Hacker News new | past | comments | ask | show | jobs | submit login
Evidence that the brain’s ability to control thinking relies on beta rhythms (news.mit.edu)
240 points by el_duderino on Jan 28, 2018 | hide | past | favorite | 63 comments



Study: Gamma and beta bursts during working memory readout suggest roles in its volitional control

Citation: Mikael Lundqvist, Pawel Herman, Melissa R. Warden, Scott L. Brincat & Earl K. Miller. Nature Communications 9, Article number: 394 (2018).

Link: https://dx.doi.org/10.1038/s41467-017-02791-8

DOI: 10.1038/s41467-017-02791-8

Abstract: Working memory (WM) activity is not as stationary or sustained as previously thought. There are brief bursts of gamma (~50–120 Hz) and beta (~20–35 Hz) oscillations, the former linked to stimulus information in spiking. We examined these dynamics in relation to readout and control mechanisms of WM. Monkeys held sequences of two objects in WM to match to subsequent sequences. Changes in beta and gamma bursting suggested their distinct roles. In anticipation of having to use an object for the match decision, there was an increase in gamma and spiking information about that object and reduced beta bursting. This readout signal was only seen before relevant test objects, and was related to premotor activity. When the objects were no longer needed, beta increased and gamma decreased together with object spiking information. Deviations from these dynamics predicted behavioral errors. Thus, beta could regulate gamma and the information in WM.


I can’t tell from this, but does this demonstrate a causal link or is the headline somewhat disconnected from the paper?

That is to say, does the brain truly rely on beta rhythms, or are they just a symptom of something else?


I think the answer is necessarily "no," because the rhythm is a sign we observe of some functional system in the brain, not the functional system itself. Similar to how your heart rate is just a sign of the beating of your heart, only the latter of which is a real functional mechanism. Something is happening in the brain that generates these patterns and which also has to do with the cognitive faculties described in the paper; I don't think anyone has a very clear idea what that mechanism is, though.

I'm not sure if this answers your question, or is just pedantry. :)


The waves come from synchronicity of neurons "firing". When neurons fire there is a flow of ions that causes a potential difference. If a lot of this happens at the same time with some periodicity you can measure this. But it would be like looking at a CPU running some math operations and determining "there is periodic activity in the ALU" (which could be due to memory reads). Now this is a poor analogy because a single neuron doesn't do "one thing" it's massively distributed. My go to argumentation for "how the brain works" (and thus consciousness) has been that it's a massive coincidence detector that tries to encode these coincidences in spike train activity using spike train dependent plasticity. In doing so the brain echoes the coincidences of the neural (sensory) input, and since these encode for some "real" cause-effect you can predict the future: memory is "echoing" in a neural substrate, consciousness is exploiting those "echos" to try to predict the future. I wrote it up in 2011 https://www.dropbox.com/s/huol1vf4j1fs1ll/mind_matters.pdf?d... I've never seen anyone poke holes in this, and even hints at the details of this paper. Guess I chose the wrong career somewhere.


Thanks. That sort of answers my question. I think that means that I disagree with the headline because this research doesn’t demonstrate evidence of the brain relying on beta rhythms to control thinking.

(From my strictly lay reading).


The study's design cannot address causality. They report a correlation between beta rhythms and the subjects' behavior.

It's unclear whether local field potentials (LFPs), of which "beta rhythms" are one specific frequency band (approximately 12-30hz), are actually a mechanism that the brain uses to transmit information. Information is known to be transmitted via the discharges of single neurons; LFPs represent the aggregate activity of tens of thousands of neurons within ~3mm of a recording electrode. It's possible that these aggregate voltage fluctuations feed back onto the activity of single neurons [0], but LFPs are generally considered an emergent property of populations of neurons as opposed to an information transmission mechanism in their own right.

[0] https://www.nature.com/articles/nn.2727


Thank you for providing useful, concise, and well-formatted study citations. I'm not in academia as of this year, and it pains me to see this information omitted from posts that would greatly benifit from this. I'll do a better job at including relevant cites in my own submissions.


I wonder how much this can be applied to day to day life. For example, the Binaural app on the iOS app store acts as a wave generator that lets you listen to specific frequencies. I'd be interested to see if you could force yourself into getting beta wave activity in the brain by just listening to beta frequencies -- almost as if to tune yourself manually.


But note that in this study the beta wave activity turned on and off in task-sensitive ways. It may not be useful to have some stimulus that increases or decreases beta wave amplitude in ways that are not sensitive to exactly what your brain needs at that very moment.

I mean, imagine you were investigating a CPU and found a pin that went high when the CPU wanted to write to memory. You wouldn't think "yeah, remembering shit is awesome, let's connect this to Vcc". No, that pin is only supposed to go high at specific times depending on exactly what the CPU is doing.


Expecting your brain to operate in a way that produces certain frequencies because your ears are detecting air vibrations with those frequencies sounds extremely homeopathic.


