help
help > RE: Your advice on laterality analysis (Alfonso?)
May 5, 2016 04:05 PM | Alfonso Nieto-Castanon - Boston University
RE: Your advice on laterality analysis (Alfonso?)
Hi Francesco,
If you are looking, for example, at the analyses on 'all subjects' and comparing the connectivity with a left-hemispheric seed vs. a right-hemispheric seed (e.g. selecting left-HG and right-HG and entering a 'between-sources contrast' of [-1 1]), then a significant "positive" effect cluster in the right-superior temporal pole that would mean that, across all subjects, the connectivity between right-HG and right-STP is higher than the connectivity between left-HG and right-STP, while a significant "negative" effect would mean that the connectivity between left-HG and right-STP is higher than that between right-HG and right-STP. Keep in mind that "higher" in this context simply means a higher connectivity value (either more positive or less negative), so in order to properly interpret this you would want to look at what the individual connectivity values are (e.g. are these positive correlation values or anticorrelations?).
In terms of reporting, rather or in addition to reporting the connectivity with individual seeds it might be useful to: a) look at networks by aggregating/averaging across multiple seeds that might show similar connectivity patterns (e.g. select left-HC, left-PT, leftPP, right-HC, right-PT, rightPP and enter a [-1 -1 -1 1 1 1]/3 between-sources contrast); and/or b) test all multiple seeds simultaneously to avoid having to deal with additional multiple-comparisons corrections (e.g. select left-HC, left-PT, leftPP, right-HC, right-PT, rightPP and enter a [-1 0 0 1 0 0; 0 -1 0 0 1 0; 0 0 -1 0 0 1] between-sources contrast).
Hope this helps
Alfonso
Originally posted by Fran :
If you are looking, for example, at the analyses on 'all subjects' and comparing the connectivity with a left-hemispheric seed vs. a right-hemispheric seed (e.g. selecting left-HG and right-HG and entering a 'between-sources contrast' of [-1 1]), then a significant "positive" effect cluster in the right-superior temporal pole that would mean that, across all subjects, the connectivity between right-HG and right-STP is higher than the connectivity between left-HG and right-STP, while a significant "negative" effect would mean that the connectivity between left-HG and right-STP is higher than that between right-HG and right-STP. Keep in mind that "higher" in this context simply means a higher connectivity value (either more positive or less negative), so in order to properly interpret this you would want to look at what the individual connectivity values are (e.g. are these positive correlation values or anticorrelations?).
In terms of reporting, rather or in addition to reporting the connectivity with individual seeds it might be useful to: a) look at networks by aggregating/averaging across multiple seeds that might show similar connectivity patterns (e.g. select left-HC, left-PT, leftPP, right-HC, right-PT, rightPP and enter a [-1 -1 -1 1 1 1]/3 between-sources contrast); and/or b) test all multiple seeds simultaneously to avoid having to deal with additional multiple-comparisons corrections (e.g. select left-HC, left-PT, leftPP, right-HC, right-PT, rightPP and enter a [-1 0 0 1 0 0; 0 -1 0 0 1 0; 0 0 -1 0 0 1] between-sources contrast).
Hope this helps
Alfonso
Originally posted by Fran :
Hi Alfonso,
I finally managed to start working on the analysis as you suggested.
I run analyses on "all subjects", "patients" and "controls", separately and got some interesting results.
For example, when I compared activity in the left and right primary auditory cortices (i.e., R>L), I got an effect in the right superior temporal pole. What does it mean? How can I read the direction of the effect (i.e., is activity higher in the right superior temporal pole?).
I wonder if reporting this makes sense anyway. I want to make things simple, reporting only inter-hemispheric activation differences for a given seed (i.e., I want to see if activation of the primary cortex is greater on the right or on the left hemisphere)
I attached a notepad file with the results for your reference.
Thank you in advance
Francesco
I finally managed to start working on the analysis as you suggested.
I run analyses on "all subjects", "patients" and "controls", separately and got some interesting results.
For example, when I compared activity in the left and right primary auditory cortices (i.e., R>L), I got an effect in the right superior temporal pole. What does it mean? How can I read the direction of the effect (i.e., is activity higher in the right superior temporal pole?).
I wonder if reporting this makes sense anyway. I want to make things simple, reporting only inter-hemispheric activation differences for a given seed (i.e., I want to see if activation of the primary cortex is greater on the right or on the left hemisphere)
I attached a notepad file with the results for your reference.
Thank you in advance
Francesco
Threaded View
| Title | Author | Date |
|---|---|---|
| Fran | May 5, 2016 | |
| Alfonso Nieto-Castanon | May 5, 2016 | |
| Fran | May 5, 2016 | |