 and I believe we are officially on air. Hello everyone, and welcome to another session of our Sussex Vision Seminar Series, Always Within the Worldwide Neuroinitiative. I'm George Cafetsis, former master student in Thomas Euler's lab and currently a PhD student with Tom Badden. And as your host for today, I would like to once again begin by thanking Tim Vogels and Panos Bozellos for putting forward this very initiative towards a greener and much more accessible seminar world. Having said that, please allow me, of course, to get back to the reason we all gathered here for today and introduce our guest from Lund University, Professor Eric Warant. Following a honours degree in physics at the University of South Wales, he went on and obtained his PhD in Visual Sciences from the Australian National University in Cumbera by studying the optics of arthropod superposition eyes. An invitation in 1990 from Dan Milsson took him to Sweden and Lund University at the time as a postdoctoral fellow and seems 2002, if I remember correctly, as a professor of zoology. With research interests revolving around vision in dim light environments and with a wide arsenal of techniques ranging from electro-physiological to behavioral to theoretical. In his own words, he had the privilege of studying a number of organisms among vertebrates and invertebrates from deep sea cephalopods and fish to nocturnal insects. President of the International Society of Neuroethology, recipient of many personal awards and author of a number of books including Visual Ecology, together with Thomas Cronin, Sonke Jonsson and Justin Marshall. It is with great pleasure that I'm leaving the stage for him, Professor Eric Warant, for a talk entitled Australian Bogong Moths Use a True Stellar Compass for Long Distance Navigation at Nights. So without any further ado from my side, please all welcome Professor Warant. Eric, the stage is all yours. Thank you very much, George. Can you hear me? Yes, crystal clear. Thank you very much for that very kind introduction. That was very, very nice. And that's very nice to be here. Thanks to you and thanks to Tom and all the organizers of the seminar series. So I'm very happy to be here. Thank you. Just tell me when to share and I... Yeah, you can go ahead, please, Eric. Okay, with some luck, you can see that, I hope. Is that... Is that visible? We don't see the full screen presentation mode. We see the different tabs. Oh, I'm sorry. I've done the wrong one. Indeed, that's right. So how does that look? Splendid. Right. Okay. Right, well, thank you. As I said, George, thank you, Tom. Yes, my talk is going to be about a curious Australian moth, the Australian burgong moth, and its ability to navigate over long distances. So just to begin, I'm just going to tell you a couple of things about migratory incestors, a couple of brief things. Firstly, when insects migrate, as many animals do, many animals make enormous migrations across the surface of the earth, some shorter and some longer. When it comes to insects, huge numbers of insects migrate and at night time over the British Isles, for instance, there are huge numbers, millions of trillions actually of insects which migrate over the British Isles at this time and over the summer and during the autumn. But generally, most insects when they migrate tend to migrate from one very broad geographical region to another in search of more favorable conditions, sort of a broad area from a northern part of Europe, say to a broader part of Northern Africa, but there's no specific actual destination attached to that migration. There are, however, two specific species of insects that have been reasonably well studied and there are very few insects that have this behavior that have a highly specific and geographically restricted destination. There are two such species we know of reasonably well, one in particular, and that insect we know particularly well is the Monarch Butterfly from North America. This is a long-range, day-active navigator that travels enormous distances four to 5,000 kilometers in a single direction from the south of Canada and the north of the United States. It travels across the United States southwards during the autumn, before winter, arrives in central Mexico to a very, very specific mountain area with specific type of vegetation where it spends the winter in huge numbers. And there's a picture of it in the next slide here. They congregate on trees in vast numbers. It's a very well-known phenomenon. And we know quite a lot about its migration as a day-active insect that relies on several cues, but the major cue that it relies on for navigation as a navigational compass is the sun. It relies on the sun, particularly because it's a very constant and a predictable cue every day for days on end during the migration. It rises in the same place. It sets roughly in the same place and it travels roughly across the same trajectory across the sky. And for that reason Monarch Butterflies have evolved the ability to use the sun as quite an excellent compass actually for keeping a migratory heading towards Mexico. And then back again, actually, once the time in Mexico is over in the spring. The other insect that makes such a highly-directed migration to a very specific destination is the Australian Bogong Moth, a grotesque infuser. And this is an insect that's very well-known to Australians like myself. Many people in Canberra where I did my PhD know much about this moth because it sometimes ends up off a little bit off course attracted to the lights of Canberra where in vast numbers they can invade public buildings and block ventilation systems, cause short circuits, disrupt lift operations and cause what most people would consider to be mayhem. And some people actually find these insects to be quite abhorrent as a result of this. But actually I think they're quite delightful and they've got remarkable navigational skills which I'll tell you about today. Now these insects, like Monarch Butterflies, they start from an enormous starting area throughout South Eastern Australia. This is their major migratory, sort of their migratory roots. So this is a map of Australia, hopefully you recognize here on the left. And on the right, you see these arrows showing where the moths are starting their journeys and where they end them during the spring in this particular case. So the moths during the winter are going through their egg lava and pupa stages in the breeding areas. And the breeding areas are spread throughout Western Victoria, Western and Northwestern and Northern New South Wales and Southern and South Eastern Queensland. And they emerge in the early spring as adults and then they begin a very, very long journey of about a thousand kilometers to arrive in Australia's highest mountains which are the Australian alpine areas which extend from North Eastern Victoria to South Eastern New South Wales, which you see here. And also actually in Western, Western Australian Capital Territory, the small sort of territory where Canberra exists. And so they arrive in these high mountains and when they get there from all different directions, they find high alpine caves on the mountain ridges. So this mountain over here at the top you can see with a bit of snow on the side