The heritability of fertility makes world population stabilization unlikely • Hacker News Board

Re: The heritability of fertility makes world population stabilization unlikely

#101

Post by neffy » Fri, Jul 05, 2019, 2:49 PM UTC

The Beak of the Finch by Jonathan Weiner describes (extremely well, it´s a great book to read btw.) the evolution of Finch beaks on the Galapagos Islands. There was one particular Finch/beak size that had relatively small families and population growth, and was clearly being outcompeted by other, better adapted Finches.

Until there was a severe drought, when that type of beak was able to take advantage of a neglected source of foods, which wasn´t accessible to the other Finches - who were more sucessful, but only when there was abundant easy food. After the resultant die-off this particular Finch subtype dominated of course.

A lot of the discussion of evolutionary traits implicitly assumes an adaption to an unchanging environment, but that is very unlikely to be the actual situation, especially as in adapting, the adaption itself can change the larger environment.

Re: The heritability of fertility makes world population stabilization unlikely

#102

Post by aluren » Fri, Jul 05, 2019, 2:59 PM UTC

j_wtf_all_taken wrote: post #98

xamuel wrote: post #87

Earlier quoted context omitted.

This obsession with journal impact factor is toxic in academia. It's what allows parasitic publishing companies to continue to extract rents despite contributing nothing to science for decades. The majority of papers are envelope-pushers, churned out in enormous numbers because of "publish or perish". Academics who need CV bullet-points are going to churn out quick papers that go with the status-quo, with bibliograph…

I don't get the point. If science does care about the JOURNAL impact factors too much, why then, as a consequence, would everyone publish in low impact factor journals?

Low-IF journals are easier to publish to by definition. You first try to publish in a high-IF journal, if they reject you then you try one with lower IF, and so on. I'm all about not taking IF as face value and don't need a n-th reminder about the metric's issues seeing how rehashing them is something of a favorite pastime among scientists, but in this specific case other more qualitative assessments don't line up either:

-Journal is completely obscure to the genomics community

-No causal genetic mechanism is shown, everything is shoved into a "heritability" black box that some people seem to think is like your video game character starting stats or something

-Authors don't have a genetics background

-It's not my field, but there appears to be a wealth of literature on how fertility rates decrease in history, none of them involving "heritability" and little is done to address, reconcile or unify that

-There does seem to be darker political overtones that are exacerbated by commenters, leading me to think the arguments are not being made in good faith

Journal IF is just a (flawed) heuristic but in this case it sets a low prior and the "updates" didn't help

Re: The heritability of fertility makes world population stabilization unlikely

#103

Post by xamuel » Fri, Jul 05, 2019, 2:59 PM UTC

aluren wrote: post #99

xamuel wrote: post #87

Earlier quoted context omitted.

This obsession with journal impact factor is toxic in academia. It's what allows parasitic publishing companies to continue to extract rents despite contributing nothing to science for decades. The majority of papers are envelope-pushers, churned out in enormous numbers because of "publish or perish". Academics who need CV bullet-points are going to churn out quick papers that go with the status-quo, with bibliograph…

While I agree there are plenty of issues in academia among which too much emphasis on IF is certainly a thing, it does provide a helpful filter (or prior, if you will) to quickly assess whether some material is worth your time. In this case, stuff in low-IF journals that are virtually unknown in the community (e.g. there are a few journals like Bioinformatics that are also fairly low-IF but very well respected), whos…

I'm not trying to score points in favor of a political position (please follow HN etiquette of giving posters the benefit of the doubt).

The filtering you refer to may be useful[1] when looking at bulk lists of papers (e.g. when sitting on a hiring committee), but this is an example where the article has already been curated for our attention, namely: by being upvoted on Hacker News. You can still choose to filter it out, but it's silly to write comments attacking the paper based on the journal's impact factor. Better to grapple with the paper's actual scientific content.

[1] (Useful in the same tragedy-of-the-commons way that it's "useful" to leave a picnic ground without cleaning up after yourself.)

