 And hello everyone and welcome for another online talk. I know that this talk is taking place a bit earlier than usual. And I know that it's a middle of the night for our colleagues on the American continent. But I guess today is literally on the other side of the world. So I'm sure they will watch our talk as a podcast later today. As usual, here's a quick reminder that autos are part of the worldwide neuro-initiative. I bet that you know my gimmick now. So just go on the website, look for topics you're interested in, find the topics you like and do share the podcast brand. That being said, it is my great pleasure today to receive Justin Marshall from the University of Queensland. Justin obtained his PhD in neurobiology in 1991 from our very own University of Sussex in Brighton, and then it was already working with MontyStream. After some time in Sussex, he moved in Australia where he assumed a position in the Vision and Touch and Hearing Centre at the University of Queensland, and he then joined the Queensland Brighton Institute in 2010. Justin's work came to understand how animals perceive their environment. By taking an approach to sensory system which is based around ecology, but also includes physiology, anatomy, behavioral, and neurointegration, he hopes to decode languages such as polar and polarisation. There are really many things to say about Justin's work. For me, it's time to use the accuracy with base, with collaboration with David Tamburo, with implication, with the Coral Watch programme, but I think the best remain to hear any of it directly from him. So hello Justin, how's life in Brisbane? Yeah, pretty good. It's beer time now. So you're eating into my beer time. I'm deeply offended. I have to give my talk at this time, but it's a pleasure to be here and thank you very much for asking me along. And yeah, it's going to be fun. The stage is yours whenever you want. Okay, I will share my screen. How's that? Just to check. It's perfect. You can go ahead. It's perfect. All right, very good. Yeah, it's great to be talking about what I guess are my my favourite animals, which are mantis shrimps or stomatopods and the talk is entitled from the sublime to the stomatopod, you know, sublime to the ridiculous. And it's the story which is unfolded. And I think it's worth telling it again because every time I talk to people, they're obviously a bit confused about where we're at and where we came from and how we got there. So I just hope that this is going to relate that in a way which which makes sense. So we see your presenter site, not your full presentation. Okay, so I'm going to need to swap display again. It's all right just realizing it. How's that? No, you can go. Is that better? Yep. Yep, cool. So, as usual with these things, we start off with thanks. And I'd like to thank the people that have funded this including the Australian Research Council and a couple of US forces and actually previously in the UK, BBSRC and other English funding agencies. And right up front, I would like to acknowledge my mentors, Mike Land, who is, as you know, sadly no longer with us. Here he is looking into a small aquarium full of mantis shrimps that I made for him or not. And Tom Cronin, who is the person I then went to work with in America, and much of the work that I'll be describing has been done with Tom. Can I just ask if you can see my cursor there? Going around on Tom's face, Maxime, can you see that just nod? Yeah. Yeah, good. And then a very important acknowledgement at this point is also Roy Caldwell, who is this gentleman wading through the shallows on Lizard Island. And Roy has taken many of the photographs that you'll see in this talk. He's also the person that's taught us the most about stomatopod behavior and what they actually do in the real world, you know, we poke out their eyes and work on their visual system and do all that stuff. Roy is one of the few people in the world that actually understands how mantis shrimps think. So let's blunder on and for those of you that read my abstract, you may be wondering where this quotation came from. It's actually the first few lines out of Moby Dick by Herman Melville. And here are the lines in their reality at the top. And it's basically to point out that I realized that actually doing this sort of thing would have been much easier than studying mantis shrimp vision as you're about to find out. So this is the plan to talk plan A may turn into plan B a bit later, but there's quite a bit to get through. So let's go on. And for those of you that don't know me, here's a quick introduction to the lab and what we do. So the lab. We're interested not just in mantis shrimps but also in all sorts of animals. So brief fish, alasma branks birds, paradise turtles cephalopods and deep sea fish as well. I also have this citizen science group called Carl watch which is all about trying to save the reef. And then quite recently we've started working with engineers and the train represents engineers because that's what engineers love the best. So what do we do we're really interested in animal color, how animals see their color these are some examples of reef fish colors just to show you some pretty colors. We're interested in eyeballs of course that's why we're here. We're all interested in visual systems of different animals. And we're interested in ecology so how these animals use their visual systems in their natural habitat. And the habitat that we work on mostly coral reef systems such as this one that you see there. So just a few more thanks to the people that actually did the real work. There's a few teams involved here this team anatomy and to some extent behavior as well so Hannah, Rachel and Nick Strauss felt in his team have helped in recent years, put together what lies behind the mantis shrimp retina. And Hannah in particular has been very instrumental in some of the behavioral work. Yeah, here again young has been very good within circular polarization. And then here is the engineering team again, linked with their train. Sam Powell, who is still in the lab here, and a very talented engineer that also understands biologists which takes a lot and Victor his previous PhD supervisor, who helped put the cameras together that we'll talk about later. More recently than that and again you're going to see some work from these two PhD students Amy who's just handed in Amy streets and Jing Wang who's still going. Amy has been looking in even more detail into the stomatopod retina using high resolution. And Jing Wang has been looking at the electrophysiology behind the retina so she's been making sharp electrode recordings from the laminar. Amy has also been frustrated by mantis shrimp behavior. So that's the stomatopod crew. And let's go on to just remind ourselves hopefully remind you what stomatopods really are so for those of you that haven't really thought about mantis shrimps. I'm going to go through this and I'm going to refer you back to my box talk of a few weeks ago, which was so beautiful and he really put together very nicely not just mantis shrimp vision but also worm vision, and how worms see the world. And I'd be I'll be borrowing some of my slides in a bit. But you know what are mantis shrimps what are stomatopods, whether these animals with these fearsome raptorial appendages at the front. They strike out at animals that they want to eat they've either got an array of spines, which usually hit their target this fish doesn't even know it's dead yet. And basically stomatopods love light and killing stuff that's how you can summarize what they do. Those that don't have stabbing appendages have these enlarged clubs at the front there. These clubs can be used to beat their way into a variety of hard bodied prey, or they also fight with each other over resources such as holes and cavities cavities on the