Science for the People
October 26, 2021 9:56 AM Subscribe
From Scientists to Salesmen - A Review of Noah Hutton’s In Silico.
Pervasive corporate managerialism has made it such that the number and prestige of publications in prestigious for-profit journals has become the key metric of one’s value, and the critical currency through which stable, managerial-class jobs are procured. Meanwhile, the deliberate ballooning of for-profit student programs in STEM, coupled to an ever-shrinking job market, keep the labor pool highly skilled, oversaturated, and ever more desperate to compete. Taken together, these conditions have helped precipitate a euphemistically named “replication crisis” that, seen as the natural outcome of competitive social credit-seeking behaviors within a corporate system, is less a crisis of statistical power than it is a crisis of social disempowerment: negative results, however truthful, simply do not sell.
- In Silico explores an audacious 10-year quest to simulate the entire human brain on supercomputers (regularly screened, and screenings can be booked).
- Science for the People (Magazine): Science for the People is dedicated to building and promoting social movements and political struggles around radical perspectives on science and society.
- Do No Evil: An Interview with Noah Hutton.
- Guts is a short film directed by Noah Hutton and Taylor Hess premiered by The Atlantic about Dr. Max Liboiron and CLEAR, a feminist, anti-colonial marine plastics lab in Newfoundland (YT).
If I were to find myself with many billions of dollars, boosting research into real replication of science -- and rewarding diligent and smart science which just happens to prove the null hypothesis -- would be one of my main plans to put that money to real use.
posted by tclark at 10:25 AM on October 26, 2021 [4 favorites]
posted by tclark at 10:25 AM on October 26, 2021 [4 favorites]
I share your confusion late afternoon but I think if I'm reading the review correctly (since I haven't actually seen In Silico) that it's arguing that a replication crisis is an inevitable consequence of the "corporatization" of science that rewards self-promotion, courting capital, and rapid publication above all else. I think it's saying that it's impossible to fix a replication crisis without fixing the incentive structure in scientific research.
posted by Wretch729 at 10:31 AM on October 26, 2021 [13 favorites]
posted by Wretch729 at 10:31 AM on October 26, 2021 [13 favorites]
While a lot of the critiques in this article are spot on, I'd just like to blow a BIG FUCKING RASPBERRY to this sentence, which is the worst kind of illogical guilt by association
Money is certainly being thrown at basic neuroscience research by both Democrats and Republicans alike, and, if the field’s feverish rate of publication is any indication,9 the wheels of academic research certainly are spinning. Meanwhile, we still know next to nothing about how the brain works.
We know a fuckton more about how the brain works, thankyouverymuch.
What we don't know is much more about how consciousness and other higher order phenomena emerge from this, and that's i) it's a fucking hard philosophical problem and ii) because we're studying probably the hardest scientific problem ever : the most complexly interconnected single piece of matter (yet discovered) in the fucking universe, with the same piece of matter as the studying tool.
The barrier to this problem being solved will not be dismantled by somehow 'democratizing' science, removing neoliberal culture from science funding conditions or improving the working conditions of grad students.
Those goals are moral and laudable (and imho necessary), but they are NO way meaningfully limiting our ability to understand the brain. What is absolutely limiting us is the fucking difficulty of the problem.
posted by lalochezia at 10:32 AM on October 26, 2021 [16 favorites]
Money is certainly being thrown at basic neuroscience research by both Democrats and Republicans alike, and, if the field’s feverish rate of publication is any indication,9 the wheels of academic research certainly are spinning. Meanwhile, we still know next to nothing about how the brain works.
We know a fuckton more about how the brain works, thankyouverymuch.
What we don't know is much more about how consciousness and other higher order phenomena emerge from this, and that's i) it's a fucking hard philosophical problem and ii) because we're studying probably the hardest scientific problem ever : the most complexly interconnected single piece of matter (yet discovered) in the fucking universe, with the same piece of matter as the studying tool.
The barrier to this problem being solved will not be dismantled by somehow 'democratizing' science, removing neoliberal culture from science funding conditions or improving the working conditions of grad students.
Those goals are moral and laudable (and imho necessary), but they are NO way meaningfully limiting our ability to understand the brain. What is absolutely limiting us is the fucking difficulty of the problem.
posted by lalochezia at 10:32 AM on October 26, 2021 [16 favorites]
I think it's saying that it's impossible to fix a replication crisis without fixing the incentive structure in scientific research.
