Brain tissue samples suggest we stop growing new neurons in our early teens | Hacker News Reader

Brain tissue samples suggest we stop growing new neurons in our early teens

latimes.com

139 pointsmerrier51 comments

I see this as a positive. I always lamented the idea that as we grew older we generated less and less neurons, and thus our learning capacity decreased (I thought).

But it seems even young university students are no longer generating neurons. Nor are the 30 year olds learning new languages or switching careers.

It seems neuron growth thus has nothing to do with 'plasticity' or our capability to learn.

* disclaimer: I know very little of neurology so please correct any false assumptions.

True! But someone who learns to play the guitar, or a language at 8 ~ 15 years old shows skills that are very difficult to master (if not impossible) when you are older. There's a lot of evidence of this if you talk with guitar teachers, someone who moved and start speaking a different language when adult, etc...

Can you point to any studies? I don't know if this is because our brains get less plastic, or because we just stop bothering when we get older.

I have personally noticed some drop in language acquisition skills, but I'm learning Spanish at 35 and I wouldn't say it's significantly harder than learning English at 12 (apart from the fact that English is everywhere around me all the time, but Spanish is very hard to come by).

I've got Japanese friends who live in the UK for more than 10 years now, and it's more Engrish then English, than anything else. My english sucks by the way.

Maybe this will help you?

http://www.jneurosci.org/content/36/3/755

http://pandora.cii.wwu.edu/vajda/ling201/test4materials/seco...

https://www.sciencedaily.com/releases/2012/03/120328172212.h...

https://www.sciencedirect.com/science/article/pii/S0093934X0...

https://books.google.co.uk/books?id=5vvY2BDWZjEC&pg=PA251&dq...

https://www.justinguitar.com/en/BC-101-CommonQuestions.php

Hmm, I don't see anything in there related to skill acquisition in younger vs older people, I'm afraid. Most of the links are about skill acquisition in general, and how the brain changes. One is to a Google book that won't show the page (I don't know why, it says the page is unavailable), and the last link is a teacher saying he guesses children learn more quickly because they're more plastic (but I guess they learn faster because they have more free time/willingness to practice)...

Children are blank canvas, adults have habits; repetition makes it great but it also perpetuates habits. To achieve mastery in some areas you'd have to be born again.

Pinker's The Language Instict.

https://en.m.wikipedia.org/wiki/The_Language_Instinct

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2947444/

Noam Chomsky, the world's most popular linguistic https://study.com/academy/lesson/noam-chomsky-on-language-th...

Same goes for chess.

15 is already considered very late to start serious study in chess. You simply do not develop the unconscious mastery the blink like ability to play a good move instantly.

It is not the lack of time in older players either. Tales abound of players of independent means dedicating themselves to chess and getting nowhere near master level.

By reverse, no chess player started at 25 and made Grandmaster.

A huge outlier is https://en.wikipedia.org/wiki/John_K._Shaw who was a club level player at 19.

I’m not sure these claims are entirely true.

Quick search turns up this list: https://www.quora.com/Is-there-any-famous-chess-player-or-gr...

Chigorin is from a different era, but he didn’t begin playing “seriously” until 24, and was a 2 time world championship challenger.

Getting to master (2200 rating) level play is ultimately just an investment of time and money. For better or worse, the “consensus” is that master level play is not due to some kind of innate ability. Drilling the pattern recognition inherent in all phases of play is the key determinant for reaching ~2200 rating. Do a cursory Google search, and you’ll turn up hundreds of threads about this topic.

Becoming an international master (IM) or grandmaster (GM) requires navigating FIDE’s, the world’s chess governing body’s, arcane rules around norms, or results, for the GM or IM title. I’m by no means an expert on the scientific piece of this, but I’d posit that the number of adults who have the financial means and time to travel across the United States or world to find tournaments that will earn them a norm is negligible.

When you’re young and your parents are paying the bills, chess can seem fun and rewarding. When it’s you personally who is paying the bills for tournament registration, hotels, and planes, it’s a lot less fun.