reef. So they're really nasty animals. And here are some stomatopods either beating up glaring octopus. A slow down strike here and a crab that's getting it in the neck as well. Here are some statistics that I won't read out about how fast they can strike. It is the fastest strike in the animal kingdom. And this comic strip picked up not only on their nastiness but also described very accurately their color vision system. And what I like about the oatmeal in this comic strip is that I can. I can tell my students I know about I work on the harbinger of blood so rainbow so that's what I do for a living. And the reason I'm interested in mantis shrimps is not because of their raptorial strike, but because of this thing which is that I, and in particular this mid band region of the compound I, which clearly does something very different. It has different sized omitidia. So this mid band is doing something which is unlike the other parts of the eye, the dorsal and ventral hemispheres or the peripheral parts of the stomata pot I. So in the beginning. Let's think in 1988 I first described this system by cutting into the mantis rebinding anatomical observation. This was rapidly followed with a paper with Tom Cronin, which made the front cover of nature, in which we use Tom's micro spectrophotometer to describe at least 10 spectral types of photoreceptor within the eye. And in the end of the series, we thought that they had probably exquisitely fine color resolution, based on the fact that their spectral sensitivities were very sharp. And here you can see some spectral resolution some spectral sensitivities in their eye, as plotted by Daniel Sorio, who joined us on this paper in 97 to suggest that this sort of exquisite tuning might also be used for color constancy. So they had a set of dichromases within their retina, which they could use to analyze color in extreme detail. So that was kind of a normal set of ways of doing color vision by comparing the output of different photoreceptors, but in a set of dichromases, rather than some sort of octagonal or dodecohedral color space so it seemed vaguely sensible. In 1999, we added more, working with Johannes Oberwinkler, we added more UV photoreceptors in this area of the eye. So one of the confusing things about mantis rims is you hear all sorts of numbers. So it was at least 10 with that paper with Tom, and then you'll hear 16 photoreceptor types. Well, that's correct, but that means there's 16 anatomically distinct. You'll hear 12, but that's just to do with color vision. The real number is probably there are 20 different channels of information coming into this eye. And this sparked an idea, and this is actually a terrible pun, because later on we'll be talking about how the mantis rim eye could in fact possibly be thought of as an ear. But with 20 inputting, channels of input, 12 of those to do with color, this was a very exciting beginning. And now to go on to borrow some of my box slides because he describes very well the diversity of these animals in terms of their color and their size. So there's all sorts of different sizes from a fingernail size to, you know, about the size of, you know, almost a meter long, some mantis rims. So they come in all sizes or colors. And clearly they're doing something very different because, as already mentioned, when you delve into their color vision system, here are the spectral sensitivities of mantis shrimps. Here are the spectral sensitivities of man. So they're very different visual systems, or it has a very different visual system to the one that we have. The detail is to where these different photoreceptors lie. So within those hemisphere regions, the dorsal and ventral hemispheres or peripheral regions, you find two different spectral sensitivities, a green one and a UV one. And actually this is very much like other crustaceans, such as crabs. And this is the mid band and this is where the spectral diversity lies because within the first four rows of those omitidia, you find those 12 different spectral types of photoreceptor. So that's the color vision bit. Then looking at the rest of the eye of the rest of the mid band. So you find you get into the polarization photoreceptors, and it's there that you find not just linear polarization in two different spectral bands in green and in UV. But rows five and six in this mid band of six rows conduct circular polarization vision, which is, as far as we know, completely unique in the animal kingdom. The number is 20 mantis shrimps have 20 different channels of input, coming into their tiny little brains and they have to sort that out. As I mentioned before, this is clearly very different paths to color vision, because for humans, we have this opponents system where our cones compare their output in a red green and yellow blue sort of way. And this opponent processing tells our brain, what sort of color, what sort of colors we're looking at. So we have a variety of colors across the mantis shrimp system. And the number of them suggested to us that they may be doing color vision very differently. And we came up with a sort of parallel processing idea in which each one of these individual photoreceptors acts like a bin which is stimulated, and the animal looks for a pattern of excitation across the spectrum. You get blue stuff, you get this sort of pattern. When it's looking at red stuff, you get this sort of pattern. And this is known as the parallel processing system, or also the spectral barcode scanner, as Mike put it in his talk back in April. So is the stomata body acting as some sort of bar scanning device, looking for colors in the environment. So does this now exclude the initial ideas of opponent processing that I talked about. So here as you can see if you read my abstract, I'm kind of swinging from what our initial expectations were to where we're sort of sitting now, and going backwards and forwards in history, hopefully not confusing you too much. And hopefully by the end of this talk you'll understand where we're at. So is it a spectral barcode scanner. So we went on to look at the colors which mantis shrimp use, and they use these very specific colors. And his talk went on to a paper that we're in the process of writing, where we look at this barcode scan device and pull it across difference to matter pod colors to see what these sorts of colors are probably evolved for. And as you might expect it seems to be all about sex and sexual selection. These animals are probably looking at each other and using their very advanced color vision system to detect these particular color spots and particular color areas on each other to make decisions about mating and fighting and territoriality. And this is a very nice. I scanning video that Mike also put together. And what it shows you is the remarkable eye movements that these animals make, and this supports the idea of barcode, because you can see the eyes are moving independently and slowly. And in fact, when you are just pause that when you look at the eye and look at it in cross section and look at the optics of the eye. It's in fact looking at you three times. So each one of these eyes is looking at whatever it's looking at three times. And that means that within a narrow strip of space. It's examining all of the color photoreceptors that we've seen, and also polarization photoreceptors, as I'll emphasize in a bit further in the talk. But it's a mistake to think of this compound eye as one that just looks out into the world in all different directions. And actually, the peripheral regions of the peripheral region or the peripheral regions of the hemispheres do look out into all directions. So each eye does look in 180 degrees, but 70% of the eye, including a lot of the photoreceptors within the hemispheres, and that you can see within these so called pseudo pupils. So