help
help > RE: CON_subject*.nii files
Sep 2, 2013 04:09 AM | Alfonso Nieto-Castanon - Boston University
RE: CON_subject*.nii files
Hi Hong,
Yes, the CON_Subject* files in a second-level results folder contain contrast files for each subject (these are created whenever you specify a within-subjects contrast that includes more than one element). For example, if you are comparing two conditions A and B with a between-conditions contrast [1 -1], the CON_Subject* files will contain the difference in Fisher-transformed correlation values between the two conditions for each subject. These are equivalent to the con_*.img files generated by the SPM first-level analyses. You may take these files and enter them into SPM for additional repeated-measure second-level analyses (or if you prefer you may enter the original Fisher-transformed correlation values for each condition and for each source ROI, which are stored in conn_*/results/firstlevel/ANALYSIS##/BETA_Subject*)
The contrast files named CON_Subject* files are generated when you click on 'compute results for all sources' (seed-to-voxel analyses separately for each seed ROI; these analyses require the generation of, at most, one within-subject contrast file per subject), while the contrast files named CON_Subject*_compA### are generated when you click on 'seed-to-voxel results explorer' (since these analyses may require the computation of multiple within-subject contrast for each file, indicated by the additional numbers in ###; for example if you are performing a multivariate test across multiple seed ROIs or across multiple components in a MVPA analysis you will see several CON_Subject*_compA### files for each subject)
Let me know if this clarifies.
Best
Alfonso
Originally posted by Hong Yin:
Yes, the CON_Subject* files in a second-level results folder contain contrast files for each subject (these are created whenever you specify a within-subjects contrast that includes more than one element). For example, if you are comparing two conditions A and B with a between-conditions contrast [1 -1], the CON_Subject* files will contain the difference in Fisher-transformed correlation values between the two conditions for each subject. These are equivalent to the con_*.img files generated by the SPM first-level analyses. You may take these files and enter them into SPM for additional repeated-measure second-level analyses (or if you prefer you may enter the original Fisher-transformed correlation values for each condition and for each source ROI, which are stored in conn_*/results/firstlevel/ANALYSIS##/BETA_Subject*)
The contrast files named CON_Subject* files are generated when you click on 'compute results for all sources' (seed-to-voxel analyses separately for each seed ROI; these analyses require the generation of, at most, one within-subject contrast file per subject), while the contrast files named CON_Subject*_compA### are generated when you click on 'seed-to-voxel results explorer' (since these analyses may require the computation of multiple within-subject contrast for each file, indicated by the additional numbers in ###; for example if you are performing a multivariate test across multiple seed ROIs or across multiple components in a MVPA analysis you will see several CON_Subject*_compA### files for each subject)
Let me know if this clarifies.
Best
Alfonso
Originally posted by Hong Yin:
Hello,
This is a topic related to the prior posting on corr_*.img, yet I think it may be clearer to start a separate topic.
We need to enter the contrast images from CONN second level results into SPM for more group level analysis. After performing ROI-to-voxel analysis, I see that in the second level ROI folders, there are files like CON_Subject001.nii, or CON_Subject_compA001.nii, etc. I assume that the CON_Subject* files are contrast files at each subject level but would like to get some clarification on it.
Are the CON_Subject* files equivalent to the con_*.img files generated in SPM? In another word, it is valid to enter the contrast files into SPM for further group level analysis?
What are the CON_Subject_compA* files?
Is there documentation on the types of files generated by CONN?
Many thanks,
Hong
This is a topic related to the prior posting on corr_*.img, yet I think it may be clearer to start a separate topic.
We need to enter the contrast images from CONN second level results into SPM for more group level analysis. After performing ROI-to-voxel analysis, I see that in the second level ROI folders, there are files like CON_Subject001.nii, or CON_Subject_compA001.nii, etc. I assume that the CON_Subject* files are contrast files at each subject level but would like to get some clarification on it.
Are the CON_Subject* files equivalent to the con_*.img files generated in SPM? In another word, it is valid to enter the contrast files into SPM for further group level analysis?
What are the CON_Subject_compA* files?
Is there documentation on the types of files generated by CONN?
Many thanks,
Hong
Threaded View
| Title | Author | Date |
|---|---|---|
| Hong Yin | Jul 30, 2013 | |
| Alfonso Nieto-Castanon | Sep 2, 2013 | |
