Question: Brain As A Computer/Memory Storage

[Havok]

Not sure if this has ever been calculated, but what is the amount of memory in the human brain if it were considered a hard drive or a better question, what would the information absorbed in a life time equal out to in gigabytes? (Or I'm sure gatrazillionabytes or whatever the biggest storage unit is)

[SilverWingedSeraph]

I don't think that's really quantifiable, since as far as I know the human brain, while functioning in a computer-esque sort of way, doesn't run off a string of ones and zeros and thus can't be compared to memory or hdd space on a computer, where the measurement of filesize is basically how many 1s and 0s something is made of. A bit is a single 1 or 0, a byte is 8 bits, etc. So without being able to break down the way human memory works into 1s and 0s, which we're not presently capable of doing (as far as I know?), I don't think this is quantifiable without a massive amount of guesswork.

But as far as guesswork goes, I can vaguely recall a scientist of some description saying somewhere in the region of hundreds of gigabytes, but I sure as hell can't find any source for that.

[madd0ct0r]

I remember a stat suggesting the human mind encounters more information in a Sunday newspaper then a caveman did over a lifetime, but since I'm pretty sure that stat was in aforementioned newspaper t's of dubious veracity.

[Omeganian]

Wikipedia gives 10 terabit at one page, and 2.5 petabyte at another. Both are sourced.

[SilverWingedSeraph]

Considering those are radically different numbers though, with many zeroes seperating them, at least one of them is completely wrong!

10 terabits = 10,099,511,627,776 bits
2.5 petabytes = 225,179,981,000,000,000 bits, roughly.

One is much, much larger than the other. Both are suspect. Link to the sources?

[Omeganian]

http://www.scientificamerican.com/artic ... y-capacity

The other is The Singularity is Near. Although the Wikipedia page is not quite clear whether it is the total memory.

[Rabid]

I heard somewhere that the neurones' capacity to "stock" information operated in an holographic, distributed fashion all across the brain, with nuggets of a memory dispersed all around it in a lot of the brain's inter-neuronal connections AND in some of the brain matter that-is-not-neurones-or-connections ; so I'm sure this contribute to the difficulty of determining the "size" of the Human Memory.

[SilverWingedSeraph]

http://www.scientificamerican.com/artic ... y-capacity

The other is The Singularity is Near. Although the Wikipedia page is not quite clear whether it is the total memory.

That source is no source at all, unfortunately, it's just the statement of a number with no link to any experiment or data to support it. Irritating. Also, a point against wiki for allowing shit like that as a source. Hey, I can spout numbers too, but they don't count for shit without evidence or at least an explanation as to how you came up with said number.

[Starglider]

Human brains contain around 200 trillion synapses, so the question of raw information content is essentially 'how many bits is needed to represent a synapse'. This covers the geometry and connectivity of the parent neuron's dendrite tree and the chemical properties of the synapse itself. Reasonably detailed biochemical models use around 1000 bits per synapse (specification, the working model may be more complex), which puts a complete brain specification on the order of 25 petabytes. This number isn't terribly useful for comparison though; there's a lot more raw structural detail down to the atomic level, but most of this isn't thought to be relevant to the signal processing characteristics of the neurons (just as a PC has structural complexity down to that atomic level that isn't relevant to its programs). An actual 25 petabyte raw dump would probably be quite compressible (lossy and losslessly) due to locality and self-similarity.

The question of how much storage capacity the brain can actually use for long-term memory is completely different, and a direct comparison to technological recording mediums is not really possible. Different bits of the brain specialise in storing different kinds of information, and it is stored in a highly compressed, lossy and self-refential way (because most memory recall works by activating shared prototypes and concepts, which change over time, and because holographic storage in neural nets inherently suffers from interference). There's no sensible way to compare that to a hard drive which will perfectly (until device failure), losslessly and individually store a set number of bits that can represent any sort of information in any of a vast number of possible encoding schemes.

So in summary, a little under 1.44 Megabytes, as proven in Space Quest 4 when the villain and the hero's son were downloaded onto and safely recovered from ordinary floppy disks.