Study finds a difference between neurons of humans and other mammals | Hacker News Reader

Study finds a difference between neurons of humans and other mammals

news.mit.edu

184 pointsdootah76 comments

Here’s the actual article, since the people that write these press releases somehow think that the actual paper is unimportant:

https://www.nature.com/articles/s41586-021-04072-3

>They were able to obtain brain tissue from 10 mammalian species: Etruscan shrews (one of the smallest known mammals), gerbils, mice, rats, Guinea pigs, ferrets, rabbits, marmosets, and macaques, as well as human tissue removed from patients with epilepsy during brain surgery.

Offbeat question: Could it be that epilepsy is a disorder caused by this missing ion channel, and this isn't true for all humans?

It's not clear that every case of epilepsy has a common cause, but this is certainly possible.

If true, that would be an embarrassing oversight in the study as well as possibly #1 in one of those "Top 10 accidental scientific discoveries that changed everything" articles.

It is an oversight unless they identify it as a limitation in the study.

> To limit the potential confound of disease states associated with our human brain samples, we compared the biophysical properties of neurons from patients with hippocampal sclerosis; tumour; and other conditions, including gliosis and trauma. We found no significant differences in somatic outside-out currents between the three groups (Extended Data Fig. 7). We also directly tested the effect of epileptic seizures on somatic ionic conductance in rats and found no significant differences (Extended Data Fig. 7). Together, these results suggest that disease aetiology is unlikely to underlie the distinct features of human L5 neurons such as their low conductance.

Disclaimer: I have not read this article. However, there are multiple known genetic epilepsy disorders that have specific ion channels that are either missing/mutated.

Epilepsy is a really wide range of disorders with many causes that simply manifest in similar ways.

So it is possible this is a cause of a single type of epilepsy, but not all.

My own epilepsy was the result of spinal damage as a teen that left me temporarily, partially paralyzed. I regained movement but have frequent seizures from the damage.

Similarly a friend had a growth on his brain. Surgery removed it but he was epileptic for about a year after.

I wondered the same thing.

Or the ion channel deficit was a side-effect of anti-seizure drugs often given to people with epilepsy, or a side-effect of having seizures, or ...

Presumably there is some "gotta be ultra-fresh" reason why they aren't taking just tissue from cadavers donated to medical schools. But this study sounds like it was pretty crappy science, even before the MIT press office folks got hold of it.

I think it should be obvious that you cannot study electrical excitability in dead, fixed cadaver tissue.

This is top-notch science, but might be hard for people to appreciate.

Maybe not missing ion channels, but it does seem to be caused by over sensitive ion channels, and several different ones. They used to think it was just sodium channels but now they find the calcium channels (which are also implicated in Bipolar Disorder) and the potassium channels can play a role.

https://www.intechopen.com/chapters/65083

They use sodium channel blocker's as both anti-epileptics and mood stabilizers.

I have Bipolar Disorders and have the CACNA1C, SCN2A, and KCNQ1 risk alleles for the disorder. Depakote used to work well for me, Lamictal worked even better, but Benzodiazepines as needed seem to help me the most. They are all used to treat epilepsy as well.

This study does not report that that layer V cortical neurons are missing a particular ion channel. They examine currents through HCN and voltaged-gated potassium channels and find differences in the total current carried that depends on species.

In general, epileptic disorders are observed across species. It's common in dogs, for instance: https://www.akcchf.org/canine-health/top-health-concerns/epi...

A ton of papers examining ion channel expression in epilepsy that should be more relevant: https://pubmed.ncbi.nlm.nih.gov/?term=human+ion+channel+epil...

Since it’s about density of ion channels, surely you’d want to check brain samples from animals with brains larger than humans’ to compare?

This is a good point. There are several other metrics by which larger-massed animals are more efficient than smaller ones, such as oxygen usage per kg of body mass. Perhaps a similar effect is applicable to brains?

I heard in a lecture that populations of human neurons are able to phase-lock or resonate to an oscillating stimulus that exceeds 1000s of Hz while mouse neurons top out around 200 Hz. I haven’t been able to find citations for this. Hypothetically, the increase in temporal sensitivity might be related to the lower density of ion channels. Would a lower density of ion channels increase the capacitance of neurons?

