Re: [SPM] A question about % signal change in MarsBaR

Hi,

I've Cc'ed the marsbar mailing list; it's the best place for marsbar questions.

> Event_signal is a function which I thought to be the key to my question. So
> I read the source code.
>
> 0053 for e_i = 1:e_s_l
> 0054   es    = e_spec(:, e_i);
> 0055   ss    = es(1);
> 0056   Yh    = event_fitted(D, es, dur);
> 0057   d     = pr_ev_diff(Yh, diff_func, varargin{:});
> 0058   s     = s + d ./ s_mus(ss,:);
> 0059 end
> 0060 s = s / e_s_l * 100;
>
> Question1. So I think by default 'd' is the absolution of max(fitted event
> time course).

This code gets the fitted time course, and then uses the pr_ev_diff
function to calculate the value for the top half of the signal change
from the event time course.  The default method is 'abs max' which
returns the value that has the absolute maximum, that is, the most
positive value if that is greater than the abs of the most negative
value, otherwise the most negative value.

> But what does 'ss' mean? Is it the number of session according
> to the definition of 'e_spec'? But it is said that s_mus is the mean of
> blocks. So 'ss' is the number of block, am I right? I am confused.

Yes, 'ss' is the the session number - for example if you have two
sessions ss can either be 1 or 2.

> Question2. For ER Design, there is no block. How can I use this method?

Well, there are sessions in a block or event related design.

> Question3. Is 'e_s_l' the number of session?

It's the event spec length - the number of events the routine is working on...

Best,

Matthew

------------------------------------------------------------------------------
Throughout its 18-year history, RSA Conference consistently attracts the
world's best and brightest in the field, creating opportunities for Conference
attendees to learn about information security's most important issues through
interactions with peers, luminaries and emerging and established companies.
http://p.sf.net/sfu/rsaconf-dev2dev