This is exceptionally cool!
It looks like this post isn’t getting much love though. I’ll see if I can get this post added to the second chance pool[1] and get it added to the front page!
All: if you seen a particularly good submission that has fallen through the cracks, please email hn@ycombinator.com so we can take a look and maybe put it in the second-chance pool (https://news.ycombinator.com/pool, explained at https://news.ycombinator.com/item?id=26998308), so it will get a random re-upment on HN's front page.
(Yes, it's permissible to request this for your own stuff, but we like it better when it's something you just ran across randomly and realized was interesting—as mmulet did in this case!)
I'd maybe make a hypothesis that a large portion of the space is "bad" smelling stuff: smoke or garbage. When people had covid-induced parosmia, it almost always seemed to be bad smelling stuff.
Super interesting! That would make sense, because a lot of the nose is presumably dedicated to smelling evolutionarily-relevant smells, most of which are "smells bad, avoid this". The method is very crude right now, but maybe with more fine-grained targetting we could better tune the smell profile.
Still have it, intermittently. A sort of nameless-but-familiar "chemical" smell that comes and goes, along with any sense of taste. That is, I have bad days with no taste, just a chemical smell. Other days I have a pretty good sense of smell, generally with a good sense of taste.
Intriguingly some of the really unpleasant smells never get through to me - I could probably work at a sewage works now. Worryingly I have next to no ability to smell burning, though I do now get the smell of natural gas (or the additive used to make it smell).
There has been promising work on olfactory training, which you can do very inexpensively at home. If you can, I would consider seeing any ENT first to rule out polyps, etc.
This is something I'm still testing, so take it with a bucket of salt, but I've found that exposing myself to very strong samples of things that I was unable to smell made something click again and I started to smell them again. Seems like something in there needs to be retrained to odors.
That's the basic method of retraining. I've got a bunch of essential oils in tiny jars and I regularly take a 20 second sniff of each whilst thinking strongly about the smell in context. For example, when I smell the lavender oil I recall the garden at my grandma's house which obviously was full of lavender. It's definitely helping, but there are still a lot of gaps.
I'm convinced that over the decades we'll continue to be a little surprised but just how much of our body-machinery is doing jobs of self-calibration, regulation, and safety-interlocks.
That was exactly my thought when reading the article and my personal experience with Covid. For a couple weeks, I perceived a persistent smell of something burning.
A smell of burning was how I suspected I had COVID the last time around. I was around some machines, I had to have someone else sniff around and let me know that nothing was actually burning.
Certainly! We didn't get a chance to test it on more people before we had to take it apart, but we thought the result was too cool to share. Would love to see other folks run with the idea!
It’s just a blog post. No academic is going to read it as more than a very promising early result.
The issue is that lay people read every paper or post as if it were a final proclamation. They’re not. Even a peer reviewed paper on the cover of Science or Nature is still not “proof” of anything, science doesn’t produce positive confirmation. It produces evidence that taken together suggest one prior is more likely than another.
Bayes Rule is very intuitive. We update the prior by the likelihood of the evidence under a given prior divided by the likelihood of the evidence. That’s all it is.
Unfortunately, there is a very strong motive to flag plant. Academia is a water full of sharks.
Oh god, do we really want to have the smells of sex when watching porn?
On the other hand, I see an opportunity even without tech: porn star perfume collabs: Spray some Gukki Bloom and press play on that video to smell what the star was wearing on the day.
But I guess high-end perfume brands don't want to be associated with actors of the flesh.
You're getting downvoted, likely for prudish reasons, but in all seriousness it doesn't seem unlikely that you're right.
The porn industry has historically been very quick to adopt new technologies, it's easy to see how this could benefit that industry, so it's a logical enough conclusion to draw. They'll very likely be the first commercial application of this, once viable.
The scene from Torrente comes to mind where the protagonist gives a set of anal beads to a gay guy who then identifies it as ar*e smell, hence it was used. Or smth like that :D.
Woah I didn't know about that theory, that's really interesting! If I understand correctly, it's that the ligand needs to both fit "physically" and also have the right vibrational mode to have high binding affinity / trigger the receptor? Sounds like the relevant frequencies would be in IR range, or roughly 10-100 terahertz. We're at 300 kHz, so 9 orders of magnitude lower, so we're likely not activating the receptors directly with that mechanism. But, maybe the acoustic radiation force from the ultrasound gives existing molecules in the air enough energy to increase the coupling? And nobody seems to really know how ultrasound neurostimulation really works, so who knows—maybe something similar even happens with neurotransmitters in cortex...
