I've seen a good number of boondoggles. At the center of each was a personality of such magnetism that people simply fell in line behind them. There are people who have the ability to spout utter nonsense, but to do so with such alpha authority that nearly everyone in the room just glazes over and nods. Whatever it is that our brains take as a signal of alpha status, these people know how to crank it to 11.
When are people going to learn that primate dominance gestures are not a reliable proxy for much else?
I will say that when actual merit and this level of dominance actually align in the same individual, amazing things can happen. But I see no sign that this is more common than would be predicted by the random assortment of traits.
That sort of thing really doesn't affect nearly everybody. But those people which it doesn't affect, leave the project, or are pushed out. So it's a self-selecting/reinforcing thing, which is what makes it problematic, I think.
The marketing and salesmanship of the Human Brain Project may have been a little too successful for its own good. On one hand, the idea that simply building a giant brain simulation would be possible or scientifically useful is not currently realistic.
However, I don't think they (Markram et al.) are that naive. In reality, the project is composed of a lot of different subprojects, in simulation, neuromorphic computing, mapping/characterizing mouse and human brains and doing actual neuroscience, theory, applications to human disease, even philosophy and ethics, etc.
see: https://www.humanbrainproject.eu/discover/the-project/sub-pr... and their list of publications so far
https://www.humanbrainproject.eu/science/publications
It's a huge umbrella for neuroscience funding that they marketed to the EU under the grand vision of simulating the brain. Whether or not this kind of marketing/funding mechanism is an optimal way to do the best science is a very legitimate issue to debate. But the implication (which the HBP brought upon themselves..) that a bunch of crackpots have been given a billion dollars to buy supercomputers and run nonsensical simulations is an oversimplification.
I was curious why there was only neuroscientists and biologists quoted so I went looking for another source. I think this link gives a different, broader overview of the situation and doesn't only focus on personal attacks.
Specifically neuroscience subproject funding was removed... This seems like angry academics that lost funding are trying to derail it through the media to my uneducated observance...
I think the overall problem is that traditional neuro scientists and biologists bring virtually none of the required skills to the table.
Moreover while neuroscience is able to show pretty pictures of brain activity, they have made insufficient progress in understanding it. On the other hand the physicists/mathematicians that have invaded the field have a background in computational methods, simulation and hardware development. So for example a group at the University I studied had previous experience designing analog chips for feature detection in High Energy Physics, a number of years ago they re-branded themselves as a neurophysics group and recently landed a ~100 million euro+ grant in the context of the Human Brain Project.
It is also much easier to learn the jargon and read some of the softer phenomenological articles than to develop a sound understanding of the underlying mathematics and physics.
It is also much easier to conjure up some entertaining but biologically meaningless mathematical model than to spend years doing real biological experiments.
I think the overall problem is that the physicists and the biologists don't have respect for each other's skills and knowledge.
Obviously a biologist will have lots of knowledge and skills a physicist doesn't have. Mathematical models can be very useful even if they don't model the real world faithfully, as long as they capture part of the behavior to some precision. Physicists spend an extraordinary amount of time studying toy models to build up intuition, which together with an understanding how to proceed in more complicated solutions can be very useful. For example even though most real world EM problems won't have the symmetries required to come up with a closed form solution, the intuition gained from studying toy examples is still invaluable.
In the case of neurons you can model the membrane of the neuron and the interior and exterior to various degrees of accuracy as an EM problem, if you wish you can even add surface defects. Since the temperature is high, all quantum effects are essentially washed out, as are the need to model individual calcium ions etc., at least to first approximation. After the dust settles you can come up with a few effective parameters that model one neuron and an ODE that models its dynamics. As it turns out the ODE indeed exhibits dynamics similar to real world neurons in response to electric stimuli.
If you then couple those ODEs into larger system and try different coupling configurations (the coupling coefficients model the synapses between neurons), you have come up with a simplified model for parts of a brain.
The entire point of the "Future and Emerging Technologies" FET Flagship program (where the HBP got it's money) was to fund risky, daring and ambitious technology projects.
"The critics argue that the human genome project has been sold on hype and glitter, rather than its scientific merits, and that it will drain talent, money and life from smaller, worthier biomedical efforts."
"They also doubt that the project can be completed in anything close to its original deadline and budget."