is Mount Kosciosco which is Australia's highest mountain. It's sort of round and unobtrusive compared to a lot of other countries' highest mountains. But it's our highest mountain. And all around here, including on the side of Mount Kosciosco, there are a number of piles of huge stones which create cave-like hollows and rocky interiors where moths arrive literally in their billions from all over South Eastern Australia. They arrive in these caves which are dotted along the ridge tops and then they move into these caves and they line the walls of the caves in huge numbers like scales on a fish that you see in this picture on the bottom left. There are about 17,000 of these moths on every square meter of cave wall. So any given cave can hold probably millions or at least hundreds of thousands of moths in each cave stretched from the Australian Capital Territory down to Victoria. We don't know exactly how many caves there are but about 50 to 60 caves are currently known in New South Wales. Not every location where you might expect moths to be found is occupied. They seem to come back to the same locations every year and these are very specific locations, specific caves. Now they spend now the summer here in these caves in this state and this state that they're in is a dormant state known as estivation. It's a little bit like hibernation except over the summer. So they go into a state of dormancy. They stay in this state in these cool caves for about three to four months and the same individuals that arrived months earlier then leave the caves and return to where they were born. To again, when they get there they lay their eggs and then they die and then they're offspring developing in the following winter and in the spring that follows the adults emerge again and start the entire cycle afresh. Now the interesting thing about the moths is that they've never been to the mountains before. Nobody's ever showed them how to get there. Their parents have been dead for two or three months. So when they make this journey they make it for the first time without any assistance. Their ability to find the right direction and travel the right distance is thus inherited from their ancestors. And what exactly they use to find the way to know which direction to go in and when to stop are questions which are very interesting to us and I'll come back to that in a moment. Oh, I come to it now actually. So in other words, how is this remarkable migratory feed achieved? How do they find their way? They are having never been there before. How do they know the direction to fly and how do they know when to stop? And to answer these questions we've been looking over the last few years at migratory moths both in the spring and in the autumn. So in the autumn they're on their way down to the mountains from various parts of South Eastern Australia. And in spring, if that's in spring and then in autumn they're on their way back. So we've had two main locations where we've been studying these moths at a place called Mount Caputah which is a lovely area in the very north of New South Wales. I can actually go back a couple of shots here and show you. Can you see my arrow moving around here? Mount Caputah is just up here. That's where we've studied them very often in the spring. And in autumn we've studied them down here at a place called Adaminabee which is just below the ACT and we've studied them during autumn there. And what we've done is we've put up very powerful light traps in these locations and caught moths during their migration. And once we've caught the moths often in very large numbers then we subject them to our experiments directly which are both behavioral and electrophysiological and come to the electrophysiology a little bit later. But the standard method we use is to take moths we've captured and then attach a tungsten tether to their backs with contact cement. So it's not invasive. It's attached in such a way that as you see up here in the right-hand corner that the moth is able to fly on the other end of the tether. We're going to mount that tether with the moth into a special kind of flight arena which allows the moth to fly. I'll come to that in just a second. But the point of this apparatus is that the moth is free to move in any direction 360 degrees around its dorsal body axis but it can't move from this location. So it can move in any direction relative to north here and it turns out we can track that with our arena. And this is behavioral work that's been done by my post-doc David Dreyer who's done an excellent job on the behavioral work. And so this is the type of apparatus that we are using. So as I said, we place the moth on the end of a stalk here, a tungsten stalk. You can see that it rotates around. You'll also see underneath, there's a pattern of optic flow that's flowing always from the nose of the animal to its tail irrespective of what direction the moth is flying. So you see it rotating around there as the moth rotates around. It's very, very bright here. So you can see it. But normally in the experiment when it's inside the arena, this optic flow is at starlight intensities. It's extremely dim. We can't see it easily ourselves. This arena with the moth is sitting on top of a small table. In our first experiments, which I'll tell you about first, that we've done this out under the open sky outside. So it's sitting on a table underneath. There's a 45 degree mirror which reflects from a projector, reflects up onto the underside of the arena onto a piece of tracing paper here on this perspex top to table. It projects that optic flow underneath the arena and that's where we generate it actually. And then we have in addition, a thing up here at the top of the arena known as an optical encoder. And that optical encoder eventually is attached to that shaft. And the optical encoder's job is to register five times every second the actual heading of the moth relative to north. So five times a second, we know exactly what the heading of that moth is relative to north. So over a period of time, it's possible for us to build up a kind of virtual trajectory, if you like. So here's one such trajectory over five minutes. So one minute, two minutes, three minutes, four minutes and five minutes. That's a relatively well-directed moth that was flying for the most part just east of north. We can characterize that trajectory by a vector. And I'm gonna show you these vectors in other diagrams. So take note of this now. This orange vector has a length and a direction. The direction obviously is a reflection of the main direction that the moth flew during its five minute trajectory. The length of the vector is a measure of its directedness. That is its tendency to want to fly in that direction. The longer the vector, the more directed the flight. Of course, some moths aren't very directed at all. And might have a trajectory that looks more like that. In which case the vector has a very short length and is going to indicate some direction between the start and the end of that trajectory. But whether it significantly represents the average direction is questionable. So this would probably be representative of a highly undirected moth, in fact, which this trajectory would suggest. So this is the type of individual