Well, if it’s waves, perhaps principles of consonance and dissonance might apply.
Robert Hooke thought so…
“Now as we find that musical strings will be moved by Unisons and Eighths, and other harmonious chords, though not in the same degree; so do I suppose that the particles of matter will be moved principally by such motions as are Unisons, as I may call them, or of equal Velocity with their motions, and by other har∣monious motions in a less degree.
I do further suppose, A subtil matter that incom∣passeth and pervades all other bodies, which is the Menstruum in which they swim which maintains and continues all such bodies in their motion, and which is the medium that conveys all Homogenious or Har∣monical motions from body to body.
Further I suppose, that all such particles of matter as are of a like nature, when not separated by others of a differing nature will remain together, and strengthen the common Vibration of them all against the differing Vibrations of the ambient bodies.
According to this Notion I suppose the whole Universe and all the particles thereof to be in a con∣tinued motion, and every one to take its share of space or room in the same, according to the bulk of its body, or according to the particular power it hath to receive, and continue this or that peculiar motion.
Two or more of these particles joyned immediately together, and coalescing into one become of another nature, and receptive of another degree of motion and Vibration, and make a compounded particle differing in nature from each of the other par∣ticles.
All bulky and sensible bodies whatsoever I suppose to be made up or composed of such particles which have their peculiar and appropriate motions which are kept together by the differing or dissonant Vibrations of the ambient bodies or fluid“
I cant wait for the day when the perfume and food shops in the mall use this for truly targeted advertisement. Cue rise of ultrasound-proof hats and lawsuits by people who report feeling sick due to such ads.
That's a really interesting concept! The wiki page right now ends in "example of a combination of smells that neutralise each other", which makes me wonder if the "olfactory white" combinations are actually tuned to neutralize? I suspect what we're hitting is a bunch of receptors, and the brain is interpretting it as a common strong and evolutionarily important smell (which I assume has stronger pathways by default).
I was an organic chemist, and as such worked I've worked in various "wet" laboratories. All of them had store rooms and cabinets with hundreds or thousands of bottles filled with horribly smelling goop. Besides the occasional terpenoid (naturally occurring things smelling like menthol, cloves or cinnamon) nearly everything there was liquid death.
These smells have everything: Harsh solvent-like stuff like strong alcohol or glue, rotten fish amines, off-sweet halocarbons, things like burnt plastics, excrement, or stuff that defies description as to their lingeringly terrible sensation of olfactory wrongness (selenium compounds).
There is actually a thing called "cadaverine", that should tell you enough.
Still, every sufficiently large storage space I rememebet had this identical, not unpleasant, thickly sweet, but not easily defined smell.
So to conclude, I think it's a brain glitch when we input everything, all the smells, at the same time.
I had the same thought - I guess it's similar to that idea that if you had someone else's eyes, you might not perceive specific colours to be the same?
But actually it sort of makes sense since (from what I understand) is stimulating an external interface (the receptors), so you're mimicing what the effect a smell would have on you rather than the electrical signal created by the response to a stimulus?
Sorry it's unclear in the post, they weren't exactly the same! The numbers reported were on Lev, and we swept them around that range for me (Albert). But we didn't take down the exact values, so unfortunately I don't know how similar the maps were. iirc they were pretty different.
Woah, that would be wild! It seems like most neonatal ultrasound reaches peak internal pressures of few-hundred kPa to 2 MPa. We ranged from 150-250 kPa. So, a little lower than the lower end of prenatal diagnostic imaging.
So, the pressures are high enough to be stimulating them! But most diagnostic imaging happens at 1-20 MHz, while most neurostimulation seems to occur at few-hundred kHz (we were at 300 kHz, on the mid-high end). So I don't think it's likely that babies are being sent smells?
Haha, luckily not! It's a very speculative link, so we didn't want to talk about "AI" too much in the main post. But we originally got interested in this concept because we are interested in other forms of input to the brain (other than the classic reading, listening, watching, etc). The nose is interesting because it seems to have many independent basis vectors and very sharp discrimination ability, so it might be a sensor into which you can pack many inputs. LLMs are just a proof-by-example that ~1k input dimensions is enough to really encode semantic meaning.
