 |
 |
Thanks Steve for you reply. A couple of more questions... On Sat, 28 Oct 2006 08:09:01 +0100, Steve Smith <[log in to unmask]> wrote: >Hi, there could be several reasons for this. > >Firstly - it's probably just a typo, but did you mean for contrast 3 >to be [0 0 1] ? > lol. Yes, it was a typo. >Next, the most likely difference is that fstats don't distinguish >between negative and postitive activation - both show up as >"positive" - this is the nature of fstats. Hence your zstats only >show positive, the fstats show both. > So my fstats setup was correct then? And why does the contrast setup: >> EV1 EV2 EV3 F1 F2 F3 >> OC1 * 1 0 0 * 0 0 >> OC2 * 0 1 0 0 * 0 >> OC3 * 0 0 1 0 0 * only represent or detect positive activation ONLY as opposed to the F-test which detects both positive and negative activation? I would think they would do the same thing, no? >More subtle: if for example your regressors are somewhat correlated, >then each individual contrast may be insensitive, but the f-test >still could have greater sensitivity. > >Hope this helps - cheers, Steve. > > > > >On 27 Oct 2006, at 21:36, Michael wrote: > >> Hello, >> >> I have been running some analyses and I have some questions about F- >> tests >> >> and contrasts. >> >> When running FEAT on individual subjects in first-level analysis, >> clickin >> g >> on the stats tab, and selecting the Full-model setup, there is a >> section >> >> for setting up F-tests and contrasts. >> >> In my experiement I have 3 tasks of interests, i.e., 3 EV's in my >> GLM mod >> el. >> >> So I set up my contrasts and F-tests as such: >> >> EV1 EV2 EV3 F1 F2 F3 >> OC1 * 1 0 0 * 0 0 >> OC2 * 0 1 0 0 * 0 >> OC3 * 1 0 0 0 0 * >> >> (* indicating yellow clicked box) >> >> The trouble I'm finding is that when I look at the resulting >> activation >> >> maps for zstat and zfstat maps, I see two completely different >> pictures. >> I >> don't understand why. For the zstat activation map, I barely see any >> activation, but for the zfstat activation I see several clusters of >> voxel >> s >> in ROI's we typically see from our experiments. Is there something I'm >> >> doing wrong? Then when I take the individuatl subjects and make a >> group m >> ap >> of a one-sample t-test and I include and F-test with it, I get the >> same >> >> picture: >> >> hardly any activation in the zstat group map, but I do see >> activation in >> >> the several ROI's in the zfstat map. >> >> What is going on? I would think I would see the same, if not similar >> results. Is there something wrong with my contrast and F-test set- >> up at t >> he >> first-level analysis? >> >> Your help would greatly be appreciated > > >------------------------------------------------------------------------ >--- >Stephen M. Smith, Professor of Biomedical Engineering >Associate Director, Oxford University FMRIB Centre > >FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK >+44 (0) 1865 222726 (fax 222717) >[log in to unmask] http://www.fmrib.ox.ac.uk/~steve >------------------------------------------------------------------------ >--- >=========================================================================