Psychologists need to understand the fundamentals of the imaging methods they plan to use. After a brief historical review, highlighting the joint contributions of fundamental physics and studies of cerebral metabolism, the lecture describes the theoretical and practical principles of the current dominant method, functional magnetic resonance imaging (fMRI).
In physical terms, nuclear magnetic resonance involves the absorption by an atomic nucleus with a non-zero spin (frequently the proton) of electromagnetic radiation of a particular frequency, in the presence of a magnetic field. The return to spin equilibrium results in the emission of a remotely measurable electromagnetic wave, whose relaxation constants (T1, T2) characterize different tissues. The use of magnetic field gradients makes it possible to select a slice, and to encode in frequency and phase the spatial origin of signals within this slice.
These principles are at the origin of anatomical MRI, which has been used for decades in many hospitals. But how to make the magnetic resonance signal sensitive to brain activity? Ogawa and colleagues exploited the fact that deoxyhemoglobin is paramagnetic and disturbs the NMR signal (effect on apparent T2 or T2*). The work of the Sherrington, Sokoloff and Raichle groups had long since shown that changes in hemodynamic parameters accompany the onset of cortical activity. Cerebral activity can thus be measured indirectly by the so-called "BOLD effect" (Ogawa, Lee, Kay, & Tank, 1990), a complex causal chain whose biophysical and molecular mechanisms remain imperfectly understood, in which cerebral activity leads to :
- overconsumption of oxygen, but also an increase in blood flow ;
- an increase in the oxy-/deoxy-hemoglobin ratio ;
- a decrease in magnetic susceptibility ;
- an increase in the T2* parameter and therefore the MRI signal.
