We Induced Smells With Ultrasound
Mood
informative
Sentiment
neutral
Category
research
Key topics
Ultrasound
Olfactory
Neuroscience
Perception
Discussion Activity
Very active discussionFirst comment
1d
Peak period
88
Day 2
Avg / period
49.5
Based on 99 loaded comments
Key moments
- 01Story posted
Nov 21, 2025 at 3:02 PM EST
2d ago
Step 01 - 02First comment
Nov 22, 2025 at 7:20 PM EST
1d after posting
Step 02 - 03Peak activity
88 comments in Day 2
Hottest window of the conversation
Step 03 - 04Latest activity
Nov 23, 2025 at 11:39 PM EST
3h ago
Step 04
Generating AI Summary...
Analyzing up to 500 comments to identify key contributors and discussion patterns
Doing this via ultrasound might lower the barrier to treatment.
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?
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?
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.
is this like some second meaning or smth?
why is there a knife on the headset?
It's objectively cool, but very curious about the safety as well.
https://human.research.wvu.edu/fda-regulated-devices-used-in...
---
9. Does FDA require IRB review and approval of off-label use of a legally marketed device?
No, when a physician uses a legally marketed device outside its labeling to treat a patient and no research is being done, IRB review is not required. Note: Although not required by FDA, an IRB may still decide on its own initiative to review such use. Yes, when the off-label use of a legally marketed device is part of a research study collecting safety and effectiveness data involving human subjects, IRB review and approval is required (21 CFR 812.2(a)). For additional information on the off-label use of devices, see the FDA Information Sheet guidance, “ ‘Off-label’ and Investigational Use of Marketed Drugs, Biologics and Medical Devices.”
......
...OH you probably mean for the purposes of stimulating things OTHER THAN SMELLS
> "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.
What are the chances baby ultrasounds are doing this unintentionally?
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?
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.
Could I shoot you guys a message when I make my way down to Caltech to try this out someday? :)
If you get in contact with the researchers, please let us know how it went.
If they _lost_ their sense of smell, they had something to compare it with.
E.g. smelling something rancid for the first time - how strong of a negative reaction will it be compared to others who grew up with this negative association. Would I ever even be able to map most of the smells others have memorized to their origin? I really hope I get to find out someday.
Don't get me wrong, I would love this finding to be replicable, it would be pivotal as what other nerves could we stimulate to change perception (think pain, mental health issues, loss of senses).
Also, I wonder if this could take us closer to understand a little bit more of how the brain works. Like this could be a great way for normalizing 'inputs' and see how different brains react to it.
Very very exciting news, but I will hold on my hype until someone else can replicate this result.
These are exactly the types of smells people report when they get head CT scans (I've experienced it myself). Always thought it was ozone forming but perhaps it's more interesting than that.
This reminds me a bit of the escherian staircase video from 10+ years that went viral. A bunch of college students walking down the stairs, acting amazed when they found themselves back at the top. It was great acting and video editing, but it was fake and all part of, if I recall correctly, an art project.
I don't want to dismiss it outright either, seems cool as hell. But it's remarkable to me that all it takes is a blogpost to get this amount of uncritical acceptance of a demonstration.
I think this is cool, plausible and warrants investigation, but not suspension of disbelief. There needs to be a better way to go about this than responding "what about Galileo!?" to any principled application of critical thinking.
One has to set prejudice to the side and examine the claim being made to apply criticl thinking.
These are big scientific claims, but the work is clearly too premature to make those conclusions, and it lacks the connection to prior work and peer review needed for making priority claims. It's really great hacker-tinkering work though, and it could turn into solid science if they take more care with it.
If this effect is real and truly novel, my cynical expectation is that someone already established in focused ultrasound will read this, apply a more rigorous approach, and get the recognition that they are hoping for through more establish channels.
To be frank the reason that make me question it the most is how repetitive the redaction is. Seems LLM-like.
However, that's not a valid reason to discard an interesting result.
There's quite a lot of research in this direction (stimulation, be it ultrasound or otherwise) to tackle exactly things you mention. Not completely sure but probably stimulators to suppress epilepsy are the most common. It has been proven in animal research stimulating the right area induces visual stimuli - IIRC this has been tested and confirmed in humans as well, i.e., make people see again. And there's more going on.
In the end: everything in the brain is electrical current. Meaning that in theory stimulating the right bits can do pretty much everything.
It looks like independent hackers with a strong technical background and little regard for decorum.
Their methodology seems reasonable, and their results are plausible.
I’m reserved about the final part of the post where they moot about applications, but the core result seems solid.
1. https://en.wikipedia.org/wiki/Transcranial_focused_ultrasoun...
https://www.nhs.uk/pregnancy/your-pregnancy-care/ultrasound-...
> The scans are painless, have no known side effects on mothers or babies, and can be carried out at any stage of pregnancy.
If you read the linked article, you'd see that most of it focused on how extremely hard it was to get the ultrasound to do anything - it required an MRI and exact positioning of the ultrasound transducer. I doubt that 5 minutes of being gently prodded through the skin and fat is going to harm a child. Also, ultrasounds (and waves and radiation of all sorts) are passing into your body at all times, so it's not like they are exposing the fetus to something rare or unusual.
“Of particular interest in the present context are the observations made on patients whose middle ear had been opened in such a way that a cotton electrode soaked in normal saline solution could be placed near the cochlea. A total of 20 surgically operated ears were studied. Eleven patients heard pure tones whose pitch corresponded to the frequency of the sinusoidal voltage applied to the electrode… One patient reported gustatory sensations.”
Me wonders if this can be applied to other parts of the brain, perhaps recalling long buried memories? In my case, "a 12‑lexeme mnemonic constellation that operates as a cognitive entropy incantation, each syllabic particle mapping onto a quantized shard of an authorization singularity’s randomness reservoir. This ordered cascade of linguistic quanta serves as a deterministic bootstrap constant, re‑materializing access to a distributed transactional continuum avatar via recursive derivation algorithms. In practice, it’s a compact neuro‑linguistic checksum spell, a bridge where human cortex patterns resonate with machine‑level information‑integrity archetypes, conjuring identity from chaos like a linguistic particle accelerator, aka ₿" ;)
Translation: We’re very concerned that the only projects getting funding right now have to use AI.
[1] https://en.wikipedia.org/wiki/Long-term_nuclear_waste_warnin...
Our fingertips feel using low frequency sound generated by our fingerprints passing over things.
This kind of tech should be developed as open-source projects, even for the firmware and hardware. A sufficiently advanced version of this, if widely deployed as proprietary blackboxes like smartphones are, would allow one consciousness to take over multiple bodies without their original owners knowing.
Want the full context?
Jump to the original sources
Read the primary article or dive into the live Hacker News thread when you're ready.