Electrical properties of dendrites help explain our brain’s computing power

http://news.mit.edu/2018/dendrites-explain-brains-computing-power-1018

by magoghm • 8 years ago

180 48 8 years ago

news.mit.edu

There's quite a lot of fluff here, but I think the actual story is that some MIT researchers have been using novel techniques to do electrophysiology on dendrites?

Which is cool; I just wish the article had gone into more detail on that. (and maybe included references :-) )

gumby• 8 years ago

It’s not an article, it’s a press release from MIT’s publicity office full of stuff reporters can pull out to write their own articles without attempting to understand the paper.

The actual paper is here: https://www.cell.com/cell/pdf/S0092-8674(18)31106-1.pdf

Kim_Bruning2• 8 years ago

thank you! :-)

lamename3• 8 years ago

Dendritic recordings per se aren't novel, but the human brain aspect here is.

hliyan3• 8 years ago

electrical signals weaken more as they flow along human dendrites, resulting in a higher degree of electrical compartmentalization, meaning that small sections of dendrites can behave independently from the rest of the neuron

Reading this, I just realized that all this time I was using the wrong computing analogy to understand the brain: neurons are not the transistors of the brain, dendrites are!

teekert• 8 years ago

To be honest, using a transistors analogy may keep you on the wrong path for a long time as well :) There is not a lot of "digital" going on in those cells...

JoeAltmaier3• 8 years ago

Wait - transistors are not digital they're analog. And neurons are terribly digital, acting as rate controllers and pulse counters and dividers and so on.

buboard4• 8 years ago

It is and it isn't. There is a two-layer model of neuron that captures some properties like firing rates of hippocampal neurons, but there are also alternate models. E.g: https://static1.squarespace.com/static/5267aed6e4b03cb52f5e0...

hliyan3• 8 years ago

Thanks for this! I'm not in the neuroscience field, but the title itself tells me that I've a misconception that needs correcting.

michrassena• 8 years ago

Does this imply that the neuron is more of a power supply, and the properties of sections of dendrites (not the whole dendrite) and their connections is what does the actual computation? If so, that adds a layer of messy complexity which is fairly typical in living things.

achileas3• 8 years ago

Both. Because neuroscience is hilariously complex and a big "fuck you" to neuroscientists.

Computation happens at many levels - among dendrites, among axons (some neurons have multiple axons as well!), axonal branches, within the neuron, and also among various networks that arise (behavioral and structural networks). It's all a bit much, if you ask me.

protonfish2• 8 years ago

Dendrites are just a specific part of a neuron (1), so you are no more correct.

More importantly, neurons are nothing like transistors. Each neuron contains significant computing power. For example, in the retina of a frog there is a neuron that accepts as inputs dendrites from an area of light-sensitive cells (rods and cones) and only fires when a small, round, moving object is in the field of view. These neurons are colloquially called "bug perceivers." (2)

1 https://en.wikipedia.org/wiki/Neuron

2 "What the Frog's Eye Tells the Frog's Brain" (1959)

ivan_ah• 8 years ago

Link to journal article: https://www.cell.com/cell/pdf/S0092-8674(18)31106-1.pdf

buboard4• 8 years ago

Oddly we know a lot more about the electrophysiology of neurons of mice than of humans. Compartmentalization of synaptic integration is known to have computational consequences, e.g. see Bartlett Mel's work : http://www.pnas.org/content/111/1/498.short

raverbashing• 8 years ago

Well, as the article mentions, getting a viable sample "to play with" seems to be hard.

lamename3• 8 years ago

I don't find it odd, I don't plan to volunteer my living brain for patch clamp experiments :)

buboard4• 8 years ago

i dont think you can do that

ItsMe0000013• 8 years ago

Wouldn't one use two-photon microscopy? Patch clamp is quite limited in comparison.

There also is the combination of both methods - not surprisingly, they write how hard it is in a living brain (the clamping, the two-photon microscopy part is much easier): https://www.the-scientist.com/daily-news/robotic-patch-clamp...

Here is an article describing "three dimensional two-photon brain imaging in freely moving mice using a miniature fiber coupled microscope": https://www.nature.com/articles/s41598-018-26326-3 (the experiments that I was aware of all still were done on a sedated animal with a fixated head).

tmsldd• 8 years ago

funny how neuroscientists use terms such as "computation" and "(more) computing power", with no proper definition

arghwhat• 8 years ago

Floating point operations, and more FLOPS.

Definitely.

anonymfus• 8 years ago

How bad these news are for studies of human neurodegenerative deseases on animal models?

achileas3• 8 years ago

Not really. Much of the research doesn't focus on the electrophysiology of single neurons, but rather how networks are impaired and what causes the impairment.

maerek• 8 years ago

Is it possible to strip the Facebook click ID from URLs submitted to Hacker News?

The link for the story includes the following: fbclid=IwAR2szOstJ6_hkoar2mo8NkXXMaOnfnIS5rFq5YNcOPf397n5HctnSUCGHjk#.W86CSfP5s-9.facebook

dtornabene• 8 years ago

For real though, this is kind of gross and it'd be nice (and super easy) if hn could strip this off.

sctb• 8 years ago

Thanks! We've added that to our list of strippy things (AMP is up next) and removed it in this case.

Symmetry• 8 years ago

They're comparing human brains to rodent brains here. Suzana Herculano-Houzel did some interesting work recently that showed that human (and primate, and bird) brains have neuron counts that scale linearly with volume while most mammals only scale something like the 3/4 power of volume. So I wonder how a crow or lemur brain would look in this regard.

Balgair2• 8 years ago

Dr. Herculano-Houzel is a real life Indiana Jones of the brain. She's an absolute jewel in science. I remember her telling a group of us about her efforts in Botswana(?) to get elephant brain specimens. Normally, to preserve a brain, you pump fixative in via the arteries at low concentrations. Even in a mouse, this takes a few hours. For megafauna, it can take a few days. To get good elephant data, she told us that you have a new problem: scavengers. You take the elephant (per game-park regulations) and then get the skull off and gain access to the massive arteries. Then you have to find a tall enough tree to hang fixative from, as you can't get a pump out into the veld that will run for days. You just have to let gravity do the work. Then you slowly fix the brain in the skull in some random place in the bush. However, the rest of the elephant is very good at attracting lions, hyenas, etc at all hours of the day. So you have to fend off the skull from very hungry predators. It's not a 'fun' experience, in her recollection. I remember something about 4-bore rifles being used a fair bit. Finally you have to get the brain out of country, which, depending on international regulations and local bribery customs, can also be difficult.

Honestly, this woman is AMAZING. Here's her website: http://www.suzanaherculanohouzel.com/lab

Nelkins• 8 years ago

Would be interesting to see if genetically engineered rodents with similar neuronal characteristics have improved intelligence (for some measure of intelligence).

hosh• 8 years ago

I kinda wonder if neuro-atypical human neurons behave about the same or differently -- for example, autism, ADHD, etc.

Electrical properties of dendrites help explain our brain’s computing power

Related Stories