Hi Alfonso,

I have been processing a large data set which has gone through the pipeline successfully. However, Irun into issues at the 2nd-level analysis which might have the same source.

1) I am running Seed-to-voxel analysis with networks being the seeds and including a couple of covariates (age and education). When I ask to compute the results, it hangs on endlessly. Basically, the process indicates that it will take 10 minutes. But when it completes the initialization and it indicates that all subjects have been processed, it goes into an endless loop (at the 4 min). No error messages are indicated in matlab or by CONN itself.
2) I also have conducted ROI-to-ROI analysis. When I press the compute button, it successfully shows me the ring connectivity results. But, it goes into an endless loop when asking to show the bar charts of the connectivity.

Please advise.
Thanks.
Reza

I hope this additional information helps. After several hours of the same, I see the following messages in matlab GUI (please see attached JPG). Thanks.

Hi Reza,

It seems that the model is simply running and taking a very long time. From the "number-of-chunks" in your example image it appears as if the number of subjects times the number of conditions in this analysis is around 50,000? If that is the case, then sorry that the original time-estimate was terribly poor but it is probably just taking a long time to process that model (particularly since it appears to be a multivariate model that requires ReML covariance estimation) (if you can let me know a bit more details about this analysis and/or your set up perhaps I can suggest ways to try to speed things up?)

Hope this helps
Alfonso
Originally posted by Reza Momenan:

Hi Alfonso,

Thanks for your reply. This is an analysis with nearly 300 subjects, four groups (health, patient, male, female), 2 covariates. I did an ROI-to-ROI between-networks (Salience, Default and Visual, Dorsal Attention, Sensory) connectivity that worked fine (even it took good a while). Then I tried to see if any of these networks changed connectivity with the rest of the brain which ran into this issue. I am fine if it is simply taking time to do some many comparisons. Just because I was not sure whether progress is being made (the GUI indicated 20 minutes), I was wondering if it has hung up or is in some unanticipated loop. Any advice would be much appreciated. Thanks.

So, I followed your advice and ran bivariate instead of multivariate and only between three of the networks: Salience, DorsalAttention, and SensoryMotor. I see the ROI-to_ROI ring connectivity and when I try the "Plot effects" the following error shows up in matlab: