Scientists create working artificial nerve networks

Jan 28, 2009

Scientists have already hooked brains directly to computers by means of metal electrodes, in the hope of both measuring what goes on inside the brain and eventually healing conditions such as blindness or epilepsy. In the future, the interface between brain and artificial system might be based on nerve cells grown for that purpose. In research that was recently featured on the cover of Nature Physics, Prof. Elisha Moses of the Physics of Complex Systems Department and his former research students Drs. Ofer Feinerman and Assaf Rotem have taken the first step in this direction by creating circuits and logic gates made of live nerves grown in the lab.

When neurons - brain nerve cells - are grown in culture, they don't form complex 'thinking' networks. Moses, Feinerman and Rotem wondered whether the physical structure of the nerve network could be designed to be more brain-like. To simplify things, they grew a model nerve network in one dimension only - by getting the neurons to grow along a groove etched in a glass plate. The scientists found they could stimulate these nerve cells using a magnetic field (as opposed to other systems of lab-grown neurons that only react to electricity).

Experimenting further with the linear set-up, the group found that varying the width of the neuron stripe affected how well it would send signals. Nerve cells in the brain are connected to great numbers of other cells through their axons (long, thin extensions), and they must receive a minimum number of incoming signals before they fire one off in response. The researchers identified a threshold thickness, one that allowed the development of around 100 axons. Below this number, the chance of a response was iffy, while just a few over this number greatly raised the chance a signal would be passed on.

The scientists then took two thin stripes of around 100 axons each and created a logic gate similar to one in an electronic computer. Both of these 'wires' were connected to a small number of nerve cells. When the cells received a signal along just one of the 'wires,' the outcome was uncertain; but a signal sent along both 'wires' simultaneously was assured of a response. This type of structure is known as an AND gate. The next structure the team created was slightly more complex: Triangles fashioned from the neuron stripes were lined up in a row, point to rib, in a way that forced the axons to develop and send signals in one direction only. Several of these segmented shapes were then attached together in a loop to create a closed circuit. The regular relay of nerve signals around the circuit turned it into a sort of biological clock or pacemaker.

Moses: 'We have been able to enforce simplicity on an inherently complicated system. Now we can ask, 'What do nerve cells grown in culture require in order to be able to carry out complex calculations?' As we find answers, we get closer to understanding the conditions needed for creating a synthetic, many-neuron 'thinking' apparatus.'

For the scientific paper, please see: www.nature.com/nphys/journal/v… 12/pdf/nphys1099.pdf .

Source: Weizmann Institute of Science

Explore further: IHEP in China has ambitions for Higgs factory

add to favorites email to friend print save as pdf

Related Stories

Ex-Qualcomm exec pleads guilty to insider trading

6 hours ago

A former high-ranking executive of US computer chip giant Qualcomm pleaded guilty Monday to insider trading charges, including trades on a 2011 deal for Atheros Communications, officials said.

Media venture creates press litigation fund

7 hours ago

The media venture created by entrepreneur Pierre Omidyar said Monday it was establishing a fund to help defend journalists in cases involving freedom of the press.

'Moral victories' might spare you from losing again

7 hours ago

It's human nature to hate losing. Unfortunately, it's also human nature to overreact to a loss, potentially abandoning a solid strategy and thus increasing your chances of losing the next time around.

Recommended for you

Chemist develops X-ray vision for quality assurance

2 hours ago

It is seldom sufficient to read the declaration of contents if you need to know precisely what substances a product contains. In fact, to do this you need to be a highly skilled chemist or to have genuine ...

The future of ultrashort laser pulses

2 hours ago

Rapid advances in techniques for the creation of ultra-short laser pulses promise to boost our knowledge of electron motions to an unprecedented level.

IHEP in China has ambitions for Higgs factory

19 hours ago

Who will lay claim to having the world's largest particle smasher?. Could China become the collider capital of the world? Questions tease answers, following a news story in Nature on Tuesday. Proposals for ...

The physics of lead guitar playing

21 hours ago

String bends, tapping, vibrato and whammy bars are all techniques that add to the distinctiveness of a lead guitarist's sound, whether it's Clapton, Hendrix, or BB King.

The birth of topological spintronics

22 hours ago

The discovery of a new material combination that could lead to a more efficient approach to computer memory and logic will be described in the journal Nature on July 24, 2014. The research, led by Penn S ...

User comments : 7

Adjust slider to filter visible comments by rank

Display comments: newest first

Yes
3 / 5 (2) Jan 29, 2009
Somehow I have the feeling they substituted semiconductor material with neurons and created a biological equivalent of a 74LS00.
Then they did a retro alimentation to make a biological oscillator and call it pacemaker. Yeah now it has a heart and it is alive. Don't make me laugh.
Then they talk about creating a many neuron apparatus (translation biological ALU Arithmetic logic unit) to do complex calculations. This ALU could be ordered with the nand gates to make a simple computer. Its just an awful expensive and stupid way to create another computer with the wrong technology. Neurons are known for being extremely slow conducting mechanisms,
Please don't call this thinking, not even with quotes around it.
I believe this study was nice, but can now be ended.
The interesting result is the fact that magnetism can fire the neurons, but that can be due to the fact that they are ordered as a wire and all currents flow in the same direction.
E_L_Earnhardt
4 / 5 (1) Jan 29, 2009
Great! Top practicioners in computer programing, physics, and electronic engineering working together could, at last, present an understanding of the living cell, its function and malfunction! Alas, only the chemist can get the financing, and he's lost!
Ashibayai
1 / 5 (1) Jan 29, 2009
So...how does turning these complex biological neurons into a simple semi-conductor get us any closer to a reasonable brain-computer interface?
RobL
4 / 5 (1) Jan 29, 2009
Amazing study.
Wicked
4 / 5 (1) Jan 29, 2009
One hundred years ago, the workings of the brain might have been compared to gearboxes instead of computer circuitry. I don't think that emulating computer circuitry with neurons helps us understand consciousness for the same reason that making a gearbox out of neurons would not help us.
gmurphy
not rated yet Jan 29, 2009
an interesting study but I'm inclined to agree with the synopsis forwarded by @Yes
neurogalactus
2 / 5 (1) Feb 02, 2009
@Yes:

I totally agree. Those guys are asking the wrong questions - or more precisely, they are mistakenly equalling "linear computation" to "thinking process". It's kind of funny that those otherwise big minds haven't somehow learned yet that human thinking is an emergent process, where individual neurons are actually doing NO pre-designed calculation whatsoever. Neurons are just part of a bigger structure, and they have very limited functions, like bees in a beehive or ants in an anthill. In a brain you can remove any neuron, or even whole assemblies thereof, and still be left with a perfectly unaltered operation of the whole brain. Try that with computers.

They "enforced simplicity" on a complex system... What a load of cr*p and waste of time. Still, even that research might yield some interesting results, e.g. in the area of neuron growth control, so let's hope it's not entirely useless.