The adult human brain contains approximately 100 billion neurons which would equate to a pc storage capacity of about 100 gigabytes.If we consider that the brain as directly addressable memory (ie like the RAM in your pc) then you would need 50 to 100 ram slots to take 1gb or 2gb RAM modules and at a minimum a 64bit operating system to address it all. Your motherboard would also be significantly bigger and need a chipset to address the memory as well as having a significantly larger power supply for the RAM and additional fans to cool it all.If you had a whole separate board for the RAM your pc might be 2 to 3 times its current size.If however a neuron is more of an on/off switch (1 bit) as opposed to a memory byte (that occupies 8 bits) we can reduce our memory requirements down to a more manageable 12.5gb.However the human brain is much more than a storage medium. It acts more like a whole pc. Storage, computation, power supply, cooling/heating, automated and ancilliary tasks, input/output and so on.The power of the brain is not so much in the number of neurons, it is in the connections of those neurons. So whilst you might have 100 billion neurons (less as you age) you have untold trillions of connections between neurons and that is something that has not been duplicated in a computer yet and won't be for a very long time.

I agree with the rough concept of visualising brain "power" by scaling up the capabilities of a single neuron (to a certain extent). However, I very much doubt that the way a neuron takes part in computation/storage is in any way similar to a digital computer. A single neuron may have connections to many neurons which menas that as the number of neurons goes up, the theoretical maximum computation/storage goes up in an exponential way.IE:1 neuron, 1 possible combination2 neurons, also one3 neurons, 4 possible combinations4 neurons, 8 possible combinationsThis is an extremely simplified view and is partially guesswork as the exact way a brain computes/stores data is not exactly fully documented.I have, however, heard it said by someone of reasonable qualifications (I forget exactly who...) that the brain processes at around 10GHz. Whether each "Hertz" is a 1+1=2 - like operation or something more ethereal (such as "emotional potential x" + emotional potential y" = "emotional response z") is anybody's guess.

This answer was selected and edited by New Scientist staffWe can only answer this question if we assume the human brain is like a computer. For example, if each neuron holds 1 bit of information then the brain could hold about 4 terabytes (4000 gigabytes).However, each neuron might hold more than 1 bit if we consider that information could be held at the level of the synapses through which one neuron connects to another. There are about 50,000 synapses per neuron. On this basis, the storage capacity could be 500 terabytes or more. But these are perhaps misleading answers because the human brain is not like a standard computer. First, it operates in parallel rather than serially. Second, it uses all sorts of data-compression routines. And third, it can create more storage capacity by generating new synapses and even new neurons.The brain has many limitations, but storage capacity is not one. The problem is getting the stuff in and, even more problematic, getting the stuff out again. We can demonstrate that storage capacity is not the problem if we consider the technique experts use to remember the order of a shuffled pack of cards. This technique, called the "method of loci", goes back to classical antiquity. It involves imagining a journey in which each card appears at a certain location.Here is an example that I found on the internet: the first card is an 8 of clubs. To memorise this you imagine going out of your front door, the first step on the journey, and finding your path blocked by a person smashing an egg timer (which is shaped like an 8) to pieces with a mallet (a club). The next card is then placed on the next step of the journey with an equally vivid image.What is striking about this technique is that the story you create to remember the order of the pack of cards contains much more information than the simple pack of cards you are trying to remember. The vivid images are necessary to get the information into our brain and to get it out again later.Chris Frith, Wellcome Trust Centre for Neuroimaging, University College London, UK

Another calculation, based on how much information the synaptic connections of the brain represent. AFAIK the neurons all behave in a standard manner, so they don't hold any information themselves - it's all in the synapses.Neurons per brain = 10^11Bits to address a neuron = log2 (10^11) = 37Synapses per neuron = 50 000Bits to specify whether synapse is positive or negative in its effect = 1Total bits to specify all synapses of a given neuron = (37+1) * 50 000 = 1.9 x 10^6 bitsTotal bits to specify all synapses of all neurons in a given brain = 1.9 x 10^6 x 10^11 = 2.375 x 10^16 bytes = 23 750 000 GB = 24 petabytes approx.This estimate is a vast oversimplification - I'm a computer scientist, not a brain scientist.

Everyone's had a good go at answering this but the correct answer is :"Incompatible mediums"."How fast can a human fly"