vii
In 1936 Pauling and Coryell discovered that hemoglobin is paramagnetic by itself,
but loses its magnetic moment when bound to oxygen (Pauling and Coryell 1936). Many
years later Ogawa would realize that since an MRI signal depends on magnetic
alignment of protons, deoxygenated hemoglobin's paramagnetic qualities would
interfere with the MRI signal where it was abundant (Ogawa and Lee 1990). He realized
it would therefore be possible to use MRI to measure the hemodynamic response by
taking a series of MRI images and Iookingfor voxels to become brighter as fresh blood
rushed in and displaced oxygen-poor blood (Ogawa, et al. 1990). His theory has been
confirmed by a large amount of data showing a positive correlation between the
amplitude of somatosensory evoked potentials and fMRI BOLD signal (Arthurs, et al.
2000; Backes, et al. 2000; Heeger and Ress 2002; Ogawa, et al. 1998)
DuringanfMRIexperiment, the subject is first given an anatomical scan
providing a single 3D image at very high resolution. The subject is then provided
stimulus, or asked to perform a task, or both depending on the experiment. During this
time the subject is continuously scanned forming a series of low-resolution 3D images.
The amount of time between each image is called the repetition time (TR), and is limited
by how quickly a scanner can produce an image at the desired resolution (generally
around Zsecondsforour purposes). Theseimages arethenoverlaidontop ofthe
anatomical image and allow activation to be measured by looking for changes in signal
intensity in each voxel over time. These changes in signal intensity can then be
correlated to the stimuli or tasks to determine which brain areas had increased or
decreased activity in response to a stimulus or task.