TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Ask HN: What data structure does our brain use?

135 点作者 Aarvay超过 13 年前
Let's have an open discussion on this. We might be able to come up with something!<p>-- Aarvay

33 条评论

gregdetre超过 13 年前
There are so many differences that one's standard intuitions as a computer scientist can be very misleading...<p>I wrote on this elsewhere:<p><a href="http://blog.memrise.com/2011/05/how-is-memory-stored-in-brain.html" rel="nofollow">http://blog.memrise.com/2011/05/how-is-memory-stored-in-brai...</a><p><a href="http://blog.memrise.com/2011/05/how-are-brains-different-from-hard.html" rel="nofollow">http://blog.memrise.com/2011/05/how-are-brains-different-fro...</a><p>For instance:<p>- Storage and parallel computation in the brain are very expansive and cheap, so the brain prefers to store rather than compute where it can.<p>- Above all, the brain's storage is highly content-addressable. Similar things in the world are stored with similar representations, so that the brain can generalize, and see commonalities. This is not a graph - graphs are discretized - this is much more flexible.<p>- Even the acts of storage and retrieval are themselves a kind of computation, a transformation, a compression and a learning experience.<p>- Memories are not clean silos. Storing a new memory can subtly (and not so subtly) affect other nearby or related memories<p>- Different parts of the brain use different storage parameters. For instance, the hippocampus is like a hash table, storing each memory relatively cleanly and in isolation, but can only be accessed with exactly the cue. In contrast, the cortex stores memories in a much more content-addressable, overlapping way that's invariant to many small differences (e.g. we can recognize a face whether it's rotated, sunny, tanned, close up, obscured).
评论 #3087920 未加载
评论 #3087891 未加载
评论 #3088047 未加载
评论 #3087900 未加载
评论 #3088490 未加载
zerostar07超过 13 年前
Our insights in memory formation are still limited. The most studied candidate is LTP/D[1], the change of weights of synapses between neurons. Memory formation is a complex process though. While LTP can be induced with a short burst of spikes, its stabilization and maintainance depends on a sizeable number of molecules and genes. We still don't know at what level the memories may be stored, it could be the level of single synapses, groups of synapses, the level of single neurons or ensembles of neurons.<p>There are experiments involving fear memory that have shown that a fearful event can be "stored" in an ensemble of identifiable neurons, and even be turned on and off [2].<p>Then there are brain rhythms and sleep. Memories would become overwritten if they were stored in the same circuits over and over, so there are various theories about how memories are transferred to various parts of the cortex via coordinated rhytmic activity or sleep.<p>Closer to your question, the data structure that our serial thinking brain uses is language. We think, reason and communicate using it. Language has a tree-like syntax, but semantics are an unsolved problem. There is even a theory that suggests that brain rhythms may encode "sentences" into thoughts via neuronal oscillations[3].<p>1: <a href="http://en.wikipedia.org/wiki/Long-term_potentiation" rel="nofollow">http://en.wikipedia.org/wiki/Long-term_potentiation</a><p>2: <a href="http://www.silvalab.com.cnchost.com/silvapapers/ZhouNN2009.pdf" rel="nofollow">http://www.silvalab.com.cnchost.com/silvapapers/ZhouNN2009.p...</a><p>3: <a href="http://osiris.rutgers.edu/BuzsakiHP/Publications/PDFs/Buzsaki2010Neuron.pdf" rel="nofollow">http://osiris.rutgers.edu/BuzsakiHP/Publications/PDFs/Buzsak...</a>
espeed超过 13 年前
The brain most closely resembles a graph. See Marko's post on "Graphs, Brains, and Gremlin" ( <a href="http://markorodriguez.com/2011/07/14/graphs-brains-and-gremlin/" rel="nofollow">http://markorodriguez.com/2011/07/14/graphs-brains-and-greml...</a>).<p>Sebastian Seung is a leading researcher in the field of neuroscience called connectomics, which studies the wiring of the brain, and he is a professor at MIT's Department of Brain and Cognitive Sciences. He is focused on mapping the connections between each neuron and calls the mappings our "connectome," which he says is as individual as our genome.<p>He says scientists have hypothesized for years that each thought, each memory is stored as a neural connection. See his TED talk "I Am My Connectome" (<a href="http://www.ted.com/talks/lang/eng/sebastian_seung.html" rel="nofollow">http://www.ted.com/talks/lang/eng/sebastian_seung.html</a>) and the Human Connectome Project (<a href="http://www.humanconnectomeproject.org" rel="nofollow">http://www.humanconnectomeproject.org</a>).
