Paola Arlotta: Brain Development from Stem Cell to Organoid | Lex Fridman Podcast #32

[@gtirsblow]

Lex you are quickly becoming my favorite podcast.

[@itsalljustimages]

I love the way how excited and delighted she is about this conversation.

[@HiberNAT]

im in love * -*

[@TimmyBlumberg]

It really is infectious. The light in her eyes is really inspiring :)

[@cuttheknot4781]

VERY Sexy ( I mean, you know - "Smart-Sexy").

[@lexfridman]

Here's the outline of my conversation with Paola. If you'd like to add other moments to it, please post timestamps as replies: 0:00 - Introduction 1:05 - How unlikely is the human brain? 5:28 - Brain development process 13:57 - Brain development organization as a distributed vs centralized system 20:08 - Nature vs nurture, hardware vs software 22:38 - Brain organoids 36:40 - Our place in the history of understanding the brain 39:45 - Building a human brain 42:45 - Ethics 49:02 - Lessons from raising kids 52:14 - How has studying the brain changed your understanding of self? 54:08 - Merging with AI

[@pleaseforgivemyinsanity2801]

Nice 👍

[@garthwoodworth3558]

55:46 Here she says, "We will be *shaped* by it." She uses strong tonal inflexion for important points. She illustrates with examples how we are shaped by our cell phones. She is aware that this relates to the earlier ethical questions discussion. It's not just what scientists can physically do to the brain, but what technology is doing to the brain. She summarizes at 57:10, "That we want it or not".

[@Pax_Et_Veritas]

13:57 Brain development organization as a distributed vs centralized system

[@UniverseSpaceSpirit1984]

lovely ✨

[@deeliciousplum]

Lex's interviews are like being treated to an exceptional meal that was prepared by chefs who are passionate about the recipes that they created. There are not that many people, dead or alive, that I would love to have a coffee with. Paola Arlotta is someone who I would love to have a coffee with and I would ask her permission to ask a gazillion questions in-between the numerous biscotti and espressos that we would be enjoying. 🌻🐈

Wow, she looks so happy talking about her children. What an incredible person! Thanks for this one, very enlightening

[@TummalaAnvesh]

This is the best interview of all you had. Bringing the people from Biology, Neuro-Sciences and getting their ideas will help in understanding AI better.

[@soulslayer59]

Oh waow ..travelling back 3 years in time and checking this podcast now ..thanks lex for spreading info and love .

[@physicsman3788]

Paola really knows things of Brain development, very clear & logical mind expression. I didn’t expect much before watching this conversation, but I’m impressed now. We’ll done, LEX.

[@obiwanbenobi4943]

This is wonderful. She's great to watch and listen to. In love with her subject and articulate enough to express it well.

[@OEFarredondo]

I’ve learned more from your talks and others then at school.

[@PecPur]

This channel has become for me a source delight!

[@christianlagares-linares9853]

Lex, I discovered your content a few days ago and now I can't stop watching it. Thank you

[@BiancaAguglia]

It's hard to watch this video and not be tempted to google "how to become a brain biologist." 😊The one reason I'd love it if we figured out how to extend the human life span by a few hundred years is that it would give each of us a chance to explore several fascinating careers. 😊 Great interview.

[@squishler]

Yes! Made me miss my science lecture halls from college. Can you imagine being a student of Paolo's? Wonder-full head-spaces.

[@BiancaAguglia]

@@honestexpression6393 I hope that’s not the impression I’m giving. I’m actually trying to encourage the KZbinrs who post videos about AI and data science. Many of these creators are doing a great job making these topics easy to understand and intriguing, even fascinating. I started this work about a year ago. I was completely clueless about both topics then and very intimidated by them. I even suffered mild anxiety because of it. 🙃It seemed that everyone talked about the inevitable doom AI will bring. I decided to learn on my own why AI was so dangerous, and I realized not everyone thought AI was something to be scared of. There are many people who see us having a great future because of AI, they’re just not as outspoken as the more pessimistic forecasters. I’ve also come to realize that, since AI is not an actual entity who has intelligence or will (it is just a bunch of mathematical functions and pattern matching algorithms running at incredible speeds), we’re the ones who determine if AI will be used for good or not. I like leaving comments to encourage KZbinrs to post more, but also to inspire people who watch these videos. My hope is they’ll start learning on their own about AI and data science. I’m sure many can fall in love with these tools and start using them to make the world better. I never mention my channel in my comments. First, I have too little content there to be of any use to anyone and I don’t want to waste other people’s time. Second, I don’t think it’s good manners to promote my channel to somebody else’s audience. Yes, I do hope that my channel can grow and be useful, but I’m not confident it will. I struggle with doubts (since I’m still a newbie in this field) and with the impostor syndrome. Plus, there’s already a lot of good content out there, maybe my time is better spent trying to help that content grow. That’s another reason I leave comments: videos with higher engagement have a better chance of growing. I’m sorry you felt I was promoting myself. I talk/comment a lot, but I was hoping my messages were helpful. I’ll be more mindful of my commenting. 🙂

[@metafuel]

Not sure what everyone else's problem with this podcast was. I thoroughly enjoyed it. Awesome and lovely guest. EDIT: Apparently all the people who were grumpy about the podcast have changed their minds and deleted complaints. Great podcast.

