I think Kent is referring to my email, excerpted below with the answer
I got from Christian, I think the same answer may be relevant here.
Does the "spike" go away if you turn off temporal filtering? Dan
----
High-pass filtering somewhat skews the beginning and the end of the
design - this is not a problem as in Feat the same high-pass filter
is applied to the design and the data, i.e. both should see the same
degree of skew. You can check the effect of the filter in the Feat
gui by simply setting 'total volumes' to some number and then
playing around with different setting of the filter - if you have the
'full model setup' window open simply press 'View design' each time
you change the setting and see what happens.
Hope this helps
Christian
On 1 Aug 2007, at 03:34, Daniel Wolf wrote:
> I'm not sure if this is relevant, but I've noticed that applying high
> pass temporal filtering to EVs sometimes produces strange dips at the
> beginning in the resulting design matrix columns...could high pass
> filtering do this to the initial timepoints of the functional data?
> Dan
>
>
On 10/3/07, Steve Smith <[log in to unmask]> wrote:
> Hi,
>
> I don't think you're allowed to call anything in the _model_ an
> 'artifact' ;-)
>
> I'm not sure what you're wanting the model to look like - do you want
> it to be very long or very short? Have you tried playing with the
> convolution options?
>
> Cheers.
>
>
>
> On 1 Oct 2007, at 17:10, Kent Conover wrote:
>
> > Hi,
> >
> > I have a question that appears related a query about "strange dips"
> > in the August 1st 2007 FSL
> > digest. In my design subject ratings are used as a Custom 1
> > predictor. The subjects are presented
> > with a long duration thermal stimulus (270 second cycle).
> > Subsequently, the subject rating is often
> > constant at the beginning of each scan. Unfortunately, a initial
> > "spike" artifact appears in the ev
> > after it is convolved with an HRF. While I can make this problem
> > "go away" by not convolving with
> > the HRF, I would like to know if there anything else that I can do.
> >
> > Regards,
> >
> > Kent Conover,
> > Research Associate,
> > Centre for Studies in Behavioral Neuribiology,
> > Concordia University,
> > Montreal, Canada
>
>
> ------------------------------------------------------------------------
> ---
> 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
> ------------------------------------------------------------------------
> ---
>
|