Bridging the gap between statistical physics and neuroscience

[CNS !MT.CNS.YRY No.4461897]

The purpose of this thread is to provide you with some insight into a framework from statistical physics which has left me fascinated ever since I first learned about it. Now that it is beginning to be applied to neuroscientific research increasingly more often, I can illustrate its usefulness to my own line of work.

In this thread, I will assume the reader has zero knowledge, but infinite intelligence. I will try my best to explain some of the more difficult concepts, but time prevents me from going into detail as much as I would like to. As such, the reader is encouraged to google terms, and ask for clarification on any concept they do not understand. Be they a physicist with limited knowledge about neuroscience, a biologist with limited knowledge about physics, or just an interested laymen. I hope to start a discussion between people from different fields, not necessarily related to the topic at hand, so don’t be afraid to join in.

[CNS !MT.CNS.YRY No.4461904]

In the next few posts, I will attempt to explain that a dynamical system poised near a critical point of a second-order phase transition comprised of constituent components that show a high degree of non-linearity can give rise to self-organized complex phenomena. Furthermore, the fundamental realization that the brain is a system that shows self-organized criticality has allowed for some spectacular applications in clinical neuroscience, of which I’ll give an example. If what you’ve just read made absolutely no sense, don’t worry. All will be explained.

Alright. Here we go.

[CNS !MT.CNS.YRY No.4461906]

Let’s begin by clarifying some jargon. ‘Complexity’ is a concept from physics. If a system is complex, this implies more than it just being highly complicated. It is a unique quality that emerges from interaction between constituent components. A complex system balances on the edge between disorder (maximum entropy) and order, between inflexibility and adaptation, stochasticity and determinism, chance and necessity, etcetera. They are in a state between two extremes, a state of phase transition.

There are a number of factors which are required for a system to be called complex. To put it briefly, a complex system is comprised of a <span class="math">large~number[/spoiler] of <span class="math">interacting[/spoiler] and <span class="math">non-linear[/spoiler] elements. Non-linearity here means that changing any single parameter of these individual elements can have consequences for function on the system-level that are hard to capture mathematically or otherwise. This leads to a high degree of unpredictability when modeling such a system. The system as observed shows self-organized characteristics, but small perturbations can potentially reshape the entire system state. This is known as criticality.

[CNS !MT.CNS.YRY No.4461910]

Complex systems poised near criticality are scale free (where no temporal scale takes primacy over average) and governed by a power-law (<span class="math">1/f[/spoiler]) distribution. Importantly, they exhibit unexpected collective spatiotemporal correlational patterns.

An archetypical example of a complex system in which such patterns become apparent is the Ising model. Basically, it is a mathematical representation of fluctuations in magnetization at different temperatures. The spin-state of individual particles within a ferromagnetic object show variation in behavior at different temperatures. At temperature <span class="math">T_c[/spoiler], criticality emerges (image). Only at <span class="math">T[/spoiler]~<span class="math">T_c[/spoiler] does the system exhibit a second order phase transition, where highly heterogeneous correlated domains are seen.

[CNS !MT.CNS.YRY No.4461913]

What’s remarkable is that the correlational structure within the brain during rest is virtually indistinguishable from the Ising model at <span class="math">T_c[/spoiler] (compare middle top to bottom).

During resting state – where there is no explicit external input or output – activity within distributed brain regions seems to co-vary. What’s fascinating is that the clusters of correlation that emerge closely overlap with a wide variety of behavioral activation conditions. Thus, the collective spatiotemporal dynamics visit the same brain regions that are activated during any given active behavior. Healthy brains can therefore (arguably) be said to be naturally poised near criticality. They are systems that have the innate functional organization that is balanced on a second-order phase transition.

This has realization has naturally resulted in formal characterizations of the autocorrelational structure that is apparent in brain activity at rest.

[CNS !MT.CNS.YRY No.4461918]

A very important discovery which has recently been made is that deviations from the autocorrelation structure are apparent in certain pathologies.

For example, Montez et al* applied the framework as described above to analyze resting-state magnetoencephalography data from AD patients, and compare it to healthy controls. They showed that the autocorrelation structure within the alpha and theta band are substantially different in AD patients (figure).

This knowledge might not seem very significant, but in the future it can lead to pattern classification algorithms that can diagnose AD based on simply analyzing autocorrelation structure, even before symptoms become apparent.