Alas that Quora tread is full of questionable answers as Quora threads tend to do.

My bias: I am a FIDE master who skipped chess study for a year when I was a teenager. (insert Brando quote from On the Waterfront here).

Notice that the main examples that would disprove my premise are from 19th century. You'd need chess historian Winters to sort the claims but Chigorin's "seriously" is something to debate.

Why aren't modern influx of adult players making masters? Plenty of adults playing in tournaments all around the world spending thousands on travel and equipment and training. Most of us are having fun but the improvement is not there.

Also GM Klovans was a master level player in his teens and GM strength player as a young adult. Source: knew him personally and cherish the only draw I had against him.

The few modern day examples started in late teeens and fit within my narrative.

If there is a "consensus" that becoming a master does not require some innate ability then where are those masters? Show me a master level player who started at age 30.

The chess.com megathread on this topic came up with an expert who achieved his title after starting at age 30 and that really is an outstanding accomplishment and a huge outlier. If "consensus" is correct there should be more numerous examples.

Pretty much all adult players have hit a serious wall this you can see by looking at rating charts.

The only exceptions are returnees who dedicate themselves to chess full time and even then the gains are miniscule 50-100 rating points at best and then they hit the wall and that's that.

Thus I take serious issue with your presumption that "Getting to master (2200 rating) level play is ultimately just an investment of time and money."

My thesis that chess skill/rating = talent * early start * dedicated study . You need all 3, if one is missing your chess skill will be seriously hampered.

If you're starting play at age 30, sure, there's not likely to be very many players who make the 2200 threshold. That would be interesting to crawl through the USCF database to check, though without birth years there'd be a good amount of manual checking. Your original question was however, who started at age 25 and made GM.

My own bias was as a competitive junior, who had several of my peers reach 2000-2100 before they were in high school (some as early as age 10). Several of these kids had languished for a few years with mediocre progress, before taking on serious study habits and effective coaching. Almost to a person, they stopped playing in high school because of academic/athletic/college time demands.

Again, maybe there's some biological/scientific reason why people late to the game of chess cannot perform the pattern recall that is central to chess. But, I'm still inclined to think that the demands of "real life" are a key reason why players starting at an older age are unlikely to reach master strength. Yes, there are many adults who devote time to the game. But, how many are able to devote the same amount of time and energy that a schoolkid is?

Considering that as we age we lose some of our neurons - even if not _because_ of aging itself but for other reasons such as drugs or disease - there still could be an observable difference in both 'plasticity' of our brain and our capability to learn.

The neurogenesis hypothesis was always more about mood/mental health rather than learning, I believe.

I think learning is more about losing neurons, not creating it. Or meatspace neural netowrks are less about changing weights and more about pruning connections.

I can't find the link, but I recently read an article that sugggested people with autism did less connection pruning during puberty than people without autism.

I mean, we already know the differences between 20- and 30- year olds from simple observation. So it's not really positive or negative.

What would you say the differences are, based on your observations?

Not my observations, observational studies.

Regardless of the exact reasons why plasticity declines with age, it's obvious that any fix is a long way off. I wouldn't take this news as positive, rather negative, actually, since it just means we have yet more work to do

Like, instead of saying "I always lamented the idea that as we grew older we generated less and less neurons" the GP can now say "I always lamented the idea that [other unknown mechanism behind declining plasticity]". Hardly good news.

This article has some good insights from other researchers and their different opinions on the study:

https://www.theatlantic.com/science/archive/2018/03/do-adult...

Adult neurogenesis in rats was my thesis focus, so it's interesting that all the rat work we did may or may not translate over to humans.

I'm gonna wait for the replication, as always.

But what about all that exercise-induced BDNF-mediated neurogenesis stuff? Was that only observed in mouse models?

Neurogenesis in the hippocampus and olfactory bulbs is ongoing trough the lifespan. I am presuming this is not about that neurogenesis.

