If researchers want to know where things are happening in the brain they typically use an MRI scanner, which is essentially a big and expensive magnet. The cool thing about this magnet is that once your head is in its field the atoms in your brain, skin, bones etc. tend to all align together. The way an MRI knows what is what is by disrupting that magnetic field (by sending a radio wave which momentarily knocks the atoms in an other direction) and recording how much time it takes for the atoms to get back into the alignment of the main magnetic field. Because different tissues allow atoms to move more or less freely (e.g., hard tissue like bones allows atoms to move less freely compared to softer tissue like grey or white brain matter) different recovery times will be registered. It is, essentially, those recovery times that are displayed in a classical MRI scan (see figure 1). The darker the quicker the atoms move.
But you may ask, ok that’s giving us a picture of the structure but how do we know where activity was happening? How do scientists make those red blobs appear on the brains?
Blooooood, sweet sweet blood!
No this is not a vampire talking but really that’s the answer to the question. One of the components of blood-cells (haemoglobin) is Iron, and as you know from bringing a magnet close to a piece of iron some cool things happen. Well, because we know this process so well and because the effect of a magnet on iron is so visible, it is very easy to also obtain data about the composition of blood in the brain, and because when the blood cells carry oxygen they are slightly different than when they don’t with respect to their Iron composition it this difference that can be recorded by the MRI.
Now, what has blood gotta do with brain activity?
It turns out that it has nothing to do directly with brain activity, since brain activity is electrical in nature, but indirectly it is a trace that some activity has happened in a certain area, because after every considerable activity, the brain cells who have worked hard to produce those electrical signals are hungry for some good O2 molecules.
So by comparing signal recorded when, say, someone is at rest not doing anything in the scanner and when someone is doing something like watching pictures of faces you can know which areas of the brain were more demanding of blood when performing the task compared to when they aren’t. And these are those red blobs that you see in most fMRI papers (see. Figure 2).
Figure 2. Brain areas activated when subjects see images of faces on a screen. These areas are thought to be the parts of the brain that specialise in processing human faces.^1
And that is how fMRI works!
^1. Image taken from: http://white.stanford.edu/teach/index.php/Neurophysiology_of_Face_Perception_and_Social_Information_Processing