 Close the door. OK, so welcome everyone to this colloquium. It's always a pleasure to have an outstanding scientist visiting us and giving colloquia. But this particular one, the abstract, at least, the title is particularly interesting. Let me just remind you, my name is Sandro Scandal. I'm representing the director of the ICTP. Fernando Quevedo is not in town today. So let me just remind you the rules of the game. At the end of the colloquium, the students will be asked to stay in the room and to have a private discussion with the speaker, only the students. In the meanwhile, there will be light refreshments served outside of this room. So I'll now give the floor to Antonio Cellani. We'll introduce the speaker. Thank you very much, Sandro. So welcome, everybody. My name is Antonio Cellani. I'm a scientist here at ICTP in the Quantitative Life Sciences section. It's a pleasure for me to be here today to introduce Ana Gallardo, our colloquium speaker today. So Ana is an ethologist. Ethologists are biologists who study animal behavior. And the animal behavior she's interested in is navigation, that is the ability of animals to find their way in space to locate some particular position. Specifically, she's interested in birth navigation. And if you don't know, you will discover during our talk that births are able to accomplish incredibly difficult tasks. For instance, the humble pigeon can retrieve its way back home from a place that it has never visited before and which could be tens or hundreds of kilometers away from his home. How can births do this thing? They use a variety of inputs. So they use internal clocks. They use the position of the sun. They use the Earth's magnetic field. They use visual cues. They use last but not least, or factory cues as well. And the sense of order and how it informs the decision-making process that births implement in order to get their way back home is one of the most fascinating, I think, and still open subjects in the domain of animal navigation. So we're looking forward to hearing more from Anna on this. Thank you. So thank you to Antonio for inviting me, for having invited me to give this talk. And thank you for coming and for being interested in this subject. I confess that I'm a bit nervous because I think that this colloquia are so, for very important people who I am not. But I try to entertain you and to give you, hopefully, interesting information about what I've been studying so far. So I have to introduce the topic because Antonio told me that you have a different background, and probably none of you or just a few are biologists. So I have to start from the beginning. So as Antonio said, births are able to find their goal, even if displaced in an area which is completely unknown, completely unfamiliar to them. So how can they do it? In order to navigate, a bird must rely on a sort of map that tells the bird where it is with respect to the goal. So there are two steps to accomplish in order to be able to navigate. The first step is a map step. So where am I? I'm here. I have to go there. OK, then the second step is the compass step. So which is the direction in space that I have to take in order to reach the goal? So there are two different mechanisms, the map step and the compass step. And this is important to understand that because very often people confuse them. A model for the study of navigation of the mechanisms underlying navigation and also the neural basis of navigation is home impedance. Home impedance are not migratory birds. They are a domestic strain, but they are important because they are able, indeed, to navigate from the compass mechanisms. There is evidence that animals, not only birds, can use a variety of absolute reference for orienting, such as the Earth's magnetic fields, the stars, the moon. However, the most widespread and well-known biological compass is the sun compass. So animals and specifically birds can determine and maintain a direction while flying using the sun azimuth as a reference. However, you can determine a direction, a compass direction using the sun azimuth only if you are aware of the time of the day because of the apparent movement of the sun in the sky. So the sun azimuth predicts a direction only if you know the time of the day. And, in fact, the sun compass is a chronometric compass. And birds, as many other animals, possess an internal clock, making the animals aware of the time of the day. You can have a demonstration, a practical demonstration of what I am saying, by shifting the internal clock of the birds. So if you take a bunch of birds, a bunch of pigeons, you put them in a room, and you subject them to a shift of their internal clock by simply switching on the light six hours, for example, six hours before the sunrise, and putting the light off six hours before the sunset. Then you subject the birds to a clock shift. And what do you observe if you release those birds? That they make an error in their orientation. So they deflect their orientation with respect to the goal of a predictable angle. So the deviation is counterclockwise if the birds have been fast shifted or clockwise if the birds have been slow shifted. The first step, I've been talking about the first step, is the map. So why I decided to talk about the compass first? Because the story of the map is much more complicated. And I'm going to talk about the map basically for the rest of the talk. So which are the cues used by the map? So however, the can determine its current position with respect to the goal, I'm sure that if I ask you, all of you will say, well, the Earth magnetic field. But unfortunately, although it's very popular to say that birds use a magnetic map, if you look at the evidences, they are not so consistent and no direct evidence for supporting the magnetic map. Because if you disturb the birds with magnetic manipulation, so you mask the Earth magnetic field, for example, by treating the birds with strong magnets, even mobile magnets, or if you put around the bird's head Elmo's coil, so to produce oscillating magnetic field, the birds home not differently from the control birds. So there is no magnetic manipulation so far being able to disrupt the navigational abilities of birds. So the support for the magnetic map comes only from experiment conducted in small cages on migratory birds that change their orientation if the magnetic field is manipulated. But no tracks, no direct evidence. This is the reality. The second hypothesis predicts that birds use environmental orders for navigation. And you may think that this is a crazy idea, which doesn't make any sense because orders are so difficult to imagine for us, for humans. And it may look quite odd hypothesis unless one consider how this hypothesis was formulated and why. So Floriano Pappi, who was a professor at the University of Pisa, in the early 70s, proposed the so-called olfactory navigation hypothesis. And one has to consider how he proposed this hypothesis and which facts, which evidence stimulated him in formulating this hypothesis. In the early 70s, the hypothesis proposed to explain birds navigation had not found any experimental support. There was the Sainarca hypothesis that failed under the experimental test. And also, other hypotheses like the magnetic map hypothesis was not actually so convincing. So at the time, Hans Varraff, who was a German