the black part of this eye is looking at you. 70% of the eye is confined to a very narrow strip in space. And that's why the animal has to make these very slow majestic eye movements which we described with Mike land a while ago. This also led to a five year prawns in space project because it turns out that the only other thing that examines the world in such spectral detail using a scanning system are in fact satellites. So satellites use a push broom photo diode array sensor, and that it appears was pre evolved by stomata pods 400 million years ago. So at that point I began working with engineers, not just on trains and train sets but on on trying to redesign satellite sensors to look at the world in the same way as mantis shrimps. So the spectral barcode scanner is lent evidence by the fact that these eyes do have to scan across things, as if they were literally scanning a barcode in the shops in the supermarket and making some sort of decision about what to do. And again just to emphasize but if they're doing this, does it also, are we now saying the opponent processing model that I first introduced is not there. Are we arguing against opponent processing. So let's look in a bit more detail about know the barcode idea in this bar scanning idea. Again, looking back to some original work in 1996, because we conducted then a behavioral experiment in which we asked these animals to look at different colors in a very much. They were called von Frisch be sort of behavior where the animal has to distinguish a color from neutral grays. The animal looks at these colors as they're presented you lift the barrier, it looks at the color, and if it then goes out and interacts with the color. And in fact these animals. We trained to pick up cubes and smash into them where they find food in the color cube and no food in the great cubes. Very rapidly, the animals learned that they would. They get food out of the red cubes, the green cubes and the yellow cubes. But they were really not very good at what we gave them in terms of blue, which was a rather pale blue for some reason. It's just how we tended to color the cube it was a pale blue and they couldn't do that. So their behavioral choice there was identical with chance. Which implied that they were seeing this blue as a sort of shade of gray, which was, you know, weird. And again it kind of lent evidence to the fact that because they're confusing blue with gray. When you look at the way in which the eye receives those different, the blue color and the neutral grays and each one of these bars represents the photon catch for different parts of the eye. And you can see that the blue appears to us to matter but pretty much as if it was a shade of gray. And when the eye scans across this, you know, behavioral choice, maybe it's setting up some sort of parallel processing, where, you know, it's just looking for this pattern of excitation across the retina. And because by chance we've chosen a blue that gives the same pattern of excitation as the neutral grays, it confused blue, which is obvious to us with the grace. And again this was sort of evidence for a parallel processing scanning system, which makes mistakes with low saturation colors, and I'll say more about that later so stand by on that one as well. Again this argues for some sort of serial processing basket barcode scanning, very different to opponent processing idea, except within the same paper. We also argued, because this was an earlier paper in 96, but these animals were still doing opponent processing. Now I'm sorry, I'm just going to have a drink here and gather my thoughts because I'm sure you're doing the same. Because I've been arguing for opponent processing, and then barcode scanning, and now I'm going back to opponent processing again. But the point is that when we examined the potential for opponent processing in the retina here of mantis rinse. And you look at the signal difference within each of these rows that a brain might compute. It turned out that the signal difference for this light blue color again was minimal for each of the mid band rows for this color, but it was substantial for the other colors. So this tended to argue that maybe there was some sort of opponent processing going on. And that's why with this minimal excitation in terms of a difference set up by the color vision rows in the mid band. We're just making a mistake with blue. So basically this paper was very much sitting on the fence it couldn't decide whether we're talking barcode scanning, or whether we're talking some sort of opponent processing. And here you can see me and Tom in the early days discussing this point. Another point to point out, and we'll come to this later is that this paper is probably wrong in some parts. But it's swiftly on and back to this idea, because another thing you may notice if you go back in the literature to about 1999 is this idea of is the stomatopod I in fact an ear. And again this goes back to the idea that the retina is picking up a pattern of excitation across the wavelength continuum, or in fact a frequency continuum. And that's exactly how the human ear with its cochlear works because the cochlear which is this snail like device in the ear receives frequency from low to high along a continuum, and sets up a pattern of excitation across that. And maybe that can be equated to the way in which mantis shrimps are seeing different colors. And again back in the literature when you see is a stomatopod visual system like an ear that's why we were thinking in those terms. The real piece of evidence that really did surprise us. I'll come a bit later in 2014 with Hannah toan and others short to for example, Martin how when Hannah went off and did a bunch of experiments which are called Delta Lambda or wavelength discrimination experiments with these animals because she trained the mantis shrimps to identify food with a color and then using in fact fiber optics you'd introduce a distractor color. And as you bring those colors closer together the mantis shrimps make confusions. And this has been done with a number of animals including honey bees humans and butterflies and what we expected with mantis shrimps was very fine discrimination. So the closer to the x axis, the data here, the better the discrimination of the animal. So you can see in humans for instance, we're pretty good our discrimination level gets down to about one nanometer. Honeybee curve here. In fact, what we found is this red curve. So mantis shrimps, it turned out have the worst spectral discrimination of any animal so far exam. Again this really argued for some sort of very different form of spectral processing and this sort of barcode scanner device where when the mantis shrimp looks at a blue fish. And this is a blue reflection or a red fish red reflection. It's seeing a pattern of stimulation such as those that you see here. So it's not combining individual spectral sensitivities in any sort of opponent processing. So this poor spectral resolution again argued against opponent processing because of course with opponent processing, you get very fine spectral discriminations. Don't you. Maybe, maybe not. And now let's have a quick aside on polarization, because this also argues for spectral scanning for the barcode hypothesis, and it's worth putting a bit more in on where the polarization photo receptors lie. So mantis shrimps along with many other crustaceans see polarized light. Otherwise like just to remind you can be broken down light can be broken down into waves which may be vertical or horizontal so these are the e vectors the electric vectors. And they can also be circular if they're out of phase and mantis shrimps, of course can see this as well. We know that light in the environment may be scattered or reflected to give us linear polarization signals which you can see