The other part of that, that's less directly and intimately connected to corporate managerialism, is that we need reviewers and publications to be more willing to publish null results. "It was reasonable to try this out, but it appears not to be the case" or "It was reasonable to try this out, but the evidence is inconclusive in either direction."
The problem-within-a-problem is that in a system where publication numbers matter and null results are publishable, lots of authors are going to omit the "It was reasonable to try this out" part and generate an endless stream of dumb shit nobody would have ever reasonably thought would be the case that turns out not to be the case. Eating cheddar cheese doesn't affect your presidential vote. People who spin around three times before answering report no different ideology than people who say "wibble" before answering.
posted by GCU Sweet and Full of Grace at 10:52 AM on October 26, 2021 [3 favorites]
The other part of that, that's less directly and intimately connected to corporate managerialism, is that we need reviewers and publications to be more willing to publish null results. "It was reasonable to try this out, but it appears not to be the case" or "It was reasonable to try this out, but the evidence is inconclusive in either direction."
The problem-within-a-problem is that in a system where publication numbers matter and null results are publishable, lots of authors are going to omit the "It was reasonable to try this out" part and generate an endless stream of dumb shit nobody would have ever reasonably thought would be the case that turns out not to be the case. Eating cheddar cheese doesn't affect your presidential vote. People who spin around three times before answering report no different ideology than people who say "wibble" before answering.
posted by GCU Sweet and Full of Grace at 10:52 AM on October 26, 2021 [3 favorites]
I don't think the statements "we still know next to nothing about how the brain works" and "we know a fuckton more about how the brain works" are necessarily incompatible. As in, the alchemists both knew a fuckton and next to nothing about chemistry.
Also, I think many would dispute that "the barrier to this problem being solved will not be dismantled by somehow 'democratizing' science, removing neoliberal culture from science funding conditions or improving the working conditions of grad students."
posted by Alex404 at 11:01 AM on October 26, 2021 [1 favorite]
Also, I think many would dispute that "the barrier to this problem being solved will not be dismantled by somehow 'democratizing' science, removing neoliberal culture from science funding conditions or improving the working conditions of grad students."
posted by Alex404 at 11:01 AM on October 26, 2021 [1 favorite]
negative results, however truthful, simply do not sell
I work in a field where reproducibility is taken so seriously that the same experiments (explicitly called "replicates", in fact) and downstream analyses are done in multiple labs around the world, to make sure we're all getting roughly the same answers, "good" or "bad".
Doing "Big Science" of this sort does require money and therefore salesmanship on the part of the PIs involved, and getting published in high-profile journals is unfortunately part of the process of asking for and getting money from government agencies, because for lack of alternatives, publication history helps show the ability to manage larger projects with uncertain outcomes.
For those who are uncomfortable with this reality, please write your legislators to prioritize basic science, such that PIs are not placed so much in competition with one another for funding.
Larger scientific projects also can require managerial oversight of the kind that may appear to be "corporate" from the outside.
However, where the money comes from and how labs are managed are separate aspects from processes that try to ensure the honesty and quality of the science being done. All of the data, procedures, software — all of it gets published openly, freely. Intermediate steps are required over years to demonstrate that the grants are being spent appropriately.
Other scientists whose work overlaps ours may question the results, for various reasons ranging from the complainants being cranks to having legitimate arguments — but it would be difficult to see how we coordinate falsifying or covering up bad data or results, just to keep the money coming in, mainly because there are steps taken outside our control to ensure it would be difficult to impossible to get away with such behaviour over the longer multi-year terms covered by our funding.
posted by They sucked his brains out! at 12:03 PM on October 26, 2021 [5 favorites]
I work in a field where reproducibility is taken so seriously that the same experiments (explicitly called "replicates", in fact) and downstream analyses are done in multiple labs around the world, to make sure we're all getting roughly the same answers, "good" or "bad".
Doing "Big Science" of this sort does require money and therefore salesmanship on the part of the PIs involved, and getting published in high-profile journals is unfortunately part of the process of asking for and getting money from government agencies, because for lack of alternatives, publication history helps show the ability to manage larger projects with uncertain outcomes.
For those who are uncomfortable with this reality, please write your legislators to prioritize basic science, such that PIs are not placed so much in competition with one another for funding.
