> We demonstrate fungal computing via mycelial networks interfaced with electrodes, showing that fungal memristors can be grown, trained, and preserved through dehydration, retaining functionality at frequencies up to 5.85 kHz, with an accuracy of 90 ± 1%. Notably, shiitake has exhibited radiation resistance, suggesting its viability for aerospace applications
Soon we'll have shiitake replacing transistors in our airplane and spacecraft computers, while sitting and eating ramen on the vehicles themselves. The future is shaping up to be interesting.
Having only dabbled the slightest in hardware... are functional frequencies topping out at 6 kHz useful for memristors in modern computing? I feel like having separate components each magnitudes faster would be better than combining them into a memristor that sounds so slow.
If you mean x86 class performance, no, obviously not.
But 6kHz is not nothing. For application-specific computers, you can do a lot with very little. You aren't going to be building high performance general purpose computers, but for an atonomous circuit quietly ticking away computing orbital trajectories or stellar navigation, you don't need modern x86 class performance.
A bunch of mushrooms as a giant 6kHz memory array with in-memory computing seems pretty pointless IRL. But it adds a nice air of plausibility to some sci-fi stories!
New Trek is so awful. There’s no depth to anything. Interweaving stories to make them seem important. It’s just bad TV.
The previous, it gets better the closer to TNG it is. Granted DS9 was a different beast than TNG or Voyager. Those shows had episodes, individual stories, as well as seasonal arcs. Back when shows were written for TV. New Trek feels like a bad movie script broken into episodes with side character filler.
What it makes me think of is 'cybernetics' research from the 1960s when it was not a foregone conclusion that transistors, especially CMOS transistors, were the future of computing. Back then there was a lot of research into alternate models of computation, something that's only becoming relevant today as CMOS may be running out of steam.
I recently read The Unaccountability Machine (which I can recommend btw), which mentions Stafford Beer's experiments with a computing pond. Who knows, maybe we'll control our factories with mushroom brains soon!
> Soon we'll have shiitake replacing transistors in our airplane and spacecraft computers, while sitting and eating ramen on the vehicles themselves. The future is shaping up to be interesting.
By the way, some people say eating meat is not going to be sustainable as more and more people become able to afford it, and fungi are a great option for providing the equivalent protein intake.
There's a theory that's been going around for a while that trees were using mycelium networks to communicate via electrical signals. Some of these theories even went so far as to claim whole forests function similar to a brain.
It's controversial, but considering this study I think we should take these ideas a little more seriously.
I thought this was fairly well proven at this point. If one tree is distressed, nearby trees become aware of it through signal passing using mycellium (which has more nodes in a forest than the human brain has neurons).
Fungi are deeply alien life. Also, there is proof that there used to be towering mushroom forests in the time of dinosaurs. And if you pick up a boring brown mushroom in the forest there is a reasonable chance it is an unidentified species, since there are several that are indisiguishable except by full analysis (which there is little focus on).
• Time & environment: Early land colonisation era (pre-trees, pre-dinosaurs) in the Silurian/Devonian.
• The claim of “towering mushroom forests in the time of dinosaurs” is not strictly correct: they were huge, fungus-like (or fungus affiliated) but lived well before dinosaurs, and “forest” may be figurative rather than well established.
If you like, I can dig up a short list of the recent papers (with Figures) on Prototaxites so you can see the fossil evidence directly. Would that be helpful, Rob Mpucee?
===
That’s a wild answer lol. Although it technically did answer the question.
What would be the evolutionary advantage of this? I know nothing about biology so I'm struggling to see the use of such a network. Some kind of synchronization?
IIRC, communication of threats, exchange of rarer nutrients (this is the main symbiosis trees have with the mycelium anyway), perhaps from healthy trees to needy trees. Solething like that? If I find a source I'll edit this comment.
We basically just haven't studied it enough, but it's very clear that massive mycelial networks are doing something.
I don't think it's absurd to hypothesize that a mycelial structure as complex and interconnected as an animal brain might have similarly complex emergent properties. It's an extraordinary claim, but really not out of the question. We just need to go and find out.
