Simple mathematical pattern describes shape of neuron 'jungle'

June 20, 2012

Neurons come in an astounding assortment of shapes and sizes, forming a thick inter-connected jungle of cells. Now, UCL neuroscientists have found that there is a simple pattern that describes the tree-like shape of all neurons.

Neurons look remarkably like trees, and connect to other cells with many branches that effectively act like wires in an , carrying impulses that represent sensation, emotion, thought and action.

Over 100 years ago, Santiago Ramon y Cajal, the father of modern neuroscience, sought to systematically describe the shapes of neurons, and was convinced that there must be a unifying principle underlying their diversity.

Cajal proposed that neurons spread out their branches so as to use as little wiring as possible to reach other cells in the network. Reducing the amount of wiring between cells provides additional space to pack more neurons into the brain, and therefore increases its .

New work by UCL neuroscientists, published today in , has revisited this century-old hypothesis using modern . They show that a simple which connects points with as little wiring as possible can produce tree-like shapes which are indistinguishable from real neurons - and also happen to be very beautiful. They also show that the shape of neurons follows a simple mathematical relationship called a power law.

Power laws have been shown to be common across the natural world, and often point to simple rules underlying complex structures. Dr Herman Cuntz (UCL Wolfson Institute for ) and colleagues find that the power law holds true for many types of neurons gathered from across the animal kingdom, providing strong evidence for Ramon y Cajal's general principle.

The UCL team further tested the theory by examining neurons in the , a part of the brain where new are constantly being formed. These neurons grow and form new connections even in the adult brain, and therefore provide a unique window into the rules behind the development of neural trees in a mature neural circuit.

The team analysed the change in shape of the newborn olfactory neurons over several days, and found that the growth of these neurons also follow the power law, providing further evidence to support the theory.

Dr Hermann Cuntz said: "The ultimate goal of neuroscience is to understand how the impenetrable neural jungle can give rise to the complexity of behaviour.

"Our findings confirm Cajal's original far-reaching insight that there is a simple pattern behind the circuitry, and provides hope that neuroscientists will someday be able to see the forest for the trees."

Explore further: Gardening in the brain: Specialist cells prune connections between neurons

Related Stories

Gardening in the brain: Specialist cells prune connections between neurons

July 21, 2011
Gardeners know that some trees require regular pruning: some of their branches have to be cut so that others can grow stronger. The same is true of the developing brain: cells called microglia prune the connections between ...

Mapping the brain: New technique poised to untangle the complexity of the brain

April 10, 2011
(PhysOrg.com) -- Scientists have moved a step closer to being able to develop a computer model of the brain after developing a technique to map both the connections and functions of nerve cells in the brain together for the ...

A 3-D reconstructed image of neural dendritic trees using the advanced electron microscope technology

September 13, 2011
Neurons in the brain play a role as an electric wire conveying an electrical signal. Because this electric wire is connected with various joints (synapse), various brain functions can occur. A neuron which has dendritic trees ...

Recommended for you

'Residual echo' of ancient humans in scans may hold clues to mental disorders

July 26, 2017
Researchers at the National Institute of Mental Health (NIMH) have produced the first direct evidence that parts of our brains implicated in mental disorders may be shaped by a "residual echo" from our ancient past. The more ...

Laser used to reawaken lost memories in mice with Alzheimer's disease

July 26, 2017
(Medical Xpress)—A team of researchers at Columbia University has found that applying a laser to the part of a mouse brain used for memory storage caused the mice to recall memories lost due to a mouse version of Alzheimer's ...

Cellular roots of anxiety identified

July 26, 2017
From students stressing over exams to workers facing possible layoffs, worrying about the future is a normal and universal experience. But when people's anticipation of bad things to come starts interfering with daily life, ...

Cognitive cross-training enhances learning, study finds

July 25, 2017
Just as athletes cross-train to improve physical skills, those wanting to enhance cognitive skills can benefit from multiple ways of exercising the brain, according to a comprehensive new study from University of Illinois ...

Brain disease seen in most football players in large report

July 25, 2017
Research on 202 former football players found evidence of a brain disease linked to repeated head blows in nearly all of them, from athletes in the National Football League, college and even high school.

Zebrafish study reveals clues to healing spinal cord injuries

July 25, 2017
Fresh insights into how zebrafish repair their nerve connections could hold clues to new therapies for people with spinal cord injuries.

2 comments

Adjust slider to filter visible comments by rank

Display comments: newest first

MrVibrating
not rated yet Jun 20, 2012
Hardly an astonishing finding - it stands to reason the network is organised by efficiency.

But of course the fascinating thing is the implications for the nature of neural information encoding, as a function of this connective efficiency.

Kauffman et al covered these issues some time ago, albeit in relation to biochemistry and autocatalysing systems. It's encapsulated by his strings'n'rings thought experiment:

Suppose we have a random scatter of small rings, a ball of string & scissors. We want to connect all rings with the minimum number of strings. So we cut a length, connect two rings at random, put them back in position and continue likewise until all rings are connected into a 'supercluster'..

We find that the optimum number of strings to rings is 1:2 because each length of string has two ends, so connects two rings. If there's 40 rings we'll need only 20 strings, beyond which more become less useful. Conversely, less, and some islands will be left from the main cluster.
MrVibrating
not rated yet Jun 20, 2012
So, following that dynamic and its informational implications, perhaps optimum connectivity equates to "no information", whereas higher or lower represent "some information"..?

Sorry, pet theory alert...

(Edit: incidently, the same may apply not just to equilibrium connective (ie. spatial) entropy, but also temporal (ie. impulse freqs)...

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.