* http://www.pnas.org/content/106/5/1614.short

[CNS !MT.CNS.YRY No.4461924]

The example paper I gave is by no means the only study to date which has applied this framework. There are many more, and most provide us with extraordinary insight into how the brain functions. This framework, in my opinion, has the potential to reshape the way we think about the brain. Systems neuroscience up until recently was concerned more with interaction between relatively segregated modular neuronal ensembles in the brain. By looking at the brain as a collective system however, we tap into a whole new world of knowledge. That is why I think it is awesome.

[Anonymous No.4461925]

Sage for too rigorous, not enough popsci, not controversial enough and generally too hard for /sci/

[Anonymous No.4461927]

So is the brain a Turing machine or not?

[CNS !MT.CNS.YRY No.4461934]

>>4461925
Yes, I was afraid of that. Really though, this looks like a lot but it is only about a page of actual text. Who knows, you might even learn something if you try reading it.

[CNS !MT.CNS.YRY No.4461941]

>>4461927
It would definitely pass the Turing test.

[CNS !MT.CNS.YRY No.4461964]

Okay, so definitely too much text. I had a similar thread before, and when I put all the text in a single image, suddenly people read it. Oh well. I'll bum again later to see if the evening crowd is interested.

[CNS !MT.CNS.YRY No.4461967]

>>4461964
>bum
bump*

[Anonymous No.4461976]

I'm tired as fuck and don't know what the fuck is going on, I'll save this thread for tomorrow and read it though. Seemed interesting

[CNS !MT.CNS.YRY No.4461979]

>>4461976
thanks

[Anonymous No.4461992]

>>4461979
Enjoyed the read but understood very little

Summary:

>the brain is like a very big complex mathematical / physical system with superpowers

[CNS !MT.CNS.YRY No.4461995]

>>4461992
>the brain is like a very big complex mathematical / physical system with superpowers
haha, yes that's pretty much the gist of it

[Anonymous No.4461998]

I would engage with this but I have a lab report to write up and I'm just on sci right now to take a break.

[Anonymous No.4462000]

>dynamical system poised near a critical point of a second-order phase transition comprised of constituent components
bad start, but the rest is relatively understandable.

I'll be honest - I think that you're wasting your time here. It's clear that your kowledge is superior to 98% of sci, and at least half won't even understand this text.

Thus, you're unlikely to have any stimulating conversation.

Thanks for the read, though. Even if it's completely unrelated to most of my interests, I shall check some parts of it in more depth later.

[CNS !MT.CNS.YRY No.4462003]

>>4461998
Ok. I'll be around, should you decide to change your mind later.

[CNS !MT.CNS.YRY No.4462012]

>>4462000
>I'll be honest - I think that you're wasting your time here. It's clear that your kowledge is superior to 98% of sci, and at least half won't even understand this text.
I'm an optimist. I really don't know all that much about physics, and I'm sure there are people here that know more about self-organized criticality than I do.

The point of this thread is also not necessarily for me to learn (that would be nice of course) but to stimulate discussion between the people reading it. I'm here to answer questions if necessary.

[SAGEGOESINALLFIELDS No.4462023]

How can it be said that with the Ising model, when T ~ Tc, it becomes self organized? Doesn't self organisation imply a drive for survival? Or at the very least sustainability?
I mean, sure, because of temperature fluxes, you will get a system that can be predicted on macroscopical level.. doesn't say anything on how neurons interact seperately. Unless I'm completely wrong. Also, I don't quite understand the second to last picture. They computed the brain at rest at temperatures up to 3000k, or what does the degree mean?

Also, I don't think the modeling will find AD more quickly than for instance certain biomarkers. A disease like AD manifests itself over a number of years, and gradually suffers from amyloid-beta buildup and other stuff, resulting in neurodegeration etc.

That's all I can give you atm. Sorry for my ignorance lol

[Anonymous No.4462024]

Can u explaim like im 5 summary...especially what the graphs are trying to convey

[Anonymous No.4462029]

>>4461897
Thanks for the read. I have been very interested in the subject of neuroscience (as a hobby) for the last few years and would love to ask you some questions.

I've never heard about "criticality" before, but having seen the image you provided, it seems pretty straight-forward.
1). What is the significance of the brain being "near criticality"?
>side note: the middle image appears to have a 2-dimensional "Brazil nut effect" going on.