"it will have generated enormous reams of uninterpretable and often useless data"
"it's hyped science"
"Everybody I talk to thinks this is an incredibly bad idea"
"Some critics have begun aggressive letter-writing campaigns"
The exact same arguments that were used 25 years ago to discredit the HGP are now resurfacing to criticize the HBP. And with genome sequencing now below 1000$, that article has become almost laughable.
Many aspects of HGP were overhyped and much of the data remains uninterpretable. There was a strong implication that knowing the genome of humans would immediately lead to trivial improvements in health, but there wasn't actually a clear path to such improvements.
That said, it's an immensely valuable resource that pays back in many ways (it's a map which can be used for discovery science).
It's not entirely clear it drained talent and money from other projects in a way that was a net negative for science and society.
Just because the HGP turned out to be worthwhile doesn't mean funding it in the right place was the correct decision. If the expected value of the HGP was below the opportunity cost it should not have been funded. That the actual value turned out to be higher than the opportunity cost is irrelevant.
To say it another way, I think that playing the lotto with the goal of earning money is a mistake. The fact that some people win it and in fact make money doesn't change the fact that they were wrong to play it in the first place if their goal was net gain of money.
That said I have no idea what the expected value of the HGP was so I have no idea whether it was a good decision. I just want to chime in in support of thinking about the probabilities involved in the right way.
Expected value is a bit of a strange concept to apply to scientific research, as no-one really knows for certain what the full, long-term impact will be. The actual value of many scientific discoveries only becomes apparent years later on.
No one knows for certain, but... suppose you have two projects and can fund one of them.
Project A wants to investigate in as much depth as possible (given their budget) the effects of nail-biting on arthritis.
Project B wants to investigate in as much depth as possible (given their budget) the effects of learning a programming language on Alzheimer's disease.
I know which of these seems like a better choice for funding, even though I don't really know for certain what the long-term impact of either will be.
Which do you think is a better choice for funding, and why?
In support of project A: fibromyalgia, which has arthritic-type symptoms, is partly linked to anxiety, of which one of the common manifestation is nail-biting. Could evidence of nail-biting be a useful tool as part of differential diagnosis, or a predictor?
In support of project B: the idea of a 'cognitive reserve' in preventing the onset of Alzheimer's disease is fairly well established, even though it has no disease-modifying therapy once the symptoms of decline are already apparent. To what extent can learning a programming language aid that cognitive reserve?
I see your point though, but I'm not sure how well it applies to funding large-scale data collection work like the HGP. Or the LHC, to use an example in another field.
Couldn't we say that you have formed an estimate of both projects in your mind, and your estimate of the present value of all net benefits from project B is higher than your estimate of project A?
More or less. That sounds like it's giving me too much credit. I wouldn't say I have two estimates so much as one intuition - but that's what my intuition corresponds to, yes. If I made those estimates, I would expect to have a higher estimate for project B than project A.
(But I should note that I deliberately chose projects where it seemed obvious which one was better, so that I didn't need to do any calculations or even think very hard about which one to prefer. In reality, sometimes it's not obvious, and you should think hard and do calculations and it still might not be obvious, and the specific project being discussed is probably one of those cases.)
I don't think we can estimate the expected value exactly, but it nonetheless exists and our only option is to estimate it as accurately as possible (future value discounting the long term impacts of course). If the project doesn't have any practical use but opens up new avenues of research or just has value in any way then those benefits should count of course.
While I generally appreciate the way you think, the problem is likely that different people evaluate the expected value function of this project wildly differently - for perfectly legitimate reasons. The path to resolve this kind of conflict is not straightforward; barring some additional (and accepted) scientific insight, the matter is resolved politically.
That's the thing, though: There's no way to calculate the value of fundamental research. We have lots of examples of breakthrough fundamental research with world-altering consequences, and we have many more examples of projects that went more-or-less nowhere.
If we want the big advances, we have to push the boundaries of what's known, but it's unclear where, exactly, the big breaks will happen. But doing these kind of mega-projects is, I think, basically a good thing, as it gets us out of the day-to-day of publishing easy papers, and into attacking the 'big' problems in new ways.
The human genome project certainly was overhypered and had underdelivered on its biomedical benefits. You should look for articles, even in science press, about the 10th anniversary of the human genome project; it was not all roses, to say the least. In fact they were quite frank about it. The main governmental proponent, the current NIH director, acknowledged it as much. HOWEVER, there is no doubt HGP is a resounding scientific success for its technological breakthrough. Genome sequencing is now a basic scientific tool in all branches of biological science. Its impact on medicine will only increase. A lot of the recent crop of newly FDA approved drugs can be traced to genomic insights.
