Methods such as EEG and MEG measure neural events directly, and are therefore a reflection of activity in the brain. With fMRI and PET this is not the case, these methods measure changes that are correlated to activity in the brain. These include chemical changes, and changes in blood flow. These features of the brain are called metabolic changes.
Neurons use energy and oxygen. Oxygen and glucose (from which energy is derived) enter the brain through the blood. When a brain area becomes more active, it needs more energy and oxygen, and blood flow to this area will increase.
PET measures these changes in blood flow to the brain. This is done by adding a tracer to the blood. A tracer is a substance that can be tracked in the brain. In the case of PET, it is a radioactive isotope, which means that it is an unstable substance. This causes the substance to fall apart and release a positron. A positron is a small particle, which can be seen as the opposite of an electron. A positron and electron have an inverse charge, so they are attracted to each other.
Electrons are already in the blood, and the positrons are released as the tracer falls apart. When an electron collides with a positron (as they are attracted to each other), two photons are created, both moving in opposite directions. Photons are particles of light, and can therefore be detected with a special 'gamma ray detector'. This detector can be used to determine where the two photons originated (where the positron and electron collided).
By mapping the collisions onto an image, it is easy to see where most of the collisions have occurred. The blood with the tracer has gone to that spot, and it can be concluded that this brain region has been active.
One problem with PET, of course, is that the blood with the tracer goes not only to the above location, but basically to every body part and brain region. So there is actually a comparison to be made between the number of collisions in a resting state, and the number of collisions while performing a cognitive task. This comparison takes place by a method called subtraction. This involves subtracting the results in the control condition from the results during the cognitive task. What remains are the points where there was more activity during the task than during rest. It can then be concluded that these points were actually more active because of the task.
The main disadvantage of PET is the radioactive substance used as a tracer. A subject may only be exposed to this tracer a limited number of times, as it is a hazardous substance. Therefore, in order to reduce noise, it is necessary to average over several subjects. Because all people have slightly different brains, it is difficult to make this average.
Author: Myrthe Princen (translated by Melanie Smekal)
Image: Marcel Loeffen