> In the previous mail you mentioned that "the optimal blocked design is one
> with a fundamental frequency close to the dominant frequency of the
> hemodynamic response (from the signal perspective; not necessarily from
> the psychological perspective), ie a block length of 10-20s, with a
> fundamental period of 20-24s." I can understand for that. However, I am
> curious that the fundamental periods are all the same in all kinds of
> experiments or just for some specific studies.
I am assuming that the blocks (conditions) are modelled as squarewave
(ie boxcar) functions. The relevant fundamental period is that of a
CONTRAST of your conditions, which is itself another squarewave. The
fundamental therefore depends on both the contrast and the block length.
Note that I am assuming for simplicity that most power in a squarewave
exists at the fundamental frequency, f - that of a sinewave with the
same period (in fact, I think a squarewave also has considerable power
at harmonics 3f, 5f, etc). More strictly, the efficiency of your design
for a particular contrast depends on the area under the product of the
power spectrum of the haemodynamic response and the power spectrum of
the contrast, as alluded to in my previous email, in which case the
harmonics may be significant and the optimal block length may differ.
The figures given above (which should have read "ie a block length of
10-20s, with a fundamental period of 20-40s..."), were therefore
based simply on my guess for the power spectrum of a "canonical"
(smoothed) haemodynamic response function. On reflection, the
dominant frequency is more like 0.03Hz, giving an optimal block
length of 16-18s. Simulations that also take into account the
harmonics may give an optimal block length that differs again
from these figures, but I doubt by much.
Rik
---------------------------8-{)}-------------------------
DR R HENSON EMAIL [log in to unmask]
Wellcome Department of
Cognitive Neurology TEL (work1) +44 (0)20 7833 7483
12 Queen Square TEL (work2) +44 (0)20 7833 7472
London, WC1N 3BG FAX +44 (0)20 7813 1420
URL: http://www.psychol.ucl.ac.uk/rik.henson/index.html
---------------------------------------------------------
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|