What will happen after 100h of evolution? Study of natural selection in an ecosystem simulator

the babits is an evolution simulation i've been working on for oh god five years now the simulation has grown a lot from its humble beginnings and i think that the project is way overdue for a showcase of how it works and what kind of complexity you can expect from it after all you can download the simulation for free on each.io and observe your own unique universe in this video i'll try to do that and let the simulation run for 100 hours and see how things develop over time because not only do their genes mutate over generations their brain too and with a lot of senses and actions at their disposal their behavior is always unique between every run we'll try to explore things from the perspective of taxonomy biology ecology and all the other science we can fit in there so without more delay welcome the fascinating world of the bibbids [Music] [Music] so the first thing to note is that you have a lot of control over the perimeters and the rules of the environment you can alter how plants grow how hard it is to move around the energy cost of producing pheromones and oh so much more in fact you have access to around 120 settings that influence the simulation in our case we'll keep things pretty normal we'll keep most settings as default but i'll play around with the plant spawning settings to make it so that we have three different and isolated plant islands on a fairly big map as a result we'll be able to see distinct populations develop on each island and i think that this will make for a more interesting video by default the islands slowly drift around but i'll make it a little slower so that the islands stay distinct and don't run it to one another that being said let's go in awesome we have our tree islands each with a large number of plant pellets ready to be colonized right at the start basic bibits with some degree of variations are spawned most are unadapted for this environment and it will take some time before real species start to develop and survive on their own in this case the islands are pretty small so it's easy for bibits to run into the void and die before they reach maturity especially if they are fast additionally their energy budget is pretty tight at the start so going too fast would be too costly for the speed at which they can digest food after approximately half an hour in we have the first biped that managed to lay an egg on our southern island they managed to do this by moving really slowly to conserve energy and prevent running into the void they also turn toward where there's the most amount of food so that's always useful as such they get the honor of being the first official species because of their beautiful cute eyes we'll name them rosoculi primitives for the initial pink eye by the way i have to tell you right now that i'm not a trained biologist everything i'm teaching you now i learned on my own and might have learned wrong that being said let me tell you things as if they were facts the binomial naming system is the system used to give scientific names to species in what we call the field of taxonomy basically all species are given a name composed of two terms the first is called the generic name and identifies the genus of the species genus just basically means group the second part is called the specific name and is used to differentiate the different species inside the genus back to it after some time other species have started taking all done the other islands using similar adaptations like being very slow and boring to watch in real time but they are less unique and not as cool as our rosoculi so we'll wait a little bit before naming them fast forward to 7 hours in on the eastern island we have a few notable species first there's these guys that are rather hot looking i think the species name stultus apare for stupid looking is fitting alongside them we have these purple eyed pipets that have a funny mutation they evolve the connection toward their immune system neuron which is unused in this simulation since i did not enable the virus system in this run so as a result we'll name them inutilus cascus for the original useless on our western island the most prevalent species is this one which has evolved a bigger size than most other species so far this seems to help them gobble up smaller plant pellets without having to bite into them which is an early form of filter feeding for them pingispa ulum for somewhat fat fits perfectly back to our southern island where rosa cooley is thriving their behavior is still relatively the same beside minor changes to their genes after all we've only just reached the eighth generation so they don't vary too much for now one thing i dislike however is that some of them have started to diverge and evolve slightly different high colors i just named the species after this exact feature so if this keeps happening it has the potential to make me look very dumb we don't want that so i'll have to play god a little bit and cleanse the members that differ too much from our ideals and a while ago i added a button to do just that but we have many means of achieving that same goal however after not even an hour i got very tired which started to put quite a strain on my carpal tunnels i had some discussions with hale hola the creator of another evolution simulation that focuses on plants he recently implemented something along those lines and suggested i do the same yeah i recently implemented something along those lines and i suggest that you do the same let me show you this this sphere in the middle is the target color and all these colors you can see around are within the tolerance range then i have created this laser tower that kills automatically any brand that is very different from the target color [Music] i will explain it better in my video i will explain how i made it to work and maybe it's you for the bibits you need to do something slightly different you're good okay alejo was right obviously the best idea was to automate selection i got to work and after some coding i ended up with an awesome result you can now place the color selector and tune its parameters to produce the exact behavior you want in this