’t know how much they remember, but they seem to not forget.

So how can we explain that three animals that are closely related — except maybe the marine one, which is a bit different — but they’re very related, and they have memory spans that are so different?

There are different possibilities, but one of the most interesting ones is that it’s probably linked to their environment. When you live in the seawater, it’s changing a lot: the environment changes a lot, you move a lot, and the prey are changing a lot. When you’re anadromous, especially living in an estuary zone, you also need to be very versatile: you need to be able to change a lot of prey, or the territory will move with a tide, the rocks will be rolled around. But in freshwater, they are living in ponds, in lakes, and this kind of stuff. So everything is very stable.

So what does that mean? It means that if you learn something one day, it might probably be still working in a year or two years. But when you’re living in the sea where those sticklebacks are, or in an estuary zone, it won’t be the case. It’s better to be able to update to new information, forget information that is useless and learn new information that is more important. So all animals don’t have amazing long-term memory, but it doesn’t mean they’re more stupid or less stupid; it just means that they need it more or less.

I can find birds who have a much better long-term memory than we do because they need it. You have Clark’s nutcrackers: they can remember for six to eight months where they’ve been hiding 30,000 seeds with an insane level of precision.

Luisa Rodriguez: Oh my gosh.

Sébastien Moro: I can’t tell you what I ate last Tuesday.

Luisa Rodriguez: Right, right.

Sébastien Moro: But I don’t need to. They need to remember, because in the winter they will have to find food, and it’s the food they’ve hidden.

Luisa Rodriguez: Wow. Yeah, that’s pretty incredible.

Sébastien Moro: Yes. So now we shall be talking about episodic memory. So what is episodic memory? It’s something really, really important for self-consciousness because you have different kinds of memories, different kinds of knowledge that gather in a unified you, in a unified being. So everything must be gathered together.

And when we remember, we usually remember something with different components, like: What was it? Where was it? When was it? How was I feeling?

I don’t know if you’ve heard about the story of Proust?

Luisa Rodriguez: No, I don’t think so.

Sébastien Moro: In French, we often talk about the madeleine de Proust. And what is it? In one of his books, [Proust] talks about a memory he had. I think it was at his auntie’s, when he was young, and he was eating madeleines in her kitchen.

And it reminded him of a very special moment of his childhood.

This is exactly what episodic memory is: it’s the totality of an event. Even if it’s not exactly what happened, because your memory is always rewriting everything. But still, it’s the whole context of where you were, not just one thing or one thing; it’s everything at once.

And it’s really interesting to know if animals can have episodic memory. Are they able to recall all the events with all their contexts? It’s been tested on many animals. Rats have amazing episodic memories, really amazing.

We have one on zebrafish, which is a more traditional way to try it. It’s been used on pigs as well. The concept is to see if the animal is noticing when something’s out of place. So we teach them a context — a what, when, why, and all that — and then we change little things to see if they’re like, “This stuff was not here.” Like if you come back home and there is a hat on the hanger, and you’re like, “This is not my hat,” or, “This hat was not here.” It’s exactly the same concept.

For zebrafish, what they did is they took a square tank that was divided in quadrants. It’s not really divided in quadrants inside; it’s just for us to understand better. So let’s imagine the tank view from top: you have the square in front of you, and the two bottom quadrants are where the fish will be put. In the two other quadrants, you have on the left side a small Lego biker. In the other one, there is a Lego knight. I hope I’m not mixing up with the pigs one there.

But then the colour of the quadrant is different. So let’s say in a yellow tank, they see the biker on the left and the knight on the right. Then we teach them the opposite. So we have a blue tank, and this time the knight is on the left and the biker is on the right. So the fish is learning these two contexts.

And then we bring them back, for example, in the yellow one — but this time there are two bikers. So the fish already saw the yellow tank. He already saw the biker in the left part of the yellow tank, but he never saw the biker in the right part of the yellow tank. And this is what we observe. The fish is much more interested by this biker on the right side. And same thing with the knight in the blue one, if you put the knight on the left and you invert everything, the fish will notice when something is not working in the context.

Which means that they can understand the what: what is biker or knight; where: left quadrant, right quadrant; and when: when we were in the yellow tank, when we were in the blue tank. So it’s one of the first and one of the only tests we have on episodic-like memory in these animals. “Biographic” memory is another term.

Luisa Rodriguez: Yeah. For a second, I was like, what is this compared to? Because I mostly think of my memory as having these features of being episodic and having the who, what, where, and why. But I do sometimes remember just a fact, and I have no idea where I learned it or how I learned it, or what I was smelling at the time.

So I guess the idea is that you could think that fish had a much simpler one-dimensional memory — in the same way that I can remember just a piece of information without anything else about it — and this is just very impressive evidence that fish have these much more complex memories with a bunch of dimensions to them.

Sébastien Moro: Totally. And we have very different kinds of memories. You have memories for words, semantic memories. You have memories for some of your body movement, for automatism, this kind of stuff. So we have different kinds of memories. And this biographic, episodic memory is really a kind of addition of different stuff. We don’t use the word “episodic” memory for animals because episodic memory normally has the feeling in it — and as we can’t really know if animals have feelings, then we use episodic-like memory. But it really looks like that.

Luisa Rodriguez: OK.

Metacognition and self-control [02:04:23]

Sébastien Moro: We know for sure now that rats, for example, do have feelings, and they have some metacognition. In fact, one of the best ways to know if an animal has feelings is a test that has been done on drugs in rats. The scientists taught rats to press a lever when they were feeling drugs or alcohol in their body. So they were either injected with a placebo or real drug, and the rats were pressing the lever for drug or alcohol, or “I feel weird” — and if it was the good answer, they had food, they had a reward.

It means that we asked the rat to assess their own feelings and report it. And this is important, because we always think that animals cannot report anything, but they can — but this kind of study has hardly been used in any animals. It hasn’t been used in any farm animal, as far as I know. And I worked a lot on this, so I don’t think there is any. But that could be a very good way to evaluate how conscious some animals are.