[@metafuel]

@@cuttheknot4781 I beg your pardon? Are you nine years old? You're in the wrong area of the internet friend. This is for intellectual adults.

[@kipling1957]

It’s like the opening scene in Godfather. This guy in the black suit looks like he means it. I’m waiting for the “offer I can’t refuse.” Love the intro, Lex. Great interviews.

[@nickking6371]

Lex youre a beast. I cant stop watching. I just changed my major in my 6th semester from healthcare to CSE and mechanical engineering. Youre guests are so eclectic. Thats the key to being great and making great advancements. Ironically the same plasticity and adaptability that make our brains ssubject to evolutionary forces providing us with survival advantages, that same concept works with your podcast and general heuristic strategy. Hows that for recursion? Lol. Basically, the more eclectic your guest line up, the broader your reach within the parameters of ideas. Progress stalls when the perameters of new ideas dont meet the needs for developing new ideas of complex interdependence. If your field of vision is too narrow you can't find broader solutions. If you can only see a field of 90 degrees, you'll never connect two dots 135 degrees apart. By having an eclectic guest line up from different backgrounds you can connect the dots between broad fields. Aka general science. The beauty of the internet is that were all connected now and only limited by personal bias. Our strength lies in objectivity and flexibility that reflects the natural tendency of our brains to adapt to eclectic interdependent new solutions. Youre the man.

[@p5742j]

you're doing amazing. thank you for everything. i will be supporting you financially too as soon as i find a job (new graduate). for real.

[@anonymously94]

You deserve so much more recognition for your work. Thank you for the great content Lex.

[@supersnowva6717]

Fascinating conversation just as our 🧠 is! Thanks Lex!

[@michelaka6836]

One of the best and most engaging interview so far. Thanks for your continued efforts on this front Lex.

[@sibonozen3876]

Thank you for interesting interview (and fascinating inputs). Grazie mille a Paola e Lex!

[@MrSativacyborg]

Thanks Lex. Keep em coming!

[@Seehart]

1:30 I would argue that the development of the complex brain capable of toolmaking has happened twice, completely independently, on Earth. The peaks of those paths are primates represented by human and and cephalopods represented by octopus. (One could argue that some birds also exhibit toolmaking skills, but they would not qualify as independent as the common ancestor of the amniote clade had a well developed nervous system). Relating to the Fermi paradox, 2 is way more than 1. The divergence between these paths was the protostome/deuterostome fork, which developmentally manifests at 8 cells, way before anything resembling a complex nervous system. Hence can be regarded as independent. Although octopuses have not reached the level of humans (i.e. they don't, to our knowledge, have a space program), they are comparable to primates (aside from social context). So the right set of environmental pressures would do it (perhaps something like the current anthropogenic mass extinction). For the human path, this phase shift happened less than a million years before space travel. Opposable thumb vs. tentacle? I'd put even money on the tentacle.

[@joepesci5845]

What a GREAT talk!!! Your podcast is the best Lex!!!

[@Seehart]

Amazing conversation! It stimulated this pipe dream in me: Can an organoid be coaxed into integrating into a living brain? Can an organoid be grown on a neural interface scaffold, such that rudimentary interfacing starts in-vitro? If a yes can be found for those two questions, can a high bandwidth BMI be grown in-vitro from host stem cells and then implanted?

[@MariaMiwkina]

All great questions! The simple answer to first two is 'Yes', BUT: (1) although organoids can be transplanted into the living brain and they form connections with surrounding structures, there are challenges to fine tune balance between number of cells vs safety of transplant; and (2) current in-vitro systems allow organoid growing on a multi-electrode array (MEA) system to read out information from it about electrical signals between neurons but programming of these signals is non existent for now - just simple stimulation of neurons is possible but not a complex encoding of a certain function for a neuronal circuit. But like Paola said the field is still very young! And high bandwidth BMI is more complex than MEA and offers more possibilities. For now though, when it comes to anything that is to be transplanted into the brain, with current knowledge of protocols to guide stem cells into neurons, it is safer to deal with a 2D neuronal cell culture than 3D organoids to reduce the number of undifferentiated cells and minimize risk of transplant overgrowth (this in fact is currently true for most types of cells not just neurons investigated in the context of cell therapy). I hope this helps! :)

[@squishler]

What an amazing conversation! Excellent questions and excellent answers.