2). What does it mean, to you, for something to exhibit a "second order phase transition"? Could you possibly explain this in more detail?
>I could google it, but I figure you'll be able to explain it better.

[Anonymous No.4462027]

>>4461897
>>4461904
>>4461906
>>4461910
>>4461913
>>4461918
>>4461924
CNS GRACES US WITH HIS PRESENCE!

[Anonymous No.4462031]

sort of off topic but I fucking love CNS, We should celebrate his existence here.

[Anonymous No.4462033]

[Anonymous No.4462048]

I don't really pay attention to trips so i don't really know if you have posted things here before, but I am happy to see that there are human beings out there like you that actually spend the time to spread the knowledge they feel is important. I am a neurobio major and I am currently in the fourth year of my second Bachelor in Physics. I understood most of it and i share your fascination with the subject. I will however, like to add that i don't really see how modelling of the brain as a chaotic, non-linear dynamical system, could have any "useful" applications apart from diagnostics. Would you care to elaborate on that a bit.

Also again thank you very much for this thread, this is what this board should be like, thank you.

[CNS !MT.CNS.YRY No.4462055]

>>4462023
>How can it be said that with the Ising model, when T ~ Tc, it becomes self organized? Doesn't self organisation imply a drive for survival? Or at the very least sustainability?
Good question. The Ising model at Tc is not self-organized, but it is critical. Criticality emerges, but only at a specific temperature. The brain, in contrast, shows criticality naturally.
>mean, sure, because of temperature fluxes, you will get a system that can be predicted on macroscopical level..
The point was that the system becomes quite <span class="math">un[/spoiler]predictable, due to non-linearity (remember that a complex system is necessarily composed of non-linear elements?)
>doesn't say anything on how neurons interact seperately
This is true.
>Also, I don't quite understand the second to last picture.
k is a measure of spatial correlation (i.e. a measure of the distribution of similarly oriented spin alignments).
>Also, I don't think the modeling will find AD more quickly than for instance certain biomarkers.
The thing with biomarkers is that they are never conclusive. They might be in the future, but as it stands the approach I describe will be a lot cheaper than genetic screening for instance. It can be done with only 4 minutes of resting state data.
>That's all I can give you atm. Sorry for my ignorance lol
No worries, glad to see someones participating!

[Anonymous No.4462058]

ok so the AD brain is different than the normal brain and we have or soon to have ways to test for this. As it should be, if you have shit built up between your cells, things go wrong.
So why is this interesting again? How will this help us with anything?

[Anonymous No.4462059]

http://chanarchive.org/request_votes
Requesting archive.

[CNS !MT.CNS.YRY No.4462064]

>>4462024
Figure one (not the one in the OP):
At a certain temperature, particles behave in a very unexpected way.

Figure two:
The way the particles behave is comparable to how brain activity varies over time.

Figure three:
The way brain activity varies over time is different in people with Alzheimer's disease.

[Anonymous No.4462070]

>>4462059
bump for archive and bump for discussion.

CNS, you explain the concept of criticality a little more in depth, and by in depth I mean more generally, using simpler words?

[CNS !MT.CNS.YRY No.4462072]

>>4462029
I'll respond to your post, but I have to eat something real quick. I'll be back in about 20 minutes. Sorry for the delay.

[Anonymous No.4462076]

Is the brain a discrete dynamical system?

[Anonymous No.4462081]

>>4462070
>and by in depth I mean more generally, using simpler words?
wat

[Anonymous No.4462082]

>>4462055
You're post to him got me curious.
>k is a measure of spatial correlation (i.e. a measure of the distribution of similarly oriented spin alignments).
Aside from my apparent lack of knowledge, I'm curious about the physical differences.

1). What does the distribution of activity actually represent on a physical level of individual groups of neurons when the brain is not at rest?
>Does it look similar to "subcritical"?

2). If you're saying that the brain is always activating neurons in a "critical" fashion, then what does Fig2 represent physically (on a neuron/neuron bundle level)?

[SAGEGOESINALLFIELDS No.4462103]

>>4462055
>The thing with biomarkers is that they are never conclusive. They might be in the future, but as it stands the approach I describe will be a lot cheaper than genetic screening for instance. It can be done with only 4 minutes of resting state data.
I don't agree. There is extensive research for biomarkers in, for example, cancer and there are a wide variety of molecules that have come to the surface. I agree that it's still inconclusive, but the research is still going on. And with every system in which these molecules interact unfolds, more convincingly we'll be able to address certain concentrations of certain molecules to certain pathogeneses.
Also, genetic screening is becoming cheaper and cheaper by the year. I was at the PhD defense yesterday and this prof from the board said a genetic screening could be done for ~5 euro's in the next 3 years.

