If you prefer instead to compute QC_ValidScans directly you may do so by using in step 2 above the following options instead:

consider covariate scrubbing: raw values
summarize across timepoints: user-defined (and when prompted enter "sum(1-x,1)" -without quotes- as your user-defined function)
summarize across dimensions: sum
summarize across dimensions first: checked
condition-specific measures: checked
and that will create two new second-level covariates, each summing the number of valid/non-outlier scans for one condition (i.e. like QC_ValidScans but separately for each condition).

Hi Alfonso,

I have conn 18.a and am trying to compute QC_ValidScans for 3 different sessions, but when I try to use those parameters, I cannot use user-defined on "symmarize across timepoints". Is this a new update to conn? How would I be able to do this on 18.a?

Best,
Omar