And same for metacognition, which is being able to evaluate if you answered a question well or not. It allows us to know whether an animal is conscious that he doesn’t know. Like, can you reflect on your own knowledge? And we’ve shown that rats seem to be able to do that as well.

Luisa Rodriguez: I just want to pause on the interpretation. So the rats can recognise in themselves a sensation that is weird. And we still don’t really know what that feels like — we don’t know with confidence that alcohol makes rats feel more relaxed or sillier or more tired — but we at least know that they can distinguish between what it normally feels like to be them, and what it feels like to be them on alcohol.

I’m trying to think, is there some alternative interpretation where they’re reporting something else? But really, this just feels like the most likely interpretation is that there is something it’s normally like to be a rat, and then there’s something it’s like to be a rat with alcohol injected into them, and they can report the difference.

Sébastien Moro: And we don’t have anything like that in most of the animals that have been tested. We don’t have anything on fish like that. But it could be doable. We could try.

So now, I’m going to talk about the one where the cleaner wrasses are doing something I couldn’t, I think.

Luisa Rodriguez: OK, great.

Sébastien Moro: I ask for some concentration, because it’s quite complicated to explain without a graph or a drawing or something.

So you have plastic plates. You have four plastic plates with patterns on them. One is red, one is yellow, one is green, one is blue. And they each have different patterns on them, to allow the fish to recognise each one. On the red one, you have a flake that is a reward for the cleaner wrasse. This flake is made mostly of some fish flake and a small part of mashed prawns that they love. It’s their favourite meal. And on all the other ones there is mashed prawn.

At first we give the fish the red plate. The fish is coming. “OK, I like it, but not that much.” Then we bring the yellow plate. So now they totally stop eating on the red one, and they only eat on the yellow one because they prefer it much more. But there is a thing: the yellow plate is only available every five minutes. So they first present to the fish, the red and yellow. “OK, I’ll go on the yellow.” Then, 2.5 minutes later, they bring back the two plates. If the fish goes to the yellow plate, both plates are removed. You totally lost. But if the fish goes on the red one, he can eat on the red one and the yellow is removed. And then, after 2.5 minutes again, which makes five minutes, the fish can go back to the yellow one.

So what does it mean? It means that the fish has to understand that, first, he can choose his favourite food, then the non-favourite food, then the favourite food again. And they can do it easily. But the first thing we have to note is it means that these animals have self-control.

But things start to get complicated now. We bring the blue plate. The blue plate is available every 10 minutes. Then we bring the green plate. The green plate is available every 15 minutes. And now they start to mix everything. So sometimes the fish has the red one and the blue one. And he has to recall, “Have 10 minutes elapsed since the last time I ate on the blue plate?” And then the blue and the yellow, the yellow and the red, the yellow and the green — and the clean wrasse can do it. I’m not sure I could. It’s incredible.

And why can they do this that good? Remember what I explained about their way of life? Their clients have exoparasites, and the cleaner wrasses are removing the parasites. So the clients are cleaned, they leave, and then they are quickly getting parasited again, so they come back. So for a fish, it’s important to know, “How long was it since I saw this fish? Is it possible that this client has parasites, or is this client just coming back so I can rub its back?” This is why these animals are so good for this task. Because this task is totally relevant to their ecological system, their way of life.

Luisa Rodriguez: Yeah. I just also don’t think I could do this. It does feel really remarkable.

Sébastien Moro: I think it’s really pretty crazy.

Luisa Rodriguez: It really is crazy.

Sébastien Moro: Cleaner wrasses are incredible. Really, really incredible animals.

How do fish perceive the world through their senses? [02:11:22]

Luisa Rodriguez: Let’s talk about some other capabilities that we might not know fish have. What do we know about fish perception? You’ve already mentioned a few things about perception of colour, perception of ultraviolet light, you’ve pointed at olfaction. What else should we know about fish perception?

Sébastien Moro: Well, I should stop a bit on olfaction because olfaction is one of the most important senses in most of the fish. And once again, remember, we’re talking about 30,000 species or more — so when I say one fish, it doesn’t mean all the fish. But many fish communicate through smell, because smell chemicals can go pretty easily with the current of the water, so many fish have a very good smell.

For example, you have salmon: all the salmonid family, which include salmon and all that, they have an insane olfaction ability. We know that when a salmon is coming back from migration and coming into the freshwater, they can smell the place where they were born. They remember the exact chemical constitution of the water there, and they can smell it hundreds or thousands of kilometres away. It’s insane.

Their smell is so good that in an Olympic pool, they can smell it if you just put in a drop of a bitter liquid: just a drop, they will smell it. It’s that good. So they are very, very efficient with that, and they use it to learn, teach, and communicate a lot.

For example, there’s a very funny study where they were studying fish who were fighting and they noticed that they were peeing at some weird moments.

Luisa Rodriguez: Peeing? Urinating?

Sébastien Moro: Yeah, peeing. Yeah, it’s not a mistake. They’re peeing. The actual study is called “To pee or not to pee.”

Luisa Rodriguez: Good one.

Sébastien Moro: And so what they’ve noticed is the fish are showing their aggression, or communicating their aggression levels through pee at some very crucial moments. So this is how they’re building up their fight. That’s pretty amazing.

Last thing on smell, because I didn’t talk about this, but you have the biggest diversity of eyes in fish. Like, you have flat fish who have both eyes on the same side of the head. Flatfish are like that. When they’re a baby, they have eyes on both sides of the head. And as they’re growing, they will lay flat on the ground, hidden in the sand. So if they have an eye in the sand, this will be a problem. This is going to scratch, to itch. So their eyes are slowly migrating to the other side of the head, so they have both eyes on the right or left side, which make them the most asymmetric animals of the animal kind.

Luisa Rodriguez: Wow. It’s also kind of funny to imagine how their brain has to work out how to change the way that it’s perceiving vision as the eye changes location.