researcher at the Masblank, was performing experiments in order to test whether the view of the horizon was important for the development of navigation abilities in young pigeons. So by comparing the performances of birds raised confined in different screened aviaries, he discovered that, contrary to any expectation, the birds were impaired if the screens were made of glass so that they were transparent screens. And the birds were not impaired if the screens did not allow the view of the surroundings, but they were penetrated by winds. So if the view of the horizon was allowed but the winds were screened, the birds were unable to navigate. If the view of the horizon was prevented, but the winds could get through the aviary, then the birds were unimpaired. So Varraff concluded that there was some dynamic factor in the atmosphere important for the development of the navigational abilities in young pigeons. And Papi read his paper. He read this paper, and then he thought, well, the dynamic factor, the dynamic factor can be others. So he made a very simple experiment. He took a bunch of pigeons. He made them an osmic by cutting the olfactory nerves. He displaced them far away from home. He released them together with controls which have been also surgically manipulated, but with the olfactory nerves intact. And he found that an osmic pigeons were unable to home. So putting together these two facts, these two evidences that pigeons needed the exposure to the winds at the home loft, and that pigeons needed to smell in order to be able to navigate, he formulated the olfactory navigation hypothesis. The olfactory navigation hypothesis predicts that there is a learning phase in which the pigeons learn to this association between the wind direction and the odors carried by the wind. So there is an association between the directional information, which is the direction of the winds, and information as regarding the quality of the odor. So this is the learning phase. Then there is an operative phase of the map. So when you take birds and you displace them at the release site, then they recognize the local odors, they recall which was the direction from which these odors arrived home, carried by which wind. And then they orient in the opposite direction. So the olfactory map is not a topographical map. So birds are not aware about where they really are, but they are aware about the direction of displacement. So the only information that the olfactory map can provide the birds is I've been taken north from home. I have to go south. So it's the direction of displacement, or I've been taken east home, and then I have to go west. There is no, it's not a mechanism like the dog and the sausage. The dog smells the sausage and goes towards the sausage. So it's not a plume, the mechanism. The mechanism is something more complex because it requires a learning phase, but also quite rough because it's only able to give you an idea of the direction of displacement. So before going on in this talk, I would like just to give you a basic information about the method for the study of pigeon navigation in the field. Before the GPS era, many studies have been done. And in those studies, the experimenters were able only to record the vanishing bearing of the birds. Typically, one jumped up a car with the chair with a powerful binocular, then released a pigeon, and then followed the pigeon with the binocular. And then when the pigeon vanished from the observer's view, the observer could take the compass direction of the birds. And this is the vanishing bearing. Then you can take the vanishing time, and then the omen time and the success. The omen success of the birds. Nowadays, there are GPSs. So we have GPS that you have to GPS loggers that you have to get back in order to download the track. But you have also the possibility to use GSM GPS. And so you can download the tracks, even if you lose the bird. And in this case, you can make lots of analysis on the track. And of course, you can really see where the bird is going. So let's go back to the olfactory map. So after having formulated the olfactory map hypothesis, Papi made a series of ingenious and elegant experiment in order to test the role of the winds during learning. And so he could show that manipulation of the information provided by the winds during map learning affected the orientation of the birds. So for example, in this slide, you can see that if the birds were exposed to inverted winds by using some fans, so when the wind was blowing from a direction, a fan on the opposite side switched on so that the natural wind was sent inside, was blown inside the cage from the opposite direction. And this treatment produced an inversion of the initial orientation of the birds. And if you took birds from a normal aviary and you kept them for several months in an aviary provided with wind deflectors, then also the initial orientation of the birds was deflected consistently. Papi made also an experiment showing that pigeons might be able to incorporate artificial odors in their map. So using aviaries like this one, cages like this one, controlling the winds and providing odors, artificial odors, in the air currents to which the birds were exposed, it showed that pigeons oriented in the opposite direction of the odors air current when they were taken away from home in a direction which was not the direction of the winds. And then they were stimulated with the same odors that experienced during map learning. This regardless the own direction, so it showed that maybe pigeons may be able to incorporate artificial odors in the map. So the map, it has been shown by several experiments that the navigational map mechanism is a plastic mechanism. In fact, pigeons can learn a second map if they are taken to a second aviary and kept there for a sufficiently long time. And then that pigeons can update the map, so that pigeons that have a normal initial orientation consistent with the own direction can deflect counterclockwise or clockwise depending on whether they were kept in an aviary provided with deflectors. The olfactory navigation hypothesis predicts that local environmental odors perceived during transportation and at the release site constitute critical information to determine the direction of displacement. And this is consistent with the dramatic impairment observed in birds released at unfamiliar locations after they have been subjected to section of the olfactory nerves. The interesting thing is that, for example, for this experiment, if you cut another nerve, nothing happens to the birds. So it's not a no specific effect of nerve section. Because if you cut the olfactory branch of the trigeminal nerve, birds are perfectly able to orient and home. And why the trigeminal nerve? Because it has been made an hypothesis that the oftalmic branch of the trigeminal nerve has been proposed as a nerve carrying to the brain magnetic information. So we tested the involvement of the oftalmic branch of the trigeminal nerve in pigeon navigation. And we observed that there is no impairment after the section of the trigeminal nerve. I told you at the beginning olfactory navigation was raised a hot debate because people was quite skeptical. Although the large body of evidence in favor of the olfactory navigation, there is always a debate about the plausibility of olfactory navigation. And