[@huguesviens]

Once again an amazing interview. Thanks you for the awesome channel.

[@dayanandak4667]

it was a fantastic podcast.

[@Daniel-oj7bx]

things about the brain are always the most exciting ..great talk!

[@Dima-rj7bv]

I haven't expected neuroscience can have so much in common with a field of AI. How much we can learn from the human brain and human evolution. As well as how the future of brain evolution will look like. The talk inspires to dig deeper into neuroscience: to understand better human nature, to steal some ideas from nature, to better understand myself.

[@guzma242]

I can tell how excited he is to be able to speak with her. ^__^

[@fogartymb]

Great conversation! Very thoughtful questions.

[@redcactuslime]

I love podcasts about brains

[@ianevans2697]

Very interesting. The ability to arrange organoids/neurons from stem cells in a way that is anatomically identical to the homogenous structure/ circuitry of the neocortex that was mentioned in the other great interview with Jeff Hawkins and to arrange the neurons in a way that is optimally conducive to observing their function with electrodes seems like a great opportunity to study what algorithms emerge once those neurons are hooked up to different sensory inputs. Don't know what I'm talking about, just trying to piece together this and that conversation.

[@Miki736]

Rock on Lex! Regards from Poland

[@DizzyPanda]

Very interesting subject. Stupid KZbin algo should promote this podcast more.

[@huuud]

Fascinating talk! 🙏

[@koningsbruggen]

oh yes! AI and regenerative biology are my favorite topics.

[@LoisSharbel]

Fascinating interview! Thank you for introducing me to this charming, brilliant scientist, Paola Arlotta. I can't wait to hear more from her investigations.

[@Nodal468]

This was amazing!

[@renereiche]

OMG! Lex wore a tie with COLOR for this one 🤯 Also...mechanical forces on the brain direct what kind of function a part of the brain will take? That sounds... do we have good proof of that? That's hard to believe for me. The idea that the proximity to other parts of the brain that have already started to take function (for example the vision part that has started to shape because of the proximity to the nerve ending of the eyes), shape the function of new parts seems ...at least easier to grasp.

[@simonstrandgaard5503]

Great explanations.

[@paulkiat]

This is very exciting.

[@sabofx]

great talk! thanx Lex!

[@stefano4861]

it's great to see some italian in the podcast ahah

[@lambi1966]

I love Human beans :D

[@onionface5835]

I think Lex is an AGI and he's doing these interviews to gather data on all the humans that could possibly stop him from developing.

[@samuelec]

Very interesting thank you

[@LiftingHard1989]

Would have been interesting what her thoughts might have been on consciousness and free will

[@AntoineDennison]

During the interview, it occurred to me that the idea of an artificial intelligence system that mimics the brain may not be a proper method of framing the development of artificial intelligence. The brain, having inherent plasticity and evolving over millions of years in response to highly varied objectives, may mislead the development of artifical consciousness. For example, we may not need artificial intelligence systems with a sex drive. While much of the mammalian sex drive can be classified as objective oriented development, the reward mechanisms, like endorphins may not need to be developed. Also, we may not want artificial systems which act on fear, anger or any other primal emotion, all of which are built into the 'firmware' of the human brain. I suspect that we primarily want systems that are able to employ complex thought processes, which is not necessarily aspiring to develop systems which are able to 'think' like humans. When we employ the human braIn as a model for artificial consciousness, we should begin by understanding that to 'evolve' is essentially to change over time, whereas development suggests intent. In my view, the question should not be 'can we replicate the human brain', but can we artificially replicate the ability to reason.

[@meows_and_woof]

We cannot get the full code. We can never outsmart millions of years of evolution. Every embryo has a DNA which is actually an instruction for cells how to divide and what to become. It’s so sophisticated and fine tuned that if there is too much or too little of those instructions ( extra chromosome, or lack) then the baby will have deformities , many times in many parts of the body. Early development is very complex, there are chemical reactions, electric impulses, mechanical changes. And it’s done in a chain reaction manner. Everything happens one by one. Every time. If by chance one step is not complete and something has not formed, it will not form later on. Every organ has its specific time for development. We can try and use our stem cells in order to repair a damaged tissue or grow organs, but the brain is so complex that I doubt it’s possible. When brain develops it not only develop the brain in the scull but also nerves which like roots go through entire body. You cannot separate the brain from the nerves system. It’s all one. That’s why it’s so difficult to grow the full brain in the lab. May be they should try creating a matrix of cells and grow brain cells along with other tissue so it progresses into more development . I feel those organoids are small because at some points it needs to connect its nerves to other tissues but there is no way so the sequence stops.