Sébastien Moro: Holy hell! I can’t imagine it. It’s just crazy. So weird.

Luisa Rodriguez: Like if one of my eyes is slightly blurry, my depth perception gets terrible. But this is moving spatially to the other side of one’s head — and being able to see throughout that.

Sébastien Moro: If you type “flatfish” or “flounders” on Google, your mind will be blown. It’s crazy.

But you have other fish like Anableps, who are called in English “four-eyed fish,” because they are living exactly at the surface, so half of their eye is underwater and the other half is on top of the water. So the cornea and all their eyes inside are designed to see outside of the water and inside of the water. The cornea has a deformation, the retina has a deformation, and the colour cones.

Because to perceive colours, you have cones in your eyes. So all the cones for green light are mostly on the bottom of the back of the eye, because it’s what gets everything in height: the trees, the vegetation, everything. And the top of the eye, which is pointing at the bottom in the water, has most of the blue cones. So when you type “Anableps,” you will see these eyes. They’re so weird. Really, they have two eyes split in two, hence “four-eyed fish.”

You even have fish who can see through their skulls. It’s a deep-sea fish, and the only light is coming from the top because they are living very deep. So they have a transparent skull, and the eyes are barrel eyes, and they can point them up and see through the skull.

Luisa Rodriguez: That’s amazing. And bizarre.

Sébastien Moro: I don’t remember the name.

Luisa Rodriguez: I think Pacific barreleye fish.

Sébastien Moro: Yes, that’s it. Barreleye fish. And if you check on Google, it’s crazy.

Luisa Rodriguez: Oh, god. That’s pretty creepy.

Sébastien Moro: It’s been filmed only once or twice. I think it’s really an unknown animal.

Luisa Rodriguez: Oh, wow.

Sébastien Moro: Another thing that I didn’t point out before, but it’s important… Oh, man. Your face. You seem terrified.

Luisa Rodriguez: It’s pretty haunting.

Sébastien Moro: Yeah. Many fish have never been studied. Like, we know about maybe a handful of species that we know very good. Let’s say, 20 or 30 maximum of 30,000. We don’t know anything about fish. It has to be clear. And the studies are really, really recent. I’m just making up this number, but it’s pretty close to the truth: maybe half or even two-thirds of the studies on fish cognition are from the last 20 years, and many of them are from the last 10 years.

Luisa Rodriguez: Wow.

Sébastien Moro: It’s really impressive. As I said, the picture one, with the cleaner wrasse, is from last year. And when we talk about electricity in fish, I have one that has been released a few weeks ago.

Luisa Rodriguez: Yeah, that’s pretty wild.

Sébastien Moro: Yeah. We were talking about smell. I came back to the eyes because we have the same thing about nostrils in fish. In humans, we have the nose and the mouth that are connected. So for us, it’s pretty close. It means when your nose is blocked because you have a flu or something, you have way less taste in the mouth.

In fish, that’s not the case at all. The nose is really dedicated only to smell. So you often have an entry nostril and an exit nostril. I don’t mean that they have two nostrils; they have four: on the right side they have one entrance and one exit, and on the other side they have one entrance and one exit. And inside, you have very small organs that are directing the flow in the kind of tube nostril, to smell precisely — which explains why sharks and these kinds of animals have amazing smells.

And hammerhead sharks, we don’t really know why they have this weird head, but the biggest probability is to be able to discriminate the direction of smell — because their nostrils are on each end, so it’s easier to know if it’s coming from the right side or left side.

And taste: another interesting thing about taste is when your mouth is dry, you can’t taste anymore — because you come from the same ancestor as fish, and you brought some of the water with you inside your mouth. But fish are living in the water, so they don’t need to have the mouth especially watery or anything — which means that you can have gustative cells outside your mouth on your body. Which means that you have, for example, some catfish who have buds, gustative buds all on their body.

Luisa Rodriguez: Like tastebuds, basically?

Sébastien Moro: Yeah. Which means they can taste at a distance. So when you’re going to swim in a lake or a river, more a river that has catfish, they can taste you. So please do not pee in the water. They can taste it.

Luisa Rodriguez: Yeah. The impression I’m getting is just… I mean, not only do we not know what it’s like to be a fish — if there is anything it’s like to be a fish, kind of in their head — but if there is, if the lights are switched on and they are experiencing things, their experiences would be so incredibly different to ours. Really foreign in some ways, fascinating and incredible. I found that really interesting.

Sébastien Moro: There is something else that is reminiscent of this taste in the distance: they can also touch in the distance. Something that all fish have is a lateral line, which is an organ that goes from the head to the tail. It’s a kind of tunnel that goes all along their body and all around their face, and there are very small cells in it. When the flow is getting inside, they can feel the hydrodynamic pressure, and they can feel the direction of the flow — which means that if there is a rock on the side, it will make a kind of pressure on their side, and they will feel it at a distance.

This is one of the mechanisms of shoaling: when all the fish are all together very close and reacting very fast, it’s not only, but also because they can feel each other. If one is moving, they feel it through the lateral line, which is also something used to listen. You can try to understand this by thinking of when you’re going to a live show and the bass is kind of moving your chest. It’s vibrating.

Luisa Rodriguez: I see.

Sébastien Moro: This is the same kind of stuff, like it’s a very low sound. They’re feeling it, hearing it through the lateral line. So it’s pressure, it’s a feeling of touch and hearing at the same time. But it’s very precise. For example, you have fish like the Mexican cave fish, which is also called the blind cave fish. It’s fish who evolved in a dark cave, so they’ve lost their eyes. They have no more eyes, but they can just go around in their environment without bumping into anything, because they are doing all of that with the lateral line.

So you have these kinds of things. You have a hearing system. Fish have the biggest diversity in hearing systems — which is weird, because they are hardly communicating through sound. Some are, but not that many. But they are listening. But we don’t know what they are listening to.