One good thing coming from HGP for the general public and journalists, I have noticed, is that they have more discernment regarding scientists' promises. Trust, but verify. There is a snowball's chance in hell that NIH will see a substantial budget increase in this decade, or the next for that matter.
BTW, that $1000 human genome sequencing claim is not believed by most scientists. Lawyers can argue it, especially Illumina's, but not according to common sense.
It's around 5k for whole genome sequencing atm and some googling showed that you can do exome sequencing for a lot cheaper (it's a very small fraction of human DNA).
This guy seems to think that the flattening of the tail on that graph is due to Illumina not having enough competition rather than being due to technical limitations at the moment.
The Human Brain Project is a brute force attempt at solving AI (and biological) research problems, and as such it's certainly not alone. I recall supercomputers being put to use in simulating just one neuron at the molecular level. The difference being of course that we do have enough modeling data to make the single neuron simulation fly in principle, whereas the entire brain is ridiculously out of scope.
The main problem with the project though is not the missing scientific data - that just makes it unfeasible for now. What makes it unfeasible in any context is the detail level at which the simulation is supposed to be carried out. This quote from the article expresses it better than I could have done:
Eero Simoncelli, a neuroscientist at New York University. "Would you try to understand the universe by simulating every molecule? What would you have achieved? It’s going to be just as complicated as the real thing and you won’t understand it any better."
Agreed that it's unfeasible for now, of course, but I'm a little amazed people aren't seeing the uses of this.
We can edit, directly observe, and record/playback simulated brains. We can test ten million different models on top of a recorded simulation and see which one fits best. Eero doesn't think we could learn about how the brain works by simulating a trillion slightly different permutations on a brain, or (maybe more ethically) small subsystems of a brain, and observing how each behaves?
I think you might be misunderstanding the intent of my argument a bit.
> I'm a little amazed people aren't seeing the uses of this.
That's not the issue - I imagine everyone here would like to see these goals reached. Nobody has to sell anyone on the rewards of neuro research. The question is merely whether this specific project can deliver them.
> We can edit, directly observe, and record/playback simulated brains.
The idea itself is a good one. However, the key issue becomes choosing an appropriate level of detail for the simulation. I believe a blanket choice of "let's just do the entire brain" is computationally infeasible right now, plus we don't have good enough models to actually program the thing - but most importantly even in a future where these problems are solved the device seems like a blunt and unwieldy instrument that won't give up its data easily.
> or (maybe more ethically) small subsystems of a brain
That's what's already happening all over the world right now, in thousands of independently scoped simulations and experiments.
If you're just saying that this specific project is a poor use of resources at the moment, then yeah, agreed. I don't really feel justified in suggesting that any particular level of abstraction will have so little to tell us that it'll never be a good use of resources.
>That [simulating small subsystems of a brain] is what's already happening all over the world right now, in thousands of independently scoped simulations and experiments.
Woah! Links? The searches I can come up with aren't turning up anything besides that simulation of a rat cortical column and the various attempts at nematode uploading.
Any university with a neuroscience department does this. Go to your local university's website and browse what they're doing in that area: more likely than not you'll find something interesting. Basic research on neurons has become very common, and computer-based modeling is a fundamental part of it. The perception problem here is that, say, modeling the signaling behavior of locust neurons seems like a very inconsequential piece of the puzzle - but in reality it's what we need to do to figure this stuff out.
The fundamental problem in neuroscience research is not a lack of complexity, for now we need to move away from complexity in order to observe the behavior of basic building blocks. It may seem embarrassing how we're still at that stage, but it's where we stand.
There are also some companies that are led by people with academic experience doing spiking neural network simulation and seem to be trying to commercialize it:
Not a neuroscientist (never thought I'd use NANS), and this may well be the boondoggle of a charismatic man who's using it to buy yachts and laugh at the EU.
That said, the one benefit I could see is that it's a billion+ motivation for folks to think about a hard problem. They likely won't accomplish a 100 billion neuron sim, but 100's to 1000's of people will hopefully be thinking about that goal, trying to decompose it, and producing useful sub-advances. Maybe like the HGP or the LHC or space flight, intense money will lead to enabling technologies that accelerate what is possible.