I'll follow the other discussions so I might ask more insightful questions later hjehjehje.

[CNS !MT.CNS.YRY No.4462113]

>>4462029
>1). What is the significance of the brain being "near criticality"?
It has consequences for how we can investigate large-scale neural dynamics. For example, in analogy with a tiny spin-perpetration causing a change in the alignment of the entire system, action of single neuromodulators (e.g. nor-adrenalin) can have effects over wide cortical areas.
>2). What does it mean, to you, for something to exhibit a "second order phase transition"? Could you possibly explain this in more detail?
It means that the brain is in a state just in between chaos and order. Apparent neural dynamics might seem disorganized at first glance, but there are temporal patterns in activity that emerge in a coherent fashion.

[Anonymous No.4462114]

>>4462103
>genetic screening could be done for ~5 euro's in the next 3 years
That is interesting.

[SAGEGOESINALLFIELDS No.4462117]

>>4462114
They're called exon chips, but that's off topic hah.

[CNS !MT.CNS.YRY No.4462121]

>>4462048
Thanks, glad to see it's appreciated.
>I will however, like to add that i don't really see how modelling of the brain as a chaotic, non-linear dynamical system, could have any "useful" applications apart from diagnostics.
I think that modeling the brain in this way is a more accurate representation of what it really is: a highly intricate system in which no single component can be seen as separate from anything else that is part of the system. It is hard to predict what fundamental changes in thinking like this might lead to, but better simulations of brain function will certainly be one of them.

[CNS !MT.CNS.YRY No.4462125]

>>4462058
This was only meant to be one example of a potential application. What this specifically can help us with though, is that it will enable to catch the disorder early on, so treatment can be targeted at preventing symptoms rather than fighting them.

[Anonymous No.4462126]

>>4462114
Full genome sequencing is down to a thousand dollars in cost and can be performed in one day.

http://www.medicalnewstoday.com/articles/240145.php

I imagine in 5 years it will be pretty standard to have your genome sequenced.

[CNS !MT.CNS.YRY No.4462131]

>>4462076
It's a continuous dynamical system.

[CNS !MT.CNS.YRY No.4462136]

>>4462070
Sorry, I seem to have skipped your post.

I'd like to be able to explain the whole thing in simple words, but I'm afraid that would take a very long time. Alternatively, I'd encourage you to ask for clarification in specific concepts.

[CNS !MT.CNS.YRY No.4462149]

>>4462082
>1). What does the distribution of activity actually represent on a physical level of individual groups of neurons when the brain is not at rest?
It means that during the engagement of a particular task (let's say, focusing attention on a particular object) engages a particular neural network within the brain (e.g. the dorsal attention network). Interestingly, activity within the brain regions of this network is also highly correlated over time during rest. This indicates that we can distinguish functionally connected networks from resting state activity based on the correlational distributions.

[CNS !MT.CNS.YRY No.4462159]

>>4462082
whoops, pressed submit too soon.
>2). If you're saying that the brain is always activating neurons in a "critical" fashion, then what does Fig2 represent physically (on a neuron/neuron bundle level)?
This means that the activity over time within the brain shows scale-free dynamics. So, there is no real temporal window in which activity in distributed networks is more strongly correlated than on any other scale.

[CNS !MT.CNS.YRY No.4462166]

>>4462103
True, and I don't feel strongly about this argument. It was only a potential application and more are likely to arise. :)

For now though, a pattern classification algorithm can be programmed with relative ease. Thus, for the moment at least, it might be a cheaper solution.

[CNS !MT.CNS.YRY No.4462168]

I see the thread got archived. That's great!

[Anonymous No.4462185]

OP TEACH ME HOW TO HAVE INFINITE MOTIVATION AND HOW TO USE MY BRAIN INTENSIVELY FOR LONG PERIODS OF TIME WITHOUT CRASHING!!!

[CNS !MT.CNS.YRY No.4462187]

Alright, enough typing for one day. Thanks a lot to everyone contributing. I'll check in later, so don't hesitate to post any additional questions you might have.

[CNS !MT.CNS.YRY No.4462190]

>>4462185
Work hard, play hard.