And there are a lot of different kinds of ways to listen. They have a swim bladder in their body, which is a gas pocket that allows them to rise in the column of water or to go deeper. And it’s acting like a transductor, because when fish are making sound — I won’t go into much detail, because it’s complicated — but basically they’re doing a wave sound, and they’re also doing a kind of wave of particles of the water: they’re pushing the water.

This is why fish probably aren’t communicating orally or by sound that much: because it’s hard to push in water. For whales or dolphins it’s not that hard, because they’re big animals. A small fish needs more power. But some fish are. So the sound, if it was just a waveform, would just go through the fish that are all made of water. But as it has a kind of gas pocket inside, the gas pocket is transmitting the sound to the inner ear. So it’s the vibration of the swim bladder that transmits sound.

And you have lots of different kinds of hearing. For example, goldfish have some mechanism that we call the Weberian apparatus, Weberian ossicle. It’s kind of like ossicles in the inner ear that works a bit like ours. They’re hearing a bit the same as we do, so a goldfish has a hearing range pretty close to ours — a bit less, but pretty close. It means that when you have goldfish in your house, and you are playing TV, they are hearing the TV with all the reverberation of the sound inside the tank, which could be a problem.

Luisa Rodriguez: Wow. Yeah.

Sébastien Moro: You have some fish that can hear in ultrasound, because they are chased by dolphins, so they have to hear the ultrasounds of dolphins.

Luisa Rodriguez: That’s really cool.

Sébastien Moro: So they are hearing in a small range, then nothing, nothing, nothing — and when you reach the ultrasound, the precise range of the dolphin, they can hear again. They have a big gap of no hearing, and they can hear normal and very very high, but nothing in between.

And you have some fish who just can hear through bones’ vibrations and lateral lines. So they can only hear very low sounds. Many sharks are along that, and we are just discovering how it’s working. For example, sharks’ and skates’ systems of hearing: the first review that has been published on that, and the first time we’ve noticed that they were also communicating by sound, it is really recent. We’re just discovering it.

And then there are other senses that are really amazing, like electromagnetism. It’s using a compass, a geomagnetic compass to orient in space.

Luisa Rodriguez: Oh, wow. That’s incredible.

Sébastien Moro: Some fish can produce electricity as well. And we can go on that. It’s long, because it’s one of my favourite topics. First, I will try to explain the families of electric fish, because there are two big groups. So when we’re thinking about electric fish, you’re thinking about which fish?

Luisa Rodriguez: Eels.

Sébastien Moro: Electric eel. Which is not an eel, it’s actually a Gymnotiform, which is the American family of freshwater electric fish.

So first, why do we have electric fish and not electric cows or electric dogs? Because electricity can go through water and not through air. So you have fish that can produce electricity — which is the case of electric eel — and some who can receive it, and some who can do both.

Those who can receive it, like sharks, have Lorenzini ampullae that are very, very sensitive. For example, a shark can perceive the electricity produced by your muscles’ contraction. But this is absolutely not amazing compared to what I’m going to explain now.

Now, we have two big families of electrogen fish who can produce electricity. You have the American family, the Gymnotiformes, and the African family, the Mormyridae. So the elephantnose fish is an African one. And in these fish, most of them are producing very low current. They are not producing a big electric shock as you find in electric eels, because electric eels are really special animals, very one of a kind, because they can do both. They can make small electricity bursts and very powerful ones — like very, very powerful ones.

And electric eels actually are animals that need to go back to the surface to breathe. They are not obligate air breathers, but they need some air. This is why when the first European explorers came there, they often got shocked by electric eels, because they were just hiding in the grass. They were not in the water, so they were just walking on them.

So in these fish from both families, you have two types of fish: the wave fish and the pulse fish. Wave fish are fish that are always producing an electricity field around them that they use to do electrolocation. They’re gathering information about their immediate environments. This is especially the case for fish who are living in very fast water. The more you produce an electric field, the more information you have. You can try to imagine that as image per second for vision: the faster the number of images per second, the smoother the video is. It goes the same with electricity. So when they’re living in very fast water, they’re always producing electricity, and they use that to check their environment through electrolocation.

And the pulse fish — and the elephantnose fish is a pulse fish — they’re just sometimes pulsating and sending a burst of electricity around them. So for example, an elephantnose fish sleeping is just doing like bip… bip… bip: it’s like taking pictures, right? And as soon as the security system will pick something, it will start to do bip bip bip bip — so it becomes an animation.

And we’re talking, especially for wave fish, about something like thousands of pulsations per second.

Luisa Rodriguez: Wow.

Sébastien Moro: Yes, per second. No, you can’t understand what it means. I can’t either. So this perception, this electrolocation, is something that is limited to around one body length of the animal. So it’s very close. But they can see through things. Like if an elephantnose fish is going close to the bottom, the electric field will go inside the [substrate of the ground], and if there is prey hiding there, the fish will find out.

There is a publication that was released in [March] this year. They found out that some elephantnose fish were often staying together in a parallel way. And they made a lot of calculations and tests, and they found out that they can use the electric field of another fish to enhance their perception: it’s collective sensing of the environment through the electric system. We don’t know of these kinds of things in any other animals.

Luisa Rodriguez: Wow. There’s so much there. Can I ask a few questions?

Sébastien Moro: Yeah, yeah, go on.

Luisa Rodriguez: So first, just to make sure I understand, can I think of electrolocation kind of like echolocation? You send out electric pulses, and something about how they bounce back gives you a sense of what they’re hitting?

Sébastien Moro: Yes and no. But it’s a very good question, because I didn’t explain how it works. So I won’t explain how it works in all fish, but I will explain how it works in an elephantnose fish, because it’s very easy to understand. So at the base of the tail, they have a kind of battery, a very small battery that is bursting. It makes a kind of circle around them, and it’s coming back to the whole body. So imagine when you see an image of a battery and you draw the electric field of this battery: it’s going from the minus pole and going to the plus one. It’s doing exactly the same thing on the fish.