There's an interactive story that has elements of this[0]. Many of the simpler objects don't have much capacity to think or feel on their own, but the corru equivalent of elevators are fully sentient beings capable of conversation and problem solving, and they're just kind of built to be quite satisfied helping move people around. Corru computers are capable of hosting entire communities of distinct intelligences, each program sentient and (mostly) dedicated to its role. Not all of them can be chatted up, the authorization/access control program understandably isn't very chatty, but it is an intelligent being.
It's a pretty enjoyable experience, and all of the graphics are ordinary HTML elements with 3D CSS transformations, which makes it super hackable and fun to crack open in an inspector.
All that to say, if the best chairs required intelligence, it'd be in everyone's best interest to make that intelligence real thrilled about ass.
In one of the Hitchhiker's Guide to the Galaxy books there was a bit about the company that makes Marvin, the depressed robot, also making sentient elevators. These elevators had issues with getting depression, and the companies using them in their office buildings would hire psychology students to talk to them on the basement levels and convince them to go up again.
Most wet organic or biological materials will produce memristive (pinched hysteresis) IV curves when measured with macroelectrodes (1–20 mm contact scale) due to inherent interfacial electrochemistry, ion migration, or redox processes at the electrode-tissue interface.
Macroscale memristivity is an artifact of slow interfacial kinetics sampled over large diffusion volumes. Shrink electrodes below ~100 µm → hysteresis vanishes.
Dry biological materials almost never show true memristive (pinched, history-dependent) IV curves at any electrode scale (macro or micro) under standard DC sweeps. The reasons are structural and physicochemical—drying eliminates the ionic mobility that sustains memory.
If we're ever going to colonize space or even do automated manufacturing in space on any scale we need to build a system which can manufacture "anything" that can be sent in a small number of launches and watched over by just a few people.
Eric Drexler's "assembler" concept has been stuck for the last 25 years, but biological systems are a good model because if they can build you out of a cell they could build just about anything else out of a cell. This kind of mycelium network is running fast compared to the neurons in your brain.
Do we have to believe this will scale and be "generally useful" (whatever that means) in order for it to be interesting enough to talk, discuss and think about?
My reaction is more, how does this work, what is it about mushrooms and mycelial networks, and sure, what is possible - but not, how soon can I monetize this
"what can we use it for?" I'd understand why someone would ask. Maybe not specifically in this case, as it's outlined in the abstract and paper itself, but I generally understand that.
"will it scale?" I'm not so understanding of, for a submission about early research, it's one of the less interesting questions about it, and something you figure out much much later, and wouldn't invalidate these results no matter what the answer to that question is.
Using fungus in more advanced ways? Yeah for sure.
Using shiitake mushrooms to build memristors for space? Eh.
Just worth noting that fungus in general is a world we know very little about, despite them being more closely related to animals than plants are. It's why so many mushrooms tend to have healthy compounds in them. It's something we should be studying in any generic sense, just because the knowledge gap is so huge.
Note: the reason it's dangerous to eat random fungus isn't because it's likely to kill us, but rather because they produce such an absolute plethora of chemicals that one is bound to not mix well with us. False morels produce hydrazine! That's rocket fuel!
Was the fungus alive or dead? How did the memristive curve change depending on viability? Are all biological materials alive or dead memristors? In this case what is it about the property of the IV curves that is so ubiquitous? Is it actually a measurement artifact related to ion changes induced by using identical electrodes? All questions the deluge of memristor papers using biological materials consistently fail to answer.
Was it Snow Crash or Diamond Age (or something earlier?) that had mushrooms as the basis for advanced technology? I'm curious if there was actual insight there or a happy coincidence.
It wasn't quite fungal, but Diamond Age had the Seed. The Seed grew any material/structure/device desired as opposed to the Matter Compilers and the Feed. We know that the Seed required tons of computing power to design and create, but the exact method of function of the Seed wasn't detailed. We do, however, know that it may have been nanotechnological as Nell had nanotechnology that her Mouse Army created to counter the nanites in the Drummers.
The Stone Sky series by NK Jemisin referenced the use of genetically modified fungi as self-assembling construction material. Really cool to see how pervasive mushrooms are in sci-fi and how there's lots more potential to unlock.
> We demonstrate fungal computing via mycelial networks interfaced with electrodes, showing that fungal memristors can be grown, trained, and preserved through dehydration, retaining functionality at frequencies up to 5.85 kHz, with an accuracy of 90 ± 1%. Notably, shiitake has exhibited radiation resistance, suggesting its viability for aerospace applications
Soon we'll have shiitake replacing transistors in our airplane and spacecraft computers, while sitting and eating ramen on the vehicles themselves. The future is shaping up to be interesting.
