Show all posts by user
Dear AFNI users-
We are very pleased to announce that the new AFNI Message Board framework is up! Please join us at:
https://discuss.afni.nimh.nih.gov
Existing user accounts have been migrated, so returning users can login by requesting a password reset. New users can create accounts, as well, through a standard account creation process. Please note that these setup emails might initially go to spam folders (esp. for NIH users!), so please check those locations in the beginning.
The current Message Board discussion threads have been migrated to the new framework. The current Message Board will remain visible, but read-only, for a little while.
Sincerely,
AFNI HQ
History of AFNI updates
> I have 35 subjects that have undergone 2 different scans in which they have completed a task following 2 different conditions
These 2 conditions are Ex and Rest? What are those 2 different scans?
> The input file is the activity of contrast 1 - 2
What are 1 and 2?
> I would like to identify this with 3dLME
Cond, Order, and Hand are all within-subject factors? I suggest t
by
gang
-
AFNI Message Board
Research Associate or Postdoc Fellows for Image reconstruction, or fMRI signal processing
(Immediately available through Sep 2022)
Hosting Institute: University of Maryland Baltimore (UMB). UMB and UMD are the two flagship research campuses of the University of Maryland and have become a joint research entity since 2016 which has been ranked among top 10 research institutions in the National
by
gang
-
AFNI Message Board
Jonathan, try specifying your output filename as something like Jonathan.nii instead of Jonathan.
by
gang
-
AFNI Message Board
You seem to assume that the brain should not be highly synchronized among most regions under a task/condition. Is it possible that your results are showing something closer to the reality rather than an abnormality? Most of the time people focus on the differential effect or contrast between two conditions/groups, which are usually much more subtle and less dramatic.
by
gang
-
AFNI Message Board
Dane, what result are you showing: the seed-based correlation under one particular condition, or the contrast between two conditions?
by
gang
-
AFNI Message Board
> I am still wondering if it would be incorrect to use all 3 values as input in the first example.
It is not about correctness. Rather, it's more about the convenience of model specification and computational cost. For instance, you can run
-model 'grp*video+(0+video|Subj1)+(0+video|Subj2)' \
The gltCode specification would become more complex.
by
gang
-
AFNI Message Board
> I am actually using the z values sub-brick to threshold, and then the sub-brick before as overlay. Is that what you mean?
Yes, this is usually preferable.
> we found that values of -0.07 (for example) are very hard to grasp on their own.
Depending on the context, that value may have some physical meaning and is interpretable, unlike statistical value (which is dimensionless).
by
gang
-
AFNI Message Board
> Does the mathematical test in 3dLMEr, which produced the z values for my above mentioned group*time interaction effect, have a name?
group:time is the interaction between group and time. It assesses the difference of the slope (or marginal) effect for time between the two groups.
> Since you said chi-square results for group:time and z values for [-gltCode Int-Ctlr_over_Time
> &
by
gang
-
AFNI Message Board
> What exactly do you mean by "similar model formulations"? Do you mean using same polort# and
> Pearson's vs Spearman's correlation, or something else? e.g. perhaps "modeling" like stim_time_IM
> vs AM, and block vs. tent? If so, how would you specify it in 3dfim+?
For task-related data, better switch to 3dDeconvolve. It's inconvenient to deal wit
by
gang
-
AFNI Message Board
> help file for 3dfim+ appears to say its old and not being maintained. Is 3dDeconvolve now the preferred approach these days?
Even though 3dfim+ will not be updated, it can still be used as long as you adopt the same or similar model formulations. In fact, you can directly obtain the output for correlation values from 3dfim+ without conversion like 3dDeconvolve.
> What is proper polo
by
gang
-
AFNI Message Board
> if i wanted the overall effect regardless of video, should i input all 3 values or take the mean after r-to-z
> transformation and use that as input (i.e. rInput = (rA+rB+rC)/3) ?
That seems reasonable to me.
> if I wanted a contrast of condition X > condition Y where rA and rB belong to conditon X and rC belondgs
> to condition Y, would it be correct to do (rA+rB)/2) - r
by
gang
-
AFNI Message Board
Thanks for the information, Taylor! The recent change in R 4.2.x is causing the trouble. I'll try to adapt to the change soon in 3dLMEr.
by
gang
-
AFNI Message Board
> I noticed that when summarizing the information for the posterior samples of Subject pairs, you apply some sort of
> standardization to all values based on the mean of all samples for that subject and the square root of 2
Vasco, are you stealing my secret codes? :) The rationale for that adjustment is due to the fact the data are used twice in the whole process.
by
gang
-
AFNI Message Board
> If I am including age, is the age the sum of ages of the 2 participants a given r corresponds to? Am I also
> correct in assuming that this summed age needs to be mean-centered across all pairs?