评论 #3089327 未加载
评论 #3087707 未加载
suki超过 13 年前
Geoffrey Hinton "Next Generation Neural Networks"<p><a href="http://www.youtube.com/watch?v=AyzOUbkUf3M" rel="nofollow">http://www.youtube.com/watch?v=AyzOUbkUf3M</a><p>-It is more biologically plausible then any other NN algorithm I've seen<p>-It results in creativity (in the video he has the computer "imagine the number 2")<p>-It pretty much explains why we need to sleep/dream. The network has to be run both forward (accepting sensory input) and backwards (generating simulated sensory input) in order to learn<p>-It emphasizes the point that the brain is NOT trying to do matrix multiply (or any other deterministic calculation) with random elements (if it was trying to be an analog computer it would be). The randomness is an essential part of the algorithm.
评论 #3087821 未加载
strayer超过 13 年前
The question is fundamentally flawed, in that "data structure" is a concept used by programmers to communicate with computers (or between programmers).<p>A comparable question would be wondering whether the computer you are using right now, at this point in time, is inside a for loop, or a while loop, or it's just using tail recursion.<p>Sure some cognitive scientist can make my point more explicit, sorry that all I can only offer is a counterexample.
评论 #3087806 未加载
gmt2027超过 13 年前
Short answer: A graph.<p>All data structures are simplified graphs.<p>The state of the physical universe is a massive graph in which interconnected objects are themselves massive assemblies of graphs of atoms and the atoms are graphs of subatomic particles. It's graphs all the way down. The properties of all systems - physical, chemical, economic, biological - emerge from the interactions between simple connected elements.<p>My opinion is that all knowledge is representable as a connected graph. The disconnect between our computers and our minds arises from the fact that brains are categorically not numerical machines but graph processing and pattern recognition engines. Neural networks are the underlying hardware and, with the typical elegance of nature, these are also graphs.<p>It should be possible to build a graph based language. The basic "Elements" [SICP] are easy to realise:<p>1. Primitive Expressions are graph nodes. They have identity and not much else.<p>2. Means of combination. Graphs can be added, subtracted etc.<p>3. Abstraction. A graph can be abstracted into a single node. We have no problem looking at a complex assembly of components as a single entity.<p>Since Google and Facebook are two massive platforms whose value arise from direct interaction with planet-scale graphs with billions of nodes, would these platforms be easier to build if our computers were more graph oriented? I would like to believe so.
评论 #3087813 未加载
评论 #3087782 未加载
评论 #3087789 未加载
评论 #3087911 未加载
评论 #3087792 未加载
cydonian_monk超过 13 年前
Along a similar subject, should we someday be able to replace sets of neurons in our brains with nanomachines (built to replicate neurons), would we be able to replicate these data structures? Or are they something intrinsic to our organic brain? Similarly, if we could build even a larger scale version of such external to the brain, could we interface it with our existing consciousness? (Could we then 'share' memories?)<p>Working backwards from this, could we build (today) a graph-based (or whatever-structure) recording device that will store data in much the same way we build memories? Such would obviously require a greater understanding of the interconnection of neurons and the storage of memory, as discussed elsewhere in this topic. [Small edit: Knowing the data structure is nice; being able to use it is golden.]<p>I personally believe the next great "hack" our species should embark upon is the brain and the body. We need to be more robust if we are ever to escape this rock. (And the dreamer in me wishes we could keep our consciousnesses and/or memories around eternally, but that introduces entirely different problems, and is probably an unrealistic ideal.)
