Hi,
The basis functions are automatically derived from running a PCA on a
large number of HRF samples, and so do not necesarily have any fixed
'meaning'. However, in practice it is generally the case that basis
function 2 looks very like the first derivative of the primary one
(and therefore represents a temporal shift between EV1 and data) and
basis function 3 looks like the second derivative (and therefore
represents a difference in HRF width, or smoothing amount, between the
EV1 and the data).
Cheers, Steve.
On 15 Feb 2008, at 09:21, Martin M Monti wrote:
> Hi y'all FSLers,
>
> I'm running a series of first-levels in FEAT, and using FLOBS for
> HRF convolution. I have a question re the interpretation of the
> thresh_zscores I get from each basis function (for any given EV).
> The first componnet is pretty clear, and seems to capture overall
> the same variance that would go into an analysis using only a
> Gaussian convolution -- maybe with increased final statistics due to
> the decrease in model error (compared to using a simple gaussian
> approach)?. What I'm unsure about is how to interpret activations in
> the other two zstats/basis-functions for the same EV -- rather I'm
> uncertain on what the other two basis functions try to capture.
> Can I please ask for a little clarification?
>
> cheers
>
> martin
>
---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------
|