Larger scientific projects also can require managerial oversight of the kind that may appear to be "corporate" from the outside.
However, where the money comes from and how labs are managed are separate aspects from processes that try to ensure the honesty and quality of the science being done. All of the data, procedures, software — all of it gets published openly, freely. Intermediate steps are required over years to demonstrate that the grants are being spent appropriately.
Other scientists whose work overlaps ours may question the results, for various reasons ranging from the complainants being cranks to having legitimate arguments — but it would be difficult to see how we coordinate falsifying or covering up bad data or results, just to keep the money coming in, mainly because there are steps taken outside our control to ensure it would be difficult to impossible to get away with such behaviour over the longer multi-year terms covered by our funding.
posted by They sucked his brains out! at 12:03 PM on October 26, 2021 [5 favorites]
GCU Sweet and Full of Grace: The other part of that, that's less directly and intimately connected to corporate managerialism, is that we need reviewers and publications to be more willing to publish null results.
The Berkeley Initiative for Transparency in the Social Sciences has a bunch of lectures and seminars on this topic. I found this one especially interesting, where they chased down the fates of a bunch of preregistered experiments. It seemed like everything from the incentive structure of the system as a whole to the psychology of individual scientists conspires to limit the publication of null results. Scientists who don't find anything interesting in their data seem to find it psychologically difficult to write up the results, let alone try to find a journal that'll publish them. It'd be interesting to know how much the psychological incentives would change if the systemic incentives did.
This video from the same group mentioning a journal which tried a null results special edition and said they'd probably never do that again was also interesting for the reasons given by the editors.
posted by clawsoon at 12:15 PM on October 26, 2021 [5 favorites]
The Berkeley Initiative for Transparency in the Social Sciences has a bunch of lectures and seminars on this topic. I found this one especially interesting, where they chased down the fates of a bunch of preregistered experiments. It seemed like everything from the incentive structure of the system as a whole to the psychology of individual scientists conspires to limit the publication of null results. Scientists who don't find anything interesting in their data seem to find it psychologically difficult to write up the results, let alone try to find a journal that'll publish them. It'd be interesting to know how much the psychological incentives would change if the systemic incentives did.
This video from the same group mentioning a journal which tried a null results special edition and said they'd probably never do that again was also interesting for the reasons given by the editors.
posted by clawsoon at 12:15 PM on October 26, 2021 [5 favorites]
I feel like “replication crisis” is not scare quotes but a euphemism for “complete bullshit getting published”
posted by Jon_Evil at 12:28 PM on October 26, 2021 [6 favorites]
posted by Jon_Evil at 12:28 PM on October 26, 2021 [6 favorites]
We know a fuckton more about how the brain works, thankyouverymuch.
This kind of reminds me of Mukherjee's discussion of cancer research in The Emperor of All Maladies. To get money for basic research in the 1950s and 60s, they had to make ridiculous promises. A moonshot for cancer. Ten years, billions of dollars, cure for cancer. Nobody who said, "50 years and hundreds of billions of dollars, actually," got funding. A whole bunch of absolutely necessary basic research got done, but none of it immediately translated to cures for cancers.
...but of course in order to get more funding it couldn't be sold as "we got a lot of basic research done but our cancer treatments are still shit," so results would be oversold to begin the next cycle of basic research.
posted by clawsoon at 1:07 PM on October 26, 2021 [4 favorites]
This kind of reminds me of Mukherjee's discussion of cancer research in The Emperor of All Maladies. To get money for basic research in the 1950s and 60s, they had to make ridiculous promises. A moonshot for cancer. Ten years, billions of dollars, cure for cancer. Nobody who said, "50 years and hundreds of billions of dollars, actually," got funding. A whole bunch of absolutely necessary basic research got done, but none of it immediately translated to cures for cancers.
...but of course in order to get more funding it couldn't be sold as "we got a lot of basic research done but our cancer treatments are still shit," so results would be oversold to begin the next cycle of basic research.
posted by clawsoon at 1:07 PM on October 26, 2021 [4 favorites]
I work in a field where reproducibility is taken so seriously that the same experiments (explicitly called "replicates", in fact) and downstream analyses are done in multiple labs around the world, to make sure we're all getting roughly the same answers, "good" or "bad".
Meanwhile, in EEB where my training is based, there is straight up no resources for that shit. You can't even get more than one replicate reliably in the same species, because competition for funding is intense and there are so many potential places to specialize. When there are two labs working on your species of interest and they're collaborators, as was the case for my PhD, what do you do?
I don't know. I've spent the last ten years working in science with an acute awareness of resource scarcity and the uncertain capriciousness in funding about whether basic research is deemed worth supporting. I'm stretching into NIH fields explicitly in part because I'm hoping to shed some of that scarcity mindset, and I must say that over in psychology I actually see more replication than I'm used to because people tend to be using the same three to five systems over and over again, and it's weird limiting yourself to those sensory umwelts but it is also undeniably a lot easier to get things done.
In some ways, my understanding of the world that scientists must inhabit was shaped by the 2013 shutdown that happened in my second year of PhD and the subsequent publication of this 2014 Animal Behaviour paper sharing strategies for scientists to defend their work against partisan political attack. I remember my college roommate, then working at the CDC, nervously chattering on Facebook about not being sure she was going to be able to make rent during the shutdown. (She went on to create a workable vaccine for Nipah virus during her PhD, incidentally; her work is certainly deemed more immediately valuable than mine ever was. It has not saved her from being affected directly by things like the shutdown.)
When my grandparents abandoned me on the side of the road in January 2017, the lynchpin during which the conversation became clearly not something I could diplomatically conflict-manage my way out of was my grandfather angrily telling me that [my? all?] scientific research was literally worthless unless it had military applications that directly allowed the US to kill enemy combatants.
It is really hard for me to envision a funding climate that is not a conflicting morass of contradictory demands that directly rewards salesmanship and novelty much more clearly than scientific insight. I have friends and collaborators in common with Jonathan Pruitt--or, I suppose, I had, fuck knows the man tried to collaborate with anyone who held still long enough--and I see how the incentives are fucked up. I see a lot of people trying to do good work in spite of the existing incentives, not because of them. And now that I am at the postdoc stage of my career, I am beginning to watch people I know who really care deeply about finding out the truth, whatever it looks like, leaving the field--because this job will chew you up and grind you down, and it's not worth the exploitation.
I don't know what a better system would look like. Some of this is just that it is really hard to identify what will be useful or interesting ahead of time, and for better or for worse this is a nation that seems to value all knowledge in terms of capitalistic gains and spending as little money as possible. I used to hope that the values of the American public were not in line with the way we actually spend our money, but I'm not so certain that's true any more. And it is very hard to imagine things improving in any meaningful way any time soon.
posted by sciatrix at 1:26 PM on October 26, 2021 [12 favorites]
Meanwhile, in EEB where my training is based, there is straight up no resources for that shit. You can't even get more than one replicate reliably in the same species, because competition for funding is intense and there are so many potential places to specialize. When there are two labs working on your species of interest and they're collaborators, as was the case for my PhD, what do you do?
I don't know. I've spent the last ten years working in science with an acute awareness of resource scarcity and the uncertain capriciousness in funding about whether basic research is deemed worth supporting. I'm stretching into NIH fields explicitly in part because I'm hoping to shed some of that scarcity mindset, and I must say that over in psychology I actually see more replication than I'm used to because people tend to be using the same three to five systems over and over again, and it's weird limiting yourself to those sensory umwelts but it is also undeniably a lot easier to get things done.
In some ways, my understanding of the world that scientists must inhabit was shaped by the 2013 shutdown that happened in my second year of PhD and the subsequent publication of this 2014 Animal Behaviour paper sharing strategies for scientists to defend their work against partisan political attack. I remember my college roommate, then working at the CDC, nervously chattering on Facebook about not being sure she was going to be able to make rent during the shutdown. (She went on to create a workable vaccine for Nipah virus during her PhD, incidentally; her work is certainly deemed more immediately valuable than mine ever was. It has not saved her from being affected directly by things like the shutdown.)
When my grandparents abandoned me on the side of the road in January 2017, the lynchpin during which the conversation became clearly not something I could diplomatically conflict-manage my way out of was my grandfather angrily telling me that [my? all?] scientific research was literally worthless unless it had military applications that directly allowed the US to kill enemy combatants.
