Each of these systems contribute to information. They may be closely entangled (compared to say a rock on mars) or have more entanglement due to proximity. This “entanglement” may relay universal truths or even advanced theories on itself at the speed of information.
Intrigued by the project, love the idea of exhaustively exploring significance for arbitrary input.
Think you need to provide a few example inputs and outputs on the github for the program.
Also not sure a project focused on decoding meaning and signals benefits from having AI generated interpretations divorced from the inherently human act of sign interpretation. Can be seen in the md file, such a rigidly enforced structured output has forced it to give some averaged amount of weight to different categories and examples, when many are facets of each-other, or purely just expressions of something mentioned earlier. I can see free-form high-temperature llm outputs fed to another model, which serves only to aggregate their core interpretations, providing more insight than what's within the document currently.
I'm not sure I understand your second sentence, as I don't follow "entanglement news." Is entanglement currently being seen as taking part in everything, everywhere? Specifically in brain processes?
There is a (decidedly non-mainstream) idea that brain processes involve large-scale quantum effects, that quantum effects beyond just the normally expected biochemistry are essential to the functioning of the brain. There's not really much empirical evidence for it, and about the only theoretical argument is that the brain is much more efficient than a computer. I think people tend to like it because it makes brains (or more specifically human minds) feel more special.
There's a guy in Twitter who thinks that the brain is an antenna that receives consciousness, or something like that. I'll see if I can find it, but it's kind of hard to navigate Twitter since I got banned.
Edit: I couldn't find it. Something about microtubules
> Is entanglement currently being seen as taking part in everything, everywhere?
Not necessarily what they're saying here (I couldn't grok it either), but there's a bunch of woo being purveyed about "quantum consciousness" by Deepak Chopra recently. Some debates with Sam Harris are fun to watch.
Here's a blurb from the paper that partially answers your question:
> Late-born neurons were preferentially recruited for retrieval at short latency after acquisition, whereas early-born neurons were preferentially recruited at later times. These divergent trajectories recapitulated reactivation dynamics recorded through longitudinal calcium imaging experiments, which further revealed distinct network-wide responses between subpopulations.
I don't quite follow their leap from seeing three distinct regions and types of neurons firing to the proposal that each is storing a separate copy of the memory.
Anyone have an idea of how they make that leap? Or are the researchers getting ahead of their skis, assuming that neurons lighting up means the memory is directly stored there as opposed to those neurons simply being part of the process?
Maybe there is a thing like the precogs in Minority Report, the storage is presumably made in different places because memory can deteriorate and when you need to recall it there is a "vote"/synchronization between those different places ?
On a small level computers do that: cpu cache (itself 3 levels), GPU memory (optional), main memory, hdd cache.
The reason why animal brains would store multiple copies of a memory is functional proximity. You need memory near the limbic system for neurotic processing and fear conditioning. Long term storage is near the brain stem so that it can eventually bleed into the cerebellum to become muscle memory. There is memory storage near the frontal lobes so that people can reason about with their advanced processors like speech parsing, visual cortex, decision bias, and so forth.
> We might be wrong. New research suggests that our brains make two copies of each memory in the moment they are formed. One is filed away in the hippocampus, the center of short-term memories, while the other is stored in cortex, where our long-term memories reside.
> Surprisingly, the researchers found that long-term memories remain "silent" in the prefrontal cortex for about two weeks before maturing and becoming consolidated into permanent long-term memories.
So that makes four (4) copies of each memory in the brain if you include the engram cells in the prefrontal cortex.
What is the survival advantage to redundant, resilient recall; why do brains with such traits survive and where and when in our evolutionary lineage did such complexity arise?
One possibility is that the different sets of neurons have different raw capabilities and the brain is utilizing features from all to balance retention rates etc.
The paper is about how the brain seems to encode memories and re encode them over multiple groups of differently aged neurons; with younger neurons being more plastic). Nothing like saying “brains store three copies”
Your comment is too late, I already launched my startup that stores data in 3 databases calling “patent-pending human biology based neural retrieval” and raised a $15M series A
Nature loves redundancy. Milton Friedman needs to lecture nature on efficiency and economics and show off the modern marvel that is the US economy. Maybe nature would get its act together and stop being so wasteful.
Based on some of the replies here it appears that you probably just need to get re-entangled with some of your friends who were there at the same time.