that the information we can get from an individual moth. Normally, however, we are interested in looking at the population behavior of a group of moths. Here's an example of 10 moths that we've placed into our arena. And here you see the 10 orange vectors that have resulted from their 10 virtual flight trajectory which we've recorded in the arena. So here you see a group of them. So some moths are quite directed and have a particular direction. Others, as you see down here, are quite undirected and have what is probably almost a random direction, actually. And you can take, you can account then for both the directions and the directednesses of each one of those, in this case, 10 individual moths and work out an average population vector for that population that accounts for both the directedness of each moth and the direction of each moth. And when you do that, you get this nice red vector here that has a particular length and the length of that vector has statistical significance. And that statistical significance is actually shown by dotted circles in these plots, okay? So in this particular case, we have three dotted circles, each of which represents a different level of significance. P is less than 0.05, P is less than 0.01 and P is less than 0.001. And if the vector breaks through any of those circles, it will obtain that degree of significance. In this case, the average population vector for this population is very significant because it's broken through the outer, P is less than 0.001 circle and it has a direction which indicates the population was going just west of north with reasonably high significance. In another set of 10 moths here, we see that the moths are neither very directed nor do they have a common direction. And the average red vector, the population mean vector for that group of 10 moths doesn't even reach the innermost P significance circle here. In fact, the P value for that population is between 0.5 and 0.9, indicating that that population is undirected. There is no significant direction that population is moving. And you're gonna see quite a number of plots like this now in this talk. So just to note that that's the kind of basis of our behavioral data. So here we have the results of a natural experiment where we've simply had a population in this case of 18 moths at Mount Calcutta in spring up here in northern New South Wales and a population of 36 moths caught in the mountains and tested near Adaminabee in autumn. These are natural, this is just natural behavior under the open sky in our open arena viewing the sky on the table. And just as I showed you in the picture earlier, and those 18 moths caught at Mount Calcutta in spring have a population vector here, the red vector here, which is highly significantly oriented to a direction which is just west of South, which is roughly what you would expect for that, a population wanting to make it down to the mountains. So that's roughly what you'd expect, which is quite comforting to see. And then in autumn, the population that we recorded from here was moving also very significantly to the Northwest roughly. We have no a priori knowledge of which group of moths we've sampled here. They could be going in any direction back to Western Victoria or Western New South Wales or Southern Queensland. This particular population however, seemed to be going Northwest. And they did so with a highly significant mean population vector. So in other words, under an open sky with a view of the stars at night and any other cue that they might be using, such as Yoast magnetic fields we'll come back to in a moment, they seem to be doing roughly what you'd expect, which is very good news. They seem to be able to reverse their directions from spring to autumn and head where they should be during their migration in those two seasons. So the next question obviously we want to investigate then, if they have this highly directed behavior in an appropriate direction, as it would seem in spring and autumn, what sensory cues do moths actually use to find that direction? There are at night several cues available and these are cues which we already know, nocturnally migrating birds use and these are magnetic cues from the Earth's magnetic field. That's one possibility. The other possible set of cues that they could be using are visual cues and they could be either terrestrial landmarks, they could be celestial cues, such as the moon and the stars. Any of these are possible cues for long distance navigation. And indeed there, as I said, they're used by nocturnal birds actually that migrate great distances. So to do this experiment, we first decided to test their ability to possibly use the Earth's magnetic field and for this experiment, we took our arena, which you've seen before and we placed it in a set of magnetic coils which has the possibility of turning the Earth's magnetic field in different directions. So in other words, it can turn it from the natural North-South direction to other directions. It could, for instance, we could, with this apparatus, turn the field lines at a point East-West if we wanted to. In our experiment, what we did is turn the magnetic field by 120 degrees. So from zero degrees, which is North-South to a direction which is sort of 120 degrees clockwise to that sort of almost Southeast. And our experimental hope here was, in our first experiments, that if one were to turn the azimuthal direction of a natural geomagnetic field, so I forgot to point out, by the way, we're doing nothing to the field, we're preserving the Earth's magnetic field to be exactly the same field as present at this site in Australia. So the same inclination angle, that is the angle of the field lines relative to the surface of the Earth, the same magnetic field strength, exactly identical. All we're doing is turning the azimuth of the field. So the question here is, if we turn the field in this way, do burgone moths follow while they're flying in the arena? Do they follow suit and also turn with the field or do they not do that? We know already that if you do such an experiment with nocturnally migrating birds, which are restricted to these famous emlin funnels, it's a kind of a lined funnel with a white paper that can be scratched by the birds and they will scratch more on one side and leave these black marks if they want to go in that direction. And when inside magnetic coils, people have found in the past, if they rotate the Earth's magnetic field in coils, they will also turn the direction that the birds are scratching in these funnels. So we had that hope that this might work. But sadly, after two entire years of field work, that did not happen. We could not get them to turn in the field in this way. In fact, most of the time, they didn't do anything at all these moths. They just kept on heading in the same direction. If they did react, they often went in directions which were not in absolutely had nothing to do with the direction of the turned field. And after a while, we just had to suddenly go back to the drawing board and have a bit of a think. And what is actually quite surprising and this is always good in hindsight is that you should never be sold on your own experiment as we were with the magnetic one. Because one of the things that of