[@epiconvergence]

09:35 -10:20 Am I the only to think that this was love making on an intellectual level ?

[@T19R0N]

Awesome

[@---df5sr]

The new Joe Rogan

[@jakubpruher979]

Good entertainment.

[@angelmonegro7364]

Paola Arlotta te amo mi amor

[@weekysleeeece]

31:52

[@VaclavKosar]

no mention of bio electric computing?

[@raphirau]

Amazing conversation, but dude, you look super tired! Get some sleep! :D

[@keylanoslokj1806]

AGI will be not at all anthropomorphic. without the biophysiology of the human body, and the sensory input and development, any concept of an anthropomorphic AGI is nonsensical

[@piyh3962]

Biomimicry anthropomorphic

[@noteniceu]

This dude reminds me of jehova witness. Never the less a fine gentleman.

[@cuttheknot4781]

More like Eddie Monster

[@grungecrunge]

Eddie Monster?

[@Dazzer1234567]

hmmmmmmm............. frustrating politically correct answer from her regarding the question about building of a whole brain. Although she may consider it science fiction and not something that she wants or needs to do, it's undeniable that something like an artificial brain that had the complexity and compute power of a real brain would have absurdly high economic value. Therefore, if it can be done, it will be done. Perhaps not in America, but then in China, Russia or elsewhere. So to just poo-poo and laugh off the idea may serve her career well, but i don't think history will look back so kindly on this sort of attitude. Just my humble 2 cents.

[@afriedli]

I think she did kind of answer the question about the ethical implications of growing a 'true' artificial brain by saying that the specific scientific data being generated by researchers doing the work must inform any ethical discussion (so people at least properly understand what they are discussing). She also said that she believes the work scientists are currently able to perform in her area of research does not raise serious ethical concerns. Science, especially in complex areas like neurobiology, proceeds in small, distinct steps, where new insights and techniques continuously redefine what may be possible and steer the direction of further investigations. However, you cannot forecast in advance how this is going to proceed, and therefore what specific ethical questions you may be facing at any given time. The best approach, therefore, would be to take each step, see where that lands you, and then determine the ethical challenges you potentially face going ahead from there. You can't know this in advance any more that you can know your research results in advance. You raise an interesting question about whether the rules under which researchers in countries such as China operate may become very different to those in the West. Although there is competition between researchers and labs around the world, there is also a lot of co-operation and cross-pollination because everyone understands that is the most effective way of making advances in the field. Science is becoming more and more global because difficult and complex problems require global resources, global talent and global communication to solve. This trend should alleviate the risk of rogue researchers and labs emerging that operate according to dubious ethical practices because those researchers are fully integrated into the global scientific and academic community. The people capable of making significant breakthroughs are part of that system.

[@paweloneill5888]

Trump's brain formed completely in 20 min.

[@gamingforfun4554]

interesting)

[@spaceexplorer3690]

Paola if you dont have boyfriend i am here.

[@MrRayne911]

WOW your questions sucked! How about being more prepared next interview. Paola clearly seems extremely knowledgeable on everything brain related and has a lot of interesting things to say. Obviously this one hour highschool-interview style is not enough. If she could take the lead on a mini series from eli5 to her state of the art and what she thinks is important I would gladly throw 50$ as a donation to the project

[@stevenjohnson9466]

ughhh the toxicity :0 I'm cringing

[@tysaylor7079]

Sign me up for that miniseries tho

[@grahamjoss4643]

Where’s your btc address

[@LaurenBridges37]

Aww yeaaahhh! Diversity

[@LaurenBridges37]

Oh man ...let me clarify: I didn't mean diversity in the people Mr. Fridman chooses to interview -I just mean that it's nice to see some other Lady-folk in the upper eschelon of AI content -very motivating and I am a super-grateful, huge fan of Mr Fridman.

[@blink2665]

Is she Italian? Damn does Italian women age well :)

[@swapanjain892]

Yeah,I thought she was Swiss...

[@dominicklittle9828]

One of the more interesting conversations but her voice I just can’t do it it’s too up and down

[@obiwanbenobi4943]

It is musical. Italians are like that. :)

[@OutdoorPOV]

I feel like she would be able to explain this more well if she spoke in her own language