chubot超过 13 年前
A big part of the brain is the divide between conscious and unconscious. Your brain is constantly making random associations. If I read an article about Steve Jobs, I might remember something a friend said 10 years ago about him; and then I might remember that this friend lives in Brooklyn now; and then think about other people I know who live in Brooklyn. The brain just likes to make connections; perhaps synesthesia is an example of it getting slightly overworked.<p>I would say the unconscious part is basically the equivalent of a visiting a web page, Googling every term on that page, visiting those pages, and repeat nauseum (so like an inverted index perhaps on concepts/ideas/sensations rather than terms).<p>As for the conscious part, that's where the magic is. The unconscious brains generates breathtaking amounts of useless crap, but the conscious brain manages to filter it and do things like design software and make movies. I can't really speculate on how the conscious mind does this. Creativity is different than recall.<p>There is a feedback loop too. If your conscious mind starts ruminating on stuff, then the unconscious mind will generate more of it. We had a discussion about how writing down dreams causes you to produce/remember more of them. There is also the phenomena of playing a game like Tetris or Scrabble, and then your unconscious brain starts "rehearsing" all the moves in the background (sometimes against your will). It knows what you've been doing and just starts going off and making connections.<p>(If you are interested in this general subject, read "On Intelligence" by Jeff Hawkins. It will at least get you thinking and he has pretty fairly concrete ideas. He doesn't go into what I am writing about here, but as far a books that pertain to your question, I was reminded of it.)
beej71超过 13 年前
On Monday, it starts a low-speed read of a big array off magtape. Tuesday through Thursday, it continues to read the array. On Friday, a random number is chosen, and the activity stored in that array element is performed Friday night, possibly setting the REGRET flag. Saturday is spent shuffling the array, while on Sunday, a high-speed write of the data is performed back to tape. At 12AM Monday, the brain executes a GOTO 10 instruction, and the process repeats.
vga15超过 13 年前
I'd like to imagine the brain is a lot like the 'world wide web' as a data structure, than a pure graph.<p>That it isn't just a question of pure storage and retrieval. There's quite a bit of varied experience on the same root(&#38; their storage) happening multiple times and layers within.<p>Content that gets shared/liked more, gets replicated, re-iterated on, transformed, re-tagged. As time progresses, you'll find more content similar to the parent, being generated -- re-experienced via dreams and the subconscious.<p>Eventually, when its necessary to dig out the piece of content using tags or searches, it'll end up finding the most 'linked-to' piece. Possibly one that was associated with a explosion of favorable chemicals.<p>-----<p>There's some evidence to suggest that the same experience, isn't stored as a single piece of memory. During the process of consolidation [when a long term memory gets etched], the 'experience' being transcribed goes through iterations. With variations being stored as well, some of them decaying almost instantly.<p>It's possible the brain applies 'instagram' like filters while etching these memories. (a process that happens over weeks)<p>It'll be interesting if in the future, we could modify/augment these 'filter' processes. Both at the storage and retrieval stages.<p>[<a href="http://pubs.acs.org/cen/coverstory/85/8536cover.html" rel="nofollow">http://pubs.acs.org/cen/coverstory/85/8536cover.html</a>] [<a href="http://en.wikipedia.org/wiki/Engram_(neuropsychology)" rel="nofollow">http://en.wikipedia.org/wiki/Engram_(neuropsychology)</a>]
aufreak3超过 13 年前
The most recent claim in this that I've heard about and on which people are willing to bet money is Numenta's "Hierarchical Temporal Memory" = "HTM". Jeff Hawkins (of Palm's Graffiti fame) is behind this and he also wrote a book called "On Intelligence" which discusses some of these ideas.<p>It looks like Numenta is making some steady progress in trying to commercialize its HTM technology as well.
shriphani超过 13 年前
Some observations I came across in a cognition class:<p>-&#62; We can perform a visual lookup (identify the circle colored differently from these other circles in this group of circles) in O(1) time. Or if we are asked to locate a friend from an array of people (array fits in field-of-view), we can identify said friend in constant time<p>-&#62; By nature, we classify objects based on how we use them. So a table and a chair have the same physical 4-legged, flat-top structure. But we differentiate them because we use them differently:<p>So, my guess :<p>-&#62; A highly trained decision tree allowing us to perform classification of objects in our environment based on their use. (the training set is whatever is in our field of view and as such we are bombarded with large amounts of data). A hebbian-rule based ANN for training.<p>For dealing with visual stimulus at-least, I would bet that this is the model we are using right now.<p>Also, our classifier seems to be operating in parallel on all the objects available in our FOV.
