 All righty, can you hear me all right? So I'm gonna use this screen. Yeah, so thanks so much for the invite. It's a great pleasure. I think it's my first talk at FossApp, which is particularly exciting. So we've taken a bit of a detour over the past few years. So I've got the word evolution in the title. And I hope to convince you all that thinking about not just how the retina works, but how different retinas work and why is instructive. I'm going to try to link a few posters that we've brought in this talk, some of which are today and tomorrow, but some of which were yesterday and the day before, but worry or not, they're online. So if you zonk out during the talk and you want to spend your time productively, nonetheless, do click around those. Okay, so does this work? No, don't worry, I shall use the key. Okay, so don't worry. Let's put it in here. Okay, so if you open in your textbook on how the brain works, and you open the chapter on retina and you try to educate yourself how it works, you might come away with a picture, something like this. We have a huge amount of investment of neuronal real estate and computation in the encoding of space and time. And once you've got those two, you get yourself a beautiful gray scare representation of the visual world, which is basically 95% there. And then we sprinkle in the color, a little bit like spice. And then it's a little bit nicer, but honestly, who cares? So, and I think that's not how it works. I think that's a fundamentally misguided idea and we need to get rid of it. So what I'm hoping to try to convince you in the next few minutes is three things. So in the stew that is red, the coating, which looks like this, the color is not the seasoning, it is the meat and the potatoes. I'm gonna try to convince you that different cone circuits and that in different behaviors, and that we should stop thinking of the ability to use multiple cones just to give us a perception of color, we should think of it as different inputs to the same system, which is designed to drive different kind of circuits and behaviors. And finally, I think by looking across species, we've stumbled across what you might call a principle of how color works. And the idea is that color is encoded in the on circuits and the off circuits serve as a universal achromatic reference. So, but this, let me jump to the question that you've all traveled here to get addressed. Did dinosaurs see color? And yes, they did. Why so sure? Well, here's a bunch of animals, all vertebrates, representing the entire lineage starting from the Cambrian and I've taken the liberty of annotating all the animals that can see color. And the dinosaurs are some of those animals, therefore the dinosaurs saw color. But we can actually go a little bit more sophisticated than this. So, where does color come from? Well, you need more than one photoreceptor. Most of these animals have that. But what you can do with these photoreceptors, and this I think is really elegant, is you can link them across animals. So for example, what we call the green cone in the mouse, or the long wavelength cone in the primate, LWS, is homologous to the single cone, the red single cone in the bird in reptiles, in amphibians, in fish, even in sharks, and down to the lump ray. And that means there's a common ancestor before any of that evolved that also had this red cone. And we can do the same thing for the UV cone. Well, in humans we call it blue, but it is the ancestral UV cone. And it's the same as the one that's in the mouse, it's the same that's in the one in the fish, it's the same as the one in the lump ray, and it's also old. And you might ask, how old? Well, we're thinking about 800 million years. So, Cambrian explosion, first vertebrates, 550. First photoreceptor, 800. First absence, 800. They immediately diversified and they never went back. And the only major change from this ancestral state to what we might have, for example, in the fish these days, is that this R-aged component here turned into the rod and the green photoreceptor before the shark split off, but after the lump ray split off. So the lump rays have a slightly different solution. They're still rods, they're just ever so slightly different. So photoreceptors are very nicely linkable. But that also means that all these clever little animals that do clever processing are built from the same color source. And that color source must have been present in the eyes that didn't have lenses yet and that didn't have a cornea yet. And basically that weren't really used for spatial vision at all. So where does vision come from? Well, you need a photoreceptor. And that's responsive to light from any direction. And then the first thing that happens, other than it gets more sensitive over time, evolutionary time, is you start to screen the photoreceptor on one side. And that makes it directional. And already you've got directional vision. You don't have spatial vision, but you have directional vision. And that's useful. But the next thing that happens is that these photoreceptors start to clump together and you get invaginations. And as soon as