Actually, that's exactly what they're claiming to refute:

> A small number of papers had indicated that humans also possessed this capability, specifically in an interior region of the brain known as the hippocampus, which is associated with memory.

> A team from UC San Francisco had hoped to see evidence of this neurogenesis in action.

> Team members examined brain tissue samples collected from 59 human subjects who ranged in age from a 14-week-old fetus to a 77-year-old man. But they couldn't find what they were looking for.

I am floored.

Absolute pitch is something that supposedly can only be learned before the age of 6 but the drug Valproate seems to reopen this window.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3848041/

I learned it as a teenager and lost it in my 20s. Like anything I learned it through practice and familiarity. Once I stopped working in music production I slowly lost my absolute pitch too. Similar to the way I used to be able to speak French but now can only string a few words together.

Similarly, I took German and French in school, and got quite good grades. However, I've not had much occasion to use either language since then. I did read some German news articles, so that part is sort of OK still.

I've been with my (German) girlfriend for about a year now, and have visited her family in Germany a couple of times. It has really made me realize how much of my ability to put together a sentence in German has simply vanished over the last 15-16 years. I'm boning up using Duolingo now, because I really want to be able to have a full conversation with her (not very English speaking) family.

I've had the opposite experience - once I learn something like a language, I never forget it. But people around me do, and I think it's as strange as forgetting your own name or who your family was. On the other hand, I'm mystified by how efficient "normal" people's memory for places and directions is. It takes me hours of studying a map and many repetitions of practice to learn a new route, and if I don't go a certain way (even to get to my own house or in the neighborhood I grew up in) it fades just like people describe losing language. So it's like those two capabilities are exchanged in my brain.

My best friend and I took 2-3 years of German in high school, and we'd talk to each other using the German we learned. We also knew two German exchange students. In academic meets (test taking competitions) I placed in German II and my friend placed in German I.

Unfortunately we had no one to speak German with after graduating high school. In all the time after high school, I only ran into one native German speaker, at some conference. I said something like, "Ja, Ich spreche ein bischen Deutsch..." and based on my accent he said, "Oh? You've been to Germany then." I shook my head, sadly no I've never been.

Years later I saw my friend again and spoke some German to him, and he didn't understand me. So I used German phrases I remembered had been his "favorites", and I was shocked he even forgot those too. It felt like I was one of the Lost Boys in the movie "Hook" trying to get Peter Banning to remember he was Pan.

I was having the same sort of difficulty with French. On vacation, I realized my ability to speak fluently had vanished over a decade of non-use of the language. I bought a few magazines, then a fiction book, and it's done wonders.

It's much more interesting to read about something that I'm interested in than it is to go through exercises. Since what you lose over time is mostly vocabulary, reading is an efficient way to re-grok a language. You just have to get over the initial bump, where you're reading really slow because you stop frequently on unknown words.

My girlfriend has taken Danish classes for two years, and we try to speak as much Danish as possible, with me giving her feedback on grammar and word choices. In return she sends me news and other articles in German, where I try my very best to read and understand them fully, without resorting to Google translate. So far it's going pretty well, actually :-)

I suspect that valproate works by increasing BDNF levels. Exercise also increases BDNF levels, and a diet high in fermentable fiber has been linked to increases in BDNF as well, through increased short chain fatty acid production. It is worth mentioning that valproic acid is just a branched-chain short chain fatty acid, so it is likely that many of its effects are similar to those from short chain fatty acids derived via dietary fiber fermentation.

Wait... so exercise and beer? Heavenly.

Mix in some oatmeal and eat your vegetables. Who would have guessed.

I thought they used isotope analysis from nuclear testing fallout to prove that the neurons in the brain are between 15-20 years old on average? How do they reconcile this?

http://www.physicscentral.com/explore/action/nuclear-neurons...

A likely possibility: the molecules get replaced, but it's still the same cell.

Theseus Ship of cell replacement?

See The Delphic Boat [0] for a discussion of this in detail.

[0]: http://digitalcommons.dartmouth.edu/cgi/viewcontent.cgi?arti...