Totally! We think this is because the brain is hard-wired evolutionarily to interpret smells by danger level first. So maybe there's just more "bad smell" receptors, or maybe the brain treats unknown smells as "uh oh, danger". Lots of cool stuff to test!
The adult videos industry must be already closely looking at this, and I wouldn't be surprised if they don't finance related research soon in the future, it will be VHS vs betamax all over again.
Yes, to simulate pheromones and related stimulus, that too, but I'm sorry if you already don't know this but a faint smell of urine would be a big deal to a non-trivial amount of men for immersion, without going into too much detail "squirting" fans and all the ecosystem around such kind of fetishes.
My prediction is that in the not-to-distant future, we’re all going to live indefinitely in simulations that optimized for human experience. To do this, AIs will “highjack” our nervous systems and feed generated worlds to use to experience. This kind of thing makes it seem like it’s pretty realistic.
If they’re using enough ultrasound energy to create a physical reaction inside the subjects head strong enough to smell like a burning camp fire, I can’t imagine they’d survive long enough to report it. Maybe I’m misunderstanding your implication?
maybe the ionized oxygen sensation, the ozone, the garbage and smoke are like a thin film of garbage and smoke particles in the nasal cavity experiencing mechanical stimulation. the mr guidance seems like a good idea, but the actual mount and targeting sounds crude, what are the odds it's just in the nasal cavity warming up and activating the gross stuff that lives in there?
i wonder if some kind of inhalable anesthetic would be a good control. ie, if the normal sensory pathways are blocked and the lifu stim of the olfactory bulb still creates the percept, then maybe it would be evidence that it is working as it appears...
OK, I want to meet these guys. This writeup has several breathtaking (if you will) passages. Like:
> "We found different scents by steering the beam over ~14 mm (20 degrees at 4 cm radius). The distance between freshness and burning was ~3.5 mm."
> "The olfactory system potentially allows writing up to 400, if not 800 due to two nostrils, dimensions into the brain. That is comparable to the dimensionality of latent spaces of LLMs, which implies you could reasonably encode the meaning of a paragraph into a 400-dimensional vector. If you had a device which allows for this kind of writing, you could learn to associate the input patterns with their corresponding meanings. After that, you could directly smell the latent space."
This just makes me grin with total delight. Completely freaking fascinating.
This is absolutely my question as well - curious if it's legal to do this, I'm guessing yes as it's an existing ultrasound device? But is there possibility of permanent damage?
It's objectively cool, but very curious about the safety as well.
This is the coolest part! Turns out, the powers you need are actually lower than what is used for imaging babies :) We measured our probe with a hydrophone on a computer-controlled scanner to get the pressure field, and made sure that it's below diagnostic levels (the generally accepted mechanical index limit is 1.9 and ours was 0.4 peak). We also made sure to avoid the eyes and keep thermals in check.
> The olfactory system potentially allows writing up to 400, if not 800 due to two nostrils, dimensions into the brain. That is comparable to the dimensionality of latent spaces of LLMs, which implies you could reasonably encode the meaning of a paragraph into a 400-dimensional vector. If you had a device which allows for this kind of writing, you could learn to associate the input patterns with their corresponding meanings. After that, you could directly smell the latent space. A bit of ultrasound, a breath in - and you understood a paragraph.
Translation: We’re very concerned that the only projects getting funding right now have to use AI.
This is exceptionally cool! It looks like this post isn’t getting much love though. I’ll see if I can get this post added to the second chance pool[1] and get it added to the front page!
[1]https://news.ycombinator.com/item?id=26998308
Added! Thanks for requesting this.
All: if you seen a particularly good submission that has fallen through the cracks, please email hn@ycombinator.com so we can take a look and maybe put it in the second-chance pool (https://news.ycombinator.com/pool, explained at https://news.ycombinator.com/item?id=26998308), so it will get a random re-upment on HN's front page.
(Yes, it's permissible to request this for your own stuff, but we like it better when it's something you just ran across randomly and realized was interesting—as mmulet did in this case!)
Wow good to know, thank you guys!
Finally some progress towards smellovision.
Not joking, my mom has asked me every year since I became a "tech person" when I am going to make smellovision. This is perfect for her.
She should be careful what she wishes for.
A Brief History of Smell-O-Vision
https://www.wired.com/2006/12/a-brief-history-2-2/
Apple has been waiting for the opportunity to use their next moniker - iNose.