you've got something like this, it's basically a lensless eye. It gives you a spatial image. It's a terrible image, but it gives you an image. And you can use it to navigate around the world. So the sort of things that you might be able to use, and oh, I should obviously point out that the color is in here all this time. So you can certainly use it for circadian function because, of course, the world looks different. The wavelengths composition in the world changes as a function of a circadian day. And it's quite obvious that animals have always been able to do that. The other thing you can use it for, vision evolved in the water, not on land, is you can use it as a depth scorch. Because the deeper you go in the water, you lose the short wavelengths. So if you have a circuit that compares, say, UV light with long wavelength light, the stronger it's driven by the long wavelength light, you're deeper in the water. That is critical information if you're an animal trying to make it. And there's actually a beautiful demonstration of how this works without an eye. So this is a platinaris. It's a lava. So think of it as a little worm. It doesn't have an eye. These are not eyes. It has two photoreceptors. One is green and one is UV. They connect it antagonistically. And if you play with this circuit, you can make these animals go up and down in the water column. So this is an animal that has color vision. It uses it for depth, but it doesn't have an eye. So clearly, you can use color to do stuff that you don't need an eye yet. But of course, we do have an eye. So what do you need in order to go from here to an eye? Well, the first thing you need is you need a corner and a lens. So you start to get a focused and under-focused image, but at least it's a bit better. Spatial vision improves. And with this sort of thing, you can navigate around the world. You can stop bumping into stones and things. But what you can also do is you can already use the color circuits, if they exist, to tell distance. And that's a very neat trick that works beautifully underwater. Light gets increasingly monochromatic with distance, just because light gets absorbed in a wavelength dependent manner. And that means that if there is an object that reflects in a sexually dependent manner, so it's colorful, and you see it as colorful, it cannot be far away, because it would be monochromatic. So if something underwater is colorful, it is near you. If it is not colorful, it could still be near you. But at least you can use the property of colorfulness as a measure of distance. And that will work, of course, with a spatially resolved image like this, but it will also work with an incredibly blurry image. So you can work with color to tell distance, even if your spatial resolution is terrible. And then eventually, we make the eye bigger. We generously afford a bit of space to stick a bit of retina in between the lens and the photoreceptors. And we can start to build the retinas that we all study today. So that wasn't the intention. OK, I'm going to talk over this. It's a video that I stole. So this is what the Cambrian shallows might have looked like. And actually, I'm going to stop. This is too loud. So the idea is that in the Cambrian oceans, life exploded. We started with animals that didn't move around much. The sensory system of those animals were probably quite simple. And very suddenly, invertebrates and vertebrates at the same time evolved sophisticated spatial vision, means of fast locomotion, and centralized nervous systems that could match that information, such that complicated behaviors started to become possible. And the first vertebrates, or they're not even vertebrates yet, they're craniates, but they're just on the verge of becoming vertebrates. They would have looked something like this. I pushed it. It may be the common ancestor to invertebrates, but certainly something like this was the common ancestor to vertebrates. We can see in the fossil record how big they were that they had an eye. And we can guess how they moved around. And what's quite clear is that they were small. We're thinking thumbnail-sized. They must have been agile using all those fins. They were certainly prey species, because the invertebrates had a leg up. They were bigger and they were eating these guys. So at the origin of where all the vertebrates come from, running away from nasty things was fundamental to making it into the next evolutionary step. But also, probably these guys were predators, because zooplankton is older than all this stuff. And of course these guys would have been eating the zooplankton, and you have to hunt that. You have to chase it. So probably they were both prey and predator. And as I was pointing out, they were almost certainly also tetachromat using this complement for the receptors. Now, wouldn't it be lovely to be able to study this guy? Well, that's not going to happen. But what we can do is we can look at the sort of characteristics of this animal and identify animals that are alive today that have a lot of common features. And one of these animals is the zebrafish. And the reason