Re: The heritability of fertility makes world population stabilization unlikely

#104

Post by xamuel » Fri, Jul 05, 2019, 3:09 PM UTC

aluren wrote: post #102

j_wtf_all_taken wrote: post #98

Earlier quoted context omitted.

I don't get the point. If science does care about the JOURNAL impact factors too much, why then, as a consequence, would everyone publish in low impact factor journals?

Low-IF journals are easier to publish to by definition. You first try to publish in a high-IF journal, if they reject you then you try one with lower IF, and so on. I'm all about not taking IF as face value and don't need a n-th reminder about the metric's issues seeing how rehashing them is something of a favorite pastime among scientists, but in this specific case other more qualitative assessments don't line up ei…

>There does seem to be darker political overtones that are exacerbated by commenters

Are we reading the same comments section? Which specific comments are you talking about? The comments section I read seemed quite apolitical, except for this particular sub-thread.

>Low-IF journals are easier to publish to by definition.

It might be true that Low-IF journals in general are easier to publish to, but that's not true "by definition". By definition, Low-IF journals are journals whose papers are less-cited (this is an oversimplification, of course). For example, IF tends to vary a lot by field, so in your comment where you casually dismissed anything with IF=4.

Re: The heritability of fertility makes world population stabilization unlikely

#105

Post by aluren » Fri, Jul 05, 2019, 3:19 PM UTC

xamuel wrote: post #103

aluren wrote: post #99

Earlier quoted context omitted.

While I agree there are plenty of issues in academia among which too much emphasis on IF is certainly a thing, it does provide a helpful filter (or prior, if you will) to quickly assess whether some material is worth your time. In this case, stuff in low-IF journals that are virtually unknown in the community (e.g. there are a few journals like Bioinformatics that are also fairly low-IF but very well respected), whos…

I'm not trying to score points in favor of a political position (please follow HN etiquette of giving posters the benefit of the doubt). The filtering you refer to may be useful[1] when looking at bulk lists of papers (e.g. when sitting on a hiring committee), but this is an example where the article has already been curated for our attention, namely: by being upvoted on Hacker News. You can still choose to filter it…

Alright, I wasn't talking about you specifically, sorry if you took it that way.

On HN curation: while I certainly look up to this community when it comes to coding, technology and latest new software tools in general, I'm sorry to say the standards are not even remotely up to par when it comes to genomics. I assume it's because neither the mod team nor the majority of the community has a relevant background, and it's perfectly understandable. It does mean however that I, more often than not, encounter some pretty egregious stuff on here, especially when some dreaded words like "evolution" or "heritability" get mentioned. I sometimes try to chime in (and keep in mind I'm no authority beyond being a rando who happens to work in the field and knows plenty of people more qualified than me) but sometimes the disconnect between HN discussions and actual scientific community discussions is unreal.

On the paper itself: I didn't only mention its IF and voiced other concerns further in the thread. Ultimately there's a limited amount of time one may allocate to reading papers when there are literally millions of the damn things.

Re: The heritability of fertility makes world population stabilization unlikely

#106

Post by maire » Fri, Jul 05, 2019, 3:24 PM UTC

theotherview wrote: post #95

maire wrote: post #93

This is the opposite of what Hans Rosling says in his book Factfulness. According to Hans Rosling's findings family size is linked to child mortality which is linked to poverty. When mothers see their children die then they want to have more children. When they see their children grow to adulthood then they have fewer children. You can also play with the data yourself on his site https://www.gapminder.org/ . He also…

This is a 20+ year recollection...but while working on a Masters in Public Health, we learned that the only causal factor that could be linked to both number of children birthed and age at which a woman gives birth to her first child is the education level of the mother. That is, as women in a society have greater access to education, they postpones when they have their first child and in the end have fewer children…

Yes - Hans also said the same thing although indirectly.

His life work was reducing child mortality (under the age of 5.) He discovered that the biggest factor in reducing child mortality was the education level of the mother. It turns out that most child deaths are preventable if the mother understands how to prevent the death. For instance, dehydration due to diarrhea and vomiting can be prevented by sanitation. Also access to vaccines is linked to the education of the mother.