You can Apple Pay through the Apple Nose
Don't do it too much though, otherwise you're gonna need the Apple Noose, designed in collaboration with Coco Chanel.
Soon to be followed by Google Knows
iNose what you did there.
iNose iSmell, iBath
https://futurama.fandom.com/wiki/Smell-O-Scope
https://youtu.be/1zp3b6YCXqI
The smelloscope, right ?
Imagine the viruses and pranks.
Kickstarter incoming for a device that induces code smells in sync with review.
If we have negative code smells can we also make good code smell like cherry blossoms?
Vibe coding is so over
I'd maybe make a hypothesis that a large portion of the space is "bad" smelling stuff: smoke or garbage. When people had covid-induced parosmia, it almost always seemed to be bad smelling stuff.
Super interesting! That would make sense, because a lot of the nose is presumably dedicated to smelling evolutionarily-relevant smells, most of which are "smells bad, avoid this". The method is very crude right now, but maybe with more fine-grained targetting we could better tune the smell profile.
I know someone whose sense of taste was ruined by a small stroke. He said basically everything tastes like old gym socks now. That would suck so bad.
Still have it, intermittently. A sort of nameless-but-familiar "chemical" smell that comes and goes, along with any sense of taste. That is, I have bad days with no taste, just a chemical smell. Other days I have a pretty good sense of smell, generally with a good sense of taste.
Intriguingly some of the really unpleasant smells never get through to me - I could probably work at a sewage works now. Worryingly I have next to no ability to smell burning, though I do now get the smell of natural gas (or the additive used to make it smell).
There has been promising work on olfactory training, which you can do very inexpensively at home. If you can, I would consider seeing any ENT first to rule out polyps, etc.
Thanks for the info. I'm on top of it (in the ways you described) but still appreciate it and maybe someone else will see your comment.
Good luck!
This is something I'm still testing, so take it with a bucket of salt, but I've found that exposing myself to very strong samples of things that I was unable to smell made something click again and I started to smell them again. Seems like something in there needs to be retrained to odors.
That's the basic method of retraining. I've got a bunch of essential oils in tiny jars and I regularly take a 20 second sniff of each whilst thinking strongly about the smell in context. For example, when I smell the lavender oil I recall the garden at my grandma's house which obviously was full of lavender. It's definitely helping, but there are still a lot of gaps.
I'm convinced that over the decades we'll continue to be a little surprised but just how much of our body-machinery is doing jobs of self-calibration, regulation, and safety-interlocks.
That was exactly my thought when reading the article and my personal experience with Covid. For a couple weeks, I perceived a persistent smell of something burning.
A smell of burning was how I suspected I had COVID the last time around. I was around some machines, I had to have someone else sniff around and let me know that nothing was actually burning.
> Different focal spots corresponded to different smells, which we’ve replicated first-try on two people and validated with a blind trial.
So, N=2 and the people in question are co-authors. I'm not in this business, but isn't this too... early to publish?
Certainly! We didn't get a chance to test it on more people before we had to take it apart, but we thought the result was too cool to share. Would love to see other folks run with the idea!
It’s just a blog post. No academic is going to read it as more than a very promising early result.
The issue is that lay people read every paper or post as if it were a final proclamation. They’re not. Even a peer reviewed paper on the cover of Science or Nature is still not “proof” of anything, science doesn’t produce positive confirmation. It produces evidence that taken together suggest one prior is more likely than another.
Bayes Rule is very intuitive. We update the prior by the likelihood of the evidence under a given prior divided by the likelihood of the evidence. That’s all it is.
Unfortunately, there is a very strong motive to flag plant. Academia is a water full of sharks.
I predict a future where once again porn is the cutting edge with a cutting edge technology.
porn + vr + smell
Oh god, do we really want to have the smells of sex when watching porn?
On the other hand, I see an opportunity even without tech: porn star perfume collabs: Spray some Gukki Bloom and press play on that video to smell what the star was wearing on the day.
But I guess high-end perfume brands don't want to be associated with actors of the flesh.
No, you want the smell of your ideal partner.
You're getting downvoted, likely for prudish reasons, but in all seriousness it doesn't seem unlikely that you're right.
The porn industry has historically been very quick to adopt new technologies, it's easy to see how this could benefit that industry, so it's a logical enough conclusion to draw. They'll very likely be the first commercial application of this, once viable.