[Anonymous No.4462207]

I have no idea what I just read, but I somehow feel it was profound

[Anonymous No.4462211]

If I understand correctly a higher temperature would cause particles in the brain to become critical or supercritical. What effects can environment or faulty homeostasis have on brain states? Would a fever or a literally hot head induce strange thinking? Is AD caused by the brain improperly self-regulating this chaos?

Also this is one of the best threads I've seen on /sci/. Greatly appreciate your effort here CNS.

[Anonymous No.4462236]

>>4462185

yeah as a 3rd year college student ive been leaning towards that, i <3 behavioral psychology

i dont frequent sci much but you should make a thread of brain hacks for motivation sometime, maybe a more applied version of this thread

thanks

[Anonymous No.4462253]

Man this was interesting. You spoke of statistical physics. Do you think that the study of such a highly complex thing as the brain could be simplified as much as what thermodynamics do when it comes to particles? Is there any research that contradicts or encourages this possibility?

[Anonymous No.4462260]

>>4462211
higher temperatures were about the magnetic materials, not the brain! But temperature could be compared to stimulation, maybe?

[Anonymous No.4462271]

I masturbated to this thread.

[Anonymous No.4462284]

>>4462271
I masturbate to my mental image of CNS.

[Anonymous No.4462293]

My head hurts.

[Anonymous No.4462301]

>>4462236
> psychology
> a science

[Anonymous No.4462323]

Bumping for great justice.

[CNS !MT.CNS.YRY No.4462335]

>>4462211
My apologies for the confusion. Temperature in the Ising model is only used to show the point of phase transition, the point where criticality emerges. In the brain this would correspond to (and I'm sorry if this is even more confusing) the specific anatomical connectivity in the brain. Critical dynamics will only emerge if there is a special ratio between local and long range connectivity.
>>4462260
>higher temperatures were about the magnetic materials, not the brain!
>>4462271
>>4462284
Oh god what. I'm just some guy.

[CNS !MT.CNS.YRY No.4462342]

>>4462335
>higher temperatures were about the magnetic materials, not the brain!
Forgot to type my reply (which was short): Yes, this is correct!

[CNS !MT.CNS.YRY No.4462358]

>>4462236
Thanks for the suggestion, I'll keep it in mind.

[CNS !MT.CNS.YRY No.4462363]

>>4462253
Yes, I very much think this is the case. Time will tell whether this is really valid though, but it's picking up speed within the field. Importantly, no contemporary theoretical model of the brain as a system can account for the neural dynamics as we observe them without invoking criticality. That strongly suggest that we are onto something here.

[Anonymous No.4462370]

>>4462358
Oooh, if you're seriously considering doing it, I'd like to personally voice my support for such an endeavor.

[Anonymous No.4462374]

>>4462335
Interesting. So criticality in the brain is dependent on how many connections neurons have with each other and what type of connections they are, similar in concept to the Ising model? Are patients with Alzheimer's more likely to have flawed neural connections or problems with some kind of activation pathways?

I know that researchers have been able to decode visual and auditory images directly from the brain. Have there been any discoveries in decoding how specific anatomical connectivity corresponds to thoughts or mental illnesses?

[CNS !MT.CNS.YRY No.4462400]

>>4462374
>So criticality in the brain is dependent on how many connections neurons have with each other and what type of connections they are, similar in concept to the Ising model?
Exactly. The commonality between the two systems is that changing a certain parameter, at a certain point, will lead to unexpected behavior. With the Ising model it is temperature, and in the brain it is anatomical connectivity.
>Are patients with Alzheimer's more likely to have flawed neural connections or problems with some kind of activation pathways?
Yes, they are. Alzheimer's is a neurodegenerative disorder, where large populations of neurons progressively die off. This has consequences for connectivity, and thus for the behavior activity on the system level.
>Have there been any discoveries in decoding how specific anatomical connectivity corresponds to thoughts or mental illnesses?
Decoding, not that I'm aware of. There is however a substantial body of literature that concerns aberrant connectivity in certain pathologies. People with schizophrenia for instance show an excess in synaptic pruning (where redundant synaptic connections are removed), a process which naturally occurs in healthy humans, but which reduces connectivity in schizophrenics too much. Interestingly, schizophrenia is one of the disorders which is also identifiable by studying auto-correlation strucutre, similar to the example I already gave.

