Thanks for your help!

John Keltner

From: FSL - FMRIB's Software Library [[log in to unmask]] on behalf of Jeanette Mumford [[log in to unmask]]
Sent: Wednesday, October 10, 2012 3:49 PM
To: [log in to unmask]
Subject: Re: [FSL] basic help with contrasts

Hi,

All contrasts are one-sided with the null <=0 and alternative >0, so H0:EV1-EV2<=0 vs H1:EV1-EV2>0. Hence rejecting the null would imply EV1-EV2>0 or EV1>EV2.

yes, an intensity of 0.97 corresponds to a corrected p-value of 0.03.

Cheers,

Jeanette

On Wed, Oct 10, 2012 at 4:14 PM, John Keltner <[log in to unmask]> wrote:

Dear experts,

I would like to ask two basic questions about contrasts and their associated p value maps.

Does the contrast EV1-EV2 ask the question "is ev1 greater than ev2" or ask the question "is ev1 not greater than ev2"?

Also, I would like to confirm that the p value map (or p value image) such as the file GM_mod_merg_s3_tfce_corrp_tstat1.nii.gz associated with the first contrast EV1-EV2 is 1-p, so that an intensity value of .97 for a voxel in this file represents a p value of 0.03.

Thank you very much,
John Keltner