here photograph with different polarization filters. And most famously among animals this is used by for instance desert ants to navigate home by locusts also to navigate polarization is used by done Beatles to decide which direction to roll their ball of dung in. And most lately Ricky Patel and in Tom's lab is also shown that mantis shrimps use the polarization pattern in the sky to navigate away from their home borough, which is really nice. And here's animals using polarization to navigate with. It also turns out that mantis shrimps use polarization to talk to each other so they use colors and polarization signals. And here you can see polarization signals on the Europe and these antenna scales in these animals. And that's accentuated by this video that picks out the polarization parts and makes them flash. And it's also used to mantis shrimps talking to each other literally in polarization. So polarization and mantis shrimps is used for communication. Sometimes with those scales and your pods with this little mantis shrimp have to squilla, they have these blue maxillipeds, which reflect a lot in polarization. And this is the engineers work here where they've designed a camera to look at polarization. Into something that we can see. So red here is highly polarized. So here is the same animal in its borough. And you can see that these little blue parts are also highly polarized using a camera that in fact was inspired by mantis shrimp visual system in the first place. So Sam Powell and Victor Earth put this camera together. And we're now using it to look at mantis shrimps using their own visual system, if you like. I might go into the details, but if you look at it, it does look a bit like a train you could imagine. Some smoke coming out there and sometimes smoke does come out so it is very much like a train. But let's just remind ourselves where all this lies again within the mantis shrimp by so the color photo receptors are within the first four rows of the mid band. The linear polarization receptors are probably within the dorsal and ventral hemispheres set up by these photo receptors having orthogonal or opponent microvilli within the photo receptor which received polarized light in two different directions. The circular polarization most confusingly is done in row five and six of the mid band. And there's all sorts of stories in the background there about quarter wave retarders have a look at Nick Roberts papers and the way in which a spectrally flat quarter wave retarder is set up within the anatomy of the eye of these rows. So these animals are really concerned with seeing all of the polarization that's out there as well as all of the color that's out there in this really bizarre sort of way. Luckily, at one point in one of my talks I was saying well mantis shrimps must be looking for Beatles because Beatles were the only thing in nature that we knew that reflected circular polarized light. And here is a cop chaffer that reflects when you look at it through the circular polarized filters basically the same glasses that you put on when you go to a 3D cinema, go out and look for cop chaffers. And you can see there's a difference. So, luckily when we went back and looked at mantis shrimps it turned out that some mantis shrimps also reflect circular polarization. And it tends to be sometimes the males that do this. So again this is something to do with sex and sexual communication, communication between mantis shrimps in the same way as the colors. These eyes are looking for polarization in many different directions in fact they look in horizontal vertical 4545 and left and right circular, and here you can see the different parts of the eyes that do that. In fact, if you're a physicist if you get into physics this is providing all the necessary stokes vectors, which are needed to analyze polarized light in the environment in excruciating detail. So in the same way as the eye seems to be set up to analyze all of the wavelengths and colors in excruciating detail. The eye is also looking at polarization space in excruciating detail. And we've got hypotheses flying around about whether this represents you know nature's ellipsometer go and look up what that is. I had to learn a lot about physics and as Mike Land put it actually, I really didn't think I'd need to know much about circular polarization until mantis shrimps came along. So this degree of over description of polarization is parallel by the color system, or so it seems. And just to throw another interesting wrinkle into the works. These eyes are not always found in this position. They can rotate and they can do that continuously. As we saw with my box nice video, and that of course changes the angle of the polarization photo receptors within the eye relative to the outside work world. And then look at the work of Elsa daily and the group in Bristol that describe this very nicely. So back to summarize to matter of visual systems. They seem to want to do everything they want to do polarization and color. And this polarization and color is contained within this narrow strip in space. So they're waving the strip across space to analyze what they're looking at, not just for color. Not just for black and white but also in every single possible conceivable polarization input. Again, arguing for some sort of barcode scanner, not just for the spectrum now but also for polarization. Let's just then exclude, you know, opponent processing again, we're seesawing backwards and forwards, is it excluding opponent processing for polarization. We're not going to go to a part of the talk which is going to argue against barcode and against the idea of everything that I've just spent the last 20 minutes building up. At the beginning we talked about possible opponents. We built the case for yes there is some sort of barcode scanning thing going on. We're now going to beat that down. Just like a mantis shrimp. And we'll have an aside into anatomy because of course one of the things it's worth looking at when you're trying to determine what the mantis shrimp sees is not just what lies in the retina and here you can see the retinas of mantis shrimp. But also of course what lies within the brain, how are they processing all these signals. So we've been looking at that hard over the last few years. And here you can see retina. And as is found in many arthropods lamina medulla lobula. And other parts of the outer brain of the mantis shrimp including mushroom bodies and ready for bodies and all sorts of things, and then the central brain. So we've been looking at the neuro anatomy of what lies beneath the retina with the help of Nick Strausfeld, who of course is the king of all this sort of thing. And here is a nice figure from one of the papers by Hannah Toan and Rachel Templin helped with this early on. So we're looking closely not just at the retina. But what happens beneath the retina in the lamina, the medulla and the lobula. And one of the things we found early on is that thankfully the mid band photoreceptors are retina topic so there's a special bit of the lamina that deals with mid band photoreceptors. There's also a special bit of the medulla that deals with mid band photoreceptors. And there's a special bit in the lobula, although it gets more complicated that deal with mid band photoreceptors. So apparently, maybe exclusively or not because in fact, at both of these two stages you begin to get cross tool. So, although these lamina cartridges appear a bit bigger and they're different they're not. They kept separate from those in the hemispheres in the peripheral parts of the eye. When you look at them closely they're not that much different in terms of the cell makeup. They're kept separate. And that argues for some sort of separate or parallel processing in all of these systems. So, going back to some old retinal