nowadays, nobody denies that an osmic pigeons are unable to home, also because there are tracks showing that. There are tracking data. But nowadays, there is another alternative hypothesis, alternative to the olfactory navigation hypothesis, which has been called the olfactory activation hypothesis. According to this hypothesis, birds do not need olfactory cues for navigating. But they need only olfactory stimulation of any kind, so even olfactory stimulation with nonsense odors, so odors which are completely novel to the birds, in order to get their brain activated. So according to this hypothesis, there is no olfactory map. But there is a role, the other stimuli have simply a role in activating the brain. In activating the brain, and then the birds actually use a magnetic map to home, or another kind of unknown stimulus. So I wanted to test this hypothesis. We have tested this hypothesis by stimulating, by subjecting the birds to different stimulation during transportation and at the release site. Since the olfactory activation hypothesis was proposed only on the basis of vanishing bearings data, we wanted to test the olfactory navigation hypothesis with tracking data. By tracking the birds. And so it is relatively easy to control and manipulate the olfactory experience of the birds during transportation and at the release site. But it transported the birds in airtight containers so that you can deprive them of olfactory stimuli at all. Or you can, for example, allow the birds to smell environmental odors. Or you can allow the birds to smell only artificial nonsense odors by transporting them in purified air and by stimulating them with artificial nonsense odors. So according to the olfactory navigation hypothesis and to the olfactory activation hypothesis, if birds are stimulated with environmental odors, they should be unimpaired. But if you stimulated with artificial odors, then according to the olfactory navigation hypothesis, you should observe an impairment. While according to the olfactory activation hypothesis, you should not observe an impairment because you stimulate the birds with odors. So it would be sufficient for them to activate these mysterious non olfactory navigational system. So we did experiments by I show you only the data by comparing the behavior of pigeons stimulated with environmental odors and pigeons stimulated with artificial odors. So in this experiment, the birds were transported in these containers. Then they were differentially stimulated, one group with environmental odors, one group with artificial odors. And then before released, they were made an osmic with xylokine, which is a local anesthetic. You can anesthetize the olfactory mucosa with this anesthetic. So for about one hour, these pigeons don't smell. And so they have to rely on the information perceived why they were in the containers. You can see these are representative tracks. Please look at the yellow track because it refers to the track recorded during the first hour after release when the xylokine was presumably still active and still anesthetizing the mucosa. So you see that the birds exposed to environmental odors headed home. But the birds exposed to artificial odors were completely disoriented. In a second experiment, we used the same procedure. But this time we made the bird anosmic by washing their olfactory mucosa with zinc sulfate. Zinc sulfate induces the generation of the olfactory neurons. And so it's a long-lasting anosmia. So the anosmia does not last only for one hour, but lasts several months till the olfactory mucosa does not regenerate again. So these are the representative tracks of the unmanipulated controls. And these birds headed home nicely as expected. These are the birds transported to environmental air and then released anosmic after this zinc sulfate treatment. And you see that most of the birds show an oriented track. And the mean vector distribution is oriented towards home because home is included in the confidence limits of the mean vector distribution. Nevertheless, as expected, on the basis of previous data, these birds were unable to were impaired at homing. Because of course, birds need also factor information on the way. So they were oriented towards home at the beginning, but then they got lost because on the way they need to consult them up again. And these are representative tracks of the birds stimulated with artificial odors. The interesting thing is that most of the birds did not show oriented movement and therefore were impaired. So with these two experiments, we show that actually environmental odors allow homeward orientation. And artificial nonsense odors, which were not learned in association with the wind direction, do not allow homeward orientation in pigeon. Let's go back to talk again of the main feature of the olfactory map. So the proposed olfactory map mechanism can inform the birds about the direction of displacement without giving other details. For example, such as the distance from home. So the olfactory map is not expected to give information about the distance from home. So this implies that a bird can orient towards home by relying on odors. But once the home area is approached, then they have to shift to another mechanism. And Varraf proposed a model, which I showed in these slides, in which the olfactory map is inefficient close to the home area. And so he predicted that they have to use visual familiar cues learned during these spontaneous flights around the loft. We wanted to test this model by using GPS loggers. And we confirmed Varraf prediction. We did this experiment. So we tested two groups of birds. One bird was kept, was raised inside an aviary, open to winds so they were able to learn an olfactory map. But they had a limited view of the surrounding because they could not fly out. They were kept in this aviary. And then we also had a second group of birds, which were raised as free flyers. So this bird could explore the home area and have an extensive view of the surroundings. Then we took all the birds and we released them one by one at the release site, which was located north from home. So the free flyer birds reached home without any problem, reached home soon, once approached the home area. While the birds raised confined, oriented towards home. But then they missed home even if they were very close to the home area, very close to the home loft. Look at, for example, these zoomed images. Some pigeons even could not recognize the loft, even passing over the loft. So this means that they need the visual cues, learned, previously learned visual cues to reach the loft. The last phase of the homing process is not mediated by olfactory cues, but is mediated by visual landmarks. We have been able to show that pigeons know where they are even before takeoff. So even when they are there in a cage. In the early 70s, this was observed already by Kelezi and Pardi. They observed that pigeons are able to orient towards home in an arena at the release site. And we exploited this ability to study the relevance of both olfactory and visual cues at a familiar release site. So we tested the bird's orientation before takeoff in an arena like this. It's very easy because you train the birds to familiarize with this arena. And then once you take them at the release site, they exit from the arena roughly in the home direction. So we're able