Center the age first and then sum each pair.
> In my case I have 3 groups, and not interested in cross-group ISC, am I correct in assuming I can model
> this with just 2 dummy varia
by
gang
-
AFNI Message Board
> could I use the same model that I've applied previously for this purpose, and just change my post-hoc GLT codings slightly?
In that case, you code the factor "scram" with a mixture of between-subject and within-subject types among the 3 levels. I'm not so sure if 3dLMEr would work that way. Does the program complain about such a specification? Even if it does not compl
by
gang
-
AFNI Message Board
> I have a dataset where participants listen to a story twice. On the second listening, the story is either
> temporally scrambled at the sentence or word level.
What do the subjects listen to during the first session? During the second session does each subject listen to only one of the two scrambling cases or both?
> I'd like to contrast the within-subject effects of the in
by
gang
-
AFNI Message Board
Try
3dAttribute BRICK_STATAUX filename+tlrc''
The 4th (and last) number is the degrees of freedom for the t-statistic.
See more here:
by
gang
-
AFNI Message Board
In your original population-level analysis, directly obtain the contrast between the two conditions and then make inference.
by
gang
-
AFNI Message Board
> How can I calculate the cluster-size for the difference map (activation map 1 minus activation map2)?
What is your intention to show the difference map? Making inference regarding the difference between the two conditions?
by
gang
-
AFNI Message Board
Glad that it worked out for you. Thanks for reporting the success!
by
gang
-
AFNI Message Board
> If I want to compare the modulation of response to condition 1 vs condition 2 on the group level (a contrast of
> condition1[1] - condition2[1] ) do I need to adjust the way the AM events on the individual level are being de-meaned?
As "condition1[1] - condition2[1]" is the slope effect, it would have no impact as to whether to demean the modulation variable or how to de-me
by
gang
-
AFNI Message Board
> will 3dLME then demean a second time while it's running?
Yes, it would. To avoid the problem, add the following to your 3dLME script
-qVarCenters 0 \
by
gang
-
AFNI Message Board
> I meant the difference between the chi-sqare values (from group*time, generated automatically) and the z values
> from that contrast. So if they're essentially the same, can I report either one?
Yes, either one is fine since they are conceptually equivalent.
> My goal was to find out if there is a group*time effect. Do I have to report the main effects Time and Group, too?
by
gang
-
AFNI Message Board
> Yes I can easily obtain the ISC between the two levels, so I would just do that. I'm still curious what a model
> with such a factor incorporated would look like.
Maybe something like:
Y ~ group*socialness + (group*socialness | ROI) + (socialness |Subj1:Subj2) + (socialness | mm(Subj1, Subj2, weights = cbind(1, 1), scale = FALSE))
> the paper you attached is precisely wh
by
gang
-
AFNI Message Board
> Is there a recommended way to correct for the fact that I would be running 3 models (e.g. by
> adjusting quantiles when plotting and interpreting results)?
Exploratory studies are usually hard to reach a decisive conclusion. Thus, I think that one should focus on effect estimation, not decision making. In light of this consideration, I would not worry about the issue as to whether to
by
gang
-
AFNI Message Board
> I mainly looked at the result of [-gltCode Int-Ctlr_over_Time 'Group : 1*Intervention -1*Control Time :']
> - when I get a significant result there, can I call it a significant 'interaction effect'?
Yes, the chi-square value provides the assessment for the statistical evidence of the interaction between Group Time.
> What is the difference to the result of Ti
by
gang
-
AFNI Message Board
Vasco,
> First I wanted to check if this looks correct or if the reverse group comparison constrasts need to be added
The original script looks fine and you don't need to add those reversed versions because they are simply flipping the sign.
> Is there a way to run a single model with all of these effects or do I have to run separate ISC models (i.e. a model with all
> vid
by
gang
-
AFNI Message Board
Angus,
Unless you have multicollinearity issue, I suggest using 3dDeconvolve -stim_times_IM combined with 3dREMLfit instead of 3dLSS.
by
gang
-
AFNI Message Board