zebrafish, it lives in shallow water. It has pretty much the same photoreceptor complement that we think was around at this time. They hunt. They get hunted. It makes for a very nice comparison. So over the last three years in the lab, we've been trying to understand what color vision or the use of cones is all about in these animals. And I will not go into any detail of what we've really done there. Let's give you sort of a bit of a highlight. So here we've got a zebrafish, we've got our four cones. Notice I've ignored the rod. The reason I've ignored the rod is because we're working on baby zebrafish. The rod isn't finished cooking yet at that age. So there's no rod. We don't need to worry about it. However, when they get old, they have rods. And Pakeshi on his poster, which is coming today, has found the rods by polar cell of the fish. So do have a look at that. So first thing we asked is, so what's up with all these four photoreceptors? How do they respond to different wavelengths of light? We did that in this paper here. And the punchline is basically this, that the red photoreceptor is non-opponent. The UV photoreceptor is essentially non-opponent. There's tiny little bit of this. So this would be the long and the short wavelengths photoreceptors of the manals of the homologous one. In the middle two, they are strongly opponent. And they're strongly opponent to the point that they're balanced, that the depolarizing and the hyperpolarizing lobe are equivalent. And that means that if you go up and down in brightness, they don't respond. They need a chromatic stimulus to be driven. So fundamentally, these are color channels. And these two are grayscale channels. What we did, and we put a lot of effort into this, is actually to quantify just how good these are. It turns out that the red one is nearly optimal in order to do grayscale. And the green one is nearly optimal to do the primary axis that you get from nature in color. I'm not going to justify this here. I'm happy to discuss it later. There's a lot of data in this paper and also in this follow-up here. And then that gives you the other two, which gives you another color channel, perhaps to mop up a little bit of extra color that the green can't miss. And then you get this weird extra second grayscale channel. Why would you need two grayscale channels? And this question, I think, was very nicely addressed by Takeshi here. So what he did in this paper is he put a little aquarium in the sun outside our department. It is sometimes sunny in the UK. You have to be, it happens. And he put a camera and he put a red filter in the time. When you do that, you get a picture like this and it looks exactly like it would look if you stick your own head in the water. You see the underwater visual world in grayscale. If you want rocks, you see water reflections, you see bubbles. That's the sort of thing you need if you want to move around this world and not hit things. If you do the same exercise with the UV filter in front of it, it looks completely different. There's a few things to point out. A, you don't see the rocks. The reason you don't see the rocks is because the light scatters so strongly in the water that dominant visual feature is a massive brightness gradient. So Takeshi wears up, that's convenient. But on top of that, you see foreground objects accentuated. You see all these little blobs flying around. Those are zooplankton. There's nothing magic about these zooplankton. It's just they're near and they're small and they have the properties that they scatter UV light better than water. Therefore, they are bright spots that move around. So if you're a little fish interested in eating these things, how convenient to just use the UV channel. It basically does 90% of the feature detection for you. So we wanted to see if that's actually what the fish does. So this is what Nora did here with behavioral experiments. Long story short, you watch the fish, you give it UV light and then yellow light and then UV light and you switch back and forth and you just count when does it attack the food. And that each time does a tick here. And hopefully you can sort of visually pass that it's much better at hunting if you give it UV light and it's terrible at hunting if you don't give it UV light. So that suggests that UV light is being used. It doesn't tell you UV cones are used because it could be a non-UV cone giving that ability. So what she then did is she ablated genetically the UV cones and the behavior is completely toast, okay? So that shows I think really quite directly that the function or one of the key function of the UV photoreceptor in the baby's ever fish is to help it to hunt. But it also shows that we've got a photoreceptor system which is directly responsible for a key behavior that the fish needs in order to survive. We can call this maybe then the breakup tracks is here. It does other things, but this is one of the things it does. So how deep does this go? This is a bunch of cones, great. What's with the rest of the visual system? Well, we do have a lot of data on the reds and then we jump back at this but I