The other problem is that humans are horrible at predicting scientific progress, even in fields they are comfortable / experts in. Edge scientists may actually suffer from this the worst. So much of life has been devoted to becoming an expert in the field that it becomes more challenging to consider large changes. Even while most major breakthroughs are built from analogies to other fields, and more "simple" every day phenomena. Ex: Relativity, one of the weirdest advances ever, came from thought about trains.
Maybe a billion+ could be spent more wisely, but at least they're spending it. These are the same folks who decry our governments whenever they reduce funding and malign the sciences. Stop complaining and find a way to write the grant, or meet the right folks, or whatever's necessary so you can be a part of the money train and get something useful done.
Can someone explain a bit about the problems? I am not a Neuroscientist but a Machine Learning researcher, and given the recent results in Deep Learning / Artificial Neural Networks which reach Human performance in certain specific tasks, it almost looks like if you would just put an ANN together which is somewhat big enough and somewhat similar wired together as the Human brain, you would yield something similar. And the point is, it doesn't really matter at all whether that is exactly like in the Human brain, and also, you don't really need to understand in detail how it will work, like you can even not really explain the current simple ANNs. It just works anyway. So, under this view, it's just a matter of scaling up, and to wait until we have enough computing power. And then, you would add more details to make it more close to the real Human brain.
As an ML practitioner, you certainly know that the ANNs you use have not much in common with biological (real) NNs. First, ANNs tend to be mostly feedforward, while rNNs are highly recurrent. ANNs are therefore not very good in tasks where memory is needed (I know about the developments involving LSTM neurons, but they are not analogous to the implicit memory of recurrent neural networks.) Second, ANNs usually don't have a time dimension, the firing is essentially a floating point value instead of action potential. Third, recurrent neural network structures do not scale: an efficient/reliable/reduntant system of 100B neurons will probably have extremely different structures than another one with 100M neurons -- because of recurrency and other stuff a rNN is a chaotic process (in the sense of sensitivity to parameters) that should be stabilized by the structure.
And there are many-many other differences. Note that our task is not to solve problems but to figure out how the brain works.
Also, recurrent neural network simulation cannot really be scaled right now, we don't have the hardware. It is not parallelizable with our current tools because of the huge number of connections.
(Disclaimer: I was involved in a project trying to model real neural networks. It wasn't a huge success, but we learned a lot.)
This is a good informed summary, thank you. I asked the same question of my Alzheimer researcher friend, and he gave a pretty similar response including aspects like the huge computational requirements and the basic non-similarity of rNNs and ANNs (albeit with a disclaimer that he wasn't in the field).
Nonetheless, I look forward to seeing more simple rNNs being created over time (besides the C. elegans one that was modeled recently). Who knows what strange organizational rules or structures we will discover from this strand of research?
I would say that the problem with your proposal here is that those "more details" you mention consist of a vast horde of information we do not know anything about that far outweighs what we do know. At best, we know that certain areas of the brain concentrate certain functionality, and we can map rough levels of interconnectedness and activity traversal. That is not nearly enough information to build a brain. The neural nets that exist may sometimes achieve "Human performance" at narrow tasks but it's a big leap to infer that those neural nets are actually functional models of subsystems in the brain, or that their isolated functionality can be plugged into a larger group of such neural nets and have something coherent emerge. Our actual brains are far more interconnected and overlapping than is obvious from reading the latest fMRI study.
It defies understanding why one wouldn't start with a much simpler organism rather than trying to go directly to the human brain. For one thing, you can actually experiment on, e.g., a rat brain in a way that you can't on humans.
A reasonable near-term goal is to get OpenWorm [1] to work well in simulation. That's the simplest known organism with a system of neurons, and the connections have been fully mapped out. Until that works, there's little point in trying anything more complex.
The "the next step is the human brain" people have a poor track record. Rodney Brooks tried that once. He'd done a good reactive-controller insect, and then immediately tried to jump to human-brain level with Cog.[2] When he gave a talk at Stanford proposing Cog some years ago, I asked him "Why not try for a mouse next? That might be within reach." He said "I don't want to go down in history as the man who created the world's best artificial mouse". Cog was an embarrassing failure.
The Human Brain Project should be put on hold until OpenWorm works and that technology has been advanced to at least the lizard level. The Human Brain Project is likely to turn into an expensive supercomputer boondoggle.
