Abstract
In the fourth lecture, we continued our progress through the various MRI methods that can be used to refine the identification of neural codes. Among these, the priming or adaptation method offers a key advantage: it's the only one that can measure neural codes within a voxel, overcoming the mediocre spatial resolution of MRI. The idea is to build on the phenomenon of adaptation to repetition : in many, if not all, regions of the cortex, repetition of the same stimulus leads to a reduction in neuronal activity, and hence in the MRI signal. In an experimental cross-over design, we therefore compare the activation evoked by pairs of stimuli s1 and s2, either repeated (s1 followed by s1, and s2+s2) or unrepeated (s1+s2 and s2+s1). The effect is subtle : on average, the stimuli are identical,only their relationship changes (but the subject need not be aware of the relationship between the stimuli, or even of the presence of the adapting stimulus). If the brain region differentiates between the two stimuli, it should show a reduction in activity on repetition.
The priming method enables sophisticated inferences to be made about neural codes. For example, it has enabled us to demonstrate that the human parietal cortex automatically and unconsciously codes number names and Arabic numerals (" three " and " 3 ") as the same quantity. Tuning also makes it possible to draw tuningcurves, which can be compared with those obtained by neurophysiology in animals.
