Re: [Marsbar-users] How are FIR timecourses calculated in MarsBaR?

[Cindy F. <can...@gm...>]

Hi Matthew,

Thank you so much for replying. I appreciate your help!

2013/8/11 Matthew Brett <mat...@gm...>

> Hi,
>
> On Thu, Aug 8, 2013 at 7:15 PM, Cindy Fan <can...@gm...>
> wrote:
> > Dear MarsBaR experts,
> >
> > I would like to ask two questions:
> >
> > I have adopted the instructions in MarsBaR FAQ "How do I extract all FIR
> > timecourses from my design?" to analyze my data, but I would like to know
> > what it means if I change the options to "opt = struct('single', 1)"
> with no
> > ", 'percent', 1" in it. Does it mean that the output timecourses are not
> > estimates in percent scale and therefore shouldn't be compared between
> > subjects?
>
> Well - yes - it means the estimates are not proportional to the
> baseline.  The 'single' option means that you are not estimating the
> FIR response per second of activity, but over the whole event.
>

So if I use "opts = struct('single', 1, 'percent', 1);", how would the
baseline be defined? by the average activity of the same events or by the
average activity of the onset of the same events?

If I don't use the 'single' option, what should I enter for estimating FIR
response per second (per TR?) of activity?

> What exactly does it operate in the function "event_fitted_fir"? Before
> the
> > estimation, does the FIR model consider the signal intensity from the
> data
> > that may vary from one data set to another?
>
> I'm not sure I understand you question.  The FIR model only has the
> data from the individual subject and the region you are estimating
> for.  It takes the data, changes the SPM design to have an FIR model
> for the event of interest, estimates the parameters for that model,
> scales them by the baseline if you asked it to do that, and returns /
> plots the results.
>
>
Yes, because if I am estimating the FIR response over the whole event, the
data multiplies the inverse matrix of the model (right?), I would like to
know how exactly the model is set up and estimated, like does it have a
initial guess of signal magnitude (that may affect the estimation) of the
data?


Best,
Cindy


> Best,
>
> Matthew
>
>
> ------------------------------------------------------------------------------
> Get 100% visibility into Java/.NET code with AppDynamics Lite!
> It's a free troubleshooting tool designed for production.
> Get down to code-level detail for bottlenecks, with <2% overhead.
> Download for free and get started troubleshooting in minutes.
> http://pubads.g.doubleclick.net/gampad/clk?id=48897031&iu=/4140/ostg.clktrk
> _______________________________________________
> Marsbar-users mailing list
> Mar...@li...
> https://lists.sourceforge.net/lists/listinfo/marsbar-users
>



-- 
Hsin-Ya Fan