to test the behavior, the orientation of both control and an osmic pigeons, released from three locations, three different familiar locations before takeoff. And we saw that if the surroundings view was lowered to the pigeons, they oriented without any problem towards the home direction. When we screened the view of the familiar release site so the pigeons could not see, which was the site we took them to, then the controls were still able to orient towards home. But the anosmic birds were completely disoriented. So they didn't know which was the site we took them to. This means that at a familiar release site, pigeons can rely on both olfactory cues and visual familiar cues. But if we don't allow the pigeons to view the landscape, the familiar landscape, then they need olfaction to understand where they are. We've seen that birds use olfactory cues for navigation. So a necessary step in the research was to investigate the role of the telephonic regions involved in processing olfactory cues. As the major projection field of the olfactory bulbs is the piriform cortex, then several studies aimed at testing the role, the involvement of the piriform cortex in pigeon navigation. And so we saw that bilateral lesion to the piriform cortex do affect pigeon's ability to navigate home. Because both lesion performed before and after map learning impair navigational abilities of the birds. But we had other evidence of the involvement of the olfactory system in pigeon navigation. And these evidence come from studies of the immediate early gene expression study. So immediate early gene, maybe all of you know, but are a variety of genes which are activated transiently and rapidly in response of several kinds of stimuli. So this can be visualized with antibodies. And therefore, we can see what the birds, what the brain was doing during a certain period of time before killing the pigeons. So in this experiment, we compared pigeons subjected to three different treatment, to three different conditions. So there was a group of birds transported to an unfamiliar release site and then released. A second group was transported to the same unfamiliar release site, but then was not released. The birds were taken home and then sacrificed. And then there was a group of birds transported 200 meters from the loft, so around home, and then released. So these birds were not supposed to be to use olfactory navigation because they were in the Omer. The results were quite clear. The piriform cortex of the pigeons released was maximally activated. We observed the highest activation in this region. While the lowest cell density of zinc-activated cell was in the released at-home pigeons, so the birds which were not transported away from home. There is one important thing to observe, that we have some quite relevant activation also in the pigeons, which were transported to the release site and then kept in the cage without being released. And this is consistent with the previous data of the circular arena because it means that even before takeoff, pigeons thought, well, I'm here. I have to go there. So they realized where they were by olfaction. Other kinds of studies we performed to investigate the neural bases of olfactory navigation were lateralization studies. So I think all of you know that the two brain hemispheres are not identical. They look identical, but they are not because there is a functional specialization of the two hemispheres. For humans, it's quite evident because most of people use better the right hand than the left hand. And there are some people using better the left hand than the right hand. So it's very rare to find people using both hands at the same level. And then there is the fact that, for example, the Broca area is only in the left hemisphere and so on. So in humans, lateralization was discovered. And it was thought to be a phenomenon restricted to humans for a long time, while instead, actually, the brains of all vertebrates are lateralized. And there is a variety of function in which the hemisphere of vertebrates are specialized. So it was interesting to study lateralization in pigeons restricted to the olfactory system because it's one of the few cases in which we can investigate olfactory lateralization in a model like home impigeons of a bird that can be released in nature so we can study this behavior in the field. It's a great opportunity to investigate functional asymmetry of the olfactory system in a natural setting. First of all, since bilateral lesion to the piriform cortex produce an impairment in navigation from unfamiliar locations, we wanted to test whether what happened if we produced unilateral lesions. And we observed that there is a functional asymmetry in favor of the left piriform cortex because if we lesion the left piriform cortex, the pigeons were completely scattered. While if we lesion the right piriform cortex, the pigeons were more oriented. There was some impairment in the homing performance, but at least the initial orientation was there. We went on and we made also experiment by simply plugging one nostril. And we expected to observe a consistent impairment to that observed with the piriform cortex because the major projection are to the y-lateral from the olfactory bulb to the y-lateral piriform cortex. And instead, we saw we had a surprise because this time the left-plugged bird were unimpaired and the right-plugged bird were impaired. We observed this impairment of the birds with the right nostril-plugged, also with GPS experiment performed with experienced birds, while the previous one were inexperienced. These were experienced birds but released at a non-familiar location, of course. And we observed that after nostril-plugged birds were as good as controls in orienting towards home and they did not show any difference with respect to controls. While the birds with the right nostril-plugged showed more tortuous path, and also you can see that they are different at first glance. And then we analyzed some parameters, like, for example, the number of stops per kilometers. And we observed that the birds released with the right nostril-plugged, stopped more frequently than controls. While those with the left nostril-plugged did not. Also that's a, sorry, I'm sorry, I went too far. Also that the right nostril-plugged birds showed a more tortuous path. So in the end, we observed this effect that the left piriform cortex was dominant with respect to the right piriform cortex. And we observed also that the right nostril, the right olfactory mucosa, was dominant with respect to the left. So if you look at the anatomy, you see that there is a group of fibers crossing. And probably that group of fibers is responsible of this behavior. I've been talking about olfactory navigation in home impignts. But as we said, the home impignts are not wild birds. They are a domestic strain. And so it's important to assess whether olfactory navigation may be important also for wild birds. First time this question was addressed was in the 70s, in the early 70s, soon after the olfactory navigation was discovered. And the test was performed on swifts. So Fiaschi and colleagues captured the swifts and made an osmic only on one nostril, all those swifts. So then a group of them was released with one olfactory nerve cut, and a walk's plug placed ipsilaterally to the nerve section. And these were