 Thank you very much for inviting me here today. Before I explore the question of whether a bee could be said to have a mind, I would like to thank my co-authors that have helped me form all the thoughts and experiments that I'm about to tell you about. And they are Catherine Wilson, Quinn Solvey, Oli Lukola and Peter Skorapski. So what I'd like to explore today is whether from what we know about bee spatial cognition we can make inferences about the question of whether bees or other insects are conscious. So what I mean by that is that there is some form of mental representation of things of space and of things in space. I also want to explore whether bees in the context of spatial cognition have an appreciation of the outcomes of their own actions and their spatial problem-solving tasks and whether there might be emotional states that are linked to spatial settings. So there is a common perception that while people realize that insects can do very complex things, that they do them by means wholly different from how humans solve them. So here is something that at first glance maybe shows a similar outcome, the famous La Sagrada Familia Cathedral in Barcelona and a large termite construction, roughly of a similar shape. And the common perception is that in the human architecture case there is a master plan, a plan for how the ultimate outcome should look, and then all the other actions are subordinate to that plan, whereas in insects everything arises by local interactions where no one has any idea of what the ultimate product should look like. And I will hopefully show you some examples that at least this dichotomy between these two different paths isn't quite so clear-cut as you might think. The discussion is quite old, it's not novel. So here is Charles Bonet, a Swiss naturalist who lived from 1720 to 1793. And Charles Bonet had the following to say about self-organization in the bee colony and the possibility of how one might construct something as marvelous as the hexagonal structure of a honeycomb shown here on the bottom left. He said this, placed together in the same room 10,000 automatons animated with a living force and all induced through the perfect resemblance of their outer and inner being. If we admit the least degree of feeling in these automatons, even only such as it's necessary for them to be conscious of their own existence, seek their own conservation, avoid noxious things, prepare useful things, etc., their work will not only be regular, well proportioned, similar, equal, but it will also have symmetry, strength, convenience to the highest point of perfection. So it's interesting here that Bonet sees the idea of there being a consciousness and feelings sort of as something primitive, as something given, but he still calls them automatons and thinks that somehow by just putting all of these conscious automatons together in the same room, something as magnificent as this honeycomb structure might arise just naturally. Now, I do not think that it's quite so simple. Of course, there have been quite a few crude human experiments and stuffing 10,000 animals together in the same room, as you can see in this chicken battery farm, and for some reason something as regularly structured as a honeycomb or any form of architecture clearly does not emerge just by itself. So you need a little more than just stuffing lots of animals, perhaps conscious animals together in the same room to arrive at any form of animal architecture. So many people have mused that there is something remarkable about this extremely regular construction of a honeycomb. You can see a honeycomb under construction. Here there are dozens or hundreds of individuals of bees participating in this process. It's a mathematically perfect solution for honey storage, for rude storage in terms of minimizing materials and space used and so on. It's also double-sided, so on both sides you can store larvae and honey and so on. And typically the structure is started at the ceiling of the hive and then gradually the bees work their way down until they've reached the bottom. But another Swiss naturalist François Huber explored the inner workings of the hive centuries ago and made some very surprising discoveries. So he experimented with glass hives to see what the bees might do inside the hive while they're constructing their beautifully regular honeycombs. And what he found was the first thing that if he introduced a glass ceiling into the hive is that the bees didn't particularly like attaching wax comb to glass. So surprisingly that didn't perturb them much because they simply then inverted the entire construction process 180 degrees. Instead of working their way downward from the ceiling they built the whole comb as a tower construction starting at the bottom and then gradually working their way up until they reached the ceiling. Now it's still a very repetitive process but if you'd programmed a robot to do exactly the same as the honey bee does working down along gravity that robot would already fail at this particular challenge unless you told the robot to be able to invert their procedure in its alignment with gravity. But the next experiment was even more surprising. So Hubert when he discovered this flexibility in bees then introduced a glass bottom in addition to a glass ceiling. And what he found in that case is that the bees would start on the side wall of the hive and extend the comb construction laterally through the cavity that he had provided the bees with. What Hubert did next was the really surprising result. While the construction was in process so the bees had been built halfway through the cavity he then put a glass obstacle in the projected path of the growing honeycomb. Okay so the bees had already built their two-dimensional construction almost halfway through that cavity and then suddenly there was an obstacle that the bees would have perceived perhaps if they explored further in the linear direction of the growing honeycomb found that it must have been suboptimal because what happened was that the bees then before actually reaching the suboptimal surface would build a 90 degree kink into the comb and attach it to the nearest side wall. So they avoided projecting from the geometry of the current construction that there might be a suboptimal outcome several days down the line. It takes a long