Rewrite the textbooks: Findings challenge conventional wisdom of how neurons operate

Feb 17, 2011
Brain diagram. Credit:

( -- Neurons are complicated, but the basic functional concept is that synapses transmit electrical signals to the dendrites and cell body (input), and axons carry signals away (output). In one of many surprise findings, Northwestern University scientists have discovered that axons can operate in reverse: they can send signals to the cell body, too.

It also turns out can talk to each other. Before sending signals in reverse, axons can perform their own neural computations without any involvement from the cell body or . This is contrary to typical neuronal communication where an axon of one neuron is in contact with another neuron's dendrite or cell body, not its axon. And, unlike the computations performed in dendrites, the computations occurring in axons are thousands of times slower, potentially creating a means for to compute fast things in dendrites and slow things in axons.

A deeper understanding of how a normal neuron works is critical to scientists who study , such as epilepsy, autism, Alzheimer's disease and schizophrenia.

The findings are published in the February issue of the journal .

"We have discovered a number of things fundamental to how neurons work that are contrary to the information you find in neuroscience textbooks," said Nelson Spruston, senior author of the paper and professor of neurobiology and physiology in the Weinberg College of Arts and Sciences. "Signals can travel from the end of the axon toward the cell body, when it typically is the other way around. We were amazed to see this."

He and his colleagues first discovered individual can fire off signals even in the absence of electrical stimulations in the cell body or dendrites. It's not always stimulus in, immediate action potential out. (Action potentials are the fundamental electrical signaling elements used by neurons; they are very brief changes in the membrane voltage of the neuron.)

Similar to our working memory when we memorize a telephone number for later use, the nerve cell can store and integrate stimuli over a long period of time, from tens of seconds to minutes. (That's a very long time for neurons.) Then, when the neuron reaches a threshold, it fires off a long series of signals, or action potentials, even in the absence of stimuli. The researchers call this persistent firing, and it all seems to be happening in the axon.

Spruston and his team stimulated a neuron for one to two minutes, providing a stimulus every 10 seconds. The neuron fired during this time but, when the stimulation was stopped, the neuron continued to fire for a minute.

"It's very unusual to think that a neuron could fire continually without stimuli," Spruston said. "This is something new -- that a neuron can integrate information over a long time period, longer than the typical operational speed of neurons, which is milliseconds to a second."

This unique neuronal function might be relevant to normal process, such as memory, but it also could be relevant to disease. The persistent firing of these inhibitory neurons might counteract hyperactive states in the brain, such as preventing the runaway excitation that happens during epileptic seizures.

Spruston credits the discovery of the persistent firing in normal individual neurons to the astute observation of Mark Sheffield, a graduate student in his lab. Sheffield is first author of the paper.

The researchers think that others have seen this persistent firing behavior in neurons but dismissed it as something wrong with the signal recording. When Sheffield saw the firing in the neurons he was studying, he waited until it stopped. Then he stimulated the neuron over a period of time, stopped the stimulation and then watched as the neuron fired later.

"This cellular memory is a novelty," Spruston said. "The neuron is responding to the history of what happened to it in the minute or so before."

Spruston and Sheffield found that the cellular memory is stored in the axon and the action potential is generated farther down the axon than they would have expected. Instead of being near the cell body it occurs toward the end of the axon.

Their studies of individual neurons (from the hippocampus and neocortex of mice) led to experiments with multiple neurons, which resulted in perhaps the biggest surprise of all. The researchers found that one axon can talk to another. They stimulated one neuron, and detected the persistent firing in the other unstimulated neuron. No dendrites or cell bodies were involved in this communication.

"The axons are talking to each other, but it's a complete mystery as to how it works," Spruston said. "The next big question is: how widespread is this behavior? Is this an oddity or does in happen in lots of neurons? We don't think it's rare, so it's important for us to understand under what conditions it occurs and how this happens."

Explore further: Researchers reveal pathway that contributes to Alzheimer's disease

More information: The title of the paper is “Slow Integration Leads to Persistent Action Potential Firing in Distal Axons of Coupled Interneurons.”… n2/full/nn.2728.html

Related Stories

What makes an axon an axon?

Nov 10, 2008

Inside every axon is a dendrite waiting to get out. Hedstrom et al. converted mature axons into dendrites by banishing a protein crucial for neuron development. The results suggest that this transformation could occur after ...

Neurons found to be similar to Electoral College

Sep 14, 2009

A tiny neuron is a very complicated structure. Its complex network of dendrites, axons and synapses is constantly dealing with information, deciding whether or not to send a nerve impulse, to drive a certain action.

Research could lead to new treatments for brain injuries

Nov 20, 2007

MIT researchers have identified a family of proteins key to the formation of the communication networks critical for normal brain function. Their research could lead to new treatments for brain injury and ...

Autopilot guides proteins in brain

Apr 21, 2009

Proteins go everywhere in the cell and do all sorts of work, but a fundamental question has eluded biologists: How do the proteins know where to go?

Study: Marijuana may affect neuron firing

Nov 29, 2006

U.S. scientists have discovered the active ingredient in marijuana interferes with synchronized activity between neurons in the hippocampus of rats.