the controls. The other group received exactly the same treatment, so one olfactory nerve cut. But this time the plug was contralaterally. And these birds were an osmic because they were unable to smell from the plug and the nostril. And they were unable to smell from the olfactory nerve cut side. And the results were quite clear because the control birds, almost all the control birds could home back. And only three birds, which all lost their plug, could home. So they observed an impairment in swifts. Later, also Barraf tested olfactory navigation in starlings. So he observed that if the distance of displacement was a medium distance from the nest, they could reach home because probably they were released in a family area, in a generally family area. While if the birds were taken away more than 100 kilometers away from the nest, or even more, then they were impaired. So these two experiments suggest that olfactory navigation might be a widespread phenomenon in birds. It might not be restricted only to home implications. Two tests were performed on birds which were challenged to home to their nest. What about migratory birds? Are they able to navigate to their migratory corridor for example? They don't have a nest yet because they release their nest and they are performing an auto migration, for example. So they have to reach their wintering grounds. Are they able to navigate if accidentally or experimentally displaced? The first scientist who tested these navigational abilities in migratory birds was Perdek about 60 years ago. He captured with the nets 200,000 starlings. And then he displaced them during the, I'm sorry, but you cannot see what I can see. There is some, I cannot see. There is some problem. But in any case, he captured this 200,000 starling during their migration and he displaced them from Holland to Switzerland. And what he observed? He observed that while the other birds, he observed by recapturing them because there are lots of ornithologists in Europe capturing birds with the nest. And he observed that adult birds were able to reach their migratory corridor. And young birds at their first flight continued in their migratory direction as if they had never been displaced. So this means that young birds at their first flight have a sort of vectorial navigation. So they don't really navigate. They have an innate migratory direction. They follow this direction. And if you displace them, they are unable to go back to their migratory corridor. They go on in following that innate migratory direction. The adult birds are able to navigate because if you displace them from their migratory corridor, then they correct and rejoin the migratory corridor of their population. So I'm sorry, I can't. For some reason, ah, yes. OK, no. Yes, exactly. That's what I was going to show you. So for a long time, all factory navigation was forgotten. In wild birds, it was forgotten by all scientists. Till a day, Richard Holland and Martin Bickelski invited me to participate in an experiment conducted in the States on cat birds. So we displaced the cat birds from Illinois to New Jersey. So it's a long, long journey, more than 1,000 kilometers. And we displaced both juvenile and adults. And we subjected the birds before displacement. We subjected the birds to two different sensory treatment. We had unmanipulated controls. We had an osmic birds made an osmic by zinc sulfate washing. And then we had birds subjected to magnetic pulse. So we performed a very strong, we gave the birds a very strong magnetic pulse of short duration. And this pulse is intended to rotate the polarity of the magnetite. And people working on the magnetic map of birds claim that if you perform this treatment on the birds, their magnetic map is affected. So we repeated this treatment on these birds. And these are the results. So the juveniles had on in the same migratory direction as predicted. So the juveniles were not able, as predicted by PEDEC experiment. So the juveniles were not able to navigate and to correct for displacement. The adults subjected to magnetic pulse. And the controls were able to deflect towards the migratory corridor and to compensate somehow the migratory direction. We provide these birds of small transmitters. And then we could relocate them by flying around with the Chesna and also by some antennas which were placed in New Jersey, in several places that could detect the presence of these transmitters. So we were able to relocate the birds for hundreds kilometers. It's not really tracking, but it's somehow a sort of a large distance vanishing bearings. Because you follow the birds hundreds kilometers away. So the controls and the magnetic pulse at birds seemed to be able to compensate. And the anosmic birds oriented as the juveniles, which are not able to navigate. So this experiment encouraged us to go on testing all factory navigation in wild birds. And this time we chose this lesser black-backed gull for as a model. So these birds have extraordinary migration from the breeding area in Finland up to Lake Victoria. So they have a very long migration. And we displaced them to Helgoland. So for these birds, the Nile Delta is a very important stopover. All the birds performing these long migrations stop in the Nile Delta to eat, to rest. And then some of them stay there all winter, while some others reach Lake Victoria. So this slide shows the spring migration towards Finland, towards the breeding area, and the autumn migration into the winter inside. We tested and we provided with satellite transmitters birds caught in Finland and displaced to Helgoland. But before displacement, we subjected the birds to sensory manipulation. So we had intact controls. We had the birds subjected to trigeminal nerve section, still because people claim that trigeminal nerve is involved in magnetoreception. And then another group subjected to olfactory nerve section. And we observed that the birds, these diagrams show the mean orientation of the birds. And so both controls and trigeminal section at Gauls oriented towards the goal, the Nile Delta, which is the first goal for these animals. So they displayed the ability to correct and to reach their migratory corridor. By contrast, the Dianosmic Gauls oriented towards the migratory direction, and most of them ended up in the desert and they were completely, completely lost. So therefore, Dianosmic birds seemed to be unable to correct for displacement. And this slide showed that the mean distance from the migratory corridor was greater for Dianosmic Gauls. So they kept themselves more distant from the migratory corridor than the other two groups. Other models to test olfactory navigation are the proscilariformes. Because these birds are seabirds using olfaction for a variety of behavior, for example, food localization. And also nest localization, because they nest in burrows or in caves in the dark. So sometimes in a cave, there are four or five nests. And in the complete dark, they follow the olfactory stimuli to recognize which is their nest, their own nest, and not the other one. They have a very well-developed olfactory apparatus. They have larger factory bulbs. They have big nostrils. They are also called tubinares, because the nostrils are made like tubes. So we asked the question, can they perform olfactory navigation? And to answer this question, we tested long-distance