What it means is they have a kind of sphere around them of electricity, and every time something gets in it, it will change the waveform and the amplitude. So, for example, if there is something that is not conductive, like wood, then it will stop the current at this very spot. So what will it do? The receptors on the body of the fish will get, “OK, electricity, electricity… Oh, there’s nothing here.” We don’t know if they perceive that as a shadow or dark spot or cold spot. We have no idea, because we don’t have the sense. And spoiler: you will never know. But they feel something.

And if you put something very conductive, then they react, and they actually try to get away from it. So it’s a negative affect feeling, which means, if we’re talking about sentience, it seems that they can have a feeling associated with a sense that we don’t have. So they have a feeling we don’t, clearly. So that’s really incredible.

And what they’ve discovered about this is it seems that the fish are able to recognise different objects through that. So they can evaluate the 3D shape, how it’s constituted. Because colours can change: if you’re in a shadow, or if there are different conditions of light, the same object will have different colours. See what I mean? But it does not happen with your electrical properties. Your body always has the same electrical properties, which means perceiving something through electricity is much more accurate than using colours.

And what are colours? Colours are when you are receiving special wavelengths of an electromagnetic wave, lights. But they’re receiving electricity. We’ve noted that they can add amplitude and waveform modification to know that a certain object, like a certain kind of worm, will always have the same waveform modification and amplitude modification. The different species of grass will do the same, or algae — which means that they have the equivalent of colours, but in the electric sense, which you cannot understand. I cannot either, but it’s the case. And they have eyes, so they can see colours as we do, but they can also see something like electrical colours.

They can even get fooled by visual illusions through their electric field. For example, if I want to draw a triangle, but I only draw the point of it. So the top, the bottom right and the bottom left. I don’t put the segments on each side. You will still see a triangle. There is no triangle. There are just three small arrows. There is no triangle. It’s a visual illusion; your brain is constructing the segments.

Luisa Rodriguez: It’s going to fill it in.

Sébastien Moro: They’ve done exactly the same thing by teaching the elephantnose fish to recognise a pyramid. What they did is they cut the segments, and to hold everything together, they put it in agar-agar.

Luisa Rodriguez: The gel, yeah.

Sébastien Moro: Exactly. Which is electrically neutral, so they can’t feel it with the field. And they recognise the pyramid. Well, the “pyramid”: there are no segments at all. There are just the sides.

And just a last thing: I’m not going to develop on it, but this electricity system allows them to communicate. They’re scanning the environment, but communicate at the same time. And every waveform, every burst has a signature. It’s a signature for a specific individual containing the sex, the age, the size, and many other things.

Luisa Rodriguez: That is just completely incredible.

Social lives of fish and fish personalities [02:36:51]

Luisa Rodriguez: If you’re happy to move on, I would love to talk about the social lives of fish.

Sébastien Moro: Exactly. We were talking about learning about the environment by two views, but we did not talk about social learning. And we have social learning, and it’s really amazing.

A very old paper on fish has shown that some species of fish are living in groups, and they can have the equivalent of traditions or cultural transmission. So how does it work? It was 1984, really old: “Social transmission of behavioural traditions in a coral reef fish.” The English name of this fish is French grunts. So these French grunts are living in an area. They are observing what they’re doing, and they see that they’re going there to sleep, there to eat, during the day…

Oh yeah, we didn’t talk about this, but fish are not migrating only in the horizontal plane, but also the vertical plane. So they’re migrating in different directions. And what they did is just write down everywhere they’re going and how they’re going there, what is their everyday life.

Then they had new animals, new French grunts who’ve never been there, and they start to use exactly the same spots as the group of residents. OK, so there are two possibilities now: either they’re following the group, or it’s the environment that pushes this kind of use — so it’s not the fish themselves, but it’s just how the environment is made.

So now, they removed all the fish and put in a new population of French grunts, and they start to use the territory in a different way. So it means that it’s not the territory itself; it’s the fish learned from the others and continued. And we know it happens.

And this is a problem especially with fishing, because fishing is focusing on big fish — and big fish usually are old fish, and old fishes are the animals who know the territory. So you break culture. Sometimes you can have a collapse of a population, just because you killed the animals with the knowledge. We found that in land vertebrates as well. We have a few papers like that, but this one is the most compelling, even if it’s old. And I said, “Old papers, meh” — but this one is still pretty impressive.

So fish can learn from each other, clearly. Not all species, not everywhere, as not all fish feel the same thing, and not all fish have the same mental capabilities. What we saw in cleaner wrasses especially is very linked to their way of life. Many fish don’t have that complex way of life, but they might have other challenges that make them better at other things and so on.

Luisa Rodriguez: Yeah, yeah. OK, so social learning sounds at least present for some species. Do fish have recognisable relationships between each other? Do they seem to form bonds? What else are their social lives like?

Sébastien Moro: Oh, I love this question. Do we know if they form bonds? There have been many species tested for that, especially guppies. We have discovered social networks in guppies.

There is one paper on wild guppies where they really built a social network. They took every animal from two ponds that are linked by some water flow. They found out there were three communities of fish always interacting together. And between those communities, there were animals doing a link between the communities; they were always switching from one to the other. You have central animals with lots of relations with lots of other animals, you have more peripheral animals, you have a lot of different kind of animals — because they have personalities, as every animal does, so it changes depending on the personality.

And in guppies, the bond doesn’t seem to be in how long the animal spends next to another one, which is usually what we use to know there is a bond, but it’s more about the frequency. They are often found one close to the other, but not for a long length of time. When, if you’re talking about cows, for example, when they have a friend, they stay with a friend all the time for years.

So we have that on guppies, and we have other fish that have stronger bonds. One very interesting one was a study about prosocial behaviour. So the idea was to take fish who are bonding for life, male and female, a bit the same kind of animal as the one I talked about in the judgement bias study. And they took the male out, and they split the male and the female.