anatomy again and this is back to 1991. Again the retinal anatomy suggests that there's a ponency going on. So now we're arguing against the serial parallel processing thing. Back to this original paper on the anatomy and in fact, this paper on polarization argued that a very good reason these rows were actual opponent pairs. And that's because it turns out that when you look at the peripheral parts of the mantis shrimp which as I mentioned are very much like any other crustacean. These photoreceptors from the retina project through to the lamina and they project through to what are called EPL 1 and 2 layers. And these are different layers which contain the photoreceptors that transmit polarized light in two different directions. And these photoreceptors, you don't need to know why but they're called Rhabdoms 145 and 236 and 7. When you look at these tiers within the mantis shrimp eye because the photoreceptors have become secondarily tiered into regions that are different sensitive to different wavelengths as we've tried to show with these colors. Those different tiers are precisely the same cells as you find here. So without rewiring what's going on in the lamina at all. You go from polarization, a potency to spectral a potency within each one of these mid band units. So we're arguing again that in fact there is a potency going on within each one of these rows. Here's another way of putting it. These are the sorts of diagrams that you see when trying to understand polarized light in any crustacean. I will show you that the photoreceptors have tears of microvilli that are oriented orthogonally to each other. And these are these impinge on an internet on which receives an opponent processing signal from the two photo receptor types. The opponent processing is transmuted if you like or translated into color processing by the way in which the stomatopod row one to four photoreceptors are set out in the retina. So remember back to this paper where we're arguing that opponent processing was occurring when the animal was looking at the different colored cubes to feed from because the blue cube when it was looked at with this opponent processing system did not give a very big differential signal to the opponent processing system. And therefore this argued for opponent processing because the signal was small, it couldn't see the blue. But when it looked at red, green and yellow, the signals were large in the difference between these photoreceptors. Therefore with thumbs up for opponent processing again. So is bar coding now bar humbug. The swings and roundabouts of working with mantis shrimps. Maybe now you can begin to realize why I wish I just set off in search of a white whale to harpoon because it would have been much easier. Anyway, blundering on and getting towards the end of the talk. So let's look at some more against the barcode hypothesis and look at some electrophysiology because as well as describing the anatomy of what lies within the lamina of mantis shrimp eye beneath the retina as performed by an example and know what's going on in the lamina in terms of electrophysiology can be put electrodes in these cells and record from them. And that's exactly what we have done. We've supported the idea of of these cells within the lamina, having some sort of opponent processing has just outlined actually a number of different methods so Jim Wayne Wang who is still a PhD student in the lab has been doing this exquisite very difficult intracellular electrophysiology within the lamina and Amy streets has been looking at the lamina cartridges in much higher detail using high resolution microscopy. So what Amy's been doing is this three view or volume scope where you take serial sections off the top of an end block, and you reconstruct the cells that lie beneath that. And you can reconstruct the way in which the cells go from the retina all the way through to the lamina, either a single cell, or many cells. So here you can see the eye across section of the eye. And these are the EM sections in serial, and you can see that. Well, it's not simple is it. It's not even a fly which does a nice sort of rotation thing. It's a dog's breakfast. And Amy has worked very hard to work out what's going on here. And in fact Rachel Temple and started this work a few years ago. And we're beginning to get an idea of what's going on. Within each of the mid band rose, you know, is there something special going on that might support very different color processing. So other photoreceptor terminals at the lamina cartridge, no joining up with different monopolar cells. We're doing this as I speak. But we're definitely really finding that there's nothing very different. They're bigger. As I mentioned the lamina cartridges are bigger. But when you look even at very high resolution and it's synaptic connections, there's not that much different between these lamina cartridge types and those of other crustaceans. And again, here's some of Amy's nice work. We're looking at the different way in which the laminate monopolar cells which take the signal from the photoreceptor terminals are laid out in different EPL one and EPL two. They are Verizon these different layers. And it's likely that they're setting up some sort of opponent processing. Amy also delved into behavior, and she is regretting it because there's a mantis shrimp showing the sort of behavior that it does. We train the animals to, as I mentioned at the beginning, go out and collect the colored, in this case, a small cube with food on board and not the neutral grades. And the idea of going back to this experiment was for Amy to show that the animals were bad at low resolution colors. So if you remember back to those original behavioral experiments from 1996, the mantis shrimps were confusing pale blue. She redid this experiment, and I won't go into the details, but basically what you see here is behavior for saturated red, unsaturated, saturated orange, unsaturated green and blue. And we really hoped that they would be bad at the unsaturated colors, because that would argue for some sort of different processing of color. But no mantis shrimps being mantis shrimps. They were both good at high and low saturation colors. So this went against those initial experiments. So in the previous 1996 behavior, what was going on? Basically what we think is that those results at the time were probably wrong. I'm an old scientist now and I can admit that I often get things wrong. Actually, as Mike Land used to say, that's a good idea just in one of my papers in nature. I published the paper the next year in which I showed how the first paper was wrong. And actually we're sort of doing that thing now, not publishing in nature because nature doesn't care anymore. But we are saying that that result was probably wrong and a result at the time of keeping the animals in captivity under room illumination. And we learned later with mantis shrimps and indeed other animals. In fact, when you do this, their color vision systems can be very plastic. They change and mantis shrimps kept in room illumination, which is how we did ignorantly without really thinking about it back in 1996. We're probably just losing their color vision sense. So it's likely that when you do this, the animals get better get worse. And in fact that's what this result shows here. The animal kept the animals in room illumination and their performance got worse over time. So it turns out that we were probably wrong. It's also slightly possible that because we're using a different species, there are different interspecies differences in the way in which they process color. But really, we think that we probably got that experiment wrong. And that the animals are equally good, whatever color you give them. Are you confused? You know, I am. What's going on? So what does this mean for