navigation in birds subjected to three different treatments. Again, unmanipulated controls. Then birds subjected to magnetic disturbance. We glued a box on the head of the birds with a very strong magnet, which was free to tumble inside the box, producing an unpredictable magnetic field. And then we performed nasal washing to the anosmic, to the birds which were intended to be anosmic. We tested the shirwaters in two different species of shirwaters in two different environments. The Atlantic Cori shirwaters live in the Atlantic, and we displaced them in the open ocean 800 kilometers away from their colony. While the Scopoli shirwaters, which is a Mediterranean species, nesting and breeding in the Mediterranean, although they are also migrating outside the Mediterranean in the Atlantic Sea along the coast of Africa, but they breed inside the Mediterranean. And so we displaced them far from the coast in a place towards the on the route to Barcelona. We took a ship to go there, of course. So this couple of shirwaters nests in the Tuscan archipelago near Livorno. Let's see what the Atlantic shirwaters did. So you can see that the tracks of the controlled birds and of the birds bearing strong magnets. So you can see that the tracks are amazingly directed towards the colony. No problem also for the magnetically treated birds. But look at what the anosmic did. So some of the birds made the anosmic show a very torturous route. Others ended up to places that had nothing to do with the Azores. So they were completely confused. And the statistics confirmed what you can see from the route that the anosmic were completely scattered, randomly scattered, the magnetically treated, and the controls were homeward oriented, and that the anosmic birds were impaired at home to the colony. In the Mediterranean, sinks were different because the controls were magnetically treated, and the anosmic birds were able to reach the colony quite easily. Also the anosmic birds could reach, most of the anosmic birds could reach the colony, although after longer time and following more torturous route. So we observed that the anosmic birds were slower at homing, although they omitted, and that when we tested the initial orientation before seeing the coast, so within 60 kilometers from the release point in the open sea, the anosmic birds were not homeward oriented, while the magnetically treated birds and the controls were homeward oriented. At the tracks in detail, we observed that for most of the anosmic birds, most of the fixes were located within 40 kilometers from the coast. So the anosmic birds tried to reach home, staying close to the coast, and probably exploiting visual familiar landmarks. And then we also observed and tested another parameter. So we asked the question, did the view of the coast potentially help the birds to get home? And so, when did the bird decide to get home? After having viewed the coast, or already in open sea? So we conducted a backward analysis of each track starting from home in order to determine the point at which the bird seemed to have decided to get home. And we call it the decision point. So the decision point is the point at which the distance from home stops increasing linearly with respect to the path length. And we observed that the anosmic birds seemed to have decided to have home, to head home, only after having approached the coast, at least once. While most of the control and most of the alpha of the magnetically treated birds displayed a decision point in open sea. So they didn't need to approach the coast to get home. While the anosmic birds seemed to be able to decide to go home only after having viewed the coast. So this data suggests that olfactory cues constitute an important source of information for seabirds to navigate in open ocean. That topographical cues constitute an important source of navigational information in seabirds. And then that magnetic cues are not used by shears waters, by patterns for navigation. The experiment suggests that shears waters use olfactory cues for navigation. But this assumption raised other opening number of important question on how these birds may use odors. So we don't know whether they use an olfactory map like home impigeons. Young shears waters after after fledging go out from the nest and stay in proximity of the colony. So it might be possible that they learn an association between the direction of the winds and the odors carried by the winds like home impigeons do. But it's also possible that they use a different map like for example an olfactory landscape. These birds wander around in the ocean for many months every year and for many years before starting to reproduce. So they have a long extensive exploration of the ocean. So they might learn the different spatial distribution of the odors in the ocean because there are odors produced by the phytoplankton which reflect the topography of the sea bottom. These odors are for example dimethylsulfide which is produced by phytoplankton grazed by the zooplankton. The dimethylsulfide is not homogeneous in the ocean. It's concentrated in patches, different patches in different parts of the ocean because it reflects the bottom topography. So there are sea mountains which where the phytoplankton is more abundant and in those places the phytoplankton attract zooplankton and zooplankton attracts fish and shears waters go there to feed. So maybe it might be possible that shears waters learn this olfactory landscape and do not have an olfactory map in the sense of woman pigeons, but we don't know. So many questions about olfactory navigation. We don't know about whether sea birds use an olfactory map or an olfactory landscape. We don't know how many species of birds use olfactory cues for navigation. We don't know which are the others. They exploit as cue, potentially every other, but we don't know. We don't know how extended is the olfactory map in home impigeons and whether, for example, its extension is affected by topographical feature like, for example, chain of mountains. Let's think of the Alps of the up and nine in Italy, for example, geographical barriers. We don't know many things. We know something. We know, for example, that performing for the first time a home in flight, a pigeon become familiar with the landscape by learning the visual feature of the overflowing area. And therefore, after a direct exploration of this area, beside the olfactory map, pigeons learn a topographical map. So a visual landmark-based map and they can use visual cues for navigation. Nevertheless, it has been frequently observed that home impigeons release that familiar location in clock shifted condition, display a deviation, and so they seem not to pay attention to the visual cues. Well, the complication is due by the fact that pigeons actually have two strategies for the use of visual landmarks. They have a pilotage strategy or a site-specific compass orientation strategy. So in a site-specific orientation strategy, pigeon recognize the local release site, the local visual cues of the release site, the local feature, the familiar feature of the release site, and then they recall the direction which lead them home from that site. With a pilotage strategy, instead they follow a chain of landmarks and then they are able to reach home. So we can set these two strategies in conflict by clock shifting the