The Spike
Published in

The Spike

Now We Can Record One Million Neurons At The Same Time

A Dialogue

Credit: Image by Gordon Johnson from Pixabay

Have you heard? Scientists at Rockefeller University have just announced they can record one million individual neurons in the cortex of a mouse — at the same time!

Wow, one million. Is that a lot?

We got very excited about being able to record 10,000 neurons in the mouse brain just two years ago. And the biggest recordings of all so far have been tens of thousands of neurons in the brains of baby zebrafish.

So this is a huge leap! Why baby zebrafish, incidentally?

Because they have see-through heads. You can literally see the brain through their skin, which makes filming their brain a lot easier.

Ah, good to know, should I ever feel the need to TikTok a small tropical fish. This one million neurons sounds like it’s quite the advance.

You don’t know the half of it: a mouse cortex has about 10 million neurons in total. So this recording is ten percent of its entire cortex. Ten percent.

Blimey, so we’ve jumped from 0.1 percent to ten percent in a single bound. Human ingenuity is a ceaseless wonder. What revelations in understanding the brain can we expect, now we can see the messages, the spikes, that one million neurons are sending each other?

Er, it’s calcium imaging, we can’t see what the neurons are sending each other.

Wait, what?

All the ways we have of recording thousands of neurons use calcium imaging: we put a chemical in neurons that fluoresces according to how much calcium is present, and film the brain to record the amount of light each neuron is giving off. So then we know how much the calcium in each neuron is fluctuating.

And why is that useful?

Ah, because the amount of calcium in a neuron’s body is proportional to how active it is: the more spikes it sends, the more calcium there will be.

Oh I see, so with calcium imaging we can know the neuron is sending spikes, just not how many.

Exactly. And, er, of course, it can be a little difficult to tell when it’s sending just one or two spikes, so many spikes at once is best.

Do many neurons do that?


Hmmm. So what you’re actually telling me is we can get something that’s a bit proportional to spikes but not really, and it’s full of noise, from one million neurons. If I wanted the spikes, how many neurons could I record at the same time?

About a thousand neurons is our current best, using electrodes. And with those we don’t know which neurons are sending those spikes. With calcium imaging we’re looking directly at the neurons, so we know which ones are active.

Life is full of compromises. Good to know neuroscience is no different. Still, a complete read-out of the changes in neural activity from one million neurons, that’s quite something.

Er, not quite complete. Imaging one million neurons could only be done at two samples each second.

But some cortical neurons fire tens of spikes each second! And don’t many neurons seem to send messages using the precise timing of their spikes? Isn’t that a bit awkward?

Yes. Like you said, life is full of compromises.

OK, so what did we learn from recording one million neurons?

Actually, um, nothing yet. The paper from the team at Rockefeller focused on their recordings of about 200,000 neurons.

Oh. Why?

Probably because analysing one million neurons is a trifle difficult. As they drolly noted in the paper “Due to the sheer amount of data, it is difficult to display the time-series of all 207,030 neurons at full resolution simultaneously”.

Fair enough. What could we learn from recording a million neurons?

Actually, two things spring to mind.

Hang on, which of us is the author’s voice?


In case one of us is about to say something controversial, then the author can claim they were the other voice.

Controversial? Like what?

Penguins are evil.

How can you be sure?

Well, in this one episode of Sarah and Duck, the penguins at the zoo were mean to them.

I hardly think a gently surreal British cartoon about a young girl and her adopted pet duck is a solid evidential basis for asserting that penguins are evil.

Fair enough. You should watch it though, it’s a joy.

If I may continue.

Sorry, do go on.

One terrific thing we could learn from recording a million neurons is how the two sides of the brain work together.

The brain has two sides, you say? You wouldn’t know it from reading most studies that record neurons.

Well, quite. And that’s because when we can only record a handful of neurons at a time, it’s a bit tricky to relate what’s going on in one side of the brain to the other side.

But with a million, we can better see both sides at the same time?

Exactly. So we can have some idea of how synchronised — or not — neurons on either side seem to be, and perhaps even get some idea about what the two sides are telling each other.

Cool. Anything else that this million-neuron breakthrough might be a game-changer for?

Yep. fMRI.

Come again?

We might get a handle on what exactly fMRI can tell us about neurons. Each little dot of colour in an fMRI image is a cube containing about 100,000 or more neurons. And the intensity of that colour shows how much excess oxygen-rich blood is flowing around those neurons.

So it’s not measuring neural activity then?

No, it measures the demand for oxygen, which we think means that the neurons demanding the oxygen want to use more energy, and they want to use more energy because they are more active.

So we don’t really know how this demand for oxygen that fMRI measures relates to the actual output of the many, many neurons in each little dot?

Exactly. But a million neurons is at the scale of fMRI. So you can imagine an experiment where the same animal doing the same, dull task has it brain recorded using both fMRI and this million-neuron scale imaging.

And then we can compare the activity in the same brain regions, and see if there is any relation?


I see! That does sound cool. And if it turns out there is no relationship?

Shush. I don’t know, go “la la la la”?

Not terribly scientific.

No, sorry.

But still. One million neurons… imagine if someday we can record all the neurons in a brain!

We already did that

What happened?

No one noticed.


You have to think of it as the start of the journey, not the end. Once the technology is in place, the real science starts. Just like being able to record one million neurons at once is the start of a scientific journey, so was being able to record all the neurons of a particular brain.

Go on then, which brain do we have a full recording of?

The Hydra’s.

Hydra? That see-through wiggly tube of a sea creature with the tentacles, the one that’s possibly immortal?

That’s the one.

You do know it doesn’t technically have a brain?

What can I tell you? Life’s full of compromises.

Like a side-order of entertainment with your deep dives into how the brain works? Then you’ll like my new book, “The Spike: An Epic Journey Through the Brain in 2.1 Seconds” published by Princeton University Press. Available now in hardback, ebook, and audiobook (listen to a sample).

For more essays on the brain, follow us at The Spike on Medium

Twitter: @markdhumphries

Enjoyed this story? Then consider signing up to become a Medium member: $5 a month gives you unlimited access to all stories on Medium, and supports all their writers. If you sign up using my link, I’ll earn a small commission:



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Mark Humphries

Mark Humphries


Theorist & neuroscientist. Writing at the intersection of neurons, data science, and AI. Author of “The Spike: An Epic Journey Through the Brain in 2.1 Seconds”