I would think capacitance increases only if the number of leak channels change.

The majority of a neuron's transmembrane capacitance is attributed to the lipid membrane itself, so a lower density of ion channels should not appreciably alter capacitance.

Action potential rates and the ability to phase-lock to inputs are significantly affected by the membrane time constant. Generally, membranes with lower conductances at rest and during spiking will have higher (slower) membrane time constants. This isn't a straightforward explanation for the ability of human some neurons to phase lock to higher frequencies.

Impedance data across species relevant to this idea is in Extended Data Figure 2 and 5 of the paper.

Consider what ion channels are evolutionarily. Gaps between cells.

The first proto brains were colonies of cells that lived communally and expressed logic through group dynamics.

Think of a colony of cells on a rock under the ocean. As the sun moves across the sky, the cells have to make a decision to move to the more exposed face of the rock to maximise the capture of solar radiation.

The cells have to communicate because the cells on one side register different quantities of radiant energy to the other side and extremities. So they conduct a poll or a vote. The data is tabulated through intercellular communications, which takes place across the gaps that eventually evolved I to what we know of as ion channels and synapses in modern multi-cellular animals.

This is, in a sense, a microcosm of the argument in Daniel Dennet’s “Consciousness Explained.”

To oversimplify (and potentially butcher): consciousness results from innumerable “drafts” or guesses arising from neural interpretation of sensory input.

From Wikipedia, describing the book:

‘The brain consists of a "bundle of semi-independent agencies";[3] when "content-fixation" takes place in one of these, its effects may propagate so that it leads to the utterance of one of the sentences that make up the story in which the central character is one's "self". Dennett's view of consciousness is that it is the apparently serial account for the brain's underlying process in which multiple calculations are happening at once (that is, parallelism).”

This squares with your observation about likely early cell communication, which led (via many routes) to modern cognitive functioning.

Picked up this book just now... Seems full of red flags. In the first few pages:

- Accepts Descartes' "I think, therefore I am" without skepticism

- Aims for a scientific theory to be "respectable"

- Concludes that because sensory input to a brain seems infeasible using 1990 simulation tech (Not considering advances, using sensors etc), simulating reality to a brain isn't possible; Makes a comparison to a ladder-to-the-moon.

Hmm... those seem like pretty shallow interpretations of his research.

What do you mean by 'accepts Descartes' "I think, therefore I am" without skepticism?'

They're shallow, but there's such a high density of questionable statements in the first few pages, it's a red flag. I love speculative science and reading about potentially new and/or controversial ideas... but it only makes sense if I can trust the author on more concrete topics.

> "I think, therefore I am". Philosophers today are less concerned with proving one's own existence as a thinking thing (perhaps because they have decided that Descartes settled that matter quite satisfactorily)

I think you’re misinterpreting. Dennet is setting up to question Descartes’ assumption, which has been “accepted,” but he spends the first 50 or so pages of the book breaking down.

I’m not saying this book has all the answers about consciousness (quite the contrary — it has been criticized for not really “explaining consciousness” as the title implies), but I think you would find the studies and observations about how our minds generate “drafts” of reality quite compelling.

I highly doubt that reducing the number of channels is a simple efficiency hack, if it was simple to just reduce the density then animals would have had lower density just like humans. The question to ask is why humans would benefit from having fewer channels while animals doesn't.

My naïve hunch tells me that this could be related to rationality in some way, that weakening the ion pump signals could make it easier for cells to learn and make alternate computations, creating humans strong rationality that is capable of overriding our natural urges. An animal without significant ability to learn rationally like humans wouldn't have a reason to weaken the signals, so to them they just optimize for signal strength and energy use.

> ... then animals would have had lower density just like humans.

My just so story would be that those with the higher benefit out-competed (i.e. killed) those primates with somewhat lesser efficiency.

The article mentioned thst reducing thr number of channels decreases the energy use, but surely this is for some kind of tradeoff? Would he nice to hear what this might be since the article did not mention or speculate it

> My naïve hunch tells me that this could be related to rationality in some way, that weakening the ion pump signals could make it easier for cells to learn and make alternate computations, creating humans strong rationality that is capable of overriding our natural urges.

That's quite a conclusion you're jumping to...

"rationality" is a word denoting a very complex concept obtaining its meaning only through a specific historic background including a lot of assumptions that are not as universal as you might think.

I would always be very careful when trying to bridge between

1) A physical phenomenon.

2) Concepts derived through a complex cultural discourse.

These are two very separate worlds.

Quite, but I don't think GP is jumping to a conclusion. More like proposing an interesting hypothesis.

To say it less politely: To me it seems more like stoner talk.

"Quantum entanglement, ... blablabla, ... collective subconscious, ... blablabla, ... everything is connected, man!"

It seems like early modernism, where they discovered something called radiation existed, and immediately decided it must have superpowered positive effects (curing all diseases) rather than being bad or useless.

I don't think you understand how science works. The aim isn't to be correct, the aim is to find potential areas of progress. Is my hypothesis here most likely incorrect? Yes, of course it is, but that doesn't matter since if it is correct it would be huge and therefore it is worth exploring.

But thanks for your comment, now I understand better how luddites thinks. So in the future when I make comments like above I'll say "this hypothesis is unlikely to be true but...", so that even luddites can understand it.

> I don't think you understand how science works

I have some understanding of how science works. The purpose remains to find plausible explanations to phenomena, explanations that show some solidity.

The same value - linked to grounds - remains valid for statements in general. This is why we refine them, in the internal "lab", where they are tried under duress of criticism. Which is scientific process.

And how is "weakening the electric signals between cells makes it easier for cells to alter the message and therefore might give rise to stronger forms of learning" not a plausible link? It costs more energy to ignore a stronger signal, making it harder for cells to learn to change input signals. Humans having weaker signals could therefore be related to our ability to learn, while animals could have stronger signals making them worse at learning but better at following instincts.

Of course I could be wrong, but can you really say there is no way my hypothesis could be right?

The learning part might be right.

I'm put off by the "rationality" part.

At least to me rationality is the same thing as ability to learn complex things and have those new learnings override your intuitive responses. If you have some other definition of rationality then maybe that is why you thought my post was strange.

> rationality is the same thing as ability to learn complex things

I think this covers only a part of what the word means. There's a strong connection to (and to some even identity with) the term "reason".

https://en.wikipedia.org/wiki/Reason

"Reason" as a concept is deeply rooted in the enlightenment era and comes with a lot of metaphysical and historical baggage.

" It costs more energy to ignore a stronger signal," - wait, what, why? Do you have any citation for that? IMHO that has absolutely no basis in how neurons work, it does not cost a cell energy to ignore signals.

"making it harder for cells to learn to change input signals." this sentence is not even wrong; it essentially presumes that the default behavior is to "not change input signals", which is definitely not how neurons work.

"Humans having weaker signals could therefore be related to our ability to learn" - perhaps, but you provide no argument about the direction of causality there (if it's related, perhaps the weaker signals are caused by a different level of learning), nor for elevation of "could be" from mere possibility to why it's likely to be the case.

I'm not saying that there is no way that your hypothesis could be right, but you do need to pass a certain (quite high) bar of clearly stating your assumptions and demonstrating that all of the assumptions are reasonable before your hypothesis would be worth reviewing and evaluating, and there is very a big gap between "possible" and "plausible".

> how science works... hypothesis

This is precisely what I would criticise:

It's not actually a hypothesis in terms of being a testable statement:

How would you even operationalize "rationality"?

Of course it is testable, psychology tests way more vague concepts all the time. It isn't as testable as concepts in physics or even biology, no, but it is still testable.

What would an experiment look like? Lets say we have a drug which gives you animal level of receptors. Inject that into babies and see if they stop developing human level thinking, and try to find out why. Of course we can't run such experiments with our current morals, but it isn't impossible to do.

Another thing you can do is look at the level of these in different humans and see if it correlates to achievements on school tests etc, see if some learning disabilities might be correlated to it etc.

