Hello all, 

I am experiencing some trouble with viewing results in spm_fnirs. I am able to follow all the steps until estimation just fine without errors, but when it comes to results I have experienced a variety of issues.

During the results step, after submitting my SPM.mat file which has had spatial registration, temporal preprocessing, first-level specification and estimation done to it, I get many repetitions of this warning.

Then, IF spm_fnirs_viewer_stat shows up, it shows a blank brain image. When I try to select 'Contrast', the contrast manager does not show my design matrix and so I cannot create a contrast. When I try to select 'Activation', I get the following error:

I have also had a similar situation in which spm_fnirs_viewer_stat shows a brain image with no activation but which is colored. In that case, which I am not able to replicate currently, I get error quite similar to that one listed above, which instead says something to the effect of

and takes issue with the line

These errors with 'ic' and 'SPM' fields are interesting because they are not mutually exclusive, as in I will choose all the same options for steps up until Results and sometimes get the ic error, sometimes get the SPM error.

LASTLY, I also occasionally run into the following error before the spm_fnirs_viewer_stat has time to show up. This error is replicable, and shows on these conditions: choosing HRF smoothing instead of no smoothing during temporal preprocessing, or not choosing AR(1) when specifying 1st level.


Has anyone in the community run into similar issues? Any pointers in the right direction would be greatly appreciated. For reference, I am running the most up-to-date version of spm_fnirs with spm12 on the most recent version of R2017b, with macOS High Sierra.

Karl V.

Hello again,

I have noticed an additional error during estimation pointed out by my colleague. I think that this might be the source of my issues, since the estimation is not being handled correctly. This is the message I get during estimation. It looks as though the estimation is not completing all iterations before saving to SPM.mat. Any ideas?