Mapping the human brain through online games | Amy Robinson Sterling | TEDxThessaloniki

July 25, 2019 posted by



the world behind us in the world in front of us are small compared to the world within us that may seem weird because you know what human body a human brain is pretty little compared to the size of the earth but from a neuroscientific or a philosophical perspective the human brain is nothing short of extraordinary it's exquisitely complex amazingly intricate beautiful wonderful the human brain is arguably the most amazing thing in the known universe and luckily for us we live in a neuroscience renaissance in a golden age of understanding ourselves and it warrants us to think some wild ideas about the future it's not inconceivable that within our lifetimes we'd be able to hack our own brains think enhancing willpower improving determination banishing worry making ourselves learn faster a lab at MIT can actually already make mice learn faster by stimulating populations of neurons in their brain to fire at a different rate than they normally would and another group of Stanford can reduce anxiety in the brains of mice using deep brain stimulation and this is the tip of the iceberg the penultimate questions in neuroscience are really understanding how a human brain creates a human mind think about it the little three pound organ that sits behind your eyes is responsible for making you think and feel it creates a person who is curious and creative someone who can learn who can love feel pain and persevere like why how does the brain do that it's crazy when you think about it these questions have been rising in my mind since college one night my junior year I found myself coming home from a party reevaluating what was in gordon's me really kind of questioning if the things that were important to me college Amy were actually that important at all and I got a notebook I started writing I started journaling jotting down ideas and one of the interesting effects of writing is that it's relatively slow compared to thinking so when you're writing you have to think slower and at least in my experience deeper I started thinking in wondering what it was for something to have meaning at all why were ideas and things important to me how did a sense of value in virtue emerge and evolve not just in my brain but in everyone's brains and then humanity as a whole what's going on when we have ideas or sparks of insight these philosophical investigations prompted me to start looking to neuroscience to researching reading writing to professor's asking them these questions and again and again I was told we don't know we don't have answers to your questions at first I was dismayed but now I'm kind of excited because I realized that that means that we get to come up with answers to those questions we get to make those discoveries it's exciting and invigorating and amazing and so now let us go on a little journey into the world within this is the human brain kind of obviously it's made of around 86 billion neurons if each neuron in your brain was a person you could populate New York City 10,000 times it's pretty complicated so what you're seeing in this visualization from Adam gives Ali's lab at UCSF is the connectivity and the functional activity in the whole brain so the little sparkly things that you're seeing on the outside those are populations of cells firing sending electrical signals and the different colors indicate different frequencies different patterns of firing those orange wires that have lights moving through them that's the signals connecting to other regional parts in the brain now this is what's going on at any given time when you're doing nothing when you're thinking hard there's lots of different parts of your brain that are active at any given time but at this level you know each of those orange wires as millions of cells it's like looking at the Earth from space which is extremely valuable but there's a lot more to see we tend to think of the brain as a single organ but that's not entirely the best way to think of it so for example this is the brain of a mouse we're going to zoom in to a substructure within the substructure of the brain called the hippocampus now this part of the brain plays a role in learning memory and your ability to recognize where you are or at least in mouses does in humans too so this little purple column that you're seeing those are cells and fuzz that you see those are the branches of cells and they're called dendrites they look like trees they receive electrical input from other cells both locally and from other regions of the brain and if they receive the right amount the right frequency the right patterns the signal propagates down through the dendrites to the cell bodies those are the blobs that you're seeing an outward down the axons to reach other cells in other parts of the brain and then this way you have directional circuitry happening happening all throughout the brain and it's packed there is it is an extremely exquisitely complicated network of cells in the brain in fact it's so complicated that you can't even really wrap your mind around how many how much wiring there is in your head in fact if you laid out all the branches of all the neurons in the brain that's like in your skull right now it would stretch from the earth to the moon ten times it's crazy it's so complicated and these branches belong to multiple cell types in fact we don't even know how many types of cells there are in the brain so I hello another great opportunity for discovery the Allen Institute has recently launched an awesome project called big neuron where they're collecting data from different labs all over the world and they're striving to bring researchers together to create a periodic table of cell types in the brain and a while there may be lots of different types of cells there are there is one main thing that these cells have in common and that's that they connect doosan one of which you're seeing here now a synapse is a junction where one cell will grip another cell and it will send electrical impulses to it and that's how neurons communicate now this is one of 100 trillion synapses in your brain 100 trillion is another huge number these are astronomical numbers so it's fitting to put it in the perspective of galaxies actually if you took all the sin up if you counted the synapses in a cubic centimeter of your brain there'd be as many of there are stars in the Milky Way and you got twelve hundred centimeters cubic centimeters in your brain it's crazy complex and it's not even it's not even a static network in fact some synapses are stable but others are changing through decades years months minutes if you remember anything that I've said in this presentation it's because your brain grew a lot of clear synapses and this Nano world this mysterious place of mostly uncharted territory is the realm in which I work now in order to map out synaptic level resolution circuitry in the brain it takes some pretty intense microscopes and technology that up until a relatively recently it didn't even