Human brain 'works like US presidential elections'

shouldn't that be the other way round?

There have been human brains on the planet far longer than the americans have been holding elections. Anyhow, empirical evidence tells us this cannot be the case: otherwise it would take over a year and cost millions just to make a simple A or B choice.

Not convinced.

News?

I thought this was known?

This is how it was explained to me in a 20 year old book on neural nets I found in the library a while back.

In other words ...

... the right side of my brain is full of crazy voices screaming at the left side?

You mean ...

It has no security and utilises an Access database? GASP! <clutches pearls>

The work like US Elections

but only if the brother of the idea is in control of the last dendrite to report and magically the idea gets through despite actually losing the vote.

RE: is this new? or news

I think this is actually new but the distinction is fairly subtle. I believe it has always been assumed that each synapse contributes equally to the overall somatic potential of the neuron (and therefore whether or not an action potential is fired).

What they're saying is that each synapse only contributes to the potential of its particular dendritic tree. The potential of each dendritic tree then contributes to the overall somatic potential in a way that is not the same as if each synapse contributed independently, though I'm not sure what the mechanism is that makes it different because I can't be bothered to read the article.

Re: is this new? or news?

There are two elements to the analogy. The first is the winner-takes-all principle of operation, which I was taught at school about 30 years ago. So *that* isn't news. The second is the analogy with the US electoral system, which gathers these into a hierarchy (which again is not news) with exactly two levels. (Voters->electors->president) This would be such an astonishing limitation on the brain's complexity that I am willing to bet that it isn't true. (I'm pretty damn certain, for example, that it is *not* true for the visual cortex, which is one of the simpler parts of the brain that is relatively well understood.) So, no, it's not news. Depending on what "it" is, it is either several decades old or it is wrong.

But as I'm not a subscriber to the journal, I cannot tell whether the fault lies with the researchers, the journal's press corps, or El Reg.

Re: Errm....

You're thinking of Gerald Ford, who took over when Nixon finally went under to Watergate.

Actually Ford went one better, having only got the Vice Presidency when Agnew quit over tax-evasion charges, so he wasn't even elected to that either (Presidents and their Veeps come as a boxed set at election time.).

From this side of the pond I recall him being quite good in the job, very statesmanlike, but the American electorate never forgave him for pardoning Nixon and replaced him with a peanut farming plonker.

@Rosco

"I think this is actually new but the distinction is fairly subtle."

Ah. Too subtle for me, but upon re-reading the article I take your point. Humble pie for lunch, then.

Boo @ El Reg..

Dear El Reg,

"In the US electoral college, people in the many states vote using various different systems - 'winner takes all', congressional district, dimpled chads etc - to choose electors who then select candidates for president."

PLEASE get your facts correct before publication. The above sentence is entirely wrong. We are daft enough to NOT vote for our electoral college. This is one of the BIGGEST problems with our national election system today.

Or to put another way, the general citizenry does NOT vote for the electoral college.

Old news

I studied neural networks at university 10 years ago and I'm pretty sure I remember the brain's decision making process being described in this way. I don't think it was revolutionary at that time, so which part of this is means to be new?