course, I should have realized having worked on nocturnal vision in insects my entire life basically, is that moths and other nocturnal insects have exquisite vision. They have incredibly sensitive vision. And so a moth in our arena, viewing the world above, we'll see an enormous number of visual cues as well. They will see these coils, for instance, from within the arena. They'll see this perspex bar most likely, hanging across the top of the arena, holding the optical encoder to the shaft to which it's attached to in the middle of our arena. So there's a lot of things that could see it, could also of course see the stars. And these are all totally invariant stimuli when we turn the field. So they are maintained. If in addition, there are wobbles and other visual cues inside the arena itself along the walls of the arena, those could potentially be seen by the moth as well. All of these could be used as orientation cues and they might prefer to look at those and keep flying in a particular direction than to take any notice at all of our magnetic field that we've applied. So to get around this problem, when we suddenly realized this, we decided to introduce our own extremely dominant visual field of visual cues and attempt to get rid of all extraneous visual cues that we could. So what we did is we lined the bottom of our arena with a black horizon and we put a little mountaintop on top of that horizon, which we could move. So we could move that mountaintop around actually to different locations. We also had a disk of UV transmission, transmissive tracing paper, which we attached to the shaft above the moth. It didn't restrict its movements at all. It was just sitting against the, we had a tube around the shaft actually to protect the shaft. So we attached it to that protective tube and we placed a stripe on there because we had no a priori knowledge of which part of the visual field the animal might be using visual cues to navigate. So we placed a black stripe on that piece of circular tracing paper as well and we pointed it at the top of the mountain. And we have the ability to rotate that disk as well. So we could move the mountaintop, we could rotate the disk and thus rotate this black stripe. So now we had the possibility to do another experiment and we now made a 20 minute experiment on each moth where we subjected them to five minutes of visual and magnetic cues. In the first experiment, the apparatus was as I showed it in the previous slide. So we placed them off on the end of the tether. We had the magnetic field in its natural direction north-south. And then we placed the visual landmarks at 60 degrees to that direction. So 60 degrees east of north. So we had the stripe and the mountaintop at 60 degrees with the stripe pointing at the top of the mountaintop for five minutes. And we recorded then the virtual fright trajectory of moths in that circumstance. After five minutes, then we switched the field by 120 degrees. We rotated the field by 120 degrees into the southeast sector. At the same time, we moved the visual cues also by 120 degrees. So now the visual cues, the mountaintop you see here as seen through the side of the arena, the stripe is also again pointing at the top of the mountain. So everything was moved. So everything changed together. In the third five minute period, what we then did was we left the visual cues where they were in the second phase, which we've called phase B. And in this phase, which we've called phase C, we turned the field back to where it was at the beginning. So in other words, back to north-south. So here what we've done is introduced the cue conflict so that the visual cues and the magnetic cues are now no longer in the state that the moth has learned them to be in at the start of the experiment. And to make sure that the moth returns to the original condition, we repeat the first phase, phase A again in phase D. These two phases have exactly the same stimulus conditions. And these are the results that we obtained. So in the first experiment with 42 moths in this case, the 42 moths of this population here, each moth shown by a black vector whose length and direction again, showing a representative of the trajectory of each moth. And here you see for this population of 42 moths in that stimulus set up, a mean population vector that is pointing, the population is pointing just to the east of that set of landmarks. If we now rotate both the magnetic field and the visual landmarks by 120 degrees, the population of moths moves exactly by the same number of degrees. So the angle between these two mean vectors, both of which are highly significant, they cross this P is less than 0.01 dotted line. They move exactly as you would predict on the basis of the movement of these two cues, magnetic and visual. In the cube conflict situation, when we move the magnetic field back to the original direction again, then the moths after a while, it takes two minutes roughly for them to lose it once we've made this change. But after two minutes, the moths become completely disoriented. So in other words, you see here, there is no visible red vector anymore that describes that population. They are completely disoriented. If we then go to the final condition, which is a repeat of phase A, then you'll see that the population returns very significantly to the situation that it was in at the start of the experiment. So in other words, they just didn't get tired by phase C and start doing silly things because they were tired. The cube conflict here, which involved nothing more than turning the magnetic field by 120 degrees back to where it was initially, that single change alone caused the moth population to become totally disoriented. So the results of this experiment suggest that, which I'll go back here first, the results of this experiment suggest two things. First, it suggests that these moths do indeed have the ability to sense the earth's magnetic field because this particular change in the, if the moths were only orienting themselves according to the visual landmarks, a change in the magnetic field alone would not have changed their behavior. But because they became totally disoriented when we changed the field, that implies that they're able to sense the earth's magnetic field and use it for steering this migratory flight behavior. That's the first conclusion. The second conclusion is that they somehow seem to be using therefore a correlation between visual and magnetic cues to hold their direction, to stay as an oriented animal. And so this is all written up in more detail with a lot of controls as well. I have to say, I don't have the time to show you all the controls, but we know that this is actually a real behavior here from the controls. But with those controls, we've come to the conclusion, or we've made the hypothesis actually, that for burgone moths, it suggests that their navigational strategy that they're using is possibly something a little bit like what we might use if we go for a hike with a map. So if we want to travel somewhere when we're hiking, if we want to say go northeast, then we take a magnetic compass out of our pocket, we work out which direction northeast is, and then we look for something in the distance, like a distant mountain peak or some