Development of fMRI tasks and scan protocols

For many functional MRI studies, tasks have to be created to activate specific
brain regions. The quality of the task can be determined to some extent during
the design of it, by using models to compute the outcome in terms of
sensitivity and selectivity. Sensitivity is determined largely by the contrast
between activation and relaxation of brain areas involved in the task.
Selectivity can be estimated by assessing the chance that brain processes that
are not of interest and confound the result of imaging. Both properties have to
be tested in real fMRI experiments. One may, for instance, create a
mathematically very well designed task with high sensitivity and selectivity,
but upon testing in an fMRI scanning session it may very well fail to produce
significant brain activity. Quite often it turns out that the task is too
tightly controlled, i.e. that brain functions cannot be *switched off* during
control conditions. In general, it is difficult to predict whether a particular
task with particular stimuli, instructions and temporal sequencing of events,
will manipulate the brain as intended.
Finding the right parameters for a task often takes several fMRI tests in
individual subjects. Once results are satisfactory, the real experiment can
commence. It may however take weeks before this point has been reached, and in
some cases the project is abandoned. It is, therefore, important to assess
whether a task produces the required results before submitting a protocol to
the ethics committee, rather than wait with piloting until approval for the
whole study has been obtained.
A second element that sometimes needs piloting is the scan protocol itself.
Notably, parameters of the (f)MRI scan sequence sometimes need to be altered,
for instance to enhance signal in in the brain near the nasal cavity, and this
requires fine-tuning in human volunteers.

Development of (f)MRI tasks and scan protocols and to define the optimal
parameters in preparation of a (f)MRI study

We intend to include a maximum of 60 subjects per year for pilot work (i.e. a
maximum of 10 pilot studies of maximum 6 subjects each). Each subject will
spend a maximum of 90 minutes per session and will be paid according to the
standard payment regiment at the Division of Human Nutrition (i.e. 7 euro per
15 minutes in the scanner, plus 7 euro per hour outside the scanner (e.g.
filling out questionnaires), plus travel costs). All subjects will sign an
informed consent form and all records will be kept just like with other ethics
protocols.

Functional MRI is an eminently safe technique; there aren't risks that have
been associated with the acquisition of fMRI data per se. Above certain
limits, warming and/or a hitching/tingling feeling (stimulation of peripheral
nerve terminations) are possible. However, the intensities used in pilot
studies are amply below these limits. Subjects will be exposed to a magnetic
field of 3 Tesla. This field is routinely used in fMRI and MRI research; it is
maybe worth to mention that scanners supporting a magnetic field more than
twice as powerful (7 Tesla) are used in The Netherlands for research purposes.
No harmful side effects have been reported.