62 comments
[ 2.9 ms ] story [ 125 ms ] threadWith entanglement they all work together to form the one memory no?
I’m not convinced the same memory is even isolated to one brain.
Let’s start with familiarizing with a large collection of sign systems we know about…
https://github.com/space-bacon/SignSystems/blob/main/Codex-o...
Think you need to provide a few example inputs and outputs on the github for the program.
Also not sure a project focused on decoding meaning and signals benefits from having AI generated interpretations divorced from the inherently human act of sign interpretation. Can be seen in the md file, such a rigidly enforced structured output has forced it to give some averaged amount of weight to different categories and examples, when many are facets of each-other, or purely just expressions of something mentioned earlier. I can see free-form high-temperature llm outputs fed to another model, which serves only to aggregate their core interpretations, providing more insight than what's within the document currently.
What are your thoughts on this from a conceptual lens?
https://github.com/space-bacon/Semiotic-Analysis-Tool
I'm not sure I understand your second sentence, as I don't follow "entanglement news." Is entanglement currently being seen as taking part in everything, everywhere? Specifically in brain processes?
I'm intrigued by your third sentence.
I look forward to more sentences.
Edit: I couldn't find it. Something about microtubules
https://www.nature.com/articles/s41598-018-30453-2
A quick search on "microtubule consciousness" brings forth an enormous amount of spilled digital ink on the subject.
Sabine Hossenfelder had a "science daily" video about it recently on Youtube. Worth a watch/listen.
https://backreaction.blogspot.com/2024/05/brain-really-uses-...
Not necessarily what they're saying here (I couldn't grok it either), but there's a bunch of woo being purveyed about "quantum consciousness" by Deepak Chopra recently. Some debates with Sam Harris are fun to watch.
The actual paper goes into more detail: https://www.science.org/doi/10.1126/science.adk0997
Here's a blurb from the paper that partially answers your question:
> Late-born neurons were preferentially recruited for retrieval at short latency after acquisition, whereas early-born neurons were preferentially recruited at later times. These divergent trajectories recapitulated reactivation dynamics recorded through longitudinal calcium imaging experiments, which further revealed distinct network-wide responses between subpopulations.
... IN MICE
Anyone have an idea of how they make that leap? Or are the researchers getting ahead of their skis, assuming that neurons lighting up means the memory is directly stored there as opposed to those neurons simply being part of the process?
It's a huge field of research though, I'm by no means an expert.
Why would the brain store multiple copies of a memory? It’s so inefficient.
The reason why animal brains would store multiple copies of a memory is functional proximity. You need memory near the limbic system for neurotic processing and fear conditioning. Long term storage is near the brain stem so that it can eventually bleed into the cerebellum to become muscle memory. There is memory storage near the frontal lobes so that people can reason about with their advanced processors like speech parsing, visual cortex, decision bias, and so forth.
"Our Brains Instantly Make Two Copies of Each Memory" (2017) https://www.pbs.org/wgbh/nova/article/our-brains-instantly-m... :
> We might be wrong. New research suggests that our brains make two copies of each memory in the moment they are formed. One is filed away in the hippocampus, the center of short-term memories, while the other is stored in cortex, where our long-term memories reside.
From https://www.psychologytoday.com/us/blog/the-athletes-way/201... :
> Surprisingly, the researchers found that long-term memories remain "silent" in the prefrontal cortex for about two weeks before maturing and becoming consolidated into permanent long-term memories.
ScholarlyArticle: "Engrams and circuits crucial for systems consolidation of a memory" (2017) https://www.science.org/doi/10.1126/science.aam6808
So that makes four (4) copies of each memory in the brain if you include the engram cells in the prefrontal cortex.
What is the survival advantage to redundant, resilient recall; why do brains with such traits survive and where and when in our evolutionary lineage did such complexity arise?
video data; audio data; title information
or...
images; sounds; emotional state; more?
Example: https://en.wikipedia.org/wiki/Recurrent_laryngeal_nerve#Evid...
https://www.science.org/doi/10.1126/science.adk0997
The paper is about how the brain seems to encode memories and re encode them over multiple groups of differently aged neurons; with younger neurons being more plastic). Nothing like saying “brains store three copies”
How else are you going to look at it from other perspectives?