other landmark that we then decide to walk towards and put the magnetic compass back in our pocket. So it's almost as though maybe the burgone moths are doing something similar, that they're using a visual, they might be using a visual cue or visual landmark as a kind of a navigational beacon, which they then check with, check the fidelity of every few minutes with a magnetic compass. This few minutes comes from the fact that it seems to take about two minutes for them to realize that the cue conflict has occurred before they start becoming extremely disoriented. That data is all in that paper, that current biology paper, his reference I showed earlier. So that would imply then that visual and magnetic cues are very important during this navigation that the burgone moths are doing to arrive in the mountains. They obviously have a magnetic sense. So the question then is, what are the visual cues that have been used by the moth? And our first question that we asked, which is what I'm going to show you now for the rest of the talk, is whether it might be possible that the skyry night sky could act as a kind of a landmark, if you like, if you'd like to refer to celestial objects as landmarks, could they function as temporary beacons for navigation in this way, where those beacons are calibrated against a magnetic compass to migrate over a longer distance? And to do these experiments, we needed much more controlled conditions. So what we did was a couple of years ago, we built a lab on land. Actually, I must admit, I already own since 20 years in the Snowy Mountains of Australia in Adaminabee. This is why we've used Adaminabee as the location for our experiments. My family and I have had a house there since the early 2000s, which you can see up here on the hill. I can see it over here as well. And next, just down the hill from the house, which we now use as a kind of accommodation and field station, we built a lab out of entirely non-magnetic materials. So everything in this building is non-magnetic, including the roof, all of the structure of the building. It's obviously made out of timber, which isn't magnetic. All of the concrete slab that is built on, all the reinforcing is made out of fiberglass and not iron. So there's no possibility to get magnetic artifacts in here from magnetic materials, which is very important for these experiments. And what we did was we set up two setups in here, one behavioral setup, which I'll tell you about first, and then, and also an electrophysiological setup, both of which are entirely non-magnetic. Electrophysiological setup is non-magnetic, and so is the behavioral setup. So here is the behavioral setup. And now we, instead of having a single axis of coils, we've got a three-dimensional set of coils, an X, Y and Z set of coils, completely surrounding our behavioral arena, which you can't see here. It's actually inside this black curtain. But here you can see it opened up. This was during construction. So I was looking a little bit messy at the moment. So normally it's a little bit tidier than this. But here is our arena barrel. Here is the tabletop with a tracing paper as normal. In this arena, we have a lid on top of the arena, which holds the optical encoder at its center. So the shaft holding the moth is attached to the encoder. And so the moth is still as normal down in the middle of this arena, flying on the end of its tether. This lid that we've put on here is a UV translucent lid with a UV translucent circular piece of tracing paper, which we use as a projector screen for a projector, which we mount on an aluminium frame on top of the coils. So here you see the projector up here. And using a free planetarium software, everybody can download onto their computer called Stellarium. We were able to project highly realistic night sky images of the outside sky at the exact date and at the exact time of our experiment inside the arena here of our behavioral apparatus. So this program Stellarium is quite amazing. You can actually project any starry sky at any location on the earth at any time of night on any date. And so we just dialed up at a minute the right date, right time and use that starry sky on the lid of the arena in the right season, obviously. So the sky during autumn and the sky during spring are different actually. So we use those two different skies. So the experiments that we did then, oh, actually there's one thing I should mention. The reason why we've got the arena inside the coils is not to stimulate with a magnetic field, but now actually to completely remove the earth's magnetic field. So we've got the possibility to make a magnetic vacuum inside the arena. And that's very important for these experiments because we don't want any confounding effect of the magnetic field and the magnetic sense. So we effectively knock out the magnetic sense by getting rid of the stimulus here. And so all of the experiments I'm about to describe to you now are done without a magnetic field. So the magnetic sense has been disabled to study the visual sense in isolation. So here we have the stimuli that we used in spring and in autumn and a control. So here is the natural orientation of the spring sky and this is what the image would look like above the moths. Here we have the natural projection of the autumn sky above the moths and is north here. Then we also have a control sky that we've made where we've taken every one of these stars and moved them around randomly across the image so that we get rid of any spatial information that might arise from the Milky Way but we maintain the brightness of the image the same. So it's the same brightness. Everything is the same except there's no spatial information from the stars anymore. Then we have the ability to rotate that image as well. So in the experiments I'm about to show you we also rotated the night sky by 180 degrees and the control sky actually as well, all sky. So we could make the sky such that north pointed south both in spring and in autumn. Then we got a population of moths and we put them in the arena. No magnetic field, completely free of all visual cues apart from the projected night sky above. And what we found is doing nothing at all just letting them fly underneath that starry night sky in spring. It's a population of 70 moths. This has done over two separate years in two spring seasons, these 70 moths. And as you'll see, they're highly significantly directed to a direction in this case of just east of south. So this is roughly the predicted migratory direction you'd expect for spring migrants attempting to come to the mountains in this case from a northward direction it turns out. If then we rotate the sky by 180 degrees that entire population turns around and goes back in the opposite direction. So in other words, I don't recall the exact angular change here but it's almost 180 degrees plus minus about five or six degrees. So the population mean vectors switch directions when we wrote by almost 180 degrees when we turn the sky by 180 degrees. If we show the control sky, the jumbled starry sky instead the population is completely disoriented. So in other words, this population of moths has been able to locate its inherited migratory direction under the starry night sky we project alone in the