Sledgehammers. They provide gross oxygen levels provided by the blood in particular brain regions. A good control has to be given to subtract the actual data recordings from - because the brain is constantly activated all over, and constantly provided blood. If the controls are wrong, the study is usually not proving anything, though sometimes the reader has to figure it out.
That said, fMRIs also assume that in different people, the brain is organized EXACTLY the same, which just isn't true - if it were, then among other things, we'd all have the same memories.
So, let's find another source before we believe anything an fMRI study suggests.
And don't even get me started on which brain region "does" what. Because even that isn't exactly well established.
For example, how is this possible: http://www.eurekalert.org/images/release_graphics/pdf/brain.pdf (PDF warning).
So, if we seem to have a cell that responds to 1, then to 2, then to 3, does it really make sense? This is like the classic "grandmother cell" argument in the visual system - that we have a single neuron that fires when we see our grandmother. This is, of course, not true. So let's think - how many cells would it take to recognize every number individually?
Too many.
This seems to be a classic case of "labeled line" theory vs "population coding". It's far more likely that a population of neurons light up to mathematical quantities and different populations represent different numbers. That way you can recognize everything in a much more compact sense.
Oh... and we're probably not wired for any particular form of math (in terms of base). That would be pretty arbitrary and unhelpful, though no one's proven it, so it could be true... but I doubt it.
Slashdot Mirror
Sledgehammers. They provide gross oxygen levels provided by the blood in particular brain regions. A good control has to be given to subtract the actual data recordings from - because the brain is constantly activated all over, and constantly provided blood. If the controls are wrong, the study is usually not proving anything, though sometimes the reader has to figure it out. That said, fMRIs also assume that in different people, the brain is organized EXACTLY the same, which just isn't true - if it were, then among other things, we'd all have the same memories.
So, let's find another source before we believe anything an fMRI study suggests.
And don't even get me started on which brain region "does" what. Because even that isn't exactly well established. For example, how is this possible: http://www.eurekalert.org/images/release_graphics/pdf/brain.pdf (PDF warning).
So, if we seem to have a cell that responds to 1, then to 2, then to 3, does it really make sense? This is like the classic "grandmother cell" argument in the visual system - that we have a single neuron that fires when we see our grandmother. This is, of course, not true. So let's think - how many cells would it take to recognize every number individually? Too many. This seems to be a classic case of "labeled line" theory vs "population coding". It's far more likely that a population of neurons light up to mathematical quantities and different populations represent different numbers. That way you can recognize everything in a much more compact sense. Oh... and we're probably not wired for any particular form of math (in terms of base). That would be pretty arbitrary and unhelpful, though no one's proven it, so it could be true... but I doubt it.