Or an even cooler thing would be to see if reducing animal levels of receptors would make it easier to teach them stuff. That would probably even pass our moral test.

yeah.. my first hypothesis i was jumping to was that reducing the ion channels reduces specialization and can work on improving distributed(across) neurons decisions/calls... It would be interesting to see a cognitive behavioural task both primates and humans can do and see if the blood flow/electromagnetic activity is more spread across the brain vs more localized in parts of the brain.

> That's quite a conclusion you're jumping to...

It isn't a conclusion, it is a hypothesis. Human rationality is what makes us unique among animals, so anything that sets our brain apart from animals could be related to it and therefore warrants further study to see if it is relevant to human intelligence.

> "rationality" is a word denoting a very complex concept obtaining its meaning only through a specific historic background including a lot of assumptions that are not as universal as you might think.

> I would always be very careful when trying to bridge between

In what way am I not careful here? It would be great if we could make a connection so we should explore any potential connection further, why are you against that?

Is your point that this is a hard problem to solve and therefore we shouldn't look for solutions? The link I posited here is very far out, yes, but it is still possible that there is a link and if there actually is one it would be a huge discovery.

> Human rationality is what makes us unique among animals

Are you sure? It seems to us that reasoning, to some degree, is a distributed gift - in fact, all brains are there to take decisions.

See for example https://www.uh.edu/news-events/stories/2017/november/1101201...

The levels to which humans can learn rationality is in a completely different league from animals, so yes I'll say I'm pretty sure about that. It is like the difference between merge sort and bogo sort, sure they solve the same problem but bogo sorting doesn't work for anything but datasets with a few elements. Animals reasoning chains breaks down extremely quickly, they can do a few steps, but humans can do hundreds or thousands of steps and the reasoning can still be solid and useful even after that many steps.

That is a quantitative difference, not a qualitative one. And it is probably related to a synergy of functions.

Quantitative differences becomes qualitative differences when they get large enough. For example, the difference between merge sort and bogo sort are qualitative, you can't sort anything with 50 or more elements using bogo sort while you can sort billions of elements with merge sort.

Nothing of the sort of "rationality" exists in the history of early humans or their ancestors.

I'm not even sure what is this rationality that you're talking about. Because the main narrative of early humans evolves through giant leaps of irrationality.

Even today, most humans are irrational.

> Even today, most humans are irrational.

Compared to what? Compared to animals basically every human has genius level rationality.

The ability to develop and spread farming methods, how to create advanced tools, language, writing etc, is what I mean when I talk about capability for rational thought. Every healthy human has that, no animal is even comparable.

I will admit what I am about to say is more philosophical, and it might not be directly to your point, but hear me out.

There might be an innate rationality in other animals that you are dismissing or overlooking. We all know that when a bird eats a berry it sends the seeds out like a like Johnny Appleseed. Do they use their rational thought to do that? No. But they do not need to.

Or how about the birds who can navigate thousands of miles using the magnetic field of the earth? I do not know many humans who could do that.

Or what about a Killer Whale that can communicate to others 10 miles away without the use of any technology?

Compared to humans, every animal has genius level innate rationality that can compare to no humans.

As humans we are so obsessed with with out own type of greatness but we really need to step back and humble ourselves on the greatness of other (and I will not call them lesser) species.

IMHO, maybe humans have dumbed down and inhibited, and now we need all this technology to survive because we lack the innate ability to do it on our own.

The poster could have meant "maybe a reduction in the density of ion channels does much more than change the mechanism: it also forces the function to adapt in different directions (say, towards the basis for more distinctive abilities)".

I was thinking somewhat similarly.

The reduction in channels could mirror a difference in attention-investment. More attention invested in the "thought realm" and less in the "physical realm". Thus our unique facility with thought.

Many animals aren't under strong pressure to increase efficiency of their brains, because these brain are small and majority of metabolism is put into very basal structures like brainstem.

Birds for example were under such evolutionary pressure and look where it got them - Corvidae have 15g brain and they are much smarter than 22g squirell or 30g cat.

Moreover, such choices have tradeoffs, improving efficiency often requires more precise regulation that can be nondesirable for small species.