absence of all other visual cues and in the absence of the Earth's magnetic field which was quite a surprising result to us. When we repeated the experiment in autumn instead under the autumn night sky, which looks a little different this time now from at Adaminabee we're studying migrants returning under the natural night sky a population in this case of 54 moths taken over two autumn over two years that population of moths is also heading in a direction that you'd expect for moths migrating to the breeding areas. If we also now turn that sky by 180 degrees the population of moths turns also by 180 degrees roughly. Again, it's 180 degrees plus minus five degrees don't have the number in my head exactly. And once again, if you have that population under a random sky the population is completely disoriented. So in other words, both in spring and in autumn it seems that this population, these Bougon moths seem to be able to use the starry night sky as a kind of a true compass if you like to find their inherited migratory direction relative to north. So in other words, the conclusion from this we've come to is that the sky the night sky is simply not just a visual landmark it seems that they really are able to use it as a kind of true compass to coarsely discern north, south, east and west and thus find their correct migratory direction in different seasons. And it turns out also, and this is work of Andrea Aden who is now doing her postdoc with Tom's partner in London, Lucia. She is responsible for this work it was part of her PhD. And the work that she did during her PhD was to now study cells in the brain of the moth that might be involved in this compass navigation this stellar compass that we've found behaviorally. And what she's found in her PhD is that there are indeed visual cells that respond exclusively to the starry night sky and these are found in the central brain and in the central complex of the Bougon moth. And just to introduce that area, the central complex for those of you who've never heard of it, it's a central region of the brain which has been a hot area of research actually over the last, I would say, roughly 20 years. This region of the brain, the very center of the insect brain it's highly conserved across different insects and it's a region of the brain which is used to process different types of sensory information which is being gathered by an animal during locomotion and to turn that information into steering commands. In other words, the central complex is highly involved in helping animals to stay on a given course by analyzing how the animal itself has changed the sensory input and then to correct any changes in course that the animal may have experienced in order to put the animal back on course again. In other words, it's involved with kind of of steering and not surprisingly, therefore the central complex houses the kind of compass circuits if you like for those animals that have different types of compasses. For instance, the Monarch Butterfly Sun Compass Network is found also in the central complex. So Andrea during her PhD went looking for the cells in the non-magnetic electrophysiological setup and I'll come to that in just a second. I'll show you that setup. But here is a slightly more detailed view of the Bogon Moth central complex. This is the entire brain, including the optic lobe regions of the brain going to each eye, which house all of the visual processing centers of the brain. In the center of the brain, as I say, we have the central complex shown here in these colored regions. Here blown up to see the different regions. I won't go into these different regions at all, apart from to say that it's a very sophisticated part of the brain which houses a great many different types of neurons that have different roles in this compass and steering network. Here is the non-magnetic electrophysiological setup during its construction. So again, it's looking a little bit messy, but you can see everything quite clearly. It too has a set of magnetic coils around it for in the case of the experiments that Andrea did to also null the Earth's magnetic field so that we have a magnetic field inside the electrophysiological setup. We have a Bogon Moth, which we place on the stand there and we impale cells in the brain with electrodes and using what you don't see in this particular diagram or this photo here is that we have not yet installed a circular screen and a projector and a mount for the projector above the coils where we can also project the starry night sky above the moth and rotate it just as we did in the behavioral setup. So the two setups are really quite similar except one does behavior and one does electrophysiology, but on the stimulus side, they're identical actually. And Andrea's work and we've got a lot of data at the moment that Andrea is still analyzing that we're hoping to write up rather soon, but here is some of the earlier data that Andrea has analyzed showing a typical, in this case, unidentified central complex. So we don't know which one this is. The die injection didn't work for this, but it shows action potentials from the cell in response to rotations of the night sky, either clockwise or anti-clockwise in this case. And what you see is a quite significant change in the frequency of action potentials during the course of that rotation. Here you see, for instance, rotations of the starry night sky versus rotations of the control sky, that random star sky which we use for the control in the behavioral experiment. Andrea has also used, even in the electrophysiological experiment, what you see here is that there's a very clear direction of rotation or particular rotational or angular position of the night sky, which gives a maximal response from the cell. In this particular case, there's another region which gives a minimal response. And Andrea's now analyzed a lot of these cells and found very many similar types of cells that do similar things. And here is a kind of a summary of some of those cells, this is an earlier slide, and here you see five such cells which have also been injected with dye and identified. And here you see their action potential, their spike frequency as a function of sky rotation angle, both for the natural starry sky being rotated in the darker curves and these paler blue curves show the response of the cell to a rotating control sky. Experiments, I haven't got time to show you now. Andrea's also done a lot of control experiments where we've looked at different types of stimuli like bars and blobs rotating above the animal as well as controls and we're still analyzing that data. But the take home message from all of that is that these cells truly seem to be tuned to milky way like stimuli and respond very strongly to them. And we think that they're somehow involved in this compass circuit that's analyzing stellar information. And we're still working on exactly what that circuitry looks like and it's probably going to be a project that we'll go for some time yet. So to finish or to conclude, Bogol moths seem to have an ability to use the starry night sky in Australia as a true stellar compass to navigate. And as far as we know, only two other groups of animals are known to be able to use stars for navigation. And