I’m explaining this, so try to visualise this in your head. You have a tank split in two parts: the part closer to us, there will be the male; in the other one, there will be different people. In the part of the male, you have two compartments: one compartment with a red [circle], the other compartment has a blue triangle. So if the fish gets in the compartment with a red circle, it will receive food, and food will be delivered in the other tank as well. If the fish takes the blue triangle, this fish will receive food, but nothing will be delivered in the other tank. So we have a prosocial choice and antisocial choice.

So when there is no one in the other part of the tank, the male is choosing randomly. If there is a male, a possible rival: antisocial — almost 100% of the time, antisocial. Now, if there is his wife — yes, this is anthropomorphising; I don’t care — there is his female, this is a prosocial choice all the time.

And now a question: what happens if you put in a new female? Is it just because this is a female or is it just for their female? Well, if their female is just next, when they’re bringing a new female, it’s the antisocial choice all the time. Now, if there is not the female of the male, it will depend on how long he’s been separated from his female. At first it will be antisocial, and after a while he will start to switch to prosocial choices.

Luisa Rodriguez: Oh my goodness. OK, so if his female is just like in the vicinity, he’ll be antisocial around the other female. But if she’s gone, and she’s gone for a while, he’ll start to become prosocial with this other female. Oh, that’s complex.

Sébastien Moro: To give an idea of how much we’re discovering these kinds of things, this paper is from 2021.

Luisa Rodriguez: Wow. So it’s very recent.

Sébastien Moro: The paper is “Prosocial and antisocial choices in a monogamous cichlid with biparental care.” It’s Amatitlania nigrofasciata. When the one about judgement bias, it was Amatitlania siquia. So pretty much the same species, or close. And yeah, it gives an idea of how strong the bonds can be.

And we have another very interesting one about reciprocal altruism. So altruism is helping someone, usually without expecting something back. Reciprocal altruism is helping someone, hoping this one will repay you later. So I’m helping and you will help me back later. I’m trusting you. It’s a question of trust.

So you have rabbitfish. Rabbitfish are fish that live in pairs. But even if it’s often a pair — male/female, a couple — it’s not all the time. It could be two male, it can be two female. And a scientist discovered something really strange. They are vegetarian fish, so when they eat, they have to put their heads inside the coral reef and get pretty deep. So when one fish is doing that, it can’t check around if there is a danger. When one of the fish is doing that, the other one is taking a 45-degree upright position and don’t move, and scan around if something is coming. If something is coming, this fish will wave and touch the other fish and the two will go hiding. So he or she is on a watch, basically.

Let’s start from that. I am the fish that starts to go eating. I go eating. The other one is waiting. When I finish, I could just leave. I have nothing to gain by taking the watch. And yet when the first fish stopped, it took the upright 45-degree position, and the other one goes eating. It works very well. This is reciprocal altruism that we observed, and there are lots of videos of this behaviour that is really interesting.

And in other fish, for example, I don’t remember the species, but it’s a species where the parents are taking care of the eggs and the babies, and they have a territory. But at the same time, there are helpers from the same species that stay on the territory, even if they are adults. Even if they could reproduce, this year, they’re not reproducing and just helping. And we don’t know exactly why they help, but we know that they pay to stay: they have to help on the territory. They have to fight against the predators; they have to dig holes, so the water is filtering some food that could be stuck in the holes.

Betta splendens — the Siamese fighting fish, I think you call them — they’re known because of their fighting. First, we have to know that the fighting version has been much selected by humans to fight. When they were domesticated at first, it was to fight, as for a cockfight. Exactly the same thing. So the males of the domestic strain are more aggressive than the wild ones.

But anyway, when they fight, the two males will try not to be too aggressive to each other. Because as every animal fighting is in the same species, it’s not evolutionary smart, so they try to avoid being injured. So there is a ritual. These fish will have frontal displays, side displays, lateral displays. And they have different kinds of movements, like flaring the fins or they have a special movement of the gills. There are some bites as well. They can bite one another.

Some of these behaviours are very aggressive and sometimes end in an attack, and others are just to impress. And so if they’re just fighting male against male, they have, let’s say, normal ritual fighting behaviour. If a male is watching, they will become more aggressive because they need to impress the opponent and the audience. If it’s a female, the female likes a dominant male, but they don’t like aggressive males — which means that this male will remove a lot of the aggressive part of the display and will enhance the demonstrative part of the display that are also used in a parade for the female.

So they show what we call an audience effect. And same if they know their rival or not, they will not fight the same way. So many things are coming with the decision of how to fight.

But it doesn’t stop here. If a male wins the fight, and you put him in the presence of two females — one who saw him winning and one who didn’t see him winning — he will try to seduce any of both. Now, you take the loser. You see me coming. The loser will only try to seduce the one who didn’t see the fight, because he knows the first one will not want him.

Now I’m going to something even more incredible, which has been studied in depth: the grouper and moray eel interspecific hunting.

So what is a grouper? A grouper is a carnivorous fish, which is massive. It can grow up to two metres long. It’s a really massive fish. It’s a Formula 1 fish. It swims very fast, it’s very big, it’s very powerful. It has a massive bite force. And it’s a diurnal hunter, so it hunts only during daytime.

Now we have the moray eel. Moray eels have very snaky-shaped bodies. They are nocturnal hunters. They don’t hunt during daylight, and they are more made to go inside crevices, inside holes and this kind of thing.

So two very different hunters. The groupers, when they are hunting a prey, it often happens that the prey is hiding inside the coral reef. And at this point, the grouper can’t get inside. Well, the grouper developed a communication with the moray eel with a kind of head shaking. We have lots of videos of it. It goes to find a moray eel and shakes its head above the moray eel to ask the moray eel to come and hunt with him. Sometimes the moray eel will accept. Sometimes it won’t, because it’s sleeping.

So let’s say it’s coming. If the moray eel finds another hole, it might go inside and sleep. So the grouper will come back and go like, “Oh, we’re going to hunt.” And then the grouper is going exactly where the prey entered inside the reef, and the grouper will point at the hole. This is one of the only referential gestures we know in animals, because it means that the fish is not communicating just a feeling, an emotion or something like that: pointing at something external from him and saying to the moray eel, “You have to get here.” And the moray eel understands it and gets inside.

