Brain boffins think they've found the data format we use to store images as memories
- Reference: 1704220687
- News link: https://www.theregister.co.uk/2024/01/02/dartmouth_visua_brain_memory/
- Source link:
The long-held belief that our grey matter turns visual signals into a non-visual form of data to encode as memories is likely false, according to [1]a paper by folks at Dartmouth College in the US and published today in Nature Neuroscience. Rather, it seems more like "retinotopic" information, 3D visual information encoded by the brain via the retina, is the base coding mechanism for memories related to visual stimuli.
Think of it as our brains storing actual visual info in memory rather than descriptions that we use to reconstruct a scene on demand. A picture on the wall, not lines of text. Is this a study from the university's Dept of the Bleeding Obvious? You might argue so, or you might argue it's worthwhile research, but in any case that's science: putting aside assumptions, figuring it out for good, and writing it down.
[2]
The academics arrived at their conclusions through a trio of functional magnetic resonance imaging (fMRI) experiments in which they scanned the brains of people looking at and recalling specific images. They observed a direct inverse relationship between sensing and memory portions of the brain and where they light up when making or recalling a memory.
[3]
[4]
"We think that the retinotopic code is used as a means of reading (and writing) visual information into (and out of) memory systems," Adam Steel, a postdoc at Dartmouth's department of psychological and brain sciences and co-author of the work, told The Register .
In addition to the correlation between areas of brain activity, Steel explained there's an "opponent suppression" dynamic, in which the two sensing and memory areas of the brain show lower activity when the other is being used. This, he said, suggests a central role of retinotopic code in translating between neural systems in the brain.
[5]
When asked to recall an image, activity in mnemonic areas of the brain showed an inverted spike in visually-evoked "population reception fields" that corresponded to the original sensing areas of the brain, suggesting that "retinotopic coding could serve as a shared substrate to scaffold the interaction between perceptual and mnemonic systems," the report posits.
[6]Idea of downloading memories far-fetched say experts after Musk claim resurfaces in latest Neuralink development
[7]Boffins harnessed the brain power of mice to build AI models that can't be fooled
[8]MIT goes down to the wire: Brain cable pipes electricity, chemicals, light straight into minds
[9]The Great Brain Scan Scandal: It isn't just boffins who should be ashamed
Of course, that doesn't mean visuals are everything when it comes to memory - semantic and language encoding is done alongside retinotopic data when the brain encodes memory, Steel explained.
Of course, this can't work in every case given capability and cognitive differences. "[Retinotopic coding] can't be the only system that is used - congenitally blind individuals also have rich memory experiences despite having no visual experience," Steel said.
Differences may also exist in the brains of people with photographic memories, those with autism, and it's likely image processing capability change as we age as well, Steel said.
"The mnemonic areas where we see negative visual coding are well aligned with areas impacted by aging and Alzheimer's disease," Steel said, suggesting the discovery, which the team described in their paper as presenting a challenge "to conventional views of brain organization," could someday be used to treat memory diseases, too. ®
Get our [10]Tech Resources
[1] https://www.nature.com/articles/s41593-023-01512-3
[2] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offbeat/science&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2ZZSVmSk7BS90zuXJZOLy-AAAAQM&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offbeat/science&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44ZZSVmSk7BS90zuXJZOLy-AAAAQM&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offbeat/science&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33ZZSVmSk7BS90zuXJZOLy-AAAAQM&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[5] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_offbeat/science&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44ZZSVmSk7BS90zuXJZOLy-AAAAQM&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[6] https://www.theregister.com/2022/01/31/neuralink_job_ad/
[7] https://www.theregister.com/2019/11/15/mouse_brain_power/
[8] https://www.theregister.com/2017/02/23/polymer_highway_promises_easier_neurological_monitoring/
[9] https://www.theregister.com/2016/07/07/the_great_brain_scan_scandal_it_isnt_just_boffins_who_should_be_ashamed/
[10] https://whitepapers.theregister.com/
Re: Ermm
There's an Open Access (early version?) of their paper at [1]bioRxiv which helps to explain this a bit, with images. In the intro, they write:
" [brain models] assume that retinotopic coding is replaced by abstract [...] coding as information propagates [...] towards memory "
raising the puzzling question:
" how can mnemonic and visual information interact [...] if they are represented using fundamentally different neural codes? "
And so they go on to investigate that with fMRI on 17 adults, through population Receptive Field Mapping (pRF), Occipital and Parahippocampal Place Area localizers (OPA, PPA, SPA), and Place Memory Area localizer (PMA). They find that retinotopic coding extends all the way from pRF to PMA (in a -++ way). I love it!