[Anonymous No.4462404]

Personally, i found this refreshing, and would like to see more threads with "this flavor".

Kewl stuff :)

Cheers, OP.

And thanks!

[Anonymous No.4462406]

>>4462159
>scale-free dynamics
After looking at Fig1's "critical" and "super critical", it seems like they are similar with their only difference being scale.

1). Could this idea of "criticality" account for the brains ability to scale input?
>Example: If you have a very large letter 'A' that takes up most of your vision and a very small letter 'A' that is barely visible, the brain still "knows" that it's an 'A'.
>By "knows", i mean that the same regions of the brain are activated with varying input space activation.

What do you think?

[CNS !MT.CNS.YRY No.4462407]

>>4462404
no, thank you!

[CNS !MT.CNS.YRY No.4462443]

>>4462406
>After looking at Fig1's "critical" and "super critical", it seems like they are similar with their only difference being scale.
This is where the limitations of missing a dimension come in. It would have been optimal if I could have shown a video, but that's not possible so I'll try to explain what the difference between critical and supercritical is in words. In the supercritical state, there are very little correlations of spin alignment over time. In other words, spin behavior is chaotic, there is a large amount of entropy. In the subcritical state, there is a very low amount of entropy, but that implies inflexibility. The critical state is a mixture between the two, forming the optimal ratio between predictability (on temporal scales, not spatial scales) and flexibility. This is where non-linearity becomes important. Only in the critical state does any single particle have a small but finite chance of changing the entire system dynamic by interacting with its neighbors.
>1). Could this idea of "criticality" account for the brains ability to scale input?
It's an interesting idea. Scale-free as I've used it here actually refers to temporal scales. There is however indeed a spatial scale-free dynamic in the brain (although it is more limited in extent), which is reflected in the hierarchical organization in the visual system.

[mtp !EKFQOBEsKo No.4462444]

ive got another question, somewhat unrelated to the thread.

would someone after doing physics phd be able to go into neuroscience research? especially looking at undergrad in theoretical physics and maths and then
post grad in theoretical physics also.

[CNS !MT.CNS.YRY No.4462456]

>>4462444
It kind of depends on what type of research you would want to get into. In my lab we have a physicist, but he mostly works on the MRI scanner, optimizing signal-to-noise and developing new pulse sequences and such. It's not unheard of for physicists to become involved in neuroscientific research itself though. I think that having a background in mathematics will definitely give you an edge in developing computational models, or system level dynamics (exactly the topic here). You'd have to do some catching up in the biology department, but it's doable.

[Anonymous No.4462470]

>>4462443
When I was working on different AI models, I came across the HTM model (numenta.com). The algorithms behind this model would inevitably create a "region of neurons" that would resemble the "critical" image in Fig1.

1). So my question is, do you have any insight on how this approach to modeling the function and organization of the brain (being non-linear, complex, and critical) could be applied to a computer-driven model?

[CNS !MT.CNS.YRY No.4462496]

>>4462470
Good question. The realization that the brain shows criticality puts forward constraints on how a good representation of a neuronal network should be structured, and how it should behave. First, elements must be non-linear. A neural network with simple on/off neurons might be useful, but it is not an accurate representation of how a brain operates. Second, it should contain a large number of elements. Non-linearity only becomes important on a large scale, and therefore it should be reflected in the model. Third, neurons (or nodes within a network) must be able to interact. As I explained earlier, there is a ratio between local and long range connectivity after which criticality emerges. Fourth, activity within the network should show power-law statistics. If this is not the case, then the model is again not an accurate representation of a true neural system. Finally, the network must be balanced between flexibility and stationarity. It must be able to perform a wide variety of functions, but with consistent performance.

[Anonymous No.4462518]

Bump, for science

[Anonymous No.4462522]

>>4462496
Having problems posting from my PC. I have two questions, but I need to fix something.

[CNS !MT.CNS.YRY No.4462527]

>>4462522
It's not just you, I'm having trouble posting as well (that's why my last post took so long, and this one). I think I'm going to call it a night, and answer your questions tomorrow if the thread is still alive.

A final note for today, I really appreciate your interest.

[Anonymous No.4462533]

best thread

[Anonymous No.4462545]

>>4461941
This made me lolwut

[Anonymous No.4462586]

let me post goddammit

[mtp !EKFQOBEsKo No.4462596]

Bumping for science and going to sleep.