You can find videos everywhere on YouTube about that. This is quite amazing, because almost everything is directed by the grouper. Then the moray eel gets inside, and the grouper is waiting outside of the reef. And there is no sharing of the prey: it’s either the moray eel gets the prey, or the prey escapes from the reef and she has a massive Formula 1 killer waiting for it. And so either the grouper wins, either the moray eel wins, and then they continue hunting like that so each one has their meal.

Luisa Rodriguez: So they’re not even sharing it. They’re just like, “One of us is going to win. But it’s worth it to both of us to do this, because then one of us gets something.”

Sébastien Moro: And they found out that groupers can hunt with pretty much any animal that will accept to hunt with it. We have videos of some species of grouper hunting with octopi. We are now talking about a vertebrate hunting with an invertebrate together and communicating. This is something we don’t have on land that easy.

And one interesting thing is, when they don’t know very well the environment, and don’t really find someone to help them, they will just point until someone finds and helps them. And sometimes, they have been observed pointing for up to 30 minutes.

Luisa Rodriguez: Oh my goodness.

Sébastien Moro: It’s really impressive. Sometimes it’s fish: there are fish with very strong jaws and they can bite inside the coral reef, so sometimes they’re hunting with these fish. It depends a lot.

And when we were talking about some comparison with other species, there has been one study on groupers selecting moray eels for hunting, and it was compared to another study on how chimpanzees select their partners for hunting. And this showed that groupers follow the same rules as chimpanzees. They can actually assess the efficiency of their partner and then select only the most efficient partner.

They’ve done that in a very easy way: they’ve made fake moray eels, and then the grouper had the choice to go to one or the other. And the scientists were controlling the fake eels, and one could go help them, and one was never moving. And really quickly, the grouper stopped going to the one that was never moving.

Luisa Rodriguez: Amazing.

Sébastien Moro: They can assess the efficiency.

Luisa Rodriguez: It’s just one mind-blowing fact after another.

OK, just because I can’t help myself: probably like 10 minutes ago at this point, you said something about fish have different personalities, and that was an unexpected comment to me. Can you say how we know that, or what exactly we know about it?

Sébastien Moro: I hoped you would come back to that. So first thing: what are personalities useful for? Why did personality evolve at any point? It did because if, in a group of animals, every animal behaves the same way, you never progress, you never learn, you never go anywhere.

And we often see two big kinds of personalities, which are bold animals and shy animals. We find that everywhere. Usually bold animals are animals who are the most adventurous. They are risk-taking animals; they are discoverers. But the problem with that is it’s often animals who are not really flexible. They have a hard time moving from something they learned when it’s not working anymore to something else.

On the other side, you have the most shy animals, who are often the most social ones and also the most creative ones. They’re often the ones with the most flexibility, who can use the same thing but in different ways.

So it immediately highlights why personalities would matter in social groups. So these two facts — bold animals, shy animals; flexible animals, inflexible animals — we have them in fish, but then we don’t exactly know in detail what a fish personality will be. In humans, we use a model most of the time called OCEAN, which has different criteria of personalities.

Luisa Rodriguez: Extraversion, openness, agreeableness…

Sébastien Moro: Exactly. And neuroticism. But what is in fish? So far only shyness, boldness, and maybe one or two others have been tested, but not that much. But it wouldn’t be that surprising.

And we find big differences between animals, actually. We can really find more differences in two individuals from one species: the differences are bigger than the less efficient and most efficient of another species. I mean, you can have giant differences. It’s clear you can. So yes, the fish have personalities, clearly.

Luisa Rodriguez: I’m just kind of interested in how it’s observed in a study. Is it like you put fish in similar environments, and some of them will just be very risk-taking and others will be much more timid?

Sébastien Moro: That’s it. And you can compare it to do they learn fast or not? Can they learn a reversal of a task faster than others? You can do this test too.

Luisa Rodriguez: Wow.

How representative are these results? [02:59:02]

Luisa Rodriguez: So I basically want to ask a question where I don’t want to discount everything you said, because it is incredibly impressive, but I am curious if you feel like you’re cherry-picking? Like, are you picking out a couple of really, really clever fish, and ignoring the fact that maybe the majority of fish don’t pass these kinds of tests? Or does it seem like lots of species of fish are performing really well on lots of different kinds of tests? Or maybe we just don’t know because we’ve only studied two species of fish? That wouldn’t surprise me either.

Sébastien Moro: So many things to answer to that. First, there aren’t that many tests. We don’t know 99% of the behaviour and cognition of fish. Really, we know nothing. So I don’t have that many studies to cherry-pick.

Then something that has to be really understood is cherry-picking means I would like to show something, and especially that fish are closer to mammals than they seem. And they’re not. And I totally embrace that, because this is precisely what I love in these animals.

Luisa Rodriguez: They’re aliens.

Sébastien Moro: Yeah. They’re very different. At no point ever I will tell you that fish are feeling pain the same way we do. They don’t. But that doesn’t mean it’s more or less important. It’s just different. But for them, it’s important.

And yes, I know I have a very positive bias towards animals. I want people to say, “Oh damn, they’re crazy!” So obviously I’m picking crazy studies, but I’m reading pretty much everything that’s released, so it allows me to also notice when I have bias and be careful about it.

I’m not a scientist myself, so I had an imposter syndrome for so long that I’m explaining to people studies made by scientists much smarter than me, who probably don’t have this very positive bias, and I would feel bad to go away from what they say. So what I say, you can almost find it word for word in their papers. But I don’t really have to cherry-pick, just because I don’t have that many papers, and also because fish are amazing — it’s not my fault!

