LLN, le 6/04/09
Dear Fabien,
Are you trying to enter all your data in a FFX analysis? If so, this
is a frequent problem. The only way to oversome it is to use 1st
level analyses at the INDIVIDUAL level, and then port the .con files
of your GROUP in a 2d level RFX analysis. Note that this is in fact
better from a statistical point of view. I quote an earlier reply of
K. Friston on this topic:
--------------------
On a non-technical note it might be worth considering the utility of
these large FFX analyses. We have tended to suggest that 1st-level
analyses should proceed one session at a time (the parameter
estimates will be identical) allowing for session-specfic error
variance hyperparameter estimates.
Data over subjects can then be combined in a 2nd-level analysis of
the contrast images, that give the same results as if you combined
the subjects at the first level.
Karl Friston
---------------------
Hope this helps,
Mauro
>Dear SPM-experts,
>
>I have an event-related paradigm with 205 dynamics, 5 regressors.
>Sixteen volunteers participated in the study and had two runs (i.e
>32 differents sessions). I have a problem during the estimation of
>the first level, with an error message of OUT OF MEMORY
>
>I use SPM5 with Matlab 7. The computer run under windows XP, with 4
>Gigas of RAM and an extended memory of 10 Gigas (i.e 6 Gigas of
>virtual memory). I tried to change the followings parameters on
>spm_defaults : default.stats.maxmem and defaults.stats.maxres,
>unsuccessfully.
>Data have also been normalized with 32 slices instead of the 69 of
>the template. Finally I tried to use Matlab with no-Java again
>without success.
>
>Is there any solution to adress this problem ?
>
>Thank you for your help.
>
>
>Regards
>
>
>
>Fabien GIERSKI
>-----------------------------------------------------------------
>Fabien GIERSKI - Neuropsychologist - PhD
>-----------------------------------------------------------------
>Fédération d'Imagerie Médicale - IRM 3T
>Hôpital Maison Blanche - 51902 Reims cedex
>FRANCE
--
_____________________________________
!!! NOUVELLE ADRESSE MAIL -- NEW E-MAIL ADDRESS !!!
[log in to unmask]
_____________________________________
Help fighting hunger: http://www.hungersite.com
Just click your mouse and sponsors of The Hunger Site donate a
serving of food to a person in need - at no cost to you.
______________________________________
Mauro PESENTI
Research Associate, National Fund for Scientific Research (Belgium)
Unite de Neurosciences Cognitives
Departement de Psychologie
Universite Catholique de Louvain
Place Cardinal Mercier, 10
B-1348 Louvain-la-Neuve
tel.: +32 (0)10 47 88 22
fax: +32 (0)10 47 37 74
E-mail: [log in to unmask]
http://www.nesc.ucl.ac.be
http://www.nesc.ucl.ac.be/mp/pesentiHomepage.htm
|