It is really hard for me to envision a funding climate that is not a conflicting morass of contradictory demands that directly rewards salesmanship and novelty much more clearly than scientific insight. I have friends and collaborators in common with Jonathan Pruitt--or, I suppose, I had, fuck knows the man tried to collaborate with anyone who held still long enough--and I see how the incentives are fucked up. I see a lot of people trying to do good work in spite of the existing incentives, not because of them. And now that I am at the postdoc stage of my career, I am beginning to watch people I know who really care deeply about finding out the truth, whatever it looks like, leaving the field--because this job will chew you up and grind you down, and it's not worth the exploitation.
I don't know what a better system would look like. Some of this is just that it is really hard to identify what will be useful or interesting ahead of time, and for better or for worse this is a nation that seems to value all knowledge in terms of capitalistic gains and spending as little money as possible. I used to hope that the values of the American public were not in line with the way we actually spend our money, but I'm not so certain that's true any more. And it is very hard to imagine things improving in any meaningful way any time soon.
posted by sciatrix at 1:26 PM on October 26, 2021 [12 favorites]
Once you've said "corporate research" you've essentially said "capitalist research" and it's clear where the problem lies. A quest for profit rather than knowledge and short-term payback over understanding and depth.
posted by CheeseDigestsAll at 2:06 PM on October 26, 2021
posted by CheeseDigestsAll at 2:06 PM on October 26, 2021
I read TFA, and then I read the David Graeber piece from 2012 on which it was partly based.
The Graeber bit is like a demonstration of why people who don't really understand technology ought not to talk about it at length. We don't have flying cars, computers that can conduct an intelligent conversation, domed cities on Mars, or jet packs because those turned out to be lots harder to make than anybody in the '50s or '60s anticipated. We don't have teleportation or phasers because those were always fantasy.
TFA's main point (that scientific research is not conducted in a way that rewards risk-taking or originality or really anything but salesmanship) feels broadly true.
posted by Aardvark Cheeselog at 2:21 PM on October 26, 2021 [3 favorites]
The Graeber bit is like a demonstration of why people who don't really understand technology ought not to talk about it at length. We don't have flying cars, computers that can conduct an intelligent conversation, domed cities on Mars, or jet packs because those turned out to be lots harder to make than anybody in the '50s or '60s anticipated. We don't have teleportation or phasers because those were always fantasy.
TFA's main point (that scientific research is not conducted in a way that rewards risk-taking or originality or really anything but salesmanship) feels broadly true.
posted by Aardvark Cheeselog at 2:21 PM on October 26, 2021 [3 favorites]
If the goal is greater progress across all fields of science, I think the most productive and salutary change would be disempowering PIs and freeing young scientists from the decade or more of peonage they currently spend as Ph.D students and post docs, most often in thrall to received ideas.
Give young investigators the power to pursue the things they are actually interested in — and stand back.
posted by jamjam at 2:58 PM on October 26, 2021
Give young investigators the power to pursue the things they are actually interested in — and stand back.
posted by jamjam at 2:58 PM on October 26, 2021
Flying cars aren't hard to make, they're just not economical. That's the big problem with a lot of futurism, they only project on one axis, technology. Once you factor in social issues and economics, it's easy to understand why we don't have nuclear-powered cars and jet packs. The energy required to make a car fly is exponentially higher than making it roll. Moving a person around in a jet pack is pointless if you can't get them there safely.
posted by CheeseDigestsAll at 3:09 PM on October 26, 2021 [2 favorites]
posted by CheeseDigestsAll at 3:09 PM on October 26, 2021 [2 favorites]
While a lot of the critiques in this article are spot on, I'd just like to blow a BIG FUCKING RASPBERRY to this sentence . . .
I read TFA, and then I read the David Graeber piece from 2012 on which it was partly based. The Graeber bit is like a demonstration of why people who don't really understand technology ought not to talk about it at length . . .
Yeah, I am strongly ambivalent about lots of this piece. There's a lot that I think is important to note about the state of science but also a lot I got the side eye.
One repeated issue is that there is a lot that is narrowly accurate that is then generalized. A good editor, or more conscientious approach to sourcing rhetorical claims, would be helpful. For example: "Meanwhile, the deliberate ballooning of for-profit student programs in STEM, coupled to an ever-shrinking job market . . . " is a quote from the piece. But the link is a discussion specifically focused on neuroscience graduate level positions--which might seem really important to a neuroscience grad student like the author, but is an incredibly niche area that reflects on the overall employment trend not at all. (It continues to increase steadily, as a visit to the BLS can tell you.)
Once you've said "corporate research" you've essentially said "capitalist research" and it's clear where the problem lies.