[1] https://www.biorxiv.org/content/10.1101/2023.05.15.540807v1.full.pdf
You didn't say it, but I will
Read the paper first, then post. Seems like the early posters here are are quick to take potshots on the researchers based on their immediate reaction to the article. Tough love, fellow commentards.
While the article hits the broad summary summary of the papers points, the details are hard material, so I don't fault the author for having a hard time communicating some of the nuanced details. You seem to have the necessary background, and thanks for the quick post.
So the paper seems to support and provide good footing for the idea that the pathway to encoding visual sensory material retains the same class of structures as were encoded off the retina all the way through the round trip from perception to storage and back to memory retrieval. The part where active recall and perceptive encoding suppress each other caught my eye. This dovetails with prior research from years ago, of interest as it was offered at the time as an explanation why traditional power-point decks cause a brain crash and are actually terrible as a method for teaching and conveying information.
because IT angle.
"Differences may also exist in the brains of people with photographic memories, those with autism, and it's likely image processing capability change as we age as well , Steel said."
You don't think, Mr/Dr/Prof/whatever Steel?
Since I had a photographic memory when I was at school and I don't now, I think it's pretty certain really.
"and it's likely image processing capability change as we age as well, Steel said. ".
Yeah, it's all JPeG encoding. And every so often the brain does a defrag, moves the images around and re-encodes them. Thus the fading memories as one grows older.
It's possible you were one of the beta testers for PNG but have had an "update" since then :-)
Since I had a photographic memory when I was at school and I don't now, I think it's pretty certain really.
...and then you do the science so you know, not just think.
What about the people who can't visualize?
People with aphantasia have no inner "mind's eye" and can't visualize what's in front of them. That's a few percent of the population, and they are otherwise normal people. Based on those stats I have to know people like that, but I don't which they are.
Heck for that matter supposedly fully HALF the population doesn't have an "inner monologue". I can't conceive of that, but I guess they can't conceive of what it is like to think to yourself "I gotta remember tomorrow to put the trash out" and hear those words "spoken" in your mind.
There is a lot of variety of how humans think, so I'm skeptical about any claims they've found out much beyond the brain region that lights up when you visualize something. They certainly haven't "found the data format we use to store images as memories!"
Re: What about the people who can't visualize?
I definitely feel like I've encountered a lot of people with no inner monologue. You never have to wonder what's on their minds because it will come out of their mouths soon enough.
Re: What about the people who can't visualize?
Some of them become very successful [1]animators.
Which still blows my mind.
[1] https://www.theguardian.com/lifeandstyle/shortcuts/2019/apr/10/aphantasia-why-a-disney-animator-draws-a-blank-on-his-own-creations
Re: What about the people who can't visualize?
"People with aphantasia have no inner "mind's eye" and can't visualize what's in front of them. That's a few percent of the population, and they are otherwise normal people. Based on those stats I have to know people like that, but I don't which they are."
I have aphantasia, and a trick to find us is to watch the reaction to common things like "counting sheep".
Ermm
Visual memory is the base coding for visually acquired stuff, wow! How long before the boffins discover that aural memories encode voice and music, olfactory recall stinky stinks, and tactile that good smack round the lug'ole for the painfully obvious? Now, about that bright idea acquired by actually thinking about e=mc 2 ....
No, sorry, I am being unkind. Congrats for doing some good and bleeding-edge science so alien-implant theorists can at last be officially passed over for promotion.