case i'll choose the right hue and define a small tolerance i'll select the right parameters and voila nice so we should be good to go i'll leave the simulation running and see where it lands us 15 hours in things have developed in an interesting manner our rosokulli primitives have separated into two species the first still has a pretty similar brain but has started to grow a little bit their size gene went up by a significant but still small amount we'll name them rosuculi paudo magus for a little more of pink eye the second species has a lower speed gene but evolved the more complex brain compared to their ancestor and cousin most notably they started to produce pheromones proportionally to how full they are they do not sense the pheromones in any way but now that pheromones are present in the environment it could evolve at any point in the future that's usually how the use of pheromones evolved too in the animal kingdom on earth first a random mutation makes a specie produce a certain alder and further down the line others might have an evolutive pressure to be able to recognize that scent and use it for a specific purpose then they also have the incentive to produce that pheromone in greater quantity essentially being a positive feedback loop of pheromone development for that reason we'll name these ones rosoculi benolet for good smelling pink eye on our eastern island things are dire all the bibbids seem to now be pink eyed too looking at their brain and jeans it seems like they are all descendants of inutilus cascus that just happened to develop pink eyes and are not at all related to the rosoculi species of the south looking around it also looks like the stultus apare species was wiped out the were stupid after all so it kind of makes sense the inutilus genus split in two notable species that look identical but have pretty different brains and behavior the first one evolved to digest things slower as it ages this seems to allow them to be more efficient as they age and need less energy for growth the name inutilus cano salvus for useless aging stomachs is good enough their cousins on their part have started to turn as their stomach fills up this seems like a helpful way to stay around places that have a good amount of food in fact after watching an old video from dvgen i recognize this behavior was a form of indirect kinesis a non-directional change in activity in response to a stimulus basically if they turn when they manage to eat food and have a full stomach they are more likely to stay in the same area and profit from the same food source in our case this seemed like a very good additional safety against running out into the void as a result they'll get the perfect name of inutilus turboplanus for useless that turns when full and finally on our western island the big pingus powlum disappeared in favor of their descendants they have grown even larger and have a mean look in their eyes we'll name them pingis avarus for greedy and greasy and whoops i forgot to pause the game overnight and we accidentally jumped 15 hours in the future we are now 30 hours in and very interesting things have happened first of all our southern island is now populated by big greenish creatures they are big and slow and seem to be the first direct observation of filter feeding so far the older and bigger individuals open their mouth as much as possible and directly swallow all the food they run into however when trying to classify them i run into a little bit of difficulty and i'm getting quite confused first they seem to share a lot of the genetic markers of rosocoli paolo magus and we could even say that they follow a natural progression as paolo magus had already started increasing in size however if we look at their brain they share a structure that is very similar to what inutilus turboplanos from the eastern island had in fact putting them side to side the similitude is pretty evident as a result i'm having a very hard time figuring out what species is the ancestor of our green giants here are they from the rosukuli genus and just a result of lucky convergent evolution or are they from the initialist genus that migrated across the void colonizing the island and wiping out the local rosocoulee population and also happening to lose their characteristic useless connection i tried calculating the probabilities of both options and both ended up being very unlikely and have similar proportions well now you can see how being a biologist and trying to classify species can be a very hard endeavor i think that based on the fact that our green giants have a gene that has the exact same value as rosocoulee paolo magus had 15 hours ago i'm going to put our giants in the rosaceae genius as descendants of the paolo magus i might also be influenced by the fact that i don't want the roseokuli genus to be extinct already but this is a very hard call to make we'll give them the name rosoculi major because of their size moving on from that conundrum and onto the western island the dominating species is still extremely similar to what we had 15 hours ago inutilus turboplanus stays pretty much unchanged as such we'll name the present one innutellus turboplanus cordi for the second inutilus turboplanus i'll probably keep that framework and give the cordy title to species that don't change much after a long time i want to say however that this is probably not standard practice i don't know what i'm doing on our western island we find that a lot of plants seem to have accumulated creating a lush and abundant environment i'm not very sure how that happened since food should grow similarly on all the islands probably a bug looking at the actual inhabitants of this island we see cute red-eyed bibits that should be the descendants of our avarus wait oh god looking into it more these ones are differently from the initialist genius they have the exact same brain structure and the same useless connection that characterizes them this proves that inutulus turboplanus has migrated at least once they managed to survive the trip through the void and reach the west island where they took hold reproduced and eventually wiped out the previous native population of pingus avarus we'll name them inutulus proculis for