| |
Introduction to mrFindBorders
39
different subjects.
·
Features of the human cortex can be derived from invasive studies done with monkeys
by mapping monkey visual areas to human visual areas. This is very useful since many
experimental approaches that are suitable for monkeys cannot be used to study the
human visual cortex due to ethical and technical reasons.
There are different criteria to define or identify visual areas in monkeys; these include func-
tion (receptive field properties, e.g., kind of stimuli and size of receptive fields), anatomy (cell
types), connections (to other areas), and topography (retinotopic maps). However, many tech-
niques that can be used with monkeys cannot be used in the case of humans. Fortunately, reti-
notopic visual areas can be identified noninvasively in humans (or monkeys) by using fMRI
measurements. Retinotopic maps can be produced by applying appropriate visual stimuli. A
rotating wedge visual stimulus creates traveling waves on the cortex, which can be recorded.
As explained before, these traveling waves have opposite directions in adjacent retinotopic
visual areas. Measurements that are obtained when applying a rotating wedge stimulus can
therefore be used to segregate retinotopically organized visual areas.
Instead of determining the boundaries of the visual areas in three-space, the cortex can be
flattened computationally. This makes the analysis easier. Please refer to chapter 3 for more
information on flattening. Once the functional flat maps (phase maps) from the rotating wedge
experiments have been created (see Figure 16), phase reversals indicate the positions of the
boundaries between retinotopically organized visual areas. The boundaries are often deter-
mined manually. This is done as follows: The flat map from the expanding or collapsing ring
stimulus can be used to determine the position of the fovea and the periphery. These areas
might also be estimated, less precisely, based on the flat map from the rotating wedge data. To
segregate adjacent visual areas, a flat map that shows the response from the rotating wedge
stimulus is examined. The user looks for phase reversals and draws lines (boundaries) be-
tween the fovea and the periphery as in Figure 15. Manual segregation of visual areas is time
consuming and the results depend very much on the performance of the user, i.e., different
users will produce different results and even the results from one user vary over time.
|  |
|
| |
|
|