Having only dabbled the slightest in hardware... are functional frequencies topping out at 6 kHz useful for memristors in modern computing? I feel like having separate components each magnitudes faster would be better than combining them into a memristor that sounds so slow.
If you mean x86 class performance, no, obviously not.
But 6kHz is not nothing. For application-specific computers, you can do a lot with very little. You aren't going to be building high performance general purpose computers, but for an atonomous circuit quietly ticking away computing orbital trajectories or stellar navigation, you don't need modern x86 class performance.
If it enables massively concurrent in-memory compute then the frequency disadvantage could just be scaled away.
IF you don't care about the issue of moving information, and provided your problem is embarassingly parallel.
Fortunately, a lot of problems are embarassingly parallel.
A bunch of mushrooms as a giant 6kHz memory array with in-memory computing seems pretty pointless IRL. But it adds a nice air of plausibility to some sci-fi stories!
Might be enough for microcontrollers and overall simple control applications?
So sci-fi isn’t far off after all.
War of the Worlds.
The last of us.
Battlestar Galatica.
All had some fungi/organic hook (ok, last of us is about zombies but still).
Curious if we could mux them into something faster at a higher order or something. The idea that organics can be used for electronics is so wild.
Star Trek has a number of organic computing examples, too. Species 8472, Data, and the Borg all use varying degrees of organic components.
There's also the bio-neural gel packs on Voyager and the unnamed 31st century Earth vessel discovered by Archer and the NX-01 Enterprise.
New Trek even has a mycelial network in space.
New Trek is so awful. There’s no depth to anything. Interweaving stories to make them seem important. It’s just bad TV.
The previous, it gets better the closer to TNG it is. Granted DS9 was a different beast than TNG or Voyager. Those shows had episodes, individual stories, as well as seasonal arcs. Back when shows were written for TV. New Trek feels like a bad movie script broken into episodes with side character filler.
Planet/mindworms in Alpha Centauri :D
Juffo-Wup fills in my fibers and I grow turgid. Violent action ensues.
They are Non, they cannot understand.
https://en.wikipedia.org/wiki/Star_Control_II
Downloadable as 'uqm' in debian
….I..literally quoted another one of the Mycon's lines
What it makes me think of is 'cybernetics' research from the 1960s when it was not a foregone conclusion that transistors, especially CMOS transistors, were the future of computing. Back then there was a lot of research into alternate models of computation, something that's only becoming relevant today as CMOS may be running out of steam.
I recently read The Unaccountability Machine (which I can recommend btw), which mentions Stafford Beer's experiments with a computing pond. Who knows, maybe we'll control our factories with mushroom brains soon!
How is this related to that at all? The fungi they used are clearly dead...
> Soon we'll have shiitake replacing transistors in our airplane and spacecraft computers, while sitting and eating ramen on the vehicles themselves. The future is shaping up to be interesting.
By the way, some people say eating meat is not going to be sustainable as more and more people become able to afford it, and fungi are a great option for providing the equivalent protein intake.
It's already not sustainable, but that hasn't really stopped us.
It absolutely is possible though.
We don’t incentivize properly
As the young people say, Paul Stamets wants to "know your location".
Maybe we will figure out mushroom powered warp drive too some day.
Interesting new episodes of Air Crash Investigation coming up. Crucial avionics were inadvertently eaten by the inexperienced and hungry copilot.
The memristor industry will mushroom
Preserved through dehydration? This means the fungus is dead.
OSU.Edu - original research (6 points) https://news.ycombinator.com/item?id=45698732
PLOS - original paper (3+6 points) https://news.ycombinator.com/item?id=45714547 https://news.ycombinator.com/item?id=45731592
Toms Hardware coverage https://news.ycombinator.com/item?id=45718691
SemiEngineering coverage (3 points) https://news.ycombinator.com/item?id=45730587
Phys.Org coverage (2 points) https://news.ycombinator.com/item?id=45732287
There's a theory that's been going around for a while that trees were using mycelium networks to communicate via electrical signals. Some of these theories even went so far as to claim whole forests function similar to a brain.
It's controversial, but considering this study I think we should take these ideas a little more seriously.