barcode versus oponency? There is an answer at the end. So hold on to your hats and we'll get there. Before we get there, we're going to look at the electrophysiology, this exquisite work that Jing Wen is doing with a setup that was actually originally designed by Short Chew, who's worked in Tom's lab and mine. Short is a wizard at everything. And he set up this E-Phys thing here where we have a rotating polarizer, an eyeball or an animal in a cube. We can shine light at it, a sharp electrode with all the sharp electrode devices and things to record from the photoreceptors as we shine light of different polarization and color. And this is what we find. So what we were sort of hoping, again, was that when we started to record from the monopolar cells that received the input from the different mantis shrimp parts of the retina, that we would not find oponency. We would support the idea of some sort of different color processing, possibly some sort of barcode-y scanning thing. These are very preliminary results, but so far Jing Wen has found the ponency in the lamina. Going back to the original idea that the mantis shrimp retina is in fact set up for oponency. Jing Wen even found it not just in the lamina, the cartridge cells as she records from those, but also within the retina. So when you record from photoreceptor terminals within the lamina, you also find color oponency. And this has been found in butterflies. This is as if these little pairs within the rows one to four of the stomata pod mid band are indeed set up for oponency. So have we finally killed barcoding? Is it dead? Well, no. As the famous Monty Python parrot sketch put it, it's really just resting and we'll get to that shortly. And we'll get to that by thinking in these terms. Whether we're talking about tacos and the kind of irritating little Mexican kid that said, okay, no, no stars. Or more seriously in terms of Hispanic bisexuals, why not both? You know, why not have both? So we're going to go back now again to finish off with some more behavior and some conclusions in which we'll go back to this experiment that Hannah ran. And if you remember, this is the Delta Lambda experiment where we were expecting stomata pods to have this exquisite discrimination of colors across the spectrum. And the fact that they didn't and they actually have this sort of discrimination argued for different color processing and not perhaps opponent processing. But, you know, can you have poor spectral discrimination? And with this color set of photo receptors, some sort of opponent processing. And if you look at where in the spectrum for this species, its best points of dichromatic potential opponent processing lie, and you overlay that on this behavioral data of sort of clear away the stuff in the background. What you see is there's actually a very good match between where the behavioral data says there is the best discrimination, even though it's very poor. And we're up around 20 or 15 nanometers of resolution here, whereas humans and butterflies are sub five nanometers resolution. But at least what we found is that there is a match in these spectral binning opponents is that lie within the stomata pod spectrum, at least from 400 to 700 nanometers. So when these animals are looking at things in the world they need to look at. When they're looking at different spectrum, they may well be setting up some sort of pattern of excitation across the spectrum. But what we're saying now is that that pattern of excitation is not even as fine as we thought it was. And in fact, it's rather even more calls. So it's not eight different little bits of pattern across the spectrum, but it's only four. There are four spectral bins for areas of opponents. But that that sets up a very poor spectral discrimination system. So you can have opponent processing that's not spectrally acute, even though you've got very sharp spectral sensitivities lying next door to each other, which is what tipped us off in that direction in the first place. And you can also have a parallel or bar scanning system that's not just parallel but contains within it underneath the retina, non opponent processing or non opponent processing. Sorry, that should be that contains within it opponent processing. You see how confused, even I am. So just to finish off. The last part of my abstract if you read it was another quote from Darwin in fact on what he thought of beauty, because Darwin pointed out that animals that are beautiful may have been rendered beautiful for beauty's sake. He was not saying that goes against natural selection, but that it argues for sexual selection, and that there's a different form of selection where color signals that become very exaggerated are there. In order for animals to make choices about each other. So I'm making the same claim for mantis shrimps. And the work that we're going to continue with Mike buck and Tom Cronin is going to try and prove that when we look at how mantis shrimps look at their own color signals. But in both these extreme adaptations whether it be driving towards color or vision, it seems to be driven by sexual selection. So I'm going to go back to Mike's talk in April for more explanation of that. Finally, I read a book recently by a nuclear physicist Frank Orchek who picked up on mantis shrimps and in fact picked up on Mike's very nice graphics of how mantis shrimp see color compared to humans. And Frank also thinks that stomatopods process color very differently. And of course being a Nobel Prize winner in physics, he's far more clever than what we are, but it's nice to have this sort of corroboration from people that believe the same as we do. So finally many thanks for listening, and back to the cartoon and mantis shrimps the future is indeed bright and dark. And of course it's very beautiful. Thanks for your attention. Thank you, Justin. That was a very, very nice talk. Should I stop sharing at this point or continue? Up to you. Here we are. Thanks for that. Were you confused? I was confused giving it. I'm confused. The first question would like to ask you is, we talked a lot about mantis shrimp, but do you know if there are stomatopods that may have a similar rainbow not rainbow detection or I mean the main question I have is, what do you think what kind of evolutionary drive and for such species to develop such a detection barcode rainbow? Kind of, yeah, that was really the point of the, if you like that last bit in which I was using Darwin's or misquoting Darwin, because he's suggesting there that the beautiful colors in animals like peacocks like butterflies and maybe like stomatopods get driven by evolution to endpoints that seem ridiculous. But it just seemed beautiful and just seemed wow. How I mean how can that just possibly be worth it. And there's all sorts of arguments in there about the handicap hypothesis and carrying heavy tales of feathers that all that stuff but you come back at the end of the day as to why can't a peacock just be red and green. And it's still beautiful. Why does it need to do all that stuff. And it probably comes down to this thing of sexual selection which tends to get pushed aside in favor of natural selection. And it's sort of the same with mantis shrimp I think at some point 400 million years ago they decided to do their color vision in a particular way. And they've been driven towards complexity or apparent complexity by the need to solve their relatively simple task, which is to spot other mantis shrimp. And they've been driven to that complexity in the visual system in the same way as animals are driven to complexity and color by the same process of evolution which appears to be sexual selection. It's not that convincing but that's, that's, you know, that's the the arm way the explanation that we have at the moment. An explanation. I forgot to say, I'd like to