[Anonymous No.4462604]

>>4462596
Have a nice night.

I will continue to decipher this, because it's very interesting.

[Anonymous No.4462645]

>>4461913
>>4461913
im confused as to what this figuring is saying? is it implying that heating a brain to 1000 kelvin makes it behave like the ising model?

why is this of any significance at all surely this is just the high amounts of iron you are detecting in the brain?

[Anonymous No.4462657]

>>4462645
>>4462645
nvm read through the posts and realised k was not temperature even though k (degrees) throws one of easily

[Anonymous No.4462949]

bump for being board related

[Anonymous No.4463242]

>>4462496
Thank you. I had figured as much from what's already been discussed (and from my own research). I'll have to get back to messing around with my model sometime...

1). What do you feel contributes to the criticality of the system?

2). I have always thought that from fetus-to-adult, our brains go from super-critical to critical. What do you think about infant-brains?

>hopefully I can post now...

[Anonymous No.4463617]

>>4463242
>expanding upon my own post since everyone's gone to sleep...

1). What do you feel contributes to the criticality of the system?
>I assume that the contributing factors relate to the inherent movement of dendritic connections between the inherent layers of neurons that are created during the gestation period. The movement of the synapses had been thought be to very slow/haphazard, but aren't there reports that they can gain/lose connections quickly?

2). I have always thought that from fetus-to-adult, our brains go from super-critical to critical. What do you think about infant-brains?
>There has been research done that shows that a 12month doesn't "understand" the danger of depth (will walk out off of a cliff) while an 18month will stop at the edge. (Can't remember sauce, at most 2008)
>An 18month understands what another human "likes" while a 15month doesn't (TED, July 11).

Please feel free to elaborate or contribute. This thread is awesome and makes me hopeful.

[Anonymous No.4463973]

Why does CNS always make the most glorious threads ever?

If only we were all like CNS...

[Anonymous No.4463984]

I'm an experimental physicist becoming interested in physics in biology, especially neuroscience. and I'd love to read this. But, I'm about to go to bed. Could you post this somewhere a little more permanent?

You should blog this maybe.

[Anonymous No.4463992]

>>4463984
It's archived...

[Anonymous No.4464112]

Bumping this thread for tomorrow. See you then, CNS. I hope you've had some coffee before replying (as I'll be having mine before coming back).

[mtp !EKFQOBEsKo No.4464253]

early morning bump, because the computer is on and im not out yet

[Anonymous No.4464302]

Holy fuck this thread is awesome

[Anonymous No.4464320]

>>4461925
This makes me mad. You are the cancer killing /sci/.

[Anonymous No.4464323]

>>4464320
>doesn't understand satire

[Anonymous No.4464619]

>>4461913
There's a region with a temperature of 1000 K in my brain??
what the fuck

[Anonymous No.4464624]

Are there any studies you can quote where this method of analysis has been used to study different patterns of connectivity (both at rest and event-related experimental designs)?

I'm interested to see how this method was used and what results they got using it.

[CNS !MT.CNS.YRY No.4464866]

Sorry for my late reply, I didn't have time to check up on the thread before I had to go to work. Anyway, I'm back now for a short period.

>>4463617
>1). What do you feel contributes to the criticality of the system?
Connectivity and the neurophysiological properties of individual neurons are what determine the system dynamics.

>2). I have always thought that from fetus-to-adult, our brains go from super-critical to critical. What do you think about infant-brains?
This is a very interesting remark, I hadn't thought about it like that. Now that you mention it, yes this makes sense. As our brains mature, the system because gradually more stable and inflexible (e.g. reduced neurogenesis and plasticity). As such, it moves in the direction of subcriticality.

[CNS !MT.CNS.YRY No.4464868]

>>4464619
k is the degree of correlation. T denotes temperature.