while to build honeycomb. They projected and anticipated or seemed to have anticipated this suboptimal outcome and took corrective action before it materialized and built this strange curve construction. And Hubert 200 years ago said, I acknowledge that I could not suppress a sentiment of admiration for an action in which the brightest foresight was displayed. So Hubert thought that bees have a kind of master plan an idea of what the desirable outcome of the kink construction might be like. So the way we nowadays study bee orientation we have a tool called harmonic radar which is the equipment that you can see rotating here on the right which contains a dish that sends a signal which in turn is bounced is sent to a bee and with this so-called transponder the bee bounces it back so the top dish can pick it up and in that way informs us about its location and in that way we can monitor bees orientation from their maiden flight the first time they ever leave the hive to their death essentially and monitor where they've been all their lives. And just to make you appreciate the general challenge of course human made environments in general compared to natural environments are relatively easy to navigate and that's because we build landmarks buildings that are meant to be unique and memorable. If on the other hand you consider the typical bees environment so here's a typical bees environment then what you see that everything looks a bit the same so a typical bees challenge might be that there is a nesting site somewhere under this particular tree and several kilometers away there might be a location over here where there's good pollen to be had and another location over there behind that hill where there might be particularly good nectar and so there's a lot more similarity between perhaps familiar landmarks than there is in any kind of city. So if you've ever navigated an environment like this without a map or without satnavs or without help from others then you can sort of appreciate the challenge that navigating insects face in natural environments. And so what we can do with our fancy raider equipment is we can essentially take the view from the cockpit of a bee and see how she sees the world as she's navigating around her environment. So this is actually a reconstructed bee view from one of our raider tracks you can see here the slightly boomerang shaped eyes of the bee and the view through them. So the view is relatively coarse pixeled and a bee sees a lot fewer pixels so to speak than a human does but on the other hand she can see more rapidly and the reason why these images are very distorted is because bees have pretty much all round vision. What we can also do in this case is we let the bees to their own devices and do not pose them any particular challenges instead we let them roam freely, we let them explore their environment as they see fit and discover flower resources to forage from. So what we see here in green is one individual's maiden flight the first flight that she's ever made in her life and you can see that there are various loops in various directions from the hive location which is marked here in blue in red by the way it's the second bee's exploration flight and she flies in various directions there's a big loop here and there's another loop there, another loop there so she flies in a variety of directions from the hive exploring her environment both to memorize the landscape as well as perhaps to discover the first few flower patches that might be useful for harvesting nectar. Also of importance we'll later realize why is that during this first exploration flight she flies to this forest edge here once so that's the bee's first day, two flights one of them lasted about two hours, the second one much shorter. Now this is what the bee does from day two she makes one more exploratory flight in this direction there's a single loop here and then she's discovered a flower patch by the forest edge up here and then for several days for up to over the 82nd bout basically over several days she only visits that one location and exploits the flowers there and shuttles back and forth between the hive and this particular foraging location nothing but she doesn't visit anything else this is a complete record of what the bee does then there were a few days of inclement weather when the bee stayed at home and after that on day 11 she departed again, visited the familiar patch from which she'd spent the last few days foraging one more time then flew back to the hive, flew out to that patch again and then suddenly changed her mind it seems and instead of flying to that recently exploited patch turns her attention entirely to a patch that she'd only visited on her very first foraging flight 10 days earlier and then spends the rest of her days just visiting that patch but the most interesting observation here in a sense to me is the sudden change of mind on a foraging bout to the patch first exploited and then halfway through a foraging flight she switches over and retrieves from a distant memory the location of another patch that she just visited once during her early explorations and then stuck with that until during one flight she just disappeared off the radar presumably having been eaten by a bird or a crab spider when I talked to colleagues who work on cognition and primates and corvette birds they were always a bit dismissive of bees learning abilities I told them about how bees can learn colours of flowers and scents and so on and they said well that's what bees do every day in visiting flowers it's not any evidence for intelligence and so they were at the time working in on things like string pulling tasks in corvette birds for example and at some point just to be provocative in a lab meeting I said well I bet our bees can solve such a string pulling puzzle and no one quite believed it but I had a few bold post-docs in my lab who were willing to give it a go and so what you can see in this slide is a task is a string pulling task in which there are three artificial flowers each connected to a string and with a central nectar well that a bee would have to get to in order to obtain a reward and what you can see in this video is the first bee that we ever trained to solve such a task here she is landing and now she's found a string and she's pulling it out she's not landing