hi,
on XP you may also try to assign a low priority to matlab via the task
manager. i know it does not make sense but what on windoze does? anyway,
this paradoxical remedy has worked for me in some circumstances.
ciao-
andreas
----- Original Message -----
From: "Marko Wilke" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Thursday, December 04, 2003 8:51 AM
Subject: Re: Matlab 6.1/SPM2 problem
Hi Mike,
> I'm experiencing a somewhat unusual problem. I'm running SPM2 under Matlab
> 6.1 on a Windows XP machine (2GB RAM), and during an analysis with VBM
data,
> Matlab shuts down completely with no error messages, if I specify p value
as
> uncorrected. With correction for FDR or FWE it works fine, but when I try
to
> use uncorrected p Matlab exhibits this strange behavior.
I think this is due to SPM/Matlab not liking the possibly huge clusters
you get in an analysis using uncorrected p-values. I only know it from
SPM99. I have fiddled with it, but don't remember being very successful
(so probably I wasn't). You could try to change the MaxMem value, but
not sure if this helps. You could also manually threshold the t-images
and use another program, like MRicro, for display purposes (it
incidentally allows for both steps if you load an overlay and specify
the lower cutoff). Other than that, it could also be taken as a hint to
use corrected p-values :)
Cheers,
Marko
--
=====================================================================
Marko Wilke (Dr.med./M.D.)
[log in to unmask]
Universitäts-Kinderklinik University Children's Hospital
Abt. III (Neuropädiatrie) Dept. III (Pediatric neurology)
Hoppe-Seyler-Str. 1, D - 72076 Tübingen
Tel.: (+49) 07071 29-83416 Fax: (+49) 07071 29-5473
=====================================================================
|