I like to choose the duals smoothing approach because on first level we
don't care too much about statistical significance, just want to see
effect sizes. Individual structures are better preserved with less
smoothing on first level which increases confidence in localization
during data exploration. This individuality becomes unimportant/unwanted
on second level, instead we want to increase statistical sensitivity.
Cheers,
Kai
Darren Gitelman wrote:
> Wouldn't the "con" images already be considered smooth if the
> functional images were smoothed at the first level? Applying a second
> smoothing of 6 mm would result in an overall smoothing kernel of
> sqrt(72). What would be the purpose of smoothing them again rather
> than smoothing the functional images initially at around 8.5 mm.
>
> -----
> Darren Gitelman
>
>
> On Thu, Jun 11, 2009 at 4:48 AM, Helmut Laufs <[log in to unmask]
> <mailto:[log in to unmask]>> wrote:
>
> ----- Original Message ----- From: "Helmut Laufs"
> <[log in to unmask] <mailto:[log in to unmask]>>
> To: "Michael Froelich" <[log in to unmask] <mailto:[log in to unmask]>>
> Sent: Thursday, June 11, 2009 11:44 AM
> Subject: Re: 2nd-level Smoothing
>
>
> Yes, you would. In addtion you may want to use imcalc to apply
> a mask afterwards in order to limit "blobs" to the brain.
>
> Bw,
>
> Helmut
>
> ----- Original Message ----- From: "Michael Froelich"
> <[log in to unmask] <mailto:[log in to unmask]>>
> Sent: Wednesday, June 10, 2009 10:32 PM
> Subject: 2nd-level Smoothing
>
>
> Dear SPM community,
>
> At the SPM8 refresher course, the question about smoothing was
> asked. IN reply, Dr.
> Friston recommended an approximate 6 mm smoothing kernel both
> for the first and second
> level. How is the smoothing accomplished at the second level?
>
> Would we smooth the ".con" images prior to entering them into
> a 2nd level analysis?
>
> Michael Froelich
>
> University of Alabama at Birmingham
> Anesthesiology
>
>
--
Dr. Kai Lutz
Neuropsychology
Institute for Psychology
University of Zurich
Binzmühlestrasse 14 /Box 25
CH-8050 Zürich
Switzerland
_________________________
Diese E-Mail (ggf. nebst Anhang) enthält vertrauliche und/oder rechtlich
geschützte Informationen. Wenn Sie nicht der richtige Adressat sind, oder
diese E-Mail irrtümlich erhalten haben, informieren Sie bitte sofort den
Absender und vernichten Sie diese Mail. Das unerlaubte Kopieren sowie die
unbefugte Weitergabe dieser Mail ist nicht gestattet.
This e-mail (and any attachment/s) contains confidential and/or privileged
information. If you are not the intended recipient (or have received this
e-mail in error) please notify the sender immediately and destroy this
e-mail. Any unauthorised copying, disclosure or distribution of the
material in this e-mail is strictly forbidden.
_________________________
|