those are humans, ourselves, who are partner stars for navigation. And also some species of night migratory songbirds are able to use constellations of stars and particularly the pole star for navigation at night. So this is pretty amazing, I think, that an invertebrate with a tiny brain and extremely small eyes, the brain has a volume of a grain of rice but despite that, it's able to not only sense the Earth's magnetic field and possibly use that as a compass, but it's even able to use the stars to navigate as well, which is a pretty amazing thing for such a tiny animal. And apparently for many people a very annoying and rather drab animal. But for me, it's one of the most amazing insects I know. So finally, and this is all just a hypothesis, of course, it could be the case that Bogong moths combine the types of sensory information that they've got access to at night, information from the stars, information from the Earth's magnetic field and even beacon information from distant landmarks which they can follow until they no longer become useful if they pass them or they lose sight of them, they can hook on to new landmarks as they become available after calibrating them at the say against the magnetic field of the Earth. So we think that these three things together, probably create an extremely robust navigational system at night when conditions are very difficult for a migratory animal. They don't have the benefit of having something like the sun as an incredibly reliable and constant type of cue that a day active animal has access to. They need to make use of things which are a little less reliable both in terms of visual reliability and in terms of having to use a magnetic compass which seems to be also a tricky thing to use as found with many other animals, but that's another story. So finally, the conclusions then that Bogong moths seem to be able to sense the Earth's magnetic field and use it in conjunction with visual landmarks to steer migratory flight. They also seem to possess a true stellar compass for navigating in their inherited migratory direction during each season, both towards the mountains in spring and away from the mountains in autumn. They also possess visual cells in the central bain and central complex which respond exclusively to rotations of the Australian night sky. And of course, those rotations, I forgot to mention, obviously the sky isn't rotating above them when they're stationary. The only reason they experience rotations of the sky is that they themselves are probably rotating around their dorsal body axis if they change direction. So those rotations are potentially useful to them to correct a course that they may have deviated from. And finally, the last conclusion is that together, the stars, the Earth's magnetic field and temporary visual landmarks very likely create a robust compass system for long distance navigation at night in this animal. And with that, I thank you all very, very much for your attention. And of course to European Research Council, Air Force Office of Scientific Research, our own Science Council here in Sweden and our local Royal Society here in Lund, the Royal Physiographic Society for their support of all of this work. And of course, a great thanks to my many co-authors who are the ones who've done the great line share of the work I presented in this talk today. But thank you very much and I'm very happy to take any questions. Thank you very much, Eric. Such a fascinating talk and such an incredible task that these small-bodied, tiny-brained guys are managing to perform. And I wonder if bio-inspired solutions that try to be neuromorphic and stick to the size constraints will ever be able to compute something like that. Just to remind to our audience that they can post their questions in the chat or they can ask them in person, maybe in 10, 15 minutes from now in the post-talk Zoom session that we always hold. I would like to thank you once again. And it looks convenient to have some pieces of land if you want to build your own lab on site, literally on site. I have a couple of questions of mine, but I will stick to the ones that appear in the chat for the time being. First one is from Tom Badden. Your migratory map suggests that different populations of moths travel to different bits of Australia with a different angle relative to all landmarks is following these angular offsets in heritage. That's an extremely good question. And actually is the topic of a PhD thesis in Australia right at this moment. So I have a student, a fellow called Jesse Waller set the Australian National University who as one of his main projects actually is to do a populations genetics study on populations of moths from different breeding areas. So he's looking very much at the two extremes. So populations from extreme Western Victoria which travel more or less eastwards to get to the mountains and populations in South Eastern, sort of North Eastern Australia, a North Eastern New South Wales and South Eastern Queensland that have to travel more or less Southwest to get to the mountains. So the answer is that that must be inherited exactly how we don't know yet, but it could be some kind of epigenetic thing as well that is inherited. But I'm just talking through my hat here because we don't know yet, but it's a good question and we're certainly following that up. Mm-hmm. Next one up is Simon Laughlin. Do the moths determine the direction of their track over the ground to compensate for transverse winds? Oh, that's a really great question. Yeah, unfortunately we didn't get the money for this but I had a post doc who is an expert on using lidar techniques for detecting insects across the ground. And we had a plan to actually look at some of these questions with vertical lidar to track them in natural conditions to try and actually answer exactly these kinds of questions. How well are these moths able to navigate in real world conditions with real winds that we know the directions of? I mean we can work out where they should be going but then we need to be able to find out whether they actually are able to hold these directions and what they do when they're faced with adverse conditions like this. So the answer is Simon, I'm afraid I don't know at the moment of hand what the answer to that question is. I assume they can, but exactly how well and how they do it, I'm not sure. Thank you very much for addressing this. I should clarify there are a lot of people congratulating you for the talk and thanking you for the talk. I assume you cannot see it because you are not logged in in the YouTube. Thank you. Next one is like actually three questions from Gregor Belucic. I will start with the first one chronologically. Amazing, congrats. Is the EFIS setup considered as non-magnetic in spite of the Thorlabs anti-vibrational table below the coils? Don't the potentially ferromagnetic parts below count? No, there is, but that isn't actually just to answer Gregor, but that isn't actually a anti-vibration table. They're made of aluminium parts. Every single piece is made of aluminium or high grade marine stainless steel. Yeah, so now we've been very careful about that. We haven't actually used an anti-magnetic, I saw anti-vibration, special anti-vibration breadboard here, which