remind the audience that if they want to come here and join us in this room and ask them to the questions I can for the link I just shared on a chat. Thank you. I have a question from the button in nature, or the shrimp body parts are actually in polarization by a signals related in nature are polarization polarizing color signals. Yeah, other colors and polarization by signals are related in mantis shrimps they are. And they can be in other parts of nature so. If you remember those little blue things that I showed you on that it's actually on haptose squilla the little blue maxillipeds they sort of hold at the front of their blue. They contain nano structures and actually Tom Cronin and his lab have worked on this extensively and shown that the, the way in which light interacts with those structures produces both blue, but also polarize life. So if you look at the, the peacock mantis shrimp the, the antenna scales the paddles they have at the front, which they sort of stabilize themselves where they look a bit like gears. They also contain a molecule which is polarized. So in mantis shrimps you often find colored areas, which are also polarized so you get a combined signal. And that's a bit weird. The flies are supposed to combine polarization and color, when they decide where to lay their eggs and there's a bunch of nice work by our McAlber and others before her, looking at that. So it, no animals, there's a, there's a precedent for animals using both signals together. There is of course the potential to confuse the two and that that causes problems but it appears that they just do it. So there. So it could integrate some misinterpretation of the signal. Exactly. You've got two different signals. So if you've got luminance and color, no varying or covariating, you don't want that. You want to be able to discount the illuminance signal in order to see the color, except if you're a butterfly or a stomata pod. Right. Hello Michael book with us this morning. Hello, Michael. How well does modeling of the four opponent beans match the delta alum that they have your results. But I get into that. So, so again, so my the, the modeling of. I mean, you showed that to the end, but I'm going to ask a question. How well does modeling of the four opponent beans match the delta alum that they have a result. How well does the modeling of the photoreceptors match the behavioral result. Yep. So don't tell them that you should I mean the last. Yeah, very, very well it's some. Questions. I won't go back but if you look at where the, the potential is for the best opponent processing discrimination within the spectrum. That match is perfectly the areas where the animal has behaviorally the best discrimination that's a ridiculous not very well discriminated w or whatever it is shape curve. Those minimum in that curve match up to the spectrally opponent beans or areas within the spectrum that each of the mid band rows appear to be concerned with. Now is that instead of each one of those photoreceptors acting as an independent bin and providing a pattern of excitation. Within that there's a sort of sub process of opponents see which is going on, but that still gives a pattern of excitation but just using a very crappy kind of opponents see. If you're worried about what we're about to do in terms of modeling. It doesn't really change it it just means that the endpoint, hopefully can be simpler. Please. Have a follow up question from a button. Can you therefore use simultaneous colorful plus polarized signal as a recognition trick to know you are looking at a mantis string, because nothing else around as this property. That's a very, a very good possibility. And if you look at some animals in the rest of the ocean, there's, there's very few that produce polarization. Actually, the cephalopods do so. Cuttlefish do and that's another area that we're looking at is polarization vision in the cephalopods because they do that and they don't do color vision they have no color vision but they have at least orthogonal polarization. So a lot of reflections they're giving a very different to mantis shrimps. And maybe it's the case that mantis shrimps are combining these signals to say hey I am a mantis shrimp, nothing else does this so pay attention. Now why they have to go for circular polarization. You know, again, our way the argument there is it's an arms race between the cephalopods and the stomata pods which have both been in the ocean for 400 million years. It's a linear polarization vision. But stomata pods only have gone for circular. Therefore that maybe they're trying to out compete the cephalopods which are actually their major predator. So when you look at what eats stomata pods. It's very often octopus and cuttlefish. Polarization vision so maybe they're spotting the linear polarization in some mantis shrimps and then the species that have circular. Maybe they're evading detection by the cephalopods. So you're suggesting a co-evolution with prediction territory with cephalopods. Where's the evidence, you know, up there somewhere. Suggestion. So just to let you know that I have a couple more of questions then I'm going to close stream. If you want to continue this discussion please join us now ish. I have a question from color look as a race. Is there any idea of why do they have such a rich number of channels. If they don't use or need I color discrimination based on the very well color discrimination environment. So any idea was a rich number of color channel. And in the end they don't discriminate against them. So that I mean that comes down to kind of the crux of the talk which was. Now why do they need so many you don't need that many you can show theoretically that in order to decode all the colors in the world you only need four. So why have 12. And that's what drove us initially to think that they have this different way of processing color. And the reason they have so many is because they need that number to sort of fill in the spectrum with this different color processing this parallel build up a pattern of excitation across the spectrum. If you're going to build up that pattern, then of course you have to fill in the spectrum in discrete areas, and quite fine areas so 12 different points in spectral space would have to be filled in, in order to build up a pattern when you're looking at a reflection of a spectrum. So that was the reason we thought that's that explains why they've got so many. You know as the talk tried to unfold, as we look closer we kept seesawing backwards towards the, well maybe it is a potency actually after all. And well maybe it's this parallel thing. And the conclusion that I'm at at the moment is well maybe it's just both. And for whatever reason they throw away the potential to have very fine spectral discrimination with those individual dichromacies. I'm just looking for a very simple answer. Is it blue, green, orange or red. And that's a very quick way of answering that question. I move on to the last question. This is from Cedric van den Berg, which I guess is from cuisine and for my life. Yep. Hi Cedric. Is the presence and location of his opponents he beans correlated with the presence of sexually dimorphic colors in multiples to multiple spaces. Yep, good question so. Now are those bins related to the colors of the animal and those in particular those merrell spots the bits that they show off on the inner surface which Mike explained more in his talk. I kind of shot through that, but they show off particular colors to each other when they meet. And yeah, I mean that was my very first thought maybe these colors are special. Maybe these colors tune into the mantis shrimp color vision sense. And actually Roy called well in the 70s showed that mantis shrimps with different colored merrell spots behave differently so they may look the same to