The work and challenges for OpenWorm may not reflect the work and challenges for the Human Brain Project. Among other things, unlike most other neurons, C. elegans neurons don't spike. Action potentials make simulation more tractable since you only have to worry about propagating spikes, and not all internal state, between connected neurons. (At least ignoring things like ephaptic coupling, which is probably not that important in mammalian neocortex.) Also worth noting that the Blue Brain project supposedly succeeded in simulating 100 neocortical columns, which is 1 million neurons, vs. 302 neurons in C. elegans.
That was about sequencing DNA. It was known how to sequence DNA before the Human Genome Project started. It just cost too much and was too slow. That was a production scaling problem.
If we ever get to a mouse brain, it's just scaling from there - all the mammals have very similar DNA. We don't know how to make even a good lizard brain yet. Or even a full insect nervous system.
As others have pointed out there is a long anecdotal history of magnetic personalities getting hype and funding based on their empty promises. However, there are large counterexamples and this article clearly contains little technical criticism (another commenter links to a more balanced view).
The Human Genome Project is cited as one such example: simultaneously not meeting the hyped goals whilst also being more successful than imagined in other ways. Perhaps what people are missing here is that the value may not be in simulating the brain but rather having the infrastructure on which to run simulations. This likely starts in parts and on a small scale but the infrastructure alone could be incredibly useful for generating new ideas and shortening feedback time.
> "Would you try to understand the universe by simulating every molecule? What would you have achieved? It’s going to be just as complicated as the real thing and you won’t understand it any better."
I'm pretty sure cosmologists would love to make simulations to a molecular precision. They currently simulate gas clouds with supercomputers in order to study the formation of galaxies and the more precise the better.
Absolutely right, simulations are critical in physics, cosmology, and pretty much every hard science. I think the nay-saying in the article is indicative of a larger enmity some academics have towards simulation based science. The harder sciences have mostly come to terms with the contributions simulations can make, but I think it will be sometime before the softer disciplines come around. In the meantime, politics and ad hominems as usual.
I'd be very surprised if they come anywhere near reaching their goal of a comprehensive whole-brain simulation, but even if the overall project aim falls short, many of the sub-projects are likely to provide worthwhile outcomes.
For example, structural and functional data generation projects described in there (SP1, SP2 and SP3) sound reasonable and reflect the kinds of topics and techniques that are being researched already. The neuroinformatics (SP5) and medical informatics platforms (SP8), while very ambitious, seem like they could be a tremendously useful resource for linking disparate data sets together into a single, more easily-accessible database.
I can see why many neuroscientists are scoffing at the rather over-hyped grand aim of the project, but that doesn't mean the entire thing should be scrapped. Personally I think that funding comparatively open-ended, long-term, risky research is a good thing. Even if it 'fails', that failure is informative in itself in helping to provide scope for future projects. And it increases the opportunity for serendipitous discoveries.
The example given by the NYU neuroscientist seems to be betray some misunderstanding/lack of vision:
"Would you try to understand the universe by simulating every molecule? What would you have achieved? It’s going to be just as complicated as the real thing and you won’t understand it any better."
At the very least you would have achieved the ability to rewind and play the universe forward again, which is something that we most certainly can't do with our real universe. You would also have achieved the ability to experiment with and measure the universe with potentially greater precision than we can in the physical world.
Simulation is a tool to help you understand, not understanding itself. I don't think Markram ever said anything to the contrary.
There are alternative approaches that attack the problem of simulating the human brain with a smarter set of approximations, e.g. http://thevirtualbrain.org
"When a distinguished but elderly scientist says that something is possible, (s)he is almost certainly right. When (s)he says it is impossible (s)he is very probably wrong." - Arthur C Clarke
The reason these scientists are upset is because we are no where near being able to do this. They're not saying we can't do it someday, just that it's a crazy waste of money right now. We don't know enough about the brain to make reasonable full scale models of it yet.
I have a book from 1968 where this was proposed as an inevitability by the 1980s. The same author (it was sponsored by the foreign policy association) assumed ray guns and antigravity were highly likely.
I'll check it out, thank you. I really hope it can somehow work out. It just seems to me that the cost of even a small amount of people being unethical makes it not worthwhile, given the exceptionally high potential for bad. Much worse than nuclear bombs, in my opinion.
A simulation of a human brain is a human being, and human experiments are generally considered unethical. To say nothing of the possibilities if they succeed.
Assuming it is, in fact, actually experiencing life, and not just appearing to be.
But if it is, imagine what happens when, in 50 years, it becomes cheap to reproduce. Any bozo can get a copy of it. The potential for mistreatment is beyond anything seen yet on Earth, as bad as that's been so far. Check out the Christmas special of Black Mirror, for just one example.
Fine, but how are you going to prevent abuse if the thing is easily copied and spread around the Internet (when home computers are powerful enough to run it)? Empathy has worked naturally, to the extent it has, because of our natural reaction to other physical humans. And, for however bad people have been treated despite this, flesh and blood human beings at least eventually die.
How are you going to prevent abuse? I understand all innovations have risk, but I think the potential for abuse is so incredibly bad, that it's not even worth going in this direction. Assuming there's any way to stop it, but there's no sense in not trying either.
I agree that empathy would be a problem, but the exact nature of the simulation matters. If it were an exact copy of a human being, right now that's not technically feasible (or if it is, not outside of a highly visible major project).
In any case, the current proposal is mostly a funding grab, you won't see an actual brain simulation come out of it. Even a nematode worm is ambitious right now. See http://www.openworm.org/
I accept that's a consequentialist argument though. If we did manage to create cheap, easy to run brain simulations, I think we're starting to run into the question of the nature of personhood and the nature of consciousness (something I hope to contribute to some day). Do AIs deserve to be free from slavery if they are not experiencing anything consciously but still have desires and creativity and can convincingly pretend that they do experience something (which a good simulation would do)? I have a feeling that regardless of the "correct" answer to that question, whatever it is, human beings' emotions are very hackable and we'll grant them rights based on emotional appeals if they get to that point. ;)
That would open an interesting technical issue though. If a human simulation was cheap and easy to run, and we wanted to legislate the conditions under which it was run (e.g. must be given reasonable sensory inputs and have hunger, sexual desire, and other wants set to near 0), we'll run into the DRM / TPM debate again, but this time with serious "life or death" consequences.
I trust that you would do the right thing. My concern is if your software gets out there on the web (even if in 50 years, I'm not saying it's an issue now) and some jackass from 4chan has a different agenda. Or maybe if "brain uploading" becomes a thing (which I have problems with on a philosophical level, but that's another discussion), maybe North Korea could use it to up the ante on their "3 generations of punishment".
That's an interesting question. We already have that problem though, ever heard of humans? ;)
Typically we deal with it by having a waiting period of 18 years from birth (balanced by death) or a green card to citizenship process that takes a while and is limited in number.
On the other hand, would simulated humans even need much of a vote? I think their primary needs would be satisfied by simulated satisfaction. The only thing they they would need-need in the physical world would be for us humans to not pull the plug on them and to service their hardware when it fails. Much easier to deal with than real live humans. Just a few real humans might be needed to maintain entire cities of these people.
It sounds like a great sci-fi story to grant a real-world franchise to a few elected officials in the sim-city and keep the truth hidden from the sim-citizens.
Why are such projects always coordinated as one big blob? Isn't it possible to just fund startups that are active in this industry, and see which one survives? Of course there should be regulation to a specific degree. VCs are doing the same, and it seems to be working (for the VCs).
> VCs are doing the same, and it seems to be working (for the VCs).
Well, you don't want that in science. The point of startups (from VC's point of view) is to make money for VC - startups themselves and their survival is irrelevant. The point of research is to gain knowledge and not to make money. That's two completely different kinds of thinking.
Because in Europe the belief that important and wide research should be carried out by "public" research is much stronger than, say, USA. Also while we have seen a somewhat invading process of privatisations ( :( ), European universities are still largely owned publicly and financed via public funding, therefore most [public] moeny for research should and will be directed to them for the time being.
You could then argue that it's not as efficient as the VC model, but:
- I haven't seen any study confirming such thing
- It doesn't seem easy to achieve a reasonable compromise between open research and R&D protectionism unless you are a big company
- Academia, even with all its politics and problems, is still a fast and active environment for scientific research
I've seen a good number of boondoggles. At the center of each was a personality of such magnetism that people simply fell in line behind them. There are people who have the ability to spout utter nonsense, but to do so with such alpha authority that nearly everyone in the room just glazes over and nods. Whatever it is that our brains take as a signal of alpha status, these people know how to crank it to 11.
When are people going to learn that primate dominance gestures are not a reliable proxy for much else?
I will say that when actual merit and this level of dominance actually align in the same individual, amazing things can happen. But I see no sign that this is more common than would be predicted by the random assortment of traits.