I wouldn't go so far as to say such an idea is on the level of homeopathy—scientist certainly do induce ("entrain") the brain to certain frequencies of operation using flashing lights. The general principle is sound.

But, I would emphasize, they usually do that by first genetically modifying the creature whose brain they want to manipulate to make some of its neurons into opto-couplers. They wouldn't be doing that if there was an easier way. :)

(It turns out that—even without the genetic modification—some frequencies of flashing light do cause the brain to do certain things. See, for example, https://cosmosmagazine.com/biology/flashing-light-clears-tox.... But the effects of this, from what I understand, aren't the same effect you would get by actually getting the brain to entrain to that frequency; it's maybe entraining the pulse-rate of the brain's glymphatic channels or somesuch, but not synaptic firing.)


This study of memory loss in mice found that exposing them to light at about 40hz triggered gamma waves and reduced plaque formation.

https://news.mit.edu/2016/visual-stimulation-treatment-alzhe...


We perceive visual stimuli as distinct events (up to a certain frequency), but when hearing, a 40 Hz sound isn't perceived as 40 individual sounds per second.


Perhaps because it isn't? Waves are not discrete structures. 40Hz means you have 40 wave cycles per second, not 40 discrete sounds.


That's kinda what I'm getting at. My understanding is that brain waves are more akin to a CPU clock, with various processes being synchronized, than to a sound wave. If that's the case, then you wouldn't have any reason to expect sound to make a difference in your brain's clock rate, but you could reasonably expect light pulses to.


Dismissing it based on nothing more than a feeling, and no data, is just as irrational as homeopaths.


No it isn't. You should default to disbelieving such extraordinary claims.


Who is the arbiter of whether something is "extraordinary"? You? Perhaps you should do some cursory investigation on published research before being so quick to judgement.


To be clear, we're talking about binaural beats. There's no such published research showing any kind of results. I haven't been "quick to judgement", binaural beats have been on the level of homeopathy for over 20 years now.


You should default to having no strong opinion at all.


Skepticism due to insufficient evidence and lack of obvious plausibility isn't usually considered a strong opinion.


There is evidence though.


Whether or not there is evidence of this particular thing is irrelevant, because this subthread is about whether skepticism is just as irrational as blind belief when confronted with unsupported claims.


Better still, default to having a strong opinion only with good reason, and the capacity to quickly give it up when presented with a more compelling view.

And then of course... we shouldn't expect to be able to sum up highly effective heuristics for understanding the world in a single sentence.


Turmeric added to coffee adds twenty years to your lifespan.

Are you completely ambivalent to such a ridiculous claim? No, you should disbelieve it unless you get a sufficient amount of evidence.


To be fair, science thrives on curiosity. I'd prefer we establish something was safe and have at the experiments.

So, disbelieve in part by trying to disprove. Win win.


Science is not being advanced by disproving crackpots. It's a waste of precious resource better spent elsewhere.


What's your definition of a crackpot? Someone who researches something varjag has a negative emotion towards?


Someone who maintains something is true while lacking any evidence of that. There's an infinity of possible untrue statements, but only finite amount of research resource. Disproving every crackpot theory doesn't advance actual science one tiny bit.


You'd be surprised. :)


Don't think I would.


If that were the norm we'd have much shorter lifespans.

edit: https://en.wikipedia.org/wiki/Iatrogenesis


No, the absence of a proper mechanism is akin to homeopathy. Pointing the absence of a proper mechanism sounds like common sense.


The burden of proof is on the person making the claim, not on the skeptic.


Actually a thing...

Auditory Beat Stimulation and its Effects on Cognition and Mood States

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4428073/


The proposition "listening to music affects your mood" is very different from the proposition "listening to a certain frequency of tone causes your brain to produce waves at a corresponding frequency."


Sure but who said anything about music? The word music doesn't appear in the original link or in any other comment on this page, or in the link I posted.


Unless you're denying that drummers are musicians, it's my impression that beats are a kind of music.


https://en.wikipedia.org/wiki/Beat_(acoustics)

If you count any sort of sound as "music", then sure.


Large amounts of sensory input with a certain rhythm affecting brain firing is known to happen; see seizures. And the idea has a pretty obvious mechanism. So your comparison is not fair at all.


Your statement is as lacking in evidence and reasoning as you suppose the one you are attacking is.


I liken it to current scientists trying to figure out how a modern motherboard or CPU works by sticking oscilloscopes onto individual circuits. Sure, you can get a readout of the fourier transforms and thus band frequency power of all the little signals going back and forth the little metal circuits and stuff. But it's really a small piece of the puzzle.

To wit - a lot of song and dance has been made regarding beta power in the subthalamic nucleus in Parkinson's patients (mostly because it is the only place we are FDA-allowed to go in a live human brain). Some ambitious researchers are claiming beta power is the "pathological signal" of Parkinson's disease...whereas it's more likely it's an epiphenomena of deeper computational issues with a damage basal ganglia circuit'


I would love to see someone reply with personal experience. I have tried similiar apps for extremely brief periods and could feel a change.


