We're starting to use 3dDespike with our data and we're trying to find the best way to examine the data pre/post despike. We used -ssave option to save the "spikiness" dataset that contains voxel-wise 's' values from the despike algorithm. My understanding that is any timepoint that has s>2.5 (as the default option) is considered a spike and will be corrected. We'd like to generate a summary statistic for each voxel of the number of timepoints that were corrected by despike and also potentially calculate the number o voxels that had at least 1 spike in their timeseries. I assume this will involve some form of 3dcalc operation.

Can you please advise on the best way to achieve this b type of a statistic

Thanks,
Michael