pigeons and releasing them from familiar location. So we can see whether they follow a chain of landmarks or they just say, okay, I've been taken to, I don't know, to south with respect to home because this place is Livorno, for example, so I have to go north and then they make a mistake because they follow the compass. We made a GPS study in order to test the preferred strategy for navigating with the familiar landmarks. And we found that the extent of the flexion displayed by the pigeons may depend on several factors. And on which there is also a characteristics of the release site but also individual preferences. For example, this study has highlighted three different patterns of response in clock shifted birds released from familiar locations. For example, look at this bird. This is a bird, these are three tracks of the same bird released at three different release sites located in different direction. Before and after clock shift. After clock shift, this bird deviate greatly at the three release sites like if the landscape didn't offer any good information to reach home. Actually, birds behaving like this are very few. This is another bird. From these three, you are not able to recognize which is the clock shifted tracks and which are the unshifted tracks. If I don't tell you, you don't know because this bird completely ignored the compass information and followed a chain of landmarks to reach home. Also this pattern is showered by very few pigeons. Most of the pigeons do like this. They, in clock shifted condition, they display the ability to correct on the basis of by relying on visual familiar landmarks in the site close to the sea. But when you take the mainland, then they shift to a site-specific compass orientation strategy and they deflect greatly. So there are two components. One is the individual variability, the individual component. So maybe what is called now is very on fashion, the personality of the bird. And then the other one is environmental factor. So there are landmarks which are more easy, more informative. So probably pigeons don't pay attention to what we think landmark can be, like a group of houses or a tree or whatever. But they pay attention to gross feature of the landscape, like the sea, like contrast of colors or whatever because the sea is very important. It's a reorienting topographical feature. These two strategies, these two strategies, there are different cognitive processes underlying this site-specific compass orientation strategy and pilotage strategy. So familiar landmark-based navigational strategy. And the hippocampal formation is specifically involved in familiar landmark-based navigation and not in the site-specific compass orientation. Perhaps one can find quite strange that hippocampal ablated birds are able to home because hippocampus is not involved in the operation of the navigation, the olfactory map. So if I ablate the hippocampus of a pigeon and then I take it away from home, the pigeon is able to orient towards home and to reach the home area. But then it's impaired in reaching the loft because when he has to shift to landmark-based navigation, then the hippocampal ablated birds add difficulties and often they can get lost, they get lost within the home area. The hippocampal lesionate pigeons, this is an experiment showing by comparing the tracks of a bird before the lesion and after the lesion from two unfamiliar release sites. And we observe that the hippocampal lesionate pigeons display more straight route, like if they didn't pay attention to the landmarks because pigeons are often affected by topographical features. There are linear landmarks so they like to follow them or they are attracted by villages so they stop and fly over these houses and so on. So hippocampal ablated birds seem to ignore the topographical feature below them. But then when they are inside the home area, then they have difficulties, more difficulties than before the lesion because they have to rely on familiar landmarks to get close to the loft. And this is consistent with what we have observed with the ZENC expression because we observe that the highest ZENC market cells in the hippocampal formation are in the pigeons which were released within the home area. And not in the pigeons, the highest were in the pigeons released, okay. But then there is a relevant ZENC expression in the pigeons released in the home area. So of course when pigeons are released away from home, the hippocampus is active because it's picking up information about the topography of the over flown area. But then it's also active when the birds are released inside the home area because also there birds need landmark-based information, visual landmark-based information. And these are two tracks of two different birds subjected to hippocampal lesion and released at familiar location after clock shift. And this is just a demonstration of what I told you that landmark-based navigation is mediated by the hippocampus because those birds do not pay attention to the most conspicuous landmarks, which is the sea. And fly over the sea for 10, 12 kilometers. And so while the sea and the coast is reorienting Q for controlled pigeons, it is not anymore for hippocampal lesionate pigeons. So they are not able to recognize the sea as a reorienting, to use the sea, the coast as a reorienting topographical feature. So let me thank some of my colleagues that made my time nice and happy during the experiments and people who shared this experience with me. And then let me think for a moment to Floriano who passed away two years ago and I want to dedicate this talk to him. Thank you for your attention. Questions? What is this? So maybe it's a detail, but you showed many of these maps where by the angle and of the birds or the single bird, you decide whether the bird is impaired or not or you compute the direction. I imagine that you have some sort of a way of analyzing this data in such a way you can tell for sure that even with a few elements of the sample you are able to, you didn't describe roughly, even roughly how you measure this direction. In this way, the direction of the track, for example. So yes, yes, I forgot to tell. So we compute a mean vector by averaging the direction taken by the bird while moving from one fix to the next. So you have several fixes because the GPS gives you a fix depending on the duty cycle of the GPS of the sampling. So you can have one fix every second or you can have one fix every minute depending on the GPS. With GPS GSM it's not possible to have one fix every second. So we have maybe one fix every 10 seconds or one fix every minute so they can last longer. So you simply make a mean and average this direction. So a pigeon is moving from here to here and it's an angle, then it's moving from here to there and it's another angle and then moving from here to there and it's another angle and so if you average this angle you obtain a mean vector which can be tested with the test, for example, the V test and so you can decide whether this vector is significantly oriented or not because if the vector length is too short then it's not significant and we're very small. Yeah, and so the center. Then you have the mean vector distribution. So if you have a mean vector distribution then you can test this mean vector distribution with the hotel in test, for example. And then you can calculate