want to jump straight into the brain here just to make the point that this is a property that's fairly fundamental I think to how this animal works. So zebrafish of course are popular specifically for the reason that you can look at the brain like this. So here this is just nuclear g-cump in all of the neurons of the zebrafish brain. It's a little microscope trick that Philipp here invented which allows us to get the whole thing. And you get something like 2,000 neurons in the scan like this. So it's quite, you get data quickly. And then what Chiara did and there was a poster in the last previous session and also data blitz, she played a very simple stimulus, bunch of stripes that sometimes move, blackness with a bunch of dots on top and then whiteness with a bunch of black dots on top of it. And she just wanted to see what's the sort of stuff that drives responses in the zebrafish brain. And you can hopefully see here in the video that all of these stimuli drives responses in some of the neurons as you would expect. And just some examples here. So here the white shield motion response so the response when the stripes are moving specifically. Here's an object motion response. Here's an off sale. You get all kinds of stuff. So what Chiara did then here is she color codes in this particular scan whether or not the cell responded to white fields to an off transition or to object motion. So if a cell is white, it responds to a three. If a cell is red, it just responds to the white field motion and so on. And hopefully what you can appreciate is that you get a bit of salt and pepper situation going on. You get quite a lot of red cells here because those are looking directly at the screen. But basically all of these visual properties are represented. And then what Chiara can do is she can go along and genetically kill one cone or the other cone and just see what happens. And I think, I don't want to dwell in this, but I think it's really quite striking just how fundamentally dependent these simple brain responses are on the different cones. So if you kill the red cone, the only thing that remains is object motion. The other two are toast. Well, you get off sales, but not so many. It's basically object motion that's left. If you kill the green cone, everything is still there but the white field motion response is strongly accentuated suggesting that in the native condition the green cones are not there to build it, they have to tune it. And then if you kill the UB cones, the only thing that's left is the off response. So these photoreceptors are directly responsible for these very different key visual encoding properties of the brain. But it goes a little bit further than this. So what we've got here is some data from Aira. There's also a poster today where she just flashes a bunch of colors and here's a fish responding to this bunch of colors. And hopefully you can visually pass that even though it does respond to all of these flashes a little bit, there's a massive shift towards UV. So basically every time the UV one comes on, the whole brain goes and the other colors they just sort of trickle a little bit more. So this is a fish that's UV greedy, it wants UV. But this is not what we see every time. You take the next fish and you might see something like this. This is a fish that's much more nuanced. It responds to all the colors a little bit and in particular, it responds to red. This is, there's nothing magic about these fish. Do you notice fish? You break them, you get them, you stick them under the scope. Sometimes you get a UV fish, sometimes you get a red fish. What's up with that? Yeah? Okay, so there's a beautiful paper that came out a couple of years ago by these guys and what they did is they filmed fish swimming around the dish spontaneously and they would see that sometimes the fish would be like swimming around, be quite keen on exploring the world and then they would stop and start hunting and then they would go, okay, I've hunted enough, let's explore some more. And they would basically switch back between these two kinds of behaviors. And those two behaviors were strongly correlated with switches in the brain state that you could observe at the level of, well, the entire brain response basically. And so what we're currently exploring is whether or not this UV to red thing is not just the fish, this is a UV fish, this is a red fish. Maybe all fish can do both and it's just the stage transition. And while this is not super easy to test because you kind of have to hope for the state transition to happen while you're imaging, most of the time it doesn't, but ever so like sometimes you can see them switch. So I think probably there's something about brain states here. And then of course the obvious interpretation would be that this guy is hungry and this guy is keen to explore the world. And maybe if you wait a bit, he gets hungry and then he switches to UV. So we then emerge at a picture, perhaps that looks a little bit like this. You've got a very early segregation of very specific kinds of channels already at the output of the photoreceptors. They're still