useless from far away the fact that we now have proof of an existing migration event might make us rethink our classification of the green giant population however the fact that all members of the intulus genius stayed this simple and didn't vary too much makes me think that it would have been less likely for them to evolve so differently in the southern island i've never been this uncertain of a choice in my life but i'll keep my decision and leave them as members of the rosa coulee family moving on and whoops again i've managed to forget the simulation running overnight we are now 20 hours later at 50 hours into the simulation with no save files in between to go to and help us make sense of how things have developed this is akin to our paleontologist might have a lot of information missing from certain periods of the fossil record and it will probably complicate our process a lot i'll try my best to be better and take more regular and frequent saves of the simulation from now on so 50 hours in and our western island still seems to be more fertile than the others the population there looks very chaotic and displays a lot of variety in color looking deeper into it i think i managed to identify three distinct species that are different enough they are not too interesting so i think that i'll name them quickly inutilus unusmultum one two and three for one of the many useless one two and three looking at the situation in our favorite southern island things are surprising we have two very different and identifiable species one displays a flashy green body and a dangerous looking mouth indicating a high-strength gene that would result in higher damage when biting the other one is whip paler but instead has started growing armor and defenses they also produce grainy pheromones when seeing other vivids and leave beautiful trails behind them just to be clear however they are both herbivores which can be seen through the shape of their mouth however they have plenty of other differences that we can look into if we now look at the growth genes we can predict that their growth curve will look pretty different the growth curve of a species just describes how fast they grow over their lifetime and while they can look quite complex for species we find in the real world in the bibits that curve is defined by those three genes and a far simpler formula if we compare our two species here we can see that while their growth start at a similar level they quickly diverge with the paler one growing approximately two times slower also it's interesting to note that both species will continue growing all their lifetime even after reaching maturity that doesn't have to be the case as the formula allows for a large variety of curves those differences surprise me because a higher strength makes growth cost more energy so in the case of our flashy ones here not only do they grow faster their growth is also more expensive this is somewhat balanced by the fact that their size gene is a little lower and they also have a slower metabolism but still i found that surprising if we compare the energy costs of two individuals that are approximately at the same stage of growth we see that the paler species uses a lot more energy overall especially for sustaining their metabolism but our flashy ones just have more energy diverted to growth so faced with the same environment they seem to be going for very different strategies in managing their energy i think it's going to be very interesting to see this dynamic develop over the next generations we'll name them accordingly rosa cooley in fences for aggressive pink eye and rosukouli rigikitis for tough skinned pink eye now going to our eastern island we see an interesting situation first we have these which are rather hard looking and somewhat familiar to a species that inhabited the same island not too long ago however studying the genome and brain of that species they are clearly from the unutulas genus they just seem to have diverged a lot from their ancestor which is not too surprising considering the huge gap since our last study as an homage to the past we'll name them inutilus stultum for useless and stupid the other species inhibiting this island also looks quite familiar in fact by reviewing their brain structure and genes they are definitely rosoculi descendants even more so i think i can safely assume that they migrated from the southern island quite a while ago before the two species we now have there started diverging they share brain structures that this migrator doesn't have they are in far greater number than the stool tombs suggesting that they have slowly been taking more space and slowly driving them closer and closer to extinction as such we'll name them rosocoli austis for conqueror pinkeye this video is already getting a lot longer than i first anticipated so i'm going to stop things here and rush the second part at the time at which this is uploaded i'm already mostly done and will do my best to finish and upload it within a week i don't want to leave you waiting for too long i'm going to announce a big event at the end of the second part so make sure to subscribe to the channel and ring the notification bell to get to see the video early and i know i left the video on a big cliffhanger with our two competing strategies on the southern island but aside from forcing you to watch the next part in order to get closure i think that this is an opportunity to let you the viewer place your bets and try to guess how this situation will develop let me know your theory in the comments and i'll draw a discord key at random amongst the comments that were right allowing you to join the private discord server where my patreons and i are active so thanks a lot for watching and i'll see you very soon in the second part of this adventure [Music] do [Music] do [Music] uh [Music] oh [Music] you

Channel: The Bibites: Digital Life

Views: 877,860

Keywords: AI, Evolution, Genetic Algorithm, Ecosystem, Simulation, Artificial Life, ALife, Neural Networks, Algorithms, Natural Selection, C#, Unity, NEAT, gamedev, game development, Life, Digital Life, Science, Biology, Genetics, devlog

Id: sEPh6bAQVP0

Length: 26min 25sec (1585 seconds)

Published: Thu May 26 2022