I assume the reasons are more visceral than prudish. I don't think I would want to know what a porno set would smell like.
This is coming from a place of physical discontent, not moral discontent.
It doesnt have to smell like the porno set. They could choose any smell that stimulates arousal.
For some reason this sent my train of thought towards the "people who think their shit doesn't stink" expression.
Is it still shitposting if it actually smells nice?
Movies don't sound like their sets during filming either.
The scene from Torrente comes to mind where the protagonist gives a set of anal beads to a gay guy who then identifies it as ar*e smell, hence it was used. Or smth like that :D.
I wouldn’t put too much stock in fake Internet points.
Prediction: Pi-hole gets a double entendre and becomes more useful than ever.
But will the smell be in 3D?
Just light a candle.
[flagged]
You made this account to reply to my joke comment? Haha weird
The year is 2032. My smart fridge started A/B testing scents to reduce snacking. I ate a carrot and felt promoted.
Reminds me of the vibration theory of olfaction.
https://en.wikipedia.org/wiki/Vibration_theory_of_olfaction
Woah I didn't know about that theory, that's really interesting! If I understand correctly, it's that the ligand needs to both fit "physically" and also have the right vibrational mode to have high binding affinity / trigger the receptor? Sounds like the relevant frequencies would be in IR range, or roughly 10-100 terahertz. We're at 300 kHz, so 9 orders of magnitude lower, so we're likely not activating the receptors directly with that mechanism. But, maybe the acoustic radiation force from the ultrasound gives existing molecules in the air enough energy to increase the coupling? And nobody seems to really know how ultrasound neurostimulation really works, so who knows—maybe something similar even happens with neurotransmitters in cortex...
Does everything come down to waves or to bits?
Well, if it’s waves, perhaps principles of consonance and dissonance might apply.
Robert Hooke thought so…
“Now as we find that musical strings will be moved by Unisons and Eighths, and other harmonious chords, though not in the same degree; so do I suppose that the particles of matter will be moved principally by such motions as are Unisons, as I may call them, or of equal Velocity with their motions, and by other har∣monious motions in a less degree.
I do further suppose, A subtil matter that incom∣passeth and pervades all other bodies, which is the Menstruum in which they swim which maintains and continues all such bodies in their motion, and which is the medium that conveys all Homogenious or Har∣monical motions from body to body.
Further I suppose, that all such particles of matter as are of a like nature, when not separated by others of a differing nature will remain together, and strengthen the common Vibration of them all against the differing Vibrations of the ambient bodies.
According to this Notion I suppose the whole Universe and all the particles thereof to be in a con∣tinued motion, and every one to take its share of space or room in the same, according to the bulk of its body, or according to the particular power it hath to receive, and continue this or that peculiar motion.
Two or more of these particles joyned immediately together, and coalescing into one become of another nature, and receptive of another degree of motion and Vibration, and make a compounded particle differing in nature from each of the other par∣ticles.
All bulky and sensible bodies whatsoever I suppose to be made up or composed of such particles which have their peculiar and appropriate motions which are kept together by the differing or dissonant Vibrations of the ambient bodies or fluid“
Page 9: https://quod.lib.umich.edu/e/eebo/A44322.0001.001/1:3?rgn=di...
Reminds me of the vibration theory of simulated chemicals. 3 Fisted Tales of Bob. They used sound to keep gunpowder from exploding or something.
I cant wait for the day when the perfume and food shops in the mall use this for truly targeted advertisement. Cue rise of ultrasound-proof hats and lawsuits by people who report feeling sick due to such ads.
I wonder if this makes you smell "laurax" or "olfactory white" if it glitches and triggers every receptor site simultaneously
https://en.wikipedia.org/wiki/Olfactory_white
That's a really interesting concept! The wiki page right now ends in "example of a combination of smells that neutralise each other", which makes me wonder if the "olfactory white" combinations are actually tuned to neutralize? I suspect what we're hitting is a bunch of receptors, and the brain is interpretting it as a common strong and evolutionarily important smell (which I assume has stronger pathways by default).
I was an organic chemist, and as such worked I've worked in various "wet" laboratories. All of them had store rooms and cabinets with hundreds or thousands of bottles filled with horribly smelling goop. Besides the occasional terpenoid (naturally occurring things smelling like menthol, cloves or cinnamon) nearly everything there was liquid death.