Wow, thanks for the link. Kind of humbling when you think you've hit on a clever reference and someone's written a book about it =)

DNA doesn't get replaced that I know of, and that's what they focused on.

The only way to get "new" DNA is for a cell to assemble a new strand during division.

DNA repair mechanisms can synthesise new segments in response to damage.

There was also a paper published last month claiming DNA synthesis following learning, but I'm sceptical:

https://www.ncbi.nlm.nih.gov/pubmed/29427086

Interesting paper. Do you work in the area? I'm working on proving that memories are in fact stored in genetic material, as was believed back in the 60s.

Drop me an email (gauravvman at gmail) if you're interested in this kind of stuff! Would like to know how you came across this paper.

Afaik nature has good stuff, but it's not like a 'research journal' it's more popular stuff. A quote from the article: "There are only a handful of studies out there that have already attempted to look at this, and they came to wildly different conclusions." So, although I didn't look into this more deeply, it seems to me there is wide room to not take this popular article as 'full confirmation'.

Nature is basically the most prestigious scientific journal, along side Science. Getting a pub in either of those journals is considered a huge win.

Nature is a research journal. In fact it is one of the premier research journals. A publication in Nature can make a career.

Nature is not a specialist journal however. Most fields publish in it.

Well I guess the 'afaik' in my answer was wrong in this case, it was based on recollection of the last time I held a nature in my hand, and a comparison with a sort of definition as listed here: https://www.library.georgetown.edu/tutorials/scholarly-vs-po....

Sorry to unjustly disparage nature, however, it does not discount the quote I lifted from the post, which is still a valid point.

Nature is ... kinda good:

http://www.scimagojr.com/journalrank.php

This worries me. First, any study of neurology it must be kept in mind is done on a specific population. The neurology of a Papua New Guinea highlander and an upper middle class white kid from the suburbs of NYC are not going to have the same brain development. Brain development does NOT proceed as a consequence of aging. This is a dangerous and destructive myth that gets perpetuated all the time. Brain development proceeds as a consequence of experience. If you eliminate experience from a persons life, you will eliminate brain development.

We can see this clearly in the very young. An infant must master binocular vision by about 3 weeks old, or else they will never master it. Putting an eyepatch on an infant because they are "not ready" for binocular vision or because their brain "isn't mature" in the region responsible for that would damage them for life. A child must be exposed to the general concepts of language before about age 5 or else they will remain incapable of learning any sort of language-based communication for their entire life. Likewise if you were to restrict them and keep them away from language exposure because their brains are 'not mature' or because they're 'not ready' for it or whatever, you would do them grievous harm by actively preventing brain development.

There are similar critical periods and developmental milestones like the Existential Crisis where children develop an understanding of death as the end to life and understand their own mortality around age 10. It was always thought that neuroplasticity pretty much disappeared after adolescence. I've always wondered if that was a consequence of the way the life experience of a person in a western society changes after adolescence. That is typically when almost all experience of novelty comes to a grinding halt. Jobs are taken up, spouses found, routines established. And without unique experience providing intense stimulus to a brain, it will stop developing. Absent those stimuli, it never would have developed in the first place. It is the origin of growth.

If we start to see neuroplasticity decreasing even earlier in members of our society, I would not be terribly surprised given the degree to which adolescence has been either criminalized or heavily regulated. And we will pay the price for that.

Unless it happens somewhere other than the hippocampus..

I was wondering as well, is it possible that there’s another part that it happens instead?

Adult neurogenesis has been seen in other species in the olfactory bulb, I don't think this has ever been observed in humans though.

Is this a blow to neuroplasticity?

Not this result in itself. Neuroplasticity refers to the ability of existing neurons of creating/pruning connections to other neurons.

However, separately, there is scientific evidence to show that neuroplasticity decreases after adolescence.

I appreciate the suggestion, but how am I supposed to do that?

And why would I listen to a brain tissue sample anyway?

Reply on news.ycombinator.com