on the top of the flower she clearly is not necessarily efficient at it but she knows what she's doing and that she has to pull that string and remove it from under the glass table to get an access to a reward not only can bees learn to pull strings by us gradually telling them they can also learn it by observing each other so in this video here we have two bees one marked with a red dot and that's an experienced individual that has done it before and a completely naive bee who has no idea what she's doing and you can see that the experienced individual with the red dot pulls out the flower and then they both imbibe the sugar solution together and then the inexperienced bee simply scrounges on the abilities of the experienced demonstrator and then after a while you can see now they're getting a bit nervous because actually the nectar droplet has been depleted there's nothing in and the red marked experienced bee just runs over to the next flower and pulls it out this one still tries there must be something more here now she's running over scrounging again on the experienced bees' efforts and this will continue until they've pulled all the four of them out but if you repeat this process multiple times then the observer bees will actually learn to pick up the technique and solve it by themselves not only that here the bees of course interact directly with one another but you can even lock up the observer bee with a little glass cage so to speak and just let it view the trained demonstrator from a distance and even from such distant observations can the observers learn the correct technique to pull a string so in the next slide we can see one dot for each individual bee so at the top there is number 31 that's the first demonstrator bee and in this video we can see how the technique of string pulling spreads through an entire colony so you can see there are little lines appearing now each of these lines marks one interaction one pairwise interaction between a trained demonstrator and an observer bee moreover you can see now some of the other dots are turning coloured and they turn coloured when the interaction has resulted in the demonstrator being able to solve the task by herself and so now there's a good number of bees that have all been trained by yellow 31 at the top there and all the ones that have turned orange actually can do it by themselves and as this goes on there are now interactions between bees who are not the generation one learner now is a purple bee that colour marks that this bee has not learnt the technique from the original trained demonstrator but a second generation bee and as this process continues more and more bees here another purple bee now learn the technique from second generation demonstrators and in fact as the process continues and gradually all bees learn the technique ultimately even the original demonstrator passes away and the technique continues spreading through the colony so there is in the end a whole colony of string pulling foragers so there is a kind of cultural spread of this string pulling technique through an entire bee colony for the next experiment I will let Jane Goodall do the introduction the most recent example which blows people's minds is the bumble bee and the bumble bee has been taught to roll a little ball it rolls it backwards it rolls the little ball until he can push it down or maybe she I don't know into a hole like a goal and as soon as this little ball gets into the goal the bumble bee is given a little drop of nectar as a reward so they learn to do this but the mind blowing thing is other bumble bees who've merely watched a taught bumble bee can do the same without being taught just by watching and that is supposed to be a mark of very superior intelligence so this is something we're learning all the time we have been far too arrogant the animal kingdom of which we are apart is filled with secrets and gosh I'd love to be young and learning about these things now because all the doors are wide open and you never know what you're going to find so the task as Jane Goodall just explained is to move the ball in this case from the periphery to the center of this horizontal area and once the ball arrives in the center the bee gets a sugar reward so they can be trained to do that just fine but the reason why the social learning by observation in this task is so particularly remarkable is this so what we can see in this slide is a task with two bumble bees here is an experienced bee that has already solved the task before that's our so-called demonstrator and then there's an observer which has never solved that task before so she's entirely naive and when the two bees get together this is what happens if you so you can see here the experience bee rolling the ball to the center and then they both get a little sugar droplet as a reward now obviously a logical way to solve this task if all you need to do is to get a ball into the center would be to pick the closest ball that's closest to the target area and move that to the center but we've played a little trick the experience bee the one that you've just seen roll the ball into the center actually knows that the two closest balls this one and that one move down so that bee knows I have to pick the furthest ball so the least optimal one to move to the target area and so for three times the naive observer gets to see that it gets to see the demonstrator be used the furthest ball moving it into the central target area so what we see in this slide now is the observer being alone the observer has seen the demonstrator solve the task three times and every time the demonstrator used the furthest ball now the observer is on her own will she pick the same furthest ball or will she solve the task in a better way and pick the closest ball and she picks the closest ball so she's not simply copying the demonstrator exactly she actually seems to have a form of understanding of what is the desired outcome of her actions and rather than simply aping the demonstrator she picks the optimal ball to solve the task so our next question is can bees picture things can they imagine things such as flower shape and so on now bees are of course famously good at recognizing flower patterns but it's actually surprisingly difficult to find out whether there is such a thing as a mental representation of a flower