These smells have everything: Harsh solvent-like stuff like strong alcohol or glue, rotten fish amines, off-sweet halocarbons, things like burnt plastics, excrement, or stuff that defies description as to their lingeringly terrible sensation of olfactory wrongness (selenium compounds).
There is actually a thing called "cadaverine", that should tell you enough.
Still, every sufficiently large storage space I rememebet had this identical, not unpleasant, thickly sweet, but not easily defined smell.
So to conclude, I think it's a brain glitch when we input everything, all the smells, at the same time.
Maybe it's just because I can't smell anything, but I have an overwhelming hatred for this concept.
I find it incredible that the same olfactory activation patterns mapped to the the SAME smells in both subjects.
I had the same thought - I guess it's similar to that idea that if you had someone else's eyes, you might not perceive specific colours to be the same?
But actually it sort of makes sense since (from what I understand) is stimulating an external interface (the receptors), so you're mimicing what the effect a smell would have on you rather than the electrical signal created by the response to a stimulus?
Sorry it's unclear in the post, they weren't exactly the same! The numbers reported were on Lev, and we swept them around that range for me (Albert). But we didn't take down the exact values, so unfortunately I don't know how similar the maps were. iirc they were pretty different.
Maybe it's the resonance frequency of the sensor molecule?
Burning Smell or Trash doesn't give me a lot of confidence. Like pushing your thumbs into your eyes can make 'color' appear.
Those few smells of extremes (garbage /clean air) make me think they are saturating the sensors.
I'm thinking it could be that we are very attuned to smelling bad smells because it's for safety.
Totally agree with both points! I would love to see what happens with more fine-grained control of the ultrasound.
The real question is that if you make it smell like garbage, will flys show up? Like is this a human experience, or a universal one?
How do the power levels here compare with say a baby ultrasound?
What are the chances baby ultrasounds are doing this unintentionally?
Woah, that would be wild! It seems like most neonatal ultrasound reaches peak internal pressures of few-hundred kPa to 2 MPa. We ranged from 150-250 kPa. So, a little lower than the lower end of prenatal diagnostic imaging.
So, the pressures are high enough to be stimulating them! But most diagnostic imaging happens at 1-20 MHz, while most neurostimulation seems to occur at few-hundred kHz (we were at 300 kHz, on the mid-high end). So I don't think it's likely that babies are being sent smells?
The angle and position of the transducer would make them leveragable by future VR headsets.
After we have Smell-O-Vision[0] we should work on the next big step for the internet:
<[SA]HatfulOfHollow> i'm going to become rich and famous after i invent a device that allows you to stab people in the face over the internet
[0]: https://en.wikipedia.org/wiki/Smell-O-Vision
If they achieve their goals; I don't know of any tech company I would trust with a direct write access to my brain.
I shudder to think what capability the execute bit would grant.
That doesn't surprise me.
Our fingertips feel using low frequency sound generated by our fingerprints passing over things.
Very cool, although I found the link to LLMs toward the end to be a little odd.
Definitely, it’s a “we know the only thing anyone cares about right now is AI, so, look this is kind of like AI, sort of, if you squint”
Haha, luckily not! It's a very speculative link, so we didn't want to talk about "AI" too much in the main post. But we originally got interested in this concept because we are interested in other forms of input to the brain (other than the classic reading, listening, watching, etc). The nose is interesting because it seems to have many independent basis vectors and very sharp discrimination ability, so it might be a sensor into which you can pack many inputs. LLMs are just a proof-by-example that ~1k input dimensions is enough to really encode semantic meaning.
So they can already implement the smell of files restored from the Recycle Bin
Interesting that the smells they were able to trigger seem to be related to basic survival. Smoke bad. Rotting food bad. Fresh air good.
Totally! We think this is because the brain is hard-wired evolutionarily to interpret smells by danger level first. So maybe there's just more "bad smell" receptors, or maybe the brain treats unknown smells as "uh oh, danger". Lots of cool stuff to test!
> At the time, all of our headsets had a knife taped to the probe
is this like some second meaning or smth?
why is there a knife on the headset?
Ha, nope. We needed something to stabilize the probe and the plastic knife from lunch was within reach :)
In case the smells make you go crazy, a failsafe system I assume.
The adult videos industry must be already closely looking at this, and I wouldn't be surprised if they don't finance related research soon in the future, it will be VHS vs betamax all over again.