present at the level of the brain. They may be linked to object versus white field motion. And those may in turn be directly represented at the level of brain states, perhaps. Okay, so then how do we get there and what about the rest of the eye? So let's jump back and look at bipolar cells. We published this, so I won't dwell on it, but basically the punchline is that bipolar cells do all of the things that the photoreceptors do plus two extra things. So you can see here, the red guy is conserved, the UV guy is conserved, these two are conserved. And we get a broad channel, which is a bit boring, so I'm not gonna talk about it. And we get an extra opponent channel that's missing from the photoreceptors, okay? So if we don't think this is your optimal color axis, this is a little bit of extra color, why would you bother building a third one? And I should point out that if you count them up, this is the one that's numerically dominant. You get a handful of these guys and quite a lot of these guys, right? So there's clearly something that goes beyond color that this thing does. And I think one clue comes from just considering, and Chiara showed this previously, what happens if you don't focus near, but far underwater. And again, you do the red versus UV trick, right? So if you do red, again, you get the gray scale scene of the visual world as you see it, if you stick your own head in the water. In UV, what you get is you get a bright background because the UV light scatters in the water and gives you basically a fog. And anything that's in front of the fog is the silhouette, dark, and anything that's hidden in the fog, for example, these little fish here that are a little bit further away, the contrast is terrible, whereas you can quite easily see them here. So what would happen if you take the opponent channel, where you get a distance channel? It's very nice, right? The bright things are near, the dark things are far, or vice versa if you do the opponent to the other way around. So it's very tempting to speculate, I think, that this guy, even though it gives a little bit of color information per se, it gives a lot of distance information and how neat to have the bipolar cells tell you how far something is away just because it's either UV dominant or long wavelength dominant. So just a little evolutionary detour, how do you turn the fish into a mouse where you kill those two cones? Right? So maybe, maybe, maybe what we have in the mouse here is that distance channel. And maybe, maybe the mouse uses it a little bit in the same way. And of course the mouse doesn't live in the water, but I would encourage you to look at some of the videos from this beautiful paper here from Thomas Euler's lab, which I think highlights a lot of opportunity for using the same spectral contrast to do some very meaningful contrast computations above the water. And of course then, if you turn the mouse into a primate, this becomes your blue-yellow axis and then you add the midgets on top to get your trichromatic vision, which is not really related in a strong way to what the fish originally had here in the middle. Now, what's next? Gungan cells. We did this a while ago. Punchliners, it's kind of like the bipolar cells, at least if you only look at color. If you start looking at time and color as well, it gets a bit complicated. I'm not gonna talk about this, but that seems to be the extra ingredient from the Gungan cells to start to mess with time and color. But then, when it really gets interesting, I think is when you start looking in the brain, because it mostly lose all these crazy opponentcies in the sense that they're not then individual cells, but they start being there in pairs of cells. So, you still see this would be a response, but it doesn't have the yellow lobes, the yellow lobes by itself. Same for these, same for these. So, you start to basically rectify them and flip them up. But what's really interesting is that in the brain, what we see is that they're not randomly distributed across the on and off space. Basically, all the broad ones are off and all the narrow ones are on. And I think that's leading us to what may end up being a fundamental principle of how color works, certainly in the fish and maybe across other species as well. So, I just wanna highlight from a paper here that came out fairly recently, just how dramatic this is. So, what Philip did, Philip battles here at the time, as he just did basically super zoomed out imaging of the zebrafish brain and fleshed a bunch of colors. And hopefully you can see that in the UV range, you almost only get on responses. If you go into the blue-green range, you almost only get off responses. And then if you go back out here, you get on and off responses, ah crap. Okay, this is what that looks like and we can actually link it to the spectrum of natural light. So, the off response, which is the gray curve here, is basically nicely superimposed on a natural light curve, meaning it is the achromatic reference, whereas the on curves look absolutely nothing like it. And that brings us to a picture, I think, that looks a little bit