It goes without saying that access to birth control also plays a big factor. Melinda Gates tells a story about how she was talking to a group of African women about vaccines and what they really wanted was birth control so they could plan their family. They couldn't use condoms because culturally it implied they had AIDs. This also seems linked to education.

Re: The heritability of fertility makes world population stabilization unlikely

#107

Post by lalaland1125 » Fri, Jul 05, 2019, 3:50 PM UTC

marliechiller wrote: post #2

admittedly i havent read the whole article, only the abstract, but: assuming that children from larger families will partially inherit the trait for having larger families themselves seems like a leap to me - arent family sizes in less developed societies generally considered to be either due to religious superstition or because more children = more workers to provide for the family a la victorian times? Once these f…

Keep in mind that religion is heritable to a certain degree in that parents usually educate their children in their own religion (bring them to church with them, etc). Sure, it might not be as strongly heritable as a genetic trait, but you can certainly see it.

Re: The heritability of fertility makes world population stabilization unlikely

#108

Post by option_greek » Fri, Jul 05, 2019, 3:59 PM UTC

maratd wrote: post #62

barry-cotter wrote: post #22

Earlier quoted context omitted.

If a trait is not strongly selected for but variation exists it will be maintained in the population. See height, hair colour, propensity to gain weight, etc. Then you change the environment so that the trait is under strong selection. The people living in Scandinavia 3,000 years ago couldn’t digest lactose. Now ethnic Scandinavians who are lactose intolerant are basically nonexistent. It’s at above 99% homozygotes.…

If it was advantageous for us to have lots of children, then women would have a dozen at a time, rather than one. There are plenty of mammals that do that. There’s clearly a tradeoff there and the selective pressures for humans tend to lean toward fewer. Not sure where we’ll end up, but time will tell.

Minor correction: Advantageous to the children (conferring evolutionary advantage). Not to us.

Re: The heritability of fertility makes world population stabilization unlikely

#109

Post by j_wtf_all_taken » Fri, Jul 05, 2019, 4:03 PM UTC

xamuel wrote: post #104

aluren wrote: post #102

Earlier quoted context omitted.

Low-IF journals are easier to publish to by definition. You first try to publish in a high-IF journal, if they reject you then you try one with lower IF, and so on. I'm all about not taking IF as face value and don't need a n-th reminder about the metric's issues seeing how rehashing them is something of a favorite pastime among scientists, but in this specific case other more qualitative assessments don't line up ei…

>There does seem to be darker political overtones that are exacerbated by commenters Are we reading the same comments section? Which specific comments are you talking about? The comments section I read seemed quite apolitical, except for this particular sub-thread. >Low-IF journals are easier to publish to by definition. It might be true that Low-IF journals in general are easier to publish to, but that's not true "b…

>> There does seem to be darker political overtones that are exacerbated by commenters

> Are we reading the same comments section? Which specific comments are you talking about? The comments section I read seemed quite apolitical, except for this particular sub-thread.

Please.

>> Low-IF journals are easier to publish to by definition.

> (...) By definition, Low-IF journals are journals whose papers are less-cited (this is an oversimplification, of course). For example, IF tends to vary a lot by field, so in your comment where you casually dismissed anything with IFAccepted. So in this specific case, do we talk about a journal in a field where an IF of 4 is low or high?

Re: The heritability of fertility makes world population stabilization unlikely

#110

Post by malvosenior » Fri, Jul 05, 2019, 4:10 PM UTC

maratd wrote: post #62

barry-cotter wrote: post #22

Earlier quoted context omitted.

If a trait is not strongly selected for but variation exists it will be maintained in the population. See height, hair colour, propensity to gain weight, etc. Then you change the environment so that the trait is under strong selection. The people living in Scandinavia 3,000 years ago couldn’t digest lactose. Now ethnic Scandinavians who are lactose intolerant are basically nonexistent. It’s at above 99% homozygotes.…

If it was advantageous for us to have lots of children, then women would have a dozen at a time, rather than one. There are plenty of mammals that do that. There’s clearly a tradeoff there and the selective pressures for humans tend to lean toward fewer. Not sure where we’ll end up, but time will tell.

The size of the human brain makes this physically impossible.