...
......
...OH you probably mean for the purposes of stimulating things OTHER THAN SMELLS
Yes, to simulate pheromones and related stimulus, that too, but I'm sorry if you already don't know this but a faint smell of urine would be a big deal to a non-trivial amount of men for immersion, without going into too much detail "squirting" fans and all the ecosystem around such kind of fetishes.
Can a machine detect smell? That bit is left unattainable.
This experiment smells like weed
That anyone would direct a device towards their brain, intending the device to cause a physical impact on it, is amazing to me.
My prediction is that in the not-to-distant future, we’re all going to live indefinitely in simulations that optimized for human experience. To do this, AIs will “highjack” our nervous systems and feed generated worlds to use to experience. This kind of thing makes it seem like it’s pretty realistic.
I think a lot of people are already living in this future, trapped in a fantasy bubble world maintained by social media algorithms.
How do you know it hasn't already happened?
that's an interesting mix of smells. i can't help but wonder if it's resulting from stimulation or the sensing of byproducts of the process itself.
If they’re using enough ultrasound energy to create a physical reaction inside the subjects head strong enough to smell like a burning camp fire, I can’t imagine they’d survive long enough to report it. Maybe I’m misunderstanding your implication?
maybe the ionized oxygen sensation, the ozone, the garbage and smoke are like a thin film of garbage and smoke particles in the nasal cavity experiencing mechanical stimulation. the mr guidance seems like a good idea, but the actual mount and targeting sounds crude, what are the odds it's just in the nasal cavity warming up and activating the gross stuff that lives in there?
i wonder if some kind of inhalable anesthetic would be a good control. ie, if the normal sensory pathways are blocked and the lifu stim of the olfactory bulb still creates the percept, then maybe it would be evidence that it is working as it appears...
Great idea for a control! Will have to try it if we set this up again..
This is so incredibly cool. Will a non-contact version be possible?
A scary concept... I think the hardest part would be coupling the ultrasound through the air. But there are probably solutions..
I wonder if — within the decade — “cheap porn smell” will be a recognisable thing
Whomever smelt it dealt it.
A properly bizarre and interesting blogpost. Wow.
What kind of probe are they using?
OK, I want to meet these guys. This writeup has several breathtaking (if you will) passages. Like:
> "We found different scents by steering the beam over ~14 mm (20 degrees at 4 cm radius). The distance between freshness and burning was ~3.5 mm."
> "The olfactory system potentially allows writing up to 400, if not 800 due to two nostrils, dimensions into the brain. That is comparable to the dimensionality of latent spaces of LLMs, which implies you could reasonably encode the meaning of a paragraph into a 400-dimensional vector. If you had a device which allows for this kind of writing, you could learn to associate the input patterns with their corresponding meanings. After that, you could directly smell the latent space."
This just makes me grin with total delight. Completely freaking fascinating.
<3 email me or dm us on twitter! links on the post.
That is an amazing thought!
What about all the other senses?
Lev also found vestibular! Email or dm him on twitter :)
...is this safe?
This is absolutely my question as well - curious if it's legal to do this, I'm guessing yes as it's an existing ultrasound device? But is there possibility of permanent damage?
It's objectively cool, but very curious about the safety as well.
This is the coolest part! Turns out, the powers you need are actually lower than what is used for imaging babies :) We measured our probe with a hydrophone on a computer-controlled scanner to get the pressure field, and made sure that it's below diagnostic levels (the generally accepted mechanical index limit is 1.9 and ours was 0.4 peak). We also made sure to avoid the eyes and keep thermals in check.
I want it the other way - I want google "search this smell" feature..
Ah reverse smell search? You'd need to upload the smell.
We are witnessing the dawn of smell-o-vision teledildonic VR tentacle porn
[flagged]
Definitely not my fantasy. Just a funny image of the future.
> The olfactory system potentially allows writing up to 400, if not 800 due to two nostrils, dimensions into the brain. That is comparable to the dimensionality of latent spaces of LLMs, which implies you could reasonably encode the meaning of a paragraph into a 400-dimensional vector. If you had a device which allows for this kind of writing, you could learn to associate the input patterns with their corresponding meanings. After that, you could directly smell the latent space. A bit of ultrasound, a breath in - and you understood a paragraph.
Translation: We’re very concerned that the only projects getting funding right now have to use AI.
[dead]