even if an insect recognizes it because often it turns out it can be flower patterns can be recognized with simple feature detectors that analyze edge orientation and so on so in our following experiment we asked if bees can actually draw on a representation of a particular shape through multiple different sensory modalities which would be more of an indication of a mental representation of shape than just recognizing a shape in the visual modality so what we did initially is we trained the bees to distinguish two different shapes balls and cubes but in this case they could not touch them they could only see them through a plexiglass lid one of the shapes in this case the ball was associated with reward that means a little reward of sugar solution and the other in this case the cubes was not and then the same bees that were trained in this manner were now presented with the same shapes in complete darkness where they couldn't see at all but in this case they were allowed to touch different shapes and in this case when the bees had first been trained to see and be rewarded over balls that however they could not touch in the light then if the same bees were tested in complete darkness then they chose the balls over the cubes only using the touch modality so they could recognize the same object in a different sensory modality and we then also reversed this experiment where we first exposed the bees to the same shapes in complete darkness as shown in this photograph here we used infrared light of course to monitor the bees behavior and you can see that the bee here is sort of hugging the ball while she's imbibing the sugar solution reward the cube in this case was rewarding and then in the same manner we switched the situation to one in which they could no longer rely on the touch modality but on vision however again in this particular experiment the bees could not touch the shapes as they had during training but only see them and sure enough the bee again after being trained to the round shape in darkness when she faced the same shapes in the light picked the balls over the cubes so in both of these scenarios they picked the correct object in a different sensory modality from which we conclude from this kind of flexible access through different sensory modality that indeed there must be a kind of mental representation of the shape of an object do insects have emotions as Turner did at least take this possibility seriously I invite you all to read the papers of this man he lived from 1867 to 1923 he was a entomologist as well as a psychologist and his literature is extremely rich with visionary ideas that unfortunately have been largely forgotten today but it's remarkable also that he actually did consider the possibility of these digger wasps having emotions so this is a photograph by him from one of his papers in 1912 and he writes this the coiled antenna the protruding mouth parts and the general attitude indicate intense excitement one who believes that insects have emotions will find much in the attitude of these true homophilus to support his view now Turner was aware that this was merely an observation it was not an experimental exploration he has very rigorous experiments in many of his other papers which I would strongly encourage you to read but how might one test emotions in insects experimentally and quantitatively well we chose to use a paradigm that is simply borrowed from vertebrate research it is basically asking whether an animal or an individual's glass is half full or half empty symbolized in this little photograph here so an optimist would judge this ambiguous situation as being glass half full and the pessimist faced with exactly the same situation would say it is half empty and so you can actually test animals emotional states by asking them the same kind of question in ambiguous situations and that sort of task that I am just going to explain is used exactly in the same kind of manner in domestic animals for example to find out whether they are happy domestic animals or unhappy ones so what did we do we have a little flight arena as you can see there on top and the B has five options either to pick the left most options option marked in purple or the right marked in green and there is only ever one option available so it is a so called no go task and what the animal learns over time is that whenever the blue left option is available there is a reward whenever the green right option is available there is no reward and after this training the B is then faced with various intermediate ambiguous options such as turquoise or greenish turquoise or blueish turquoise and we are then asking well do you judge this more likely as being a positive outcome might the glass be half full or a negative one is it unrewarding in this case so here is the training in simple forms what the B does when she has learned that the blue left option is rewarding she flies there in a straight line if on the other hand the green option is presented the B already knows well this isn't any good she fuffs about for a great deal of time and then finally says well ok I might as well go and try it but she takes a much longer delay time before accepting an option that she already knows as being unrewarding the question then is what happens with the ambiguous intermediate option so here is a turquoise one and again when we present the B with the turquoise option we then measure the delay time that it takes her to accept that option and interestingly that delay time depends on something that happened before this experiment it depends on whether we gave the B a little sucrose droplet as a surprise so here she gets 5 microliters of a surprise droplet of sucrose in the run up to the experiment before she even enters the arena and in the control group she gets nothing and it turns out that the way the B judges this ambiguous situation turquoise depends on what happened before she even started the experiment and that's shown in this graph on the right side here so when the B's face their familiar rewarding option that's P on the right the delay time is invariably very short they fly straight there if on the other hand the green right option is shown that's shown on the right here invariably the B's will delay for quite some time before accepting that option but for all the ambiguous option so turquoise