exist and the biggest hurdles and making discoveries in this area are representative of the hurdles that are faced by neuro scientists and scientists as a whole and that's that the ability to generate data sets has rapidly outpaced software that allows us to analyze that data so my journey into neuroscience began at Sebastian songs computational neuroscience lab at MIT where they build software and specifically artificial intelligence AI now it's not the kind of AI that you're thinking it's not like ex machina you know robots moving around talking compared to human levels of intelligence Rai is more of a savant not really that intelligent basically all it does all it tries to do is reconstruct 3d objects neurons from volumes of image data and that's how we map out the full 3d structure of cells so we can see where they have synapses then pair that with the activity that's flowing through the synopsis and understand how that activity results in function now up until a few years ago it took hundreds of hours to map one neuron now it takes tens of hours and pretty soon it'll be less than one hour but even if it took just a few minutes to map one cell there's 80 billion in one brain and you can't map just one brain you want to map lots of brains across lots of different populations so you can understand a range of normals and you can understand different disease states and then you also maybe could figure out states of excellence in the brain so while we continue to improve the software that underlies neuroscience discoveries we also want to still make discoveries so we looked outside the world of neuroscience for inspiration specifically to the world of games now there are over a billion gamers in the world and collectively they're spending tens hundreds of billions of hours playing games you know if we could capture even a tiny amount of that time it would amount to far more time than all of the people in all of the neuro science labs on the planet so that's the inspiration for eyewire it's a game to map the brain now if we launched I wire a little over two years ago and so far over 200,000 people from 160 different countries have signed up in the game people solve 3d puzzles but as they're solving the puzzles they're actually analyzing data and mapping neurons so this is what the game looks like there are at any given time 50 to 100 people online at this one there are 76 people that thing rotating that's one neuron it's an interesting salads by stratified you can see there's kind of two layers of branches when I turn it to the side this is one of hundreds of cells the players are saying hello from Spain hello from Indiana Oregon Finland they come from all over the world and you can choose which cell you want to reconstruct within the game so here we're going to pick one of the mystery cells it's a mystery because we don't know what type of cell it is and when you actually go to play I wire you had start playing and what you're going to see is a cube on the side of the street now that's a tiny volume of brain it's about four and a half microns on each side that little blue thing that looks like a broken off tree branch that's a piece of a branch of one neuron and what we're going to try to do is map it from one side of the cube to the other as I scroll through you can see a plane moving through the 3d cube and all we're trying to do is kind of color inside the lines color and with one click we added that 3d chunk now that's what the AI does the AI basically segments out an entire volume into 3d chunks but it doesn't know which chunks fit together so this is a great example of humans working with machines to create a force that's more powerful than either alone so when a player in the game has you know mapped a branch across the cube click Submit it cross-references what you added with what was added by other top scoring players highly accurate players in the game and then boom you proceed that one cube put me in 342 place for the day that's basically it there are lots of cubes they have different shapes different structures but that's not really it because I wire is about community it's more than a game because it has a higher purpose we're working together to figure out how the brain works and it's generally non neuro scientists it's people like high school students and sculptors painters tugboat drivers in Egypt in fact when I was recording this you know overview of the game our top scoring player of all time was not online because he's actually here right now messer off in athens hey these in the audience right now so the top neuron mapper in the world is an animator in athens I wire is the first of what will be many games and it's it's very community driven there's happy hours and competitions there are BOTS the query wall from alpha and a lot of the competitions are actually organized even by players but I wire was kind of a proof of concept and it's working amazingly it's working so the next game that will be building will actually be reading memories by mapping out circuits of neurons in the brain it's a real thing it's really going to happen and eventually we will map human brain with the help of a community of gamers from all over the world and games are not even the end we're also going to be crowdsourcing the software itself crowdsourcing better AI inviting hackers from all over the world to create better software better algorithms that are going to help labs make discoveries from their data faster and I think this illustrates an interesting shift in science as a whole I love explorers I'm obsessed with explorers whether it's exploring the brain are explorers through history who charted new routes for trade and first for adding ideas I was once reading Captain James Cook's journals he's a wonderful navigator he discovered new zealand and hawaii among lots of other things but he also took with him experimental gentlemen now it's probably not what you're thinking is actually early scientists though these were rich people who bankrolled expeditions and experiments they brought equipment and they documented what they saw this was before the word scientists even existed it wasn't even until the mid eighteen hundred that the word scientist was even a thing and throughout most of the course of science in history you've had to either have some aristocracy behind your name or in you know a master's degree or PhD but now with the advent of citizen science and games like I wire crowdsourcing even of the software that underlies science itself all you need is passion and an internet connection and an interest in solving puzzles an interest in participating and being a part of something that's bigger than yourself so now a global effort of people from all over the world are shining light and helping us understand the world within us and consequently the world around us now we have learned more about the brain in the past ten years then in the previous ten thousand imagine where we'll be in ten more thank you you

No Comments

Leave a Comment

Your email address will not be published. Required fields are marked *