zipdog超过 13 年前
It's worth noting that the brain changes as we learn - for e.g. when we learn maths the brain's structure is altered in discernable ways:<p><a href="http://med.stanford.edu/ism/2011/june/math.html" rel="nofollow">http://med.stanford.edu/ism/2011/june/math.html</a><p>So quite possibly the data storage used by a person with one particular upbrining/education would differ from a person with a different one.<p>In addition, different parts of the brain are likely storing data in different ways.
woodson超过 13 年前
There has been done a lot of work on this issue specifically to the mental lexicon, of lexemes/entities of language, and its storage. A lot of it was based on psychological testing, for example reaction times when words are presented that are somehow associated etc.<p>I guess that, while on a neurological level storage of different types of memory might work similar, the 'data structure' is perhaps different.
aangjie超过 13 年前
Well, my bet is on it being something of a mix between bloom filter and a graph. i.e the interconnectedness of graph and the probabilistic nature of Bloom filter both would be a fundamental element. Ofcourse, this will make sense only as a simulation model of the black box that's our brain/neurons/neural pathways.
gbog超过 13 年前
Interesting answers here, but from their diversity it seems to me that we in fact still don't know much about cognitive representations and their physical anchoring in the brain. I once worked in a lab, studying how the brain sees through the eyes, it wasn't close to explain how we know what is what we see.
tgflynn超过 13 年前
I think of it as a concept graph . Nodes represent concepts and weighted edges navigable relations between concepts. Further one can associate excitation levels with concepts.<p>At any given instant there is a set of excited concepts. Then there is some algorithm for time evolution of the excitations.
brettvallis超过 13 年前
Muscle memory, or engrams? How about: off CPU EEPROM with very slow burn in rate. The computation is handled remotely, or distally, but the computation only becomes overriding after a certain number of like computations. Can be unlearnt: erased.
jtchang超过 13 年前
So many different answers. We really don't know much about how our brain really encodes the data and we are so far away from actually reproducing it. Scary when you think all of us have a copy of this memory structure.
Aarvay超过 13 年前
Just did a crappy summary : <a href="http://blog.aarvay.in/brain-is-the-most-complicated-data-structure" rel="nofollow">http://blog.aarvay.in/brain-is-the-most-complicated-data-str...</a>
jamalkumar超过 13 年前
<a href="http://www.quantumconsciousness.org/documents/membytespublished.pdf" rel="nofollow">http://www.quantumconsciousness.org/documents/membytespublis...</a> I'll just leave this here
评论 #3088556 未加载
saintfiends超过 13 年前
This is an interesting topic which I have not found an adequate answer.<p>I wish we could someday know this and learn to control what we want to store for later use and what not to.
felipernb超过 13 年前
I believe it would be a graph, but there's also a "hashmap cache" with O(1) access for the most used nodes of the graph indexed by keywords or key-actions :)
kandu超过 13 年前
The right response is that science doesn't know yet.
评论 #3088276 未加载
jackylee0424超过 13 年前
A list (with less than 10 elements a time) to store easy things. I think only few people's brains can function like a graph.
drycnyc超过 13 年前
They're all wrong! It's just a massive array of Strings. No wonder I take so long to recall anything.
nicholas22超过 13 年前
There are multiple. I suppose long-term memory is a hashtable and working memory is a stack ;)
评论 #3087833 未加载
评论 #3088736 未加载
评论 #3087702 未加载
sathishmanohar超过 13 年前
I also try to make sense of this, but it is too complex, where do we store imagination?
begriffs超过 13 年前
Why are programmers so self-involved? The world is not a computer, and the human mind is not digital. It would be like a forum post on news.clockmakers.com asking, "Which wind-up clock spring does our brain use?"
Vargas超过 13 年前
An associative array: <a href="http://en.wikipedia.org/wiki/Associative_array" rel="nofollow">http://en.wikipedia.org/wiki/Associative_array</a>
afaulkner超过 13 年前
checkout numenta.com
PaulHoule超过 13 年前
a neural network
评论 #3087698 未加载
teflonhook超过 13 年前
RC circuits or delay lines