being in the middle there is a difference depending on what happened before the experiment so if the B's before starting the experiment had their little surprise reward there's a sweet that I've never seen here before they tend to accept the ambiguous option faster that's our red line here then there would otherwise so if there was a surprise reward before the experiment the B's accept the intermediate option more rapidly so they behave in this glass half full glass half empty in this ambiguous situation more optimistically if they've had a surprise reward before entering the experiment and you can do of course the same with adverse stimuli in which case the B will behave in the opposite way so there is by the same criteria by which we judge mammals domestic animals as being emotionally biased in either a positive or a negative direction by the same criteria B's could be judged to have simple emotional states now a final question might be whether insect brains aren't just too small to contain anything such as a mind aren't they just too simple to undifferentiate it and undoubtedly B brains are extremely small the size of a B head and of course the brain is substantially smaller than that roughly one cubic millimeter and indeed they contain relatively few neurons certainly compared to the 80 billion that a human might have so roughly 850,000 neurons in a B's brain here is a portrait of the frontal view of a B's brain so insect brains are small and the number of neurons are few they are by no means simple so here is the structure of a single neuron in a B's brain and you can see that it's about as finely branched as a fully grown oak tree extremely wide ramifications through all regions of the brain including the antenna lobes and the mushroom bodies and so on that is just a single neuron and it turns out that this neuron largely constitutes the reward pathway in the B's brain so each single such neuron might make connections with perhaps 10,000 other neurons and from that you can deduce that there is an extremely finely grained and detailed neural network across the B brain if you ask computer scientists how many neurons you need for consciousness like phenomena such as predicting outcomes of one's own actions then it's diminutively small perhaps 10,000 compared to even a B's brain there is at least some suggestive evidence to indicate that insect brains might support consciousness one comes from the neurobiology of a central complex whose most important and well described role is in navigation but it also integrates information from external stimuli, internal states such as motivation and past experiences and could thus represent a kind of neural model of the familiar space around the insect as well as of the self as for example suggested by Baron Klein in a 2016 PNAS paper in the same vein some jewel wasps, parasitoid insects sting cockroaches into the vicinity of the central complex and this has the effect of turning them into something like behavioral zombies that have still got reflexes and they can still walk but they lack any form of self initiated action as might be generated by for example interestingly also Bruno van Swinderen's team has shown that analogous to vertebrates insects have several types of brain waves that correlate with the state of wakefulness or sleep, several sleep phases are also well described by such brain waves and crucially more synchronized brain waves across brain areas occur in awake states perhaps facilitating the kind of cross model cognition that we've heard about earlier and finally there are neural correlates of attention, attention is a form of inner eye focusing selectively on certain stimuli but not others and these investigations also showed brain states that precede the bee's selection of one or another stimulus that's also work from Bruno van Swinderen's lab. So insect reactions are not simple and I think not too simple to contain a mind. Finally some broader conclusions that there are mental explorations of solutions to possible problems and that there are sorts of emotional states at least by the same criteria as we diagnose them in large brain mammals that there are emotional states that we can diagnose and that we should perhaps explore whether there are even emotions that we don't know in other animals such as that might be linked to unique bee minds unique bee problems such as the discovery of a nectar rich flower or the swarming process when honey bees move from one hive to the next. And finally I think it's important to consider the ethical implications of such research and the possibility that there is a mind suffering in bees and I think that is important for how we consider their conservation as well as thinking about how we treat animals in research laboratories. And so in that vein I'll play you a little song with my band The Killer Bee Queens there's someone called The Bee Keeper's Dream and it's loosely inspired by Patricia Highsmith's The Snail Watcher about a person whose pets laid their own as nightmares. Of course the song doesn't literally have to be about a beekeeper. It could also be about a scientist tormented by ethical concerns about what he does to his animals in the laboratory. The material for the music video is taken from David Blair's surrealist masterpiece Wax or the Discovery of Television Among the Bees. The first film ever aired on the internet I heavily recommend it. Anyway, enjoy the music. I knew which way to go into the darkness. Thank you very much for your attention. To questions for you, Lois I just hope that we can get through a reasonable number of them. I'm sorry apologies to anybody whose question we don't manage to get to. So I'll just start off with one of the very early questions that was asked last. This is a question about the foraging slides that you showed at the beginning when you showed the tracks of the bee changing from 11 days onward and the question is from Jim Monag and the question is could this be due to a change in the availability of food and he's thinking in also that maybe the change could be due to some transfer of information within the hive either by the bees regurgitating for each other or of course by the waggle dance itself which might transmit information. Sure, just briefly it seems that the host has turned off my video so I can't currently show myself. So yes it's