like this. You've got your broad cells. They give you an achromatic reference. They tend to sit roughly across the entire wavelength space that the fish can see with a little bit of a UV suppression. Then you've got a lot of UV stuff sitting on the end and there's a lot of nuance. If you look at Iris poster, you see some are extremely sharp, some are not so sharp. So, there seems to be a little barcode kind of thing going on. They're all on cells. You've got a lot of on cells on the other end. They do something like this as well and then sprinkle it in between. You get the middle ones and they're also all on cells. So, basically you've got the achromatic reference and lots of on, which is narrow, which may just be how it works. And since I'm running out of time and I'm gonna zoom through this rather quickly, Shin Wei here has a poster looking at armor queen cells and he sees that the armor queen cells use the on channel, not the off channel to do spectral processing. So, when you have a bipolar cell, that's spectrally broad, you block armor queen cells, it is still spectrally broad. That's off achromatic modulation. If you've got an on cell, which is not spectrally broad, this is narrow, it goes spectrally broad. So, you've got a spectrally selective on circuit suppression. And what you've got here is, you've got an, it's achromatic under control conditions. You block the armor queen cells. The off circuit is basically still achromatic, but you're gaining here these on things which effectively gives color opponents in the end. So, basically, we've got lots of examples of this. Basically, all of the spectral modulations using armor queen cells use the on channel, the achromatic ones use the off channel. And we can actually see that at the level of the armor queen cells themselves, all of the off armor queen cells are achromatic. And the entire off layer is basically black and white whereas all the color opponent armor queen cells sit in the on layer. So, I think there's something there. And since I'm really out of time, I just wanna zoom through three slides, I promise. Other animals. Okay, these are fish. These are not fish. They all have photoreceptors. They're all interesting. Let's look at them. So, this is a tadpole. I'm pibian, obviously. We have now started to establish the ability to record in the brain of a tadpole that's alive much like we do in a zebrafish. This is a tectomy. It's what it looks like when you zoom in. Here's just a response to a flash of light. So, it kind of works. The tadpoles, Xenopus in this case, have three, actually four photoreceptors for this age three. They've got a blue cone, they've got a red cone and they've got a green rod. Now, what happens if we look at the off response? It's basically rod shaped, making me think that this is the broad achromatic reference. If you look at the on response, you kind of need all three photoreceptors to explain it. And it's quite striking how big this and narrow this red on is. So, it seems that the tadpole is doing something like this as well. And we actually see it in the adult as well. So, I should point out that amphibians are special because they've got two kinds of rods. It's the frogs and the salamanders that do this. And Karola here has a poster on how that might be working. And basically what we see is, again, even for this rod, rod color vision, we've got a broad achromatic off channel, which is just the green rod. Whereas the on channel is more narrow because it seems to be suppressed by the blue rod, which is the second rod. So, even for the red rods, it seems to be doing this. We're looking at chicken. This is Marvin's poster as also today. There's a lot of stuff going on. And here's our dinosaur again because, of course, birds are dinosaurs. So, here's a very simple experiment, multi-electric recording from the chicken retina, 200 ganglion cells, not selected for anything except for the fact that they're responding to light. So, this is a population property. And we're just plotting here the on and the off response through these flashes of light. You can already see that the tuning function. But, if you look only at the off responses, they basically follow your long wave length option, just like the wood and the fish. Whereas, if you look at the on responses, they don't follow that. They are much more complicated and they need all of the photoreceptors to be explained. All right, maybe all except blue, who knows. So, it seems that the birds doing something like this. And I should point out, this is the tip of the ice break, what we see in relation with color and birds. The birds really are a little bit weird with color. And you should really go and look at Marvin's poster. And, even in sharks, we see something like this. And this is particularly striking, I think, to me, because they're supposed to be colorblind. And I just don't think that's true. So, sharks were always thought to be rot only animals or rot plus one cone, but that cone has the same spectral sensitivity as the rot. Therefore, it's colorblind. Well, I don't think that's true, because if we get the off response in the