These two things are NOT the same, and I hope that's not how she's using "corporate". Direct (private, for-profit) corporate funding for university research is a mere 6%, ten times less than the share of government funding. And while some academics do found businesses and get rich, academic researchers in general are generally amazingly clueless about commercial processes. This seldom comes through in the lay press--the sources for science journalists tend to be academics, after all--but there's a huge disconnect that I could describe in more detail if you want.
The most charitable reading is her that she's using "corporate" in the sense of large, top-down funding--the word "bureaucracy" comes up four times, and "social credit" as a reward twice (as opposed to actual cash). She talks about a cultural infection too, and I totally agree that "McKinseyism" is a cancer within scientific organizations.
But to the extent that people are coming away with the idea the problems of publicly funded science research are capitalist problems they are coming away with the wrong idea. In many ways corporate research, for all its many problems, is better at caring about replication--we need a product that works, not a paper that gets cited. It's no surprise the studies that really highlighted replication crisis in the biochemistry were from Bayer and other for-profits, where researchers got frustrated chasing down crappy results published in high tier journals.
Even if you imagine a pure socialist government funding all this, you still have vanities, egos, and fiefdoms in massive bureaucracies too. I have no clue how you get past that when allocating billions.
posted by mark k at 6:00 PM on October 26, 2021 [6 favorites]
I read TFA, and then I read the David Graeber piece from 2012 on which it was partly based. The Graeber bit is like a demonstration of why people who don't really understand technology ought not to talk about it at length . . .
Yeah, I am strongly ambivalent about lots of this piece. There's a lot that I think is important to note about the state of science but also a lot I got the side eye.
One repeated issue is that there is a lot that is narrowly accurate that is then generalized. A good editor, or more conscientious approach to sourcing rhetorical claims, would be helpful. For example: "Meanwhile, the deliberate ballooning of for-profit student programs in STEM, coupled to an ever-shrinking job market . . . " is a quote from the piece. But the link is a discussion specifically focused on neuroscience graduate level positions--which might seem really important to a neuroscience grad student like the author, but is an incredibly niche area that reflects on the overall employment trend not at all. (It continues to increase steadily, as a visit to the BLS can tell you.)
Once you've said "corporate research" you've essentially said "capitalist research" and it's clear where the problem lies.
These two things are NOT the same, and I hope that's not how she's using "corporate". Direct (private, for-profit) corporate funding for university research is a mere 6%, ten times less than the share of government funding. And while some academics do found businesses and get rich, academic researchers in general are generally amazingly clueless about commercial processes. This seldom comes through in the lay press--the sources for science journalists tend to be academics, after all--but there's a huge disconnect that I could describe in more detail if you want.
The most charitable reading is her that she's using "corporate" in the sense of large, top-down funding--the word "bureaucracy" comes up four times, and "social credit" as a reward twice (as opposed to actual cash). She talks about a cultural infection too, and I totally agree that "McKinseyism" is a cancer within scientific organizations.
But to the extent that people are coming away with the idea the problems of publicly funded science research are capitalist problems they are coming away with the wrong idea. In many ways corporate research, for all its many problems, is better at caring about replication--we need a product that works, not a paper that gets cited. It's no surprise the studies that really highlighted replication crisis in the biochemistry were from Bayer and other for-profits, where researchers got frustrated chasing down crappy results published in high tier journals.
Even if you imagine a pure socialist government funding all this, you still have vanities, egos, and fiefdoms in massive bureaucracies too. I have no clue how you get past that when allocating billions.
posted by mark k at 6:00 PM on October 26, 2021 [6 favorites]
sciatrix: people tend to be using the same three to five systems over and over again, and it's weird limiting yourself to those sensory umwelts but it is also undeniably a lot easier to get things done
Following the money means following the literature because The Money is risk averse. So you get winner-takes-all for some systems sucking the oxygen from other might-have-been orphan projects. Humans have 10 Toll-like Receptors TLRs for detecting immunological threats. Total number of Pubmed papers [27Oct21] for three of these:
TLR3 5,164
TLR4 25,461
TLR5 1,886
It's hard to believe that TLR4 is 10x or 5x more important to our health and happiness than TLR5 or TLR3. It's a lot easier to fund your talented TLR4 post-doc.