[CNS !MT.CNS.YRY No.4464883]

>>4464624
>Are there any studies you can quote where this method of analysis has been used to study different patterns of connectivity (both at rest and event-related experimental designs)?
Sure, here are a few:
http://www.jneurosci.org/content/23/35/11167.abstract
http://www.jneurosci.org/content/27/50/13882.short
http://www.ploscompbiol.org/article/fetchObjectAttachment.action;jsessionid=68DFB0BE30585F9DFD136A7E
DFA6A24A?uri=info%3Adoi%2F10.1371%2Fjournal.pcbi.1000314&representation=PDF
http://www.bio.vu.nl/enf/vanooyen/papers/poil_linkenkaer_humanbrainmapping_2008.pdf
http://www.bio.vu.nl/enf/linkenkaer/Papers/Linkenkaer._Breakdown_of_long-range_temporal_correlations
_in_theta_oscillations_in_patients_with_major_depressive_disorder._J_Neurosci._2005.pdf

This review is a good overview of how this framework can be applied to study phenomena in the cogintive domain:
http://www.cell.com/trends/cognitive-sciences/abstract/S1364-6613%2810%2900046-X

About the event-related stuff though, it is difficult to use this paradigm in such a design. Having a trial-like structure in a particular task automatically means that you are introducing rhythmicity over longer time periods, which interferes with endogenous autocorrelations. It would be possible in tasks that are subject-driven (e.g. bistable perception, binocular rivalry, and such) but there are no studies to date which have done this that I'm aware of.

[Anonymous No.4464904]

>>4461897
jesus fuck, this thread... whoa

[Anonymous No.4464925]

What do you do OP? Neuroscience major?

[CNS !MT.CNS.YRY No.4464935]

>>4464925
PhD student in cognitive neuroscience.

[Anonymous No.4465189]

>>4464935
bumping this

[Anonymous No.4465191]

>>4465189
It's archived.

[Anonymous No.4465202]

>>4465191
right, but I want more

[Anonymous No.4465211]

You are my favorite motherfucker

<3

[Anonymous No.4465220]

I think I understood the gist of this but I'm also very very dumb. Would it be correct to say that "mind" is a definite emergent property of the brain?

[mtp !EKFQOBEsKo No.4465314]

bump for sci

[Anonymous No.4465325]

CNS confirmed for is-the-rest-even-trying-tier tripfag

[Anonymous No.4465955]

Oh hey what's going on in this th- OH MY GOOD LORD SWEET JESUS ITS MAGNIFICENT

[Anonymous No.4466042]

>>4462496
>power law statistics
What is this so called power law?

[Anonymous No.4466085]

>>4465325
/sci/ shits on all the tripfags even if they help others. So bitch please, even if what you say is true, refrain from expressing it that way.

[Anonymous No.4466137]

>>4466085
No need to get mad. I was just pointing out that this thread is awesome.

[Anonymous No.4466150]

>>4466042
http://en.wikipedia.org/wiki/Power_law

[Anonymous No.4466155]

>>4466137
I know, it is awesome. I just wished people on /sci/ didn't hate on so much tripfags and encouraged them to share their actual knowledge instead by giving them the idea to create threads like this!

[Anonymous No.4466376]

>biology
>a hard science

[Theoretical Physician No.4466482]

>>4461910
In the image it looks superficially as though the only distinction is of scale.

Also having a bit of trouble telling apart critical and super-critical conceptually.

[Anonymous No.4466500]

>>4464883
Thanks for the links to the studies.

[Anonymous No.4466526]

>>4464883
Btw, do you really think that there is a state in which the brain can be said to be at rest?

There are some people who doubt this. What is more, the "default network" might be nothing but the result of gradual usage, which becomes "the default" mode of expectation of incoming stimuli. For example the posterior part of the cingulate is considered to be part of the default network while the anterior part of the cingulate is rapidly activated once attention needs to be focused on controlling how that stimuli is dealt with. It seems in this case the posterior cingulate plays the role of "background expectational awareness" for possible incoming stimuli.
Not sure I explained it very clearly, but what I'm saying is that the resting state might be only a preparatory state for possible arousal and attention. Wouldn't that qualify as rhythmicity in this case?

[Anonymous No.4466536]

>>4466526
*how the stimuli are dealt with.

[Anonymous No.4466544]

>>4464883
I'll have to read this material some time and see if I can apply any of it to my own work. Thanks.

>>4464866
>the neurophysiological properties of individual neurons
I agree with this, but I'm not as knowledgeable as I'd like to be. Can you point me in the direction of useful information?
>I'd like to analyze the functions of neurons (and subsequently the functions of groups of neurons, etc) in order to create a worth-while AI.

[Anonymous No.4466562]

>20 TED talks about the brain
http://neuroethicscanada.wordpress.com/2011/02/03/20-ted-talks-about-the-brain/

>Just posting relevant link to add to the discussion

[Anonymous No.4466721]

>>4461910
>collective spatiotemporal correlational patterns
>mfw