Recommended for you

User comments : 15

Adjust slider to filter visible comments by rank

Display comments: newest first

3.6 / 5 (7) Feb 17, 2011
Once again rewriting the textbooks on neuroscience. Looks like they might have to redesign neural nets.
2.3 / 5 (3) Feb 17, 2011
This is how computers become intelligent
3.9 / 5 (7) Feb 18, 2011
"This cellular memory is a novelty," Spruston said. "The neuron is responding to the history of what happened to it in the minute or so before."

Is this similar to how memristors work? maybe you could use them to simulate a physical neural net.
3.8 / 5 (5) Feb 18, 2011
Time to update the Blue Brain project
3 / 5 (2) Feb 18, 2011
Does this explain different cycles of action & thought?

Sometimes parts of the body react quickly to external stimuli, e.g. before we are consciously aware of the stimuli.

Is there a connection between slower axon communication and conciousness, or am I dreaming ?
2.5 / 5 (4) Feb 18, 2011
Sounds like there is still a lot we don't know about how a brain works. Which means our attempts to recreate it electronically, like electronic neurons and neural nets, are still not recreating what the brain does. That would explain why they don't simulate a brain very well yet. I would think that once we know how the brain works we can simulate it electronically, I also still don't think even that will become self aware. Even if it did how would we know it wasn't just simulating self awareness?

We may need a new term for it like apparent self awareness. If it happens there will be 2 camps that would argue about it forever, one will say it is and one will say it isn't. Just like a lot of other things.
3.8 / 5 (4) Feb 18, 2011
. Looks like they might have to redesign neural nets.

Not really since the neural nets in computer science are highly abstract.

Neural nets only model connections from node A to node B. The connections are parametrized by signal strengths and transmission times (in the simplest cases). In order to incorporate 'slow' axon action you just need to model each biological axon as two connections: one slow and one fast.
To model the 'axon computation' mentioned in the article you insert a virtual node in each connection (i.e. you treat each connection also like a node).

Is this similar to how memristors work? maybe you could use them to simulate a physical neural net.

Neural nets are modelled in software, not hardware (there used to be some dedicated hardware but it has long been superceded by software modeling). Hardware has the disadvantage that it's tricky to make/break connections to other nodes or insert/delete nodes dynamically.
3 / 5 (4) Feb 18, 2011
Neural nets are modelled in software, not hardware

Yes, but software simulation is much slower and so the more complex the simulation, the slower it becomes.
Hardware has the disadvantage that it's tricky to make/break connections to other nodes or insert/delete nodes dynamically.

Also true, which is why it's difficult to make progress. FPGA's have the reconfigurability, but simply can't approach the complexity required.
3.8 / 5 (4) Feb 18, 2011
Another problem with FPGAs is that you can't easily model signal strength nor signal time very well. Both are pretty much determined by the the hardware. In software the simulation is much easier - and with parallel processing not noticeably slower. Also you are only limited by the size of your memory as to the amount of neurons/connections you can have. with swapping (which slows stuff down, though) you're not even limited to that.
4 / 5 (4) Feb 18, 2011
electronic neurons and neural nets, are still not recreating what the brain does

Actually, it was thought for long that the backpropagation algorithm wasn't biologically plausible because it needs backwards connections. This study shows it's indeed the case and brings more biological plausibility to the Multi Layer Perceptron (just one type of artificial neural network among many).

antialias already gave excelent explanations about how artificial neural networks are abstracted from natural ones and how to simulate some behaviors. I just want to add that the slow connections could be (and already are) modeled by introducing unitary delays.

Another interesting thing is that there is a model of artificial neuron that uses "leaky integration". In other words, it reacts to stimuli history using a local decaying memory, exactly as the study has found.
4.3 / 5 (6) Feb 18, 2011

Simulating self awareness and being self aware are equal; as long as the simulation is flawless. You can only ever be sure of your own consciousness anyway.
2 / 5 (1) Feb 18, 2011

Simulating self awareness and being self aware are equal; as long as the simulation is flawless. You can only ever be sure of your own consciousness anyway.

They are no more equal than an orange is to one made of antimatter. Looks the same, acts the same, but isn't the same. As for the last statement, I've said that more than once.

Evidently considering the marks I got for my comment I was wrong about us not knowing how the brain works and the mark givers must have already known what was in this article. What's more you must already know more than researchers now know so why don't you enlighten us?
not rated yet Feb 19, 2011
They probably should have figured this out when they noticed that chemicals being processed by the brain don't travel one way through the synapse, but instead bounce back and forth until they're used up.
1 / 5 (2) Feb 22, 2011
This article is fascinating! Is it perhaps a 'mirror of neural pathways' that are yet to be revealed? I believe that this will ultimately be proven. The human mind is quantum! We are in the universe.. and the universe is inside us. We are the 'dark matter' and the 'super novas'. We are the 'black holes' of intense gravity.. and yet lighter than photons. Logic, reason, and creativity. All three are hardwired in the human mind. A quantum leap is now in order. It only requires truth and spirit to carry it forward.
1 / 5 (1) Feb 22, 2011
I'm thinking that this could lead to the discovery of where our consciousness resides.