posted by BobTheScientist at 2:05 AM on October 27, 2021 [2 favorites]
Following the money means following the literature because The Money is risk averse. So you get winner-takes-all for some systems sucking the oxygen from other might-have-been orphan projects. Humans have 10 Toll-like Receptors TLRs for detecting immunological threats. Total number of Pubmed papers [27Oct21] for three of these:
TLR3 5,164
TLR4 25,461
TLR5 1,886
It's hard to believe that TLR4 is 10x or 5x more important to our health and happiness than TLR5 or TLR3. It's a lot easier to fund your talented TLR4 post-doc.
posted by BobTheScientist at 2:05 AM on October 27, 2021 [2 favorites]
Meanwhile, the deliberate ballooning of for-profit student programs in STEM, coupled to an ever-shrinking job market . . .
This sentence also bothered me. If you actually talk to the people writing the grants, grad students are very expensive compared to postdocs or techs, because you have to pay both their stipend and their tuition, and they're usually not nearly as good at the work as postdocs or techs because they are less skilled. They also require more supervision, either from the PI or from a postdoc or tech. Nobody is profiting off of grad students.
Unless she's speaking of the "professional" master's programs that are unfunded, which most students pay for with loans? The ones that are usually just taking a bunch of courses and generally don't even involve research? But those programs really don't have anything to do with the matter at hand because, again, they generally don't even involve research.
posted by hydropsyche at 7:26 AM on October 27, 2021
This sentence also bothered me. If you actually talk to the people writing the grants, grad students are very expensive compared to postdocs or techs, because you have to pay both their stipend and their tuition, and they're usually not nearly as good at the work as postdocs or techs because they are less skilled. They also require more supervision, either from the PI or from a postdoc or tech. Nobody is profiting off of grad students.
Unless she's speaking of the "professional" master's programs that are unfunded, which most students pay for with loans? The ones that are usually just taking a bunch of courses and generally don't even involve research? But those programs really don't have anything to do with the matter at hand because, again, they generally don't even involve research.
posted by hydropsyche at 7:26 AM on October 27, 2021
(I should say I am speaking entirely from the NSF side of things, and maybe NIH researchers are somehow profiting off of grad students. But I'm not sure how that would work.)
posted by hydropsyche at 7:27 AM on October 27, 2021
posted by hydropsyche at 7:27 AM on October 27, 2021
If you actually talk to the people writing the grants, grad students are very expensive compared to postdocs or techs, because you have to pay both their stipend and their tuition
Depends on the program. I was actually very cheap for my (NSF-funded!) PI in terms of grants because I TAed every single long semester of my eight-year PhD--so every spring and every fall--which both paid my stipend and provided my tuition waiver. Admittedly, he did have to support me in the summers, which was not required when I joined the program--he was quite proud of insisting that he did that for all of his students out of his grants at the time--but is now department policy. He did not support my final summer, either; I won a departmental fellowship for that because it was that or go without income while I finished up.
As with everything, this is at least partially a matter of ethics on the part of faculty combined with departmental regulations about how grad students must be treated. As I recall, the new rules stating that faculty in my PhD department must be able to provide summer support for all students under 5 years of their PhD and also a year of RA support for all first years/much sharper limits on how much students should TA during their degrees were received with horror from my EEB department, since these rules were imposed by a new dean from a NIH background where funding was of course much easier to acquire.
There is a lot of variance in how much teaching graduate students and postdocs are expected to do over the course of their PhDs, and of course that variance accounts for quite a lot of variation in how expensive graduate students are to maintain vs postdocs, who are typically not expected to teach during their time in a lab.
posted by sciatrix at 9:01 AM on October 27, 2021
Depends on the program. I was actually very cheap for my (NSF-funded!) PI in terms of grants because I TAed every single long semester of my eight-year PhD--so every spring and every fall--which both paid my stipend and provided my tuition waiver. Admittedly, he did have to support me in the summers, which was not required when I joined the program--he was quite proud of insisting that he did that for all of his students out of his grants at the time--but is now department policy. He did not support my final summer, either; I won a departmental fellowship for that because it was that or go without income while I finished up.
As with everything, this is at least partially a matter of ethics on the part of faculty combined with departmental regulations about how grad students must be treated. As I recall, the new rules stating that faculty in my PhD department must be able to provide summer support for all students under 5 years of their PhD and also a year of RA support for all first years/much sharper limits on how much students should TA during their degrees were received with horror from my EEB department, since these rules were imposed by a new dean from a NIH background where funding was of course much easier to acquire.
There is a lot of variance in how much teaching graduate students and postdocs are expected to do over the course of their PhDs, and of course that variance accounts for quite a lot of variation in how expensive graduate students are to maintain vs postdocs, who are typically not expected to teach during their time in a lab.
posted by sciatrix at 9:01 AM on October 27, 2021
I will agree, however, that I would not call my program a "for-profit" program in any sense of the term, although I did a lot of research for "free." (One of my most vivid memories from my PhD is a senior faculty member unsolicited emailing the entire grad student listserv to inform us that we were, dollar for dollar, paid much more for our teaching than tenure-track faculty or lecturers because we were paid a full-time salary for only 20 hours per week of work, and everything else we did was of course for "our own education," complete with a link to his personal blog post on the subject.)
It's just that who is profiting directly on my research? Probably the journal publishing companies, if anyone. It isn't clear to me that my labor directly enriched anyone except maybe the career of my PI. Certainly my "tuition waiver" was a polite fiction for everyone, since the classes I took were fairly minimal. I did have to pay taxes on that bastard as if it was income for the first, oh, four years or so I was with the department, of course, but it's not as if that "tuition" money went anywhere I ever saw again either. Eventually the department stepped up to pay those taxes for us, at least. It is something of a hardship when you get slammed with a $800 bill (for holding the taxes, natch) about one or two months into the swing of your first adult job immediately following an interstate move.
posted by sciatrix at 9:09 AM on October 27, 2021 [1 favorite]
It's just that who is profiting directly on my research? Probably the journal publishing companies, if anyone. It isn't clear to me that my labor directly enriched anyone except maybe the career of my PI. Certainly my "tuition waiver" was a polite fiction for everyone, since the classes I took were fairly minimal. I did have to pay taxes on that bastard as if it was income for the first, oh, four years or so I was with the department, of course, but it's not as if that "tuition" money went anywhere I ever saw again either. Eventually the department stepped up to pay those taxes for us, at least. It is something of a hardship when you get slammed with a $800 bill (for holding the taxes, natch) about one or two months into the swing of your first adult job immediately following an interstate move.
posted by sciatrix at 9:09 AM on October 27, 2021 [1 favorite]
Unless she's speaking of the "professional" master's programs that are unfunded, which most students pay for with loans? . . . they generally don't even involve research.
I honestly just assumed she was talking about about undergrad and masters degrees at for-profit institutions, which are on the rise (AFAIK). I mean, the quote refers to STEM students and not academic research. I think the paper is in fact highly focused on research in academia, so I get your intuition that it must be PhDs. But I think the author isn't careful about the distinctions (or the implications of it being different worlds.)
posted by mark k at 10:47 AM on October 27, 2021
I honestly just assumed she was talking about about undergrad and masters degrees at for-profit institutions, which are on the rise (AFAIK). I mean, the quote refers to STEM students and not academic research. I think the paper is in fact highly focused on research in academia, so I get your intuition that it must be PhDs. But I think the author isn't careful about the distinctions (or the implications of it being different worlds.)
posted by mark k at 10:47 AM on October 27, 2021
« Older Welcome to the family, beautiful | I GIVE YOU... Sexy Saturn Devouring His Son Newer »
This thread has been archived and is closed to new comments
I'm in toxicology/epidemiology/public health/cancer biology, and the "replication crisis" is never used in scare quotes in these fields. I've honestly never seen it presented this way. In my experience, the replication crisis is acknowledging--for instance--that decades of uncritical reliance on non-human animal models of human disease have led to a body of research that is very, very much made up of one-off studies that have little-to-no relevance because, even in instances when they're repeated, the use of poorly characterized methods and materials in studies makes them very difficult to replicate and, if they're detailed enough for someone to replicate in practice, the results come out widely different from study to study.
In the main article linked above, the "replication crisis" is referenced to a single publication, which itself is drawn from an analysis restricted to "psychology, economics, and general interest journals." I feel like there's a fundamental disconnect here, and I'm not sure what to make of it. What's the insight here? Is there a real voice asserting that replicability is not a critical part of demonstrating validity in science, or at least drawing actionable conclusions from scientific inquiry?
posted by late afternoon dreaming hotel at 10:22 AM on October 26, 2021 [8 favorites]