us. But they have different colored merrell spots, and you can correlate that to their aggression level. So when they meet each other, a mantis shrimp with a purple spot is less aggressive than a mantis shrimp with a white spot. And they know that through previous conflict or through evolution. So they do know and they have this color coding system amongst themselves. And because of course they can just kill each other immediately. So maybe this is a way of avoiding conflict. It would be nice if they had more different merrell spots and that's what actually this this projected paper that Mike mentioned in his talk we're looking much more closely with Mike and with Megan Porter. So we're looking at these colors to try and determine whether they are matched in any way. They don't appear to be matched to species. But they are at least signals which are available for this color vision system to queue into, if that indeed is what it's doing. I hope that answered the question. We can talk more tomorrow Cedric if you're as confused as I am. I will just finish up with with this in one of your slide you had this in bracket. So what about UV. Yeah. I sort of brushed that aside. Just because, no, as you can tell it was a it was a complex series of events to get over. I thought I'll forget the UV bit for the moment. So the UV photoreceptors sit above those color pairs that that lie within row one to four. And with you had a sober vinkler we recorded from those a while ago and show that there's lots of them. Mike Bulk followed that up with some very nice micro spectra photometry and showed how the filtering of the light above those words. And there is that there are at least four different types of UV photoreceptor that lie in four different parts of the UV spectrum alone. So between 300 and 400 being our eight photoreceptors being the ones that lie in the top. They actually project their axons through the lamina to the medulla where it's really really confusing what's going on. But you just don't want to go there. It's awful. It's this soup of neurons. It's just like it's awful. So no that's a project for a PhD student I really hate at some point to give them that and go, oh sort it out would you know it's only for neurons for goodness sake. So what we again hypothesize is that there is some sort of UV processing that allows the animals to discriminate things in the UV and actually Mike, Mike Bulk, I think during his PhD or shortly after, I know that they can discriminate UV, different UV signals, and they seem to use this again in some sort of way to recognize each other. So UV is reflected from different stomatopods, and they may be using that part of the spectrum, very specifically for looking at other stomatopods so you know the bottom line is it's just an extension of the spectrum into the UV. So they may be doing the same thing, behaviorally with it, but with a different part of the brain. Thanks. Let me be linked to this if you have elements. I said that was the last one, but I forgot how to one on site. Following on that, we have the same kind of problem with UV light that we cannot reconstruct in facilities, the same natural light environment. So I said that some of your shrimp were losing some color discrimination based on the elimination around it. Just a couple of questions on that is. So I don't know where you bought your mantis shrimp. I don't know if you reproduce them in a facility is so therefore they lose our discrimination of a generation. And secondly, can they recover this color discrimination if you put them in a proper natural light condition. Yeah, so we have actually we've done that experiment. And now in the early days with Tom when we started doing the micro spectrophotometry, we would know prepare a section of the retina and look at it and look at the bit of the photoreceptor that you can see. And you kind of look down a tube of photoreceptor, and you place the beam of the spectrophotometer in it, and you're measuring the absorbance of that bit of photoreceptor. We started to notice it after a few weeks you put the beam of the spectrophotometer in that bit of photoreceptor, and there was nothing there. And we thought, yeah, what's going on that's a perfect preparation why can't we see stuff. And that began to make us think that maybe over time they were losing it. And then it turned out that when you look at animals that live at depth versus animals that live right at the surface. You can get the same species living 20 meters apart. So they've got 20 meters of water which absorbs light here, which these animals are still seeing. And their retinas are different, so they're photoreceptors adapt. And if you put animals from the surface in the aquarium under restricted light, over a very short space of time their photoreceptors adapt. We've now done this with fish, as well as have other people and they adapt in days. Animals are plastic with their color vision system within hours or days, not weeks or months, and certainly not over evolutionary time. So, when you say an animal has this color vision system, unless you're keeping it in full spectrum light, I bet it doesn't. And one of my brackets that I didn't mention was casting doubt on a lot of previous behavioral experiments where people have you know as we did as I did in 1996. I just got animals sent to me from a tropical marine supply place in London. Who knows how long they'd been there, I kept them for months. I did behavioral experiments and interpreted those in a way which was naive and as I mentioned probably wrong, because they'd lost it, they'd lost the ability to do the blue end of the spectrum. So, that really puts the cat amongst the pigeons in terms of behavior and color vision behavior. And what we do now here is to keep animals under as broad a spectrum as possible we've just installed downstairs strips of UV LEDs. So every time we go into the aquarium we have to put on UV absorbing glasses we've got curtains we've got sunblock we've got hats we've got everything to work indoors. Which is a bit bizarre but it's a really necessary part of keeping the animals visual system going, in a way which is as natural as possible. Because yeah, it wasn't us initially but way back when you're convulgated and other people showed that cyclids change their color vision system over a very short space of time. Karen Carlton has done this again with cyclids and it's really a matter of days, they express different ratios of visual pigment. They basically have a look up table of what they need they put it off the shelf they plug it in and off they go. Bastard. I'm giving up color I'm just giving up on color vision it suddenly become way too complicated, even more complicated than mantis shrimp so good luck. Yeah but I think it's a good point that we should keep our animals as a model in their natural light environment. Yeah I mean that's it just try and do it outside if you can. It's a good argument for going into the field, you know. Yeah bring it before the upside not a good idea. Anyway, thanks a lot. I would have to close the stream now so I would like to thank you for the invitation. Thanks for the talk. It was, as you can tell I got confused. It's just one of those things it's really hard. So I'd recommend people go back and read I tried to put words on the slide. But I couldn't say go back and read it through again. When you've gone through it about 10 times it might start to make sense. Does it. That's a beautiful. It does. Trust me. Anyway, thanks to audience. We will see you next week. Thank you. Thank you very much. Thank you. I would like to thank our team for the time we have. And we will receive. Marty Brown from Manchester University. So thanks a lot. I'm going to put the stream now. If you want to join us for the discussion. Please do that now. Bye bye. Hi. And we are now offline.