do have magnetic parts in them. That's true. What we have done though is searched on the internet for damping devices, which are non-magnetic. And we found these extremely expensive damping devices, which high-fi freaks put their highly expensive amplifiers on so that people who enter the same room as they are sitting in listening to music don't cause vibrations that disturb the music. These things are very expensive and they're incredibly damping of vibrations and they're totally non-magnetic. They're made of ceramics and high grade, marine grade stainless steel. So that's how we've got rid of the magnetic field, sort of magnetic artifacts in the center. So it truly is completely non-magnetic. Our manipulators that we had custom built by Senseipex in Finland are also completely non-magnetic. The second question is, have you ever had the chance to try the experiments in completely naive modes, just enclosed? Oh, I'm not sure exactly what's meant by that. In naiving the sense of, they're pretty naive because we just grabbed them out of the sky basically and do the experiments, maybe not the same night, but maybe they could clarify exactly what they need. Yeah, that's what I was considering, suggesting like Greg or if you want, like you can either clarify in the chat or join us. I already posted the Zoom room link and people start already to join us here. I will not stop the live transmission just yet. And before I continue with the third question, just a reminder for the audience that next week, we are hosting Nathan Morehouse for a talk about the evolution of vision in jumping spiders. And getting back to the third and last question from Gregor, you've mentioned that the projection plates were UV transparent. Any comments on this? Have you tried visible only stimulation? I assume. Okay, yeah, no, not yet. So if Gregor's question is asking whether we've investigated the wavelength region where the magnetic sense might be working or not working, we haven't specifically done that yet, but that's certainly planned. So the idea is that if it's a cryptochrome based mechanism, it's probably working mostly within a certain wavelength range, possibly blue. And so far we've given them broad spectrum light. So the starry night sky has got UV in it too. I forgot to mention that. So it's got all wavelengths from UV to visible. And not because of the projector, because that couldn't do UV, but we've provided extra UV from outside to make up for that loss. So the moths are able to get all wavelengths from UV to well to red, which probably doesn't see so well, but yeah, and all of it goes through that lead. Gregor clarified saying like the moth just out of the poopa. So like, naive in that sense, like. Oh, I see, right. No, that's not as easy to do because then you have to actually be lucky and be right where they are when they emerge from the soil out in the breeding areas. And I must admit, we haven't done that experiment. That would be a very difficult experiment to do probably because we'd have to get access to the moths as just prior to them hatching from the ground. I guess it's not impossible. We'd have to go up and dig them up, dig them up out in far Western New South Wales, drive them back to Adam Enneby and hatch them out there. It's an interesting idea actually, but I must admit we haven't done that now. Interesting, thank you very much. So there's another one from Tom that are already a lot of people here. So may I remind to the audience that I will be stopping live transmission in a couple of minutes. So in case you want to tag along for the informal part of our vision series, please make sure to follow the link that appears in the chat. Before I ask Tom's question, I have one of my own. So I would like to step in. Maybe it's naive, but you tried to have controls for the Milky Way and you tried to rotate it 180 degrees. But did you also try, like when we talk about the spring migration, did you try to present the autumn Milky Way to see if we have more sense of integration? Yeah, no, we haven't come that far. These are a lot of experiments that we've wanted to do, but because of COVID, we haven't been able to go back. So many of these things that you've just mentioned now, Georgia, are things on our books to do, but we've just missed our second two-month field season. And we're supposed to be going back in October, but I really am not sure we can. So, yeah, no, it's getting a bit frustrating. I have to admit it's a bit depressing, actually. Yeah, so the answer is no. And all of these experiments are obvious. The other experiment we want to do as well is obviously now to include the Earth's magnetic field and find out whether together the stars and the magnetic field together give a tighter distribution of moths in the setup than we've seen with the stars alone. Maybe it's the case that the moths are much better aligned and with less spread than with the stars alone. And then we can also try, if we can do this, to try and have the magnetic field alone. But that's gonna be more tricky to do because if they latch onto any tiny little visual view, then we may end up with the problem we had earlier with our two years of failed experiments. So returning back to Tom and his second question that attempts to generalize, like to contextualize also everything, you mentioned that only these moths, some birds and humans use stars to navigate. Can you speculate if this is because we haven't looked broadly enough or do you think it is really this exclusive? It could be either, I don't know. Well, now that an insect can do it, I mean, it's not necessarily out of the question that other invertebrates could do it. I guess it's a case of finding an animal that would get some use from it. So, I mean, being able to use the stars to navigate like this makes sense if you have to travel a long way at night. And so any animal that does something like that has to travel a very long way in a very specific direction at night is a candidate for this. And the question is what animals would they be? I must admit off the top of my head, I can't think of an nocturnal insect at least that is doing this. I can think of a few more day active ones. So there are some other moth species that may do it actually. There's a moth in North America that does a shorter migration of this type. And one in Japan as well, also noctuids that does a shorter version of the same thing. They may have some use of it. But again, they would be another noctuid moth. So I don't know whether you find it outside of noctuid moths or not. You'd have to find an insect that had a behavior like this that would be worthwhile having a stellar compass for. I just saw that Tom mentioned here in this Zoom room chat that Xenopus, they have huge migrations. Oh, is that right? Aha. Okay, I thought so, I assume. Yeah. Great. So I think I will be stopping the live transmission so we can stay offline like for this post-doc discussion. Thank you very much once again, Eric, for this wonderful talk. And thank you very much to everyone out there for following our Sussex Vision series. As always, within the Worldwide Neuroinitiative, make sure to join the Zoom room if you wish the link is available on your screens. Thank you. Thank you very much, George. And thanks everyone.