Thanks to Gisela for finding this bug in "seqefficiency" (part of the MGH-NMR
functional analysis software package). Seqefficiency is a program which computes the
efficiency of a particular sequence given its paradigm file. Fortunately, this in no way affects
the validity of optseq (the sequency optimization program) which has its own (correct!) code for computing the
efficiency.
Doug
> -------- Original Message --------
> Subject: efMRI designs - bug?
> Date: Fri, 14 Apr 2000 09:58:21 +0200
> From: "Gisela Hagberg" <[log in to unmask]>
> Reply-To: "Gisela Hagberg" <[log in to unmask]>
> To: "SPM mailbase" <[log in to unmask]>
> References: <[log in to unmask]> <[log in to unmask]>
>
> Dear Russ & Doug,
>
> regarding the program package you posted on the mailing list:
> there might be a bug in the seqefficiency programme. Using the option
> -monly it generates an m-file that contains the line
>
> eff = trace(inv(XtX));
>
> Shouldn't it be eff = 1/trace(inv(XtX)) ?
>
> greetings,
> Gisela
>
> ********************************************************************
> Gisela Hagberg, PhD
> [log in to unmask]
> Laboratory of Functional Neuroimaging [log in to unmask]
> Scientific Institute of Foundation S. Lucia
> via Ardeatina 306
> 00179 Rome - Italy
> voice1: +390 6 51501359
> voice2: +390 6 51501345
> fax: +390 6 51501213
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|