Dear Mona,
The reason you may fail to see active clusters in a second-level, one-sample t-test is likely related to the vastly different degrees-of-freedom (dof) available for the statistical tests on the first and second level (unless you have a very large number of participants).
On the first level, the dof depend primarily on the number of volumes you aquired during the scan (minus the number of regressors), and can therefore easily reach very large numbers (100+). On the second level, the dof of a one-sample t-test is simply the number of subjects (or con* images) minus 1, and is therefore often a lot lower. This can increase the (absolute) statistical thresholds tremendously.
Depending on the number of subjects you have, you could consider doing a "fixed-effect" analysis (all participants in a single, first-level type model). Alternatively, if you have an a-priori(!) region of interest you should consider using small volume correction on the second level. This should reduce the "penalty" you have to take for doing multiple comparisons on the second level (i.e. doing FWE-correction). The option you proposed (using individually defined regions) is also a possibility, depending on how you came to choose this region and how "topographically similar" they actually are.
Best,
Andreas
===================================================
Andreas Finkelmeyer, Ph.D.
Academic Psychiatry, Institute of Neuroscience
Wolfson Research Centre, Campus for Ageing and Vitality
Newcastle University
Newcastle upon Tyne
NE4 5PL, UK
Tel.: +44 (0)191 208 1357
Web: www.ncl.ac.uk/ion
>-----Original Message-----
>From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]]
>On Behalf Of Mona Moisala
>Sent: 14 August 2013 10:59
>To: [log in to unmask]
>Subject: [SPM] 2nd level analysis
>
>This may be a fairly simple issue, but since I'm new to fMRI and SPM, I cannot
>get my head around it.
>
>I simply want to compare experimental condition A to the zero level (which I
>understand is the mean of the implicit rest from all subjects). I have already
>calculated the contrast A against rest in the 1st level analysis, and I have seen
>clear activation clusters that are topographically similar across subjects and are
>significant after FWE-correction and with a p-value threshold of <0.05.
>However, when I enter the con* images (1per subject) into a one-sample t-
>test in 2nd level analysis, using the same p-value threshold does not give me
>significant activation clusters. The activation clusters are present only if I do
>not correct for multiple comparisons. If the data looks similar across subjects,
>wouldn't averaging just increase the effect observed in the data and the
>results should be even more prominent in the 2nd level analysis? Or is
>intersubject variability in absolute BOLD-signal the issue? Should I somehow
>normalize the amplitude of BOLD signal before 2nd level analysis? Or should I
>be looking at individually defined ROIs instead of whole brain analysis at the
>second level?
>
>Thank you so much for your help!
|