the confidence ellipse of this distribution and if the ellipse, the center of the ellipse does not include the origin of the system then it means that the distribution is oriented. And then you can calculate the confidence limit that 95% of confidence limit of the distribution and then you can decide whether the birds are oriented towards home or significant towards a direction that does not include home. So you can say, yeah, they are oriented but they don't go home. They go in another direction and then this is all, I mean it's very simple. Well, I have a question about the last part if you talk about hippocampus which I suppose the most of the learned behavior is in the memory over there. So suppose you do an experiment where you disactivate hippocampus, biostasia or some other. So in that case, I would suspect that whether the birds are migratory birds or they should not be able to use any of the cues whether it's an old faction or it's a visual or any other perceptual modality. Is it true? In the learning phase, you mean? No, after I mean even if you take the adult bird whether they are migratory birds or just normal one you simply disactivate hippocampus. So that means you are actually truncating all the learned information which they have about the environment and about. So and then you release them far away from their home. In that case, I would suspect that they will be completely confused because there's no- They are not because we don't impeach us, we, I show you- I'm hoping when you are saying there's some innate information, right? Ah, in the migratory birds? Yes. In the migratory birds? They are probably, they may be able to but otherwise for all other cases I would suspect that they will not be able to- Well, I don't know because I cannot tell with certainty because nobody did lesion to the hippocampus in migratory birds. So we have no data about that. Essentially the question is about separation of learned behavior and innate. Yeah. Because innate behavior, we don't know exactly where, because we're much more complex than and the learned behavior we know is memorized and the hippocampus, you have got this information. So if you take away that part, you just activate and there'll be a problem for those kinds of birds. That's what I suspect. I don't know. Have you done any experiments? So for migratory birds, I don't know because we never lesion at the hippocampus in migratory birds. But I can tell you what we know in pigeons. So if you take an adult pigeon who are ready to learn at the map, you release it, you make an hippocampal ablation. Then you take it 50 kilometers from home. The hippocampal ablated pigeons are able to orient towards home. And then they have difficulties in reaching the loft in the home area, but then they reach the home area. So they are able to navigate. But if you take a young pigeon and then you perform an hippocampal ablation before learning, before map learning, and you keep the bird confined in an aviary, these birds don't learn to navigate. But if you make the bird to perform together with the other birds, spontaneous flights around the loft, even those birds which have been subjected to ablation when young before map learning develop a map. So this is a mystery. Don't ask me why, because I don't know, but this is what happens. Do another experiment. Suppose you take the adult pigeon which have learned about the environment, you put them in a box. They don't have any visual stimulation, completely magnetically isolated. So no perceptual modality is getting any information from outside. And you take them far away, 60, 70, or whatever kilometers you think necessary, and then release them. Do you think they will be able to come back if it is the first time in that area? But with hippocampal ablation or without? With, sorry? After hippocampal ablation? No, just normal, normal. I'm not talking about hippocampus. Can you please tell me again? I'll take a pigeon. Experiments like you take the adult pigeon, they have information about their area. You put them in a box in such a way that they are completely isolated from all of their perceptual modalities. They're isolated, no magnetic stimulation from the magnetic field, no visual, no factory, nothing. And then you take them far away, and then you release them. Now you release them in the area which is the first time you have taken them. In that case, what would happen? They come back because it has been done. But in this way, they have been anesthetized. So they have been transported in complete anesthesia, in deep anesthesia. They have been taken away, and then they wake up and they go home. That's a mystery. No, it's not a mystery because they need the local information not really on the way. I mean, you can, better if you give them on the way. If they haven't, they've been called. It has been done by Varaf who did every experiment possible. It has been anesthetized, and then they wake up and then fly. Fascinating science, I'm also very envious of the way you do it by going out and seeing, so on. I also wondered about the hippocampus, but, but maybe I'll ask something else, very surprising that you said about the lateralization. Yeah. So you showed that there are these two main ipsilateral pathways, but then the crucial one is the... I don't know why, but I don't find any other explanation because the major projection are ipsilateral. So the major projection to the piriform cortex are from the same side of the, from the olfactory bulb of the same side. This is why these results were quite unexpected because we were convinced, well, we found an advantage of the left piriform cortex. For sure, we'll find an advantage of the left nostril. And we didn't. So we repeated the experiment for two years. We accumulated more data than I show you. I didn't show you all the data, but we published, of course. And then we had to admit that it was not, I mean, the case. So then Pazke and other colleagues made a detailed neuroanatomical study and they found that there are fiber crossing. And also the other way. Yeah. And so probably it's this bundle of fiber crossing. Also the major bundle of fiber are ipsilateral, but there is some fibers crossing. And although they are fewer, they are fewer, they maybe are important for this. I don't know. I don't find any other explanation because the lateralization results are very clear. With the hippocampal lesions, is there a lateralization? So this is a good question. We performed studies with vanishing bearings on hippocampal lateralization. We observed the lateralization in favor of the left hippocampus in map learning. So we lesioned the hippocampus in young birds. We kept the birds confined in an aviary. And the left hippocampal ablated birds did not develop navigational abilities. The right one oriented towards home. Nevertheless, we repeated the experiment with adult birds trained from familiar locations. And we lesioned one hippocampus. So we made unilateral lesion. And no difference between the two groups. Both groups performed compass orientation strategy, navigation according to a compass orientation strategy. Site specific compass orientation strategy. And we expected to see more, at least an advantage for the left hemisphere in performing a pilotage strategy. And we were unable to see it. So both sides are very important for pilotage, for landmark based navigation.