quite possible that any sort of change in allegiance to a plant species or a particular flower patches because the reward offerings in general bees are careful shoppers in the flower supermarket and adjust their preferences all the time. So the remarkable thing this particular case was that the bee appeared to okay thank you the bee appeared to change its mind halfway through a flight towards flying to a patch that she had only visited like 10 days earlier without actually retracing any familiar route but apparently retrieving the information from memory that is the remarkable observation there. In bumble bees unlike honey bees there is no communication about location so honey bees have a very elaborate spatial communication system by which they can alert hive mates to the precise distance and direction of a food source using a very highly ritualized motor display if you wish. Bumble bees can communicate that there currently is food somewhere in the environment but they do so without passing on any information about location. So since this particular bee that we tracked through its entire lifetime was a bumble bee it's not conceivable that she would have picked up that information in the hive and moreover the change of track was not initiated in the hive but while the bee was in flight to another familiar patch. Another question is about the string pooling task whether there is variation in how quickly individuals learn this task and what that variation might be due to. So for every task where we measure individual variation we can also find it. There are some individuals that are especially quick picking up on such tasks and others very slow. So specifically in the string pooling task we trained and tested way over 100 individuals and there were only two individuals of all the ones we tested that figured it out entirely by themselves. All the other ones were trained in stepwise training either by observation of other informed bumble bees or by us in a stepwise training procedure where we had to gradually push the blue disc under the glass platform. There were just two individuals that did it spontaneously. When we measure such variation in learning abilities there are not just individual effects between colony effects or colonies are families essentially because they share the same father and mother in bumble bees and so there are some colonies that are especially swift at learning colors, scents and other tasks other features of flowers whereas others are spectacularly slow at such tasks. Okay and while we're on variation in learning speed there's a question that I had of course all these workers are females and so I was wondering if the male bees, the drones who don't really seem to do very much at all are they poor learners or can they learn anything? Yes they can so there are some between species differences honey bees the males are indeed spectacularly incapable of doing anything much but mating they even have to be fed by workers because they can't feed themselves but even in the male honey bee they do remember their home so they have to navigate successfully to mating sites where they meet with queens and then find their way back so they still have a very good spatial memory in bumblebees however males also visit flowers so they have to remember which flowers are the best the most rewarding ones and the colors in the sense and we actually did some color learning experiments with bumblebee males and surprisingly they were actually as quick at picking up the contingencies between colors and the words as well workers so they're not as dense as one might think Good to know that a couple of questions in connection with the consciousness issue and the possible emotional states one question is is there any sign of any dream like states in bees in connection with the deep and if they can be said to have consciousness could we say they had free will That's a very interesting question so I mean it's very hard to know if any animal is dreaming there is some interesting evidence for example from rodents where people record simultaneously from many dozens of nerve cells while the animal is both awake and navigating the spatial environment and then they can still record from the same brain cells during the night and it seems indeed that these rodents replay spatial experiences from the day before we don't have that level of recordings in bees in part because we can't or people cannot do neurophysiological recordings very easily from freely moving animals there are however there is evidence for various sleep phases in bees and other insects including deep sleep phases and something equivalent to REM sleep in humans and Bruno van Swinderen especially has done quite a bit of work on on fruit flies and brain oscillations in various parts and there seem to be in certain sleep phases very close to actual awake states so it is plausible in that sense that there might be something like dream like states perhaps similarly to mammalians serving memory consolidation during sleep but we don't know this yet I guess that was one half of the question remind me what the other half was okay that's also a very interesting question that was actually first also discussed by Charles Turner in his case using cockroaches running mazes where he saw them pause and seemingly decide what to do next and he thought that was an indication of him pondering various different options and ultimately choosing which ones. There is a very good more recent paper by Björn Brems on the adaptive significance of free will as a kind of evolutionary trait and he thinks that many animals have it also adaptive to be deciding between available options to be able to mentally explore solutions and then form an adaptive decision and he includes insects in these in his discussion Martin Heisenberg has also written about this topic and then another another question while we're on consciousness and learning and so have you tried measuring how long it takes bees to forget things? Yes hardly anything is ever fully forgotten so we did and that applies to other animals as well so we did experiments on motor learning a long time ago where bees had to manipulate different artificial flower types sometimes they had to twist their bodies to the right side or to the left side and also link these motor patterns to colors and so on bees only live