Back to PLS HelpPosted on 09/10/14 01:17:02
Number of posts: 16
Dear all,
I have run a non-rotated, event-related fMRI PLS analysis, and the values for the CIs of the brain scores seem weird to me in that the orig_usc values are not in between the upper and lower ranges of the 95% CI (ulusc,llusc).
For example, the orig_usc value for Cond A is 865.4165, and the upper and lower ranges for this condition are 37.65 and 24.80 respectively.
This is almost precisely reversed in the values for Cond B: orig_usc = -865.4027; ulusc = -24.80; -37.65.
When I try and look at the "Task PLS Brain Scores with CI" figure in PLS, the bars are cut-off and only a single error bar is shown. You can see a screenshot of this here: http://i.imgur.com/BMx0e3G.png
Any ideas why this might be happening?
Many thanks,
Reece
Replies:
Posted on 09/10/14 01:21:07
Number of posts: 16
I should add that I am using version 6.1306040
Posted on 09/10/14 07:07:46
Number of posts: 394
I should add that I am using version 6.1306040
HI Reese - thanks for pointing this out. Its a bug in the mean-centreing process during the bootstrap estimation. I hope to have a fix in place within a week.
Login to reply to this topic.