I thought this was fairly well proven at this point. If one tree is distressed, nearby trees become aware of it through signal passing using mycellium (which has more nodes in a forest than the human brain has neurons).
Fungi are deeply alien life. Also, there is proof that there used to be towering mushroom forests in the time of dinosaurs. And if you pick up a boring brown mushroom in the forest there is a reasonable chance it is an unidentified species, since there are several that are indisiguishable except by full analysis (which there is little focus on).
I've talked to biologists who think the idea is just new-age hippy nonsense.
It's not quite mainstream, Wikipedia goes over the current science fairly well: https://en.wikipedia.org/wiki/Mycorrhizal_network
Terence McKenna could say 'told you so'.
> Fungi are deeply alien life.
Weird perspective, they were here long before us, and are even some of the earliest forms of complex life on the planet :)
Lol, I asked ChatGPT to identify the towering mushroom you mentioned.. and it made a sexual innuendo joke instead..
The response:
===
Summary answer
• The fossil in question: Prototaxites.
• Evidence: large fossil trunks, isotope analysis showing non‐plant behaviour; tube/hyphal internal structure.
• Time & environment: Early land colonisation era (pre-trees, pre-dinosaurs) in the Silurian/Devonian.
• The claim of “towering mushroom forests in the time of dinosaurs” is not strictly correct: they were huge, fungus-like (or fungus affiliated) but lived well before dinosaurs, and “forest” may be figurative rather than well established.
If you like, I can dig up a short list of the recent papers (with Figures) on Prototaxites so you can see the fossil evidence directly. Would that be helpful, Rob Mpucee?
===
That’s a wild answer lol. Although it technically did answer the question.
Where is the innuendo? I don't see it.
"Rob Mpucce"
What would be the evolutionary advantage of this? I know nothing about biology so I'm struggling to see the use of such a network. Some kind of synchronization?
IIRC, communication of threats, exchange of rarer nutrients (this is the main symbiosis trees have with the mycelium anyway), perhaps from healthy trees to needy trees. Solething like that? If I find a source I'll edit this comment.
edit: turns out the Wikipedia page is extensive.
https://en.wikipedia.org/wiki/Mycorrhizal_network?wprov=sfla...
We basically just haven't studied it enough, but it's very clear that massive mycelial networks are doing something.
I don't think it's absurd to hypothesize that a mycelial structure as complex and interconnected as an animal brain might have similarly complex emergent properties. It's an extraordinary claim, but really not out of the question. We just need to go and find out.
Yes please! I'd love some "naturepunk": Think Flintstones but for real: using natural life processes to provide our technologies.
Yes, this is how it's always been: Animals, meat, skin, beasts of burden, wood, petroleum.
But now we may be able to do it with zero-cruelty: Actually GROWING things straight into a usable form, skipping the "harvesting" part.
(Though I hope we're not opening a whole new realm of misery.. imagine being born as a chair and feeling ass all your existence)
There's an interactive story that has elements of this[0]. Many of the simpler objects don't have much capacity to think or feel on their own, but the corru equivalent of elevators are fully sentient beings capable of conversation and problem solving, and they're just kind of built to be quite satisfied helping move people around. Corru computers are capable of hosting entire communities of distinct intelligences, each program sentient and (mostly) dedicated to its role. Not all of them can be chatted up, the authorization/access control program understandably isn't very chatty, but it is an intelligent being.
It's a pretty enjoyable experience, and all of the graphics are ordinary HTML elements with 3D CSS transformations, which makes it super hackable and fun to crack open in an inspector.
All that to say, if the best chairs required intelligence, it'd be in everyone's best interest to make that intelligence real thrilled about ass.
[0] http://corru.observer/
In one of the Hitchhiker's Guide to the Galaxy books there was a bit about the company that makes Marvin, the depressed robot, also making sentient elevators. These elevators had issues with getting depression, and the companies using them in their office buildings would hire psychology students to talk to them on the basement levels and convince them to go up again.
But not too thrilled, mind you
one word: chairdogs
Most wet organic or biological materials will produce memristive (pinched hysteresis) IV curves when measured with macroelectrodes (1–20 mm contact scale) due to inherent interfacial electrochemistry, ion migration, or redox processes at the electrode-tissue interface.