for a few weeks so three to four weeks sometimes a bit longer in laboratory environments but we trained these bees and really tested them three weeks later and while there were over these three weeks they were a lot poorer in their performance then at the end of training there were still better than a completely naive animal and they were also better than they themselves had been three weeks earlier so while these memories seem to have to some extent faded when you prodded them again hey do you remember this they actually performed better than before and you find similar things in motor memories often so especially motor memories where if you put someone again on ice skates at age 50 when the last time they did it was 30 years ago they might say oh my god I can't do this and once they're actually on the ice well they're surprised themselves how much they actually do remember we also did some work on not forgetting but misremembering so that bees actually had false memories of flower types they encountered and we did that by sequentially training them to various flower types and then exposing them to flowers they'd never seen before but which bore elements of flower features they had encountered in separate incidents and they then preferred flowers with these combined features to the individual ones they had actually been rewarded on so there was a kind of false memory where they merged features of different things that they had seen in the past right and then does the learning performance decline with age no we have tried in a number of different tasks to find any age effects and we never found anything now again bees live a very short period so there might not simply be enough time to senesce and become and have declined memory performance because I guess other systems fail before their brains do but no we have never found any age effect on cognitive performance and then there are some questions about emotions could you perhaps define emotion and are there any ways of measuring it outside of the learning context yes so with with any animal of course the access to any non-human animal the access to emotional states is only indirect so we can't of course in the same way as with a human ask them are you a little sad today or are you or whether they are upbeat or and so on so we have to always use some sort of indirect measure and there are various ways of going about that one is to use the kind of standardized psychological tests that we used and which are also employed for domestic animals to ascertain whether they're happy pigs or chickens or goats or not in their farmer's environment for example another way of going about this is to evaluate hormonal states and of course you can also measure various brain states we don't do that in our laboratories we use mostly psychological measures others use also physiological or neurophysiological measures but in our behavioural psychological world at least the bees by the same criteria by which you for example diagnose pessimistic or optimistic psychological states qualify so if you reject these criteria then by analogy you would also have to reject them in goats and primates because they're equally indirect there of course also question about whether bees might have the ability to distinguish food sources that are for example being sprayed with pesticides from those which haven't is there any work on that kind of thing going on that the cognitive ability of the bee might overcome the toxins that are being applied in the environment well it's that's very tricky because neonicotinoids for example very common insecticides are in the heaviest concentration of course in the leaves where they're meant to deter or kill herbivores they leak in very low concentrations into the nectar as well so that bees are exposed to them and will carry them home to their colonies and they are documented detrimental effects and at these low concentrations the foraging bees that visit the flowers certainly seem not to discriminate against them there is even some evidence that in these low concentrations they moderately prefer them and they might in fact act in a similar kind of way as the bee does which at low concentrations can not just induce addiction like phenomena but can also actually assist in associative learning facilitating associating the floral features with reward so that the bees might actually be drawn to flowers that contain these toxins so it's quite a dangerous scenario they do not seem to cause the full loop of realizing how there's some harmful effect going on in our colony if I carry this stuff home on the contrary they're either neutral about it there is actually evidence that they preferentially collect such nectar okay and then there's also a couple of questions in connection with the difference between solitary bees and the colonial high forming bees is there a difference in their learning capacities and might there be any collective consciousness of bees in the hive that could be studied in some way so you're right the vast majority and many people don't know this so it's worth highlighting the vast majority of bee species tens of thousands of bee species are actually solitary not social these bees however share with the social bees the fact that they all have brood care so they feed their young they have nests whose locations they must remember and they must of course also forage adaptively that means remember good flower patches and so on so some of the requirements on spatial memory on flower memory and so on are shared by the social and the solitary bee species and one would not necessarily expect the solitary species to be therefore less intelligent if you wish although they lack of course the kind of communication systems and social coordination of nest construction that you find in the social bee is there a kind of carry on carry on of social consciousness I don't think so so just like in humans there are very impressive things that bees in social settings can do together but there are still in humans as many individual consciousnesses as there are individuals participating in joint endeavors such as the construction of a city or a home and so on there is no people sometimes have the sense that because insects so beautifully co-operate in constructions like termite mounds or honeycombs that there is some sort of mysterious joint