I do not understand this obsession people have with ranking the "smartness" of animals. Each animals has the intelligence it needs to survive in the niche it lives in. Each is as smart as it needs to be.

You say a Corvidae is smarter than a squirrel, but why do you think that matters? Each species survives just fine. Would it benefit the squirrel species survival if they were "smarter"? Or would there just be more of them and then overpopulation and pollution and...oh, I see I am talking about humanity now. ;)

You know, the human experiment is far from over, but everyone seems to have our evolutionary survival mapped out like some sci-fi movie that ends up having us colonize the universe.

> Each is as smart as it needs to be.

Petter Watts formulated it in different manner - "all animals are under stringent selection pressure to be as stupid as they can get away with".

> Would it benefit the squirrel species survival if they were "smarter"?

Probably yes, but this benefit doesn't necessarily outweight greater metabolic cost of sustaining brain. But I was talking about efficiency. Birds achieve much higher intelligence from the same volume of brain than mammals.

> "all animals are under stringent selection pressure to be as stupid as they can get away with".

Thanks for that quote. So maybe that is what humanity is doing by offloading our intelligence to machines, making ourselves stupid?

>Birds achieve much higher intelligence from the same volume of brain than mammals.

I just have a hard time supposing what "intelligence" means since human intelligence might be much different than bird intelligence. Not higher or lower, but different for different needs.

I think the more likely situation was that, given that humans spend so much of their energy in their brains, early humans had high selection pressure to make this process more efficient.

We spend roughly 10% of our entire energy budget on those ion channels. Keeping neurons at -70mV takes a lot of work.

Macaques were the closest primate studied; I'm very curious if this also exists in other primates or if it is indeed limited only to humans.

> The researchers are also interested in exploring whether primate species that are more closely related to humans show similar decreases in ion channel density.

This is fascinating, groundbreaking research I think! Never heard of anything like this. Any neuroscientists here who can speak to implications?

Just read the abstract but did nobody really run this experiment until now? Neuro seems to be in such an early stage as a field; it seems like Kepler-era astronomy.

I wonder how are the ion channels of Dolphins are like.

Can they please take samples from living humans and compare them -- compare the tissue from the brains of say einstein (I think its in a jar some place) to some other modern "great brain" -- and compare them between musk and bezos (as an example of two successful, 'peers' currently living)

Does having rich parents at the right place and time really put you in the same category as someone like Einstein, who changed humanity's entire view of reality and the nature of the universe?

I am more interested in seeing if, for example --

You compare the nerons of the father of einstein, with einstein -- then a peer of einstein and their father (or mother/whatever)

To determin if neural pathways are genetic in such a way that we currently dont understand - and if theyse pathways are similar in anyway between two different persons with similar level of intellect.

Bezos is more like a garage startup founder though.

Putting aside if this is a good idea at all, I think Terrance Tao might be a better comparison for Einstein?

So, basically we're more intelligent because we're better at pumping ion.

How can I improve my ion pumping?

You can give yourself Bipolar Disoder...

https://pubmed.ncbi.nlm.nih.gov/2735813/

It's a blessing and a curse.

Sounds like GP just needs to ingest some anti-lithium.

Several things can trigger calcium channel activity. One of them is Extremely Low Frequency Electromagnetic Fields.

https://www.nature.com/articles/srep21774

I like the aesthetics of interpreting this through the academically fringe lens of “Quantum mechanics/entanglement plays some role in the human brain and solving the binding problem” and hope this relates more to making quantum-scale events capable of triggering action potentials at a statistically meaningful rate, than it does to mere energy savings.

The aforementioned binding problem, which I'm now reading about: https://en.wikipedia.org/wiki/Binding_problem

Interesting stuff. I believe the OP is also refering to this theory by Roger Penrose:

https://en.m.wikipedia.org/wiki/Roger_Penrose#Physics_and_co...

It's well known for a a long time that humans have the highest density brains.

Time to broaden the research. Go nuts with all the other intelligent animals.

Pretty soon Moore’s Law too shall descend into bio-chemical computing.

Reply on news.ycombinator.com