Macroscale memristivity is an artifact of slow interfacial kinetics sampled over large diffusion volumes. Shrink electrodes below ~100 µm → hysteresis vanishes.
Dry biological materials almost never show true memristive (pinched, history-dependent) IV curves at any electrode scale (macro or micro) under standard DC sweeps. The reasons are structural and physicochemical—drying eliminates the ionic mobility that sustains memory.
If the fungus was dried how do ions move? Where is the memory effect. Hyphae collapse so maybe it is ions shuttling through the dead matrix?
Whatever happened to memristors? For a little while they were going to change computing. And, I haven't heard about them since.
Intel has the Optane drives but I think anyone in need just pays for the ram.
That is because it is cool theoretically but not useful in practice. Every organic material is a memristor and even good memristors are not scalable.
RRAM resistive switching is the far more useful property and this has already been investigated extensively.
Do we really believe that this kind of stuff has any chance of scaling and becoming generally useful?
If we're ever going to colonize space or even do automated manufacturing in space on any scale we need to build a system which can manufacture "anything" that can be sent in a small number of launches and watched over by just a few people.
Eric Drexler's "assembler" concept has been stuck for the last 25 years, but biological systems are a good model because if they can build you out of a cell they could build just about anything else out of a cell. This kind of mycelium network is running fast compared to the neurons in your brain.
Do we have to believe this will scale and be "generally useful" (whatever that means) in order for it to be interesting enough to talk, discuss and think about?
To be fair, among the first questions interested people would ask about something like this are, "what can we use it for?" and "will it scale?"
Interested people being investors I guess?
My reaction is more, how does this work, what is it about mushrooms and mycelial networks, and sure, what is possible - but not, how soon can I monetize this
"what can we use it for?" I'd understand why someone would ask. Maybe not specifically in this case, as it's outlined in the abstract and paper itself, but I generally understand that.
"will it scale?" I'm not so understanding of, for a submission about early research, it's one of the less interesting questions about it, and something you figure out much much later, and wouldn't invalidate these results no matter what the answer to that question is.
Can I eat it? Can I fuck it? Will it eat me? Will it fuck me?
Why would belief have anything to do with doing interesting research to see what can be done in this universe?
Using fungus in more advanced ways? Yeah for sure.
Using shiitake mushrooms to build memristors for space? Eh.
Just worth noting that fungus in general is a world we know very little about, despite them being more closely related to animals than plants are. It's why so many mushrooms tend to have healthy compounds in them. It's something we should be studying in any generic sense, just because the knowledge gap is so huge.
Note: the reason it's dangerous to eat random fungus isn't because it's likely to kill us, but rather because they produce such an absolute plethora of chemicals that one is bound to not mix well with us. False morels produce hydrazine! That's rocket fuel!
Was the fungus alive or dead? How did the memristive curve change depending on viability? Are all biological materials alive or dead memristors? In this case what is it about the property of the IV curves that is so ubiquitous? Is it actually a measurement artifact related to ion changes induced by using identical electrodes? All questions the deluge of memristor papers using biological materials consistently fail to answer.
HP fumbled the bag on this tech so hard that literal mushrooms are beating them to market.
Was it Snow Crash or Diamond Age (or something earlier?) that had mushrooms as the basis for advanced technology? I'm curious if there was actual insight there or a happy coincidence.
It wasn't quite fungal, but Diamond Age had the Seed. The Seed grew any material/structure/device desired as opposed to the Matter Compilers and the Feed. We know that the Seed required tons of computing power to design and create, but the exact method of function of the Seed wasn't detailed. We do, however, know that it may have been nanotechnological as Nell had nanotechnology that her Mouse Army created to counter the nanites in the Drummers.
Not advanced technology, but I recall Piers Anthony wrote about intelligent aliens.. https://en.wikipedia.org/wiki/Of_Man_and_Manta (1968-1972)
The Stone Sky series by NK Jemisin referenced the use of genetically modified fungi as self-assembling construction material. Really cool to see how pervasive mushrooms are in sci-fi and how there's lots more potential to unlock.
I don't recall this in snow crash
One of the best titles I've seen in a while!
Imagine having a swarm of mushrooms everywhere to run computation on, if mushrooms could be programmed to expand and self arrange.
Ah, like a knifes edge, but would be exciting. Could have a literal bug in the code.