consciousness but we're still of course at a debating stage of whether even individual bees have consciousness I think the probability is likely that there is a kind of conscious experience of being the swarm or the group of bees okay there's a few questions people are interested in the waggle dance and you know it's role in communication amongst the bees but also on whether the dance is learned or and is it the same for all colonial bees or the dance different are there local variants for example of the dance yes that's a good question so there are about a dozen species of honey bees all of the ones that are not the western high bee that's been distributed through the America and so on live originally in Asia and they all have the waggle dance while all the other bees that are non-honey bees so bumble bees, stingless bees, all the solitary bees do not have this dance language there are differences in the dance language between honey bee species and it seems also between between local varieties of the western high bee there also seem to be some subtle differences however the basic features in all of these species are the same so the basic features are that there is a very recognizable motor display where the dancing bee I should perhaps briefly explain how the waggle dance works because not everyone knows this so what you see if you look into the darkness of the hive with infrared light or by some other means it's that on the vertical honey comb some individuals that have recently discovered a good food source run in a straight line for a few centimeters then they run around in a half circle then another straight line then another half circle another straight line another half circle and so on so it's a figure 8 shaped display and the duration of this straight ahead run the longer that is the further away is the food and the angle of that waggle run indicates to other bees the direction of the food source relative to the sun so to the extent that if the waggle run is straight up inside the hive that tells other bees fly to the sun if it's straight down it tells other bees fly away from the sun and if it's 90 degrees to the right of the direction of gravity inside the hive that means to other bees fly to 90 degrees to the right of the sun so this basic figure 8 shaped design is common to all species of honey bees but there are some more subtle differences in terms of just how much duration of waggle run corresponds to which distance and there are also some honey bee species where this whole dance is performed on a horizontal not a vertical surface and so on so there are some subtle differences in terms of how many bees that this behavior is largely innate so all bees can do it without further encouragement or training and bees can also understand the waggle dance without the need for being trained in it however there's been some fascinating work where people have merged two different species of honey bees that have two different distance codes and the first one is the 1.2 cm of waggle run means flying 1.2 km away from the hive and only 800 meters and another species so they should misunderstand each other but it seems that over time of mixing these two species together they can actually learn to read each other's waggle dances so that apparently there must be some recalibration perhaps bees the target find okay but I must have misread that and then zero in on the correct location perhaps after some search and then recalibrate their reading of that other species dance so there is there can be some learning evolved even involved even in this largely hardwired routine okay and when we're now that we're on the hardwired issues there's also questions about the genetics and whether there might for example be differences among color inherited differences amongst colonies in the speed of learning and the learning capabilities are there really fast learning colonies and slow learning colonies for example yes so there are in bumblebees we have done experiments where we measured multiple individuals per colonies learning speed and then also took after we had done these laboratory experiments took the same color bee colonies and let them forage in the wild and it turned out that the faster learning ones were also superior in their foraging performance at natural flowers there are other researchers Rob Page from the University of Arizona for example who have analyzed the genetics of bees in this case in honeybee colonies and it turns out that it is strongly heritable okay we're sort of running out of time last but I'll give you one last question from a beekeeper who says that this beekeeper has been struck by a growing impression that honeybees recognize their beekeeper and the question is do you think that's just wishful thinking and imagination on the part of the beekeeper or do you think this could be possible it is at least plausible so we did a slightly tongue in cheek experiment 15 years ago where we trained bees to recognize black and white images of human faces in fact we took a test set of photos used to diagnose a condition called prosopernosia in humans so that the disorder where people can't recognize individual faces and the bees were perfectly good at this task and learned to associate these images of human faces with reward so it is possible whether indeed bees can recognize a human like that in three dimensions and life we haven't tested yet but it's not impossible I'll just apologize to the other people who put in questions that unfortunately we don't have time to answer so before drawing things to a close I'd also just like to remind people that the next talk is on the 10th of February and it will be given by Misha Glene and the title is the human scale technology and leadership in times of crisis I'm sure people are absolutely astonished by the cognitive capacities of bees and also by the musical ability of a bee scientist so and it was very good of you to make the video for us it was very nice and it was dramatic I know we had to sort out a few things at the beginning to make sure we could show it as well as we can so thank you so much on behalf of the society and its members and those who've joined who aren't members thank you so much for a fascinating talk thank you so much for inviting me it was very kind thank you very much all right have a nice evening everyone and you bye