shark, it follows the rot. If we look at the on response, it doesn't follow the rot. It follows the rot-ish, but it sort of needs a long wavelength something in order to achieve the spectral tuning function. And we do have some immuno data that suggests that there is a cone. We don't know what the wavelength selectivity is, but I bet it's red. So, we'll see. And then maybe we have our first non-colorblind shark also using the on channel in order to do this. And with this, I hope I've convinced you that color is important. And that different concepts do underpin different behaviors and that that might be something that's going on even in mammals. And maybe we have a general principle here that the off pathway is broad, the on pathway is narrow, and that seems to just work across species. So with this, I want to thank everyone in the lab that did this, our funders, and of course, you for your attention. Thank you. Is this work? Can you hear me? So that, okay. So, it's just so much fun. Thanks so much. It just was entertaining and really interesting. I just want to draw attention to one concern that I have. And this is about what we're doing as well. And that is how you decide what is something that is constant. And what is something that looks constant, but is actually conversion. Or what circuits get modified because of the particular behaviors and environments of the animals. Now, we can all agree that the photoreceptor pigments, okay, rod and cone phototransection, with a very minor variation, is the same as lamprey as it is in a mammal. You put the rod and cone responses of a lamprey on one side and a frog on the other, and no physiologist could tell the difference. Adaptation is even identical. But then once you get to the bipolar cells, even the bipolar cells, they're on and off, okay. That's very old. But deciding whether this type of bipolar cell and this animal is the same as that type in another, I think that's where things get really sticky. And the one example I would give you is again going back to lamprey, there is a rod bipolar cell lamprey, but it's off. So where does that come from? Okay, so I think we just have to be a little bit careful. And I don't want to single you out because the same problem occurs in harbor search too. It's a general problem in the field. Yeah, so thanks for this. So I completely agree with you. So what I didn't mean to imply is that everything is evolutionary related one to one. What I'm trying to argue is that there's something fundamental about animals evolving to process the information in this way. So for example, if we think about this on versus off thing that we see in the zebrafish, which I think is fairly clean. And then we look at frogs, but they do the same thing using the rods. The zebrafish don't use the rods for this. So it can't be the same circuit, right? It has to be different circuit, yet it functions in a way that's very reminiscent of. So I think what's happening there is conversion evolution that just over and over again, animals are doing this. And I think, I mean, what Rudi was pointing out in the previous talk, I think very nicely points this out, these hue cells, they look like the brain cells of the zebrafish. There's no way they're related, right? They're completely independently evolved. And yet, I don't know where Rudi are, but are these on cells with it? Well, there you go, right? So I don't know, yeah. Thanks Tom, that was super fun. So one of the quotes, I think it was from a paper of yours that just kept coming to my mind during this was the dumber of the animal, the smarter the retina. And on one hand, I wanna ask, how do the more recent studies in other species that you've been looking at make you think about that quote? But the one that really kept bringing it up was that switch between the red cone preference and the UV cone preference in the brain during the exploratory versus hunting behavior. That seems really smart to be able to pick the channel that you want depending on your biologic state. Yeah, so the two questions there. So the one about the dominance of the animal. So when we, I've got a fun answer to that. So when you look at the complexity of the retina in terms of anatomy, so we're just counting up there and seeing how big they are, you can divide vertebrates into two piles, the simple ones and the complicated ones. The complicated ones are all the ones that are not mammals or not sharks or not lump rays, right? So by that logic, the smartness of the primate should be linked to the smartness of the shark, right? So I think that argument breaks the smartness of the retina. I think that there's more factors to play. Then the, about brain state switching. Yes, I think it's about economizing resources. What we don't yet know is if it's the same neurons that switch between UV and red or if it's just different neurons coming on. I think that would be key. But it could just be economizing resources. If you've only got a small brain, you gotta maybe can't do everything at once. So we are actually in the coffee break time now. So maybe you can sort out your questions in the break. Okay.