Especially pain in fish: this is a topic that I’ve been following very closely, and I read everything, and there is a big debate around that. But I always try to explain the two sides. Especially at the present day, the very few people criticising pain in fish really harshly are more or less linked to fisheries, first — but it doesn’t mean that the scientists who are favouring the possibility of pain in fish aren’t influenced by their own opinions. It’s OK, we all are. It’s something I have to take into account. But today, we have way more information going in the direction that fish can consciously feel pain than not.

For sharks, we don’t know if sharks can feel pain, because sharks don’t have the type C fibres and they don’t have that many nociceptors, so we don’t really know. And all sharks are not the same animals, once again. And sometimes, in some species, when they are reproducing, they can bite each other very strongly, and maybe they don’t feel this and they feel something else. Maybe they can’t feel.

We don’t know exactly even if sharks feel pain, but mostly because it hasn’t been researched, we don’t have much. We have like two papers maybe, and very old ones. Everything that is before the ’80s or ’70s, you can just barely take it. But we have a lot of weird things, like fishermen tagging fish and then they see fish with massive infections and the fish doesn’t seem to have changed their behaviour. But it’s not really studies, it’s just observation. So we don’t know. We have no quantifying results to use, so we can’t really use that.

But I’m trying my best not to cherry-pick, so I hope I’m not. And this is one thing that really helped me in overcoming the imposter syndrome, is my work is really appreciated in the universities world. And actually I’m counselling for a vet school about fish and for a French governmental organisation, and I learned there that my book on fish is given to their students. So I think it should be pretty OK.

And I don’t know if you’ve noticed, but when there is only one paper on it, I say it. For example, the Atlantic cod one with the self-feeder and the tag and everything, it hasn’t been replicated. And this is something else, this is a problem there is in science generally: the way that publishing journals are working. Usually what you want is to publish some amazing stuff because you get funds, because you get a publication in top-rated journals. And sometimes the abstract, the resume of the paper, is not as amazing as what you have inside. And the other thing is that negative results are not published.

Most of the time when I work, I try to know as best as I can the whole field. So when I read something, I can find out, is this coherent with what we know about this species, or is it not? And sometimes it’s not, or I have a doubt. And then in these cases, either I don’t talk about it, or I talk about it if it’s really amazing — but I am precise that this is the only paper, so it has to be taken with caution. But, for example, cleaner wrasses: goddamn it, cleaner wrasses, we have so many things on them, there is just no doubt. Just none.

Luisa Rodriguez: Yeah. OK, so to the extent that it’s possible that there are some capabilities that might not turn out to be real or very common, it might be true of these papers where there’s one study and they might not turn out to replicate. But you’re generally pretty clear on when there’s one paper. And when there are multiple papers, that’s just reason to think that those things are a real phenomenon.

Sébastien Moro: Yeah.

Sébastien’s TV and movie recommendations [03:06:22]

Luisa Rodriguez: OK, I’ve now been asking you questions for something like five hours. So let’s end with just one more. Do you have a movie or TV recommendation for our audience, or a genre that you particularly love?

Sébastien Moro: So if I just stay on the topic, BBC Blue Planet II is incredible. And some of the things I’ve been talking about, you will see them — like collaborative hunting, tool use. This is a very, very good documentary. It’s really good.

Otherwise, I’m a big fan of Indian cinema. So if I can help people discover Indian cinema — which is totally unrelated to what we’re talking about — I’d say maybe start by S. S. Rajamouli cinema, especially RRR, which is short for “Rise, Roar, Revolt,” which is kind of a bromance in colonial India. It’s an insane movie. It’s a blockbuster movie, like an action movie, but it’s the most craziest movie you’ve ever seen. It’s the best movie I’ve ever seen.

And he’s done another one really good named Eega, which means “fly” in Hindi. Maybe not in Hindi actually, because it’s not Bollywood cinema. It’s from the south. And Eega, the pitch is one guy’s in love with a woman — I’m not spoiling; it’s the beginning of the movie — this woman is working for an NGO. And there is a big businessman mafia guy, who wants to have her in his bed. So he tries to seduce her, finds out she’s in love with someone else. He kills the other guy, and the other guy is reincarnated as a fly. And now the thing is, how is he going to protect his lover now? He’s just a fly. The movie is amazing. It’s one of the best movies I’ve seen.

Luisa Rodriguez: That sounds great.

Sébastien Moro: Indian cinema is so creative. I love it so much. I can’t watch American movies anymore. I just watch Indian cinema.

Luisa Rodriguez: Just totally converted. Amazing.

Sébastien Moro: And maybe for some more serious people, you have Rajkumar Hirani, who is another director who’s made very interesting movies, more about the social position of people in India. Like his latest movie is named Dunki, with the most famous Indian actor, Shah Rukh Khan. And it’s about Indian people who want to go to England. And it’s really good because it’s funny, it’s really well played, it’s very emotional — but also it’s hot topics, and he’s treating them with a lot of care, with a lot of sweetness.

So I really love his cinema as well. So if you don’t like blockbusters, Rajkumar Hirani movies are really good. 3 Idiots from the same director is awesome as well.

Luisa Rodriguez: I think I’ve got my plans for the rest of my evening. Thank you so much. My guest today has been Sébastien Moro. It’s been an absolute joy.

Sébastien Moro: Thank you for inviting me. It was a pleasure.

Luisa’s outro [03:09:50]

Luisa Rodriguez: In addition to shaping my views about fish experiences in particular, this interview was yet another conversation that’s forced me to reckon with how ubiquitous sentience and consciousness might be in our world — which I think is one of the most important shifts in my worldview from the last few years.

If you want to learn more about sentience in nonhuman animals, I highly, highly, highly recommend our interview with Meghan Barrett on challenging our assumptions about insects. It’s a long one, but I think it’s super worth it.

All right, The 80,000 Hours Podcast is produced and edited by Keiran Harris.

Audio engineering by Ben Cordell, Milo McGuire, Simon Monsour, and Dominic Armstrong.

Full transcripts and an extensive collection of links to learn more are available on our site, and put together as always by Katy Moore.

Thanks for joining, talk to you again soon.