Personal experience is the bane of science. And plural of “anectode” is “anectodes”, not “data”.

Having worked in the field for years, I am pretty sure that listening to some sound at some frequency might induce that rhythm in the appropriate part of the brain (I have worked on SSVEP which is the same thing for visual stimuli, and is easily detectable, there is also some evidence that it could work for sensory motor stimuli).

However, inducing some frequency somewhere in the brain is pretty useless. Beta activity has been known for some time as the “activity” indicator. But this is simply because that is the working frequency of the brain. Active means just that, is it good is it bad? That depends.

One thing that strikes me in the field is how sure people are that anything “induced” in the brain will be positive. If the binaural beats thing really worked, then you would have as much chance to create depression as a sense of calmness.


> Personal experience is the bane of science. And plural of “anectode” is “anectodes”, not “data”.

And yet, eugenics was kind of heavily touted by scientists precisely because of an inability to accept as valid the personal experiences of the populations it targeted.


The placebo effect is still an effect.


Yup. I use this app sparingly, and I feel a mild effect. And actually I don’t care if it’s placebo (likely) or not, since it kind of works as I want


>I'd be interested to see if you could force yourself into getting beta wave activity in the brain by just listening to beta frequencies -- almost as if to tune yourself manually.

Does it actually carry over? From what I understand about the speech recognition, our brain must be doing some equivalent of the Fourier transform fairly early in the pipeline (i.e. instead of a beta-wave, our brain will just get a more or less constant signal that this part of the audible spectrum is currently active).


I just finished reading Wait But Why's excellent article on NeuralLink https://waitbutwhy.com/2017/04/neuralink.html

I wonder if these top-down beta rythmes would force out implanted data similar to white blood cells and viruses.


> There are millions of neurons in the brain

Why would they use a different order of magnitude. Surely they know that there are billions of them (100 bil in fact).


If anyone here is interested in learning more about Neurotechnology, I co-founded an organization called NeuroTechX which brings together neurotechnology enthusiasts from across the world. You can find out more about us at http://neurotechx.com/


Radiolab had an interesting episode on the gamma frequency, possibly stimulated by light or sound and a possible relation to Alzheimers.

https://www.radiolab.org/story/bringing-gamma-back/

Makes one wonder about the effects of raves.


It seems almost definitional that this effect is correlative and not causative unless the authors are positing that there are structures in the brain whose purpose is to generate brain waves or that we just massively misunderstand the function of neurons.

I read the previous paper summary linked to and I didn't see a way these waves are supposed to affect the neurons. Is there absolutely any theory as to the mechanism? Seems like bad science and reasoning even if the topic is interesting.


This was my reading of the paper/summary too - a demonstration of correlation but not the evidence of a causal link that the headline suggests.

My understanding of neuroscience is limited however. Can someone with a better understanding point me at what I’m missing?


Your comprehension seems spot on to me, though maybe you feel like you're missing something given the apparent discrepancy between the hype vs. value of this research. I think there are two relevant factors here:

1) MIT is very good at publicity and press.

2) Causal neuroscience experiments are still technically challenging to implement in complex behaving organisms (e.g. monkeys and humans). We have methods such as microstimulation (invasive) and transcranial magnetic stimulation (non-invasive), but these methods are pretty ham-fisted. They indiscriminately activate (or deactivate) all the neurons in and passing through a given volume in the brain (optogenetics is promising in this regard, but it's still a nascent technology in monkeys). So while this study may appear weak because it's correlational, it may actually represent a meaningful step forward in the field.


Is the current thought that brainwaves affects the brain? What's the mechanism for this? They aren't even really waves like we think of radio waves or magnetic waves. Saying that brain waves have a "function" like a Scientific American article seems so backwards. I'm a little confused why the language used around them is of casuality.


Thank you. That’s pretty much spot on for my thought process and interesting, possibly even positive for the future. At least I’m not going crazy. Or if I am, I’m not alone :)


Is there any data about the tendency towards of beta rhythms vs not in different cultures around the world?


Since brainwaves are associated with brain disorders and brain disorders are associated with genetics, it would be worth (albeit dangerous politically?) to investigate the link between brainwaves (and by extension, different cognitive faculties) and ethnicity.


This assumes that brainwaves function the same in everybody and that their measurement means anything more than phrenology.


Do you base this claim off of any scientific evidence? Measuring schizophrenia based off of gamma oscillations is standard clinical practice.

https://www.ncbi.nlm.nih.gov/pubmed/27732692

https://www.ncbi.nlm.nih.gov/pubmed/16571550

https://www.nature.com/articles/npp201617.pdf?origin=ppub


I didn't mean to imply they were totally useless just that too much might be read into them. I guess i did use a bad comparison.


Also that "brainwaves" have meaning beyond the behavior, brain region, and context in which they're being examined.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: