Dear Anderson,

thanks for the clarification. I didn't realize that, at the first level, the calculation for the COPE %change can still be influenced by other EVs. Thanks also for the reference links - very helpful.

Best
Udo
From: FSL - FMRIB's Software Library [[log in to unmask]] on behalf of Anderson M. Winkler [[log in to unmask]]
Sent: Sunday, July 31, 2016 3:30 AM
To: [log in to unmask]
Subject: Re: [FSL] Difference between PE and COPE %change

Hi Udo,

Thanks for sending. What happens is that, for the PE percentage change, what is used are the EVs themselves as entered in the design, whereas for the COPE percentage change, what is used are the EVs of the model partitioned into EVs of interest and nuisance. The main difference is that, if the EV of interest isn't perfectly orthogonal with the nuisance, its amplitude (peak to peak height) changes, and this amplitude is used in the computation of the percentage. The partitioning isn't explicitly shown, but it is necessary to account only for the variability of the contrast itself, thus ignoring sources of variability that might be in the other EVs. If the other EVs were perfectly orthogonal to the EV tested with this contrast, or if there were only one EV, then the result should be the same (this is what might be going on at the higher levels).

The partitioning is described in the Appendix B of Smith et al (2007), and again in the context of randomise in the Appendix A of Winkler et al (2014):
- http://www.sciencedirect.com/science/article/pii/S1053811906009475
- http://www.sciencedirect.com/science/article/pii/S1053811914000913

All the best,

Anderson


On 29 July 2016 at 23:38, Will, Udo <[log in to unmask]> wrote:
Thanks, Anderson. I have uploaded three items the design/contrast.png, a featquery folder with the non-converted PE and Cope values, and a featquery+ folder with the %change values.

Btw. in the meantime I found that there is no discrepancy between PE and Cope %change when I use FEATQUERY on higher level analyses, i.e. when I feed cope.feat directories.

Best
Udo
From: FSL - FMRIB's Software Library [[log in to unmask]] on behalf of Anderson M. Winkler [[log in to unmask]]
Sent: Friday, July 29, 2016 3:33 AM
To: [log in to unmask]
Subject: Re: [FSL] Difference between PE and COPE %change

Hi Udo,

I don't have any example that would help to comment on this. Could you compress your Featquery directory, plus design and contrasts, and send? You should have just received an automated OxFile email with instructions.

Thanks.

All the best,

Anderson


On 27 July 2016 at 15:17, Udo Will <[log in to unmask]> wrote:
Hi,

I have difficulties finding an explanation for the difference between PE and COPE %change in the following case: I am examining one EV and have a contrast [1 0 0 0] where the 1 indicates the EV in question. Examining the PE and COPE values for this EV they are identical (as is to be expected, I think), but the %change values (as reported by Featquery) are different. What is the reason for this difference?Are the two measures based on different references, and what is the interpretation of the cope %change in this case. Thanks.

Udo