Hi,
> 1. Page3 last paragraph: “It is always important to look at the registration summary report page very carefully, to check that all lower-level registrations succeeded” How can I tell if a registration succeed by looking at the report page?
Check if the red borders align well with the structural boundaries that you can see in the grayscale image.
> 2. Page4 2nd paragraph: “the “time course” outputs in these higher-level results no longer refer to “time” (despite the heading). They refer to subject (or session) number.” Do you mean t = 0 is subject 1 (ac_left), t = 1 is subject 2 (at_left), t = 11 is subject 6 (eg_right).
Yes.
> 3. In the Higher-Level FEAT Report, what is: “Sum of all input masks after transformation to standard space”? How to interpret the brain images?
If any major misregistrations occur then you can see results less than the maximum value across large parts of this mask.
> 4. In the same report mentioned in Q3,
> a. What is “Unique missing-mask voxels: This shows voxels where only one mask is missing”?
> b. What so special about having one mask missing?
> c. How to easily identification of single gross registration problems?
> d. How to interpret the brain images?
See above. This image will show where one brain image does not overlap with the others, which normally indicates registration problems.
> 5. Could you please explain how to interpret the results webpage? The practical asked us to make sure that I understand what is being shown.
>
> In “Results for lower-level contrast 1 (index)” webpage,
> a) In zstat 1 – C1 (left > right), there is no activation (colored dots) over the matrix of brain images. What does this mean? There is no evidence that the left hand motor task produces more activations than the right hand motor task?
Correct.
> b) In zstat 2 – C2 (right left), there are activations over most of the brain images. What does this mean? There are more evidences (in various amounts) that the right hand task generates more activations than the left hand task?
Correct.
> c) In the time-series plots (actually the subject-effect size plots),
> i. What are "full-model fit" and "cope partial model fit"?
The former shows what happens when all EVs are used in the fit, whereas the latter shows what happens when only some EVs are used (those involved in the contrast).
> ii. What do negative effect size mean? E.g. for t=9 subject 4 df-right, the effect size is -20. What does that mean?
It depends on the contrast. For example, for C1 (left > right) it would indicate that right was 20 units more than left.
> iii. What do we use the numbers in these plots for?
It is to get a feeling for what is in the data. You will also see some similar plots in papers.
All the best,
Mark
|