Dear FSL experts:

    

2011/12/23 Xu Wang <[log in to unmask]>

     Dear Smith,Thanks very much for your response, I think I almost find the problem:

     After I got the fsf file, I mamually modified the 6 EVs' settings of motion parameters like this:

                  set fmri(deriv_yn9) 0   --> set fmri(deriv_yn9) 1

     Does this error come from this operation? And it's not allowable for feat_model to derivate the motion parameters?

     If I want to add these 6 derivative variables of motion parameters to be the regressors of the model, how could I do? 

     And how could feat get the derivative variable from custom EVs?  Does this mean the present value minus the previous value?

    Thanks very much!

    Xu Wang

2011/12/23 Stephen Smith <[log in to unmask]>

Hi - no, 9 isn't in general too many…not sure what's going on - if the matrix is rank deficient, FEAT should warn you….check the efficiency calculator in FEAT, and that your input data is valid?

Cheers.

On 23 Dec 2011, at 08:36, Xu Wang wrote:

Dear FSL experts:

    When I run feat_model command on a 9 regressors(each 240 time points) and 9 regressors' derivates(set fmri(deriv_yn9) 1) model, 

the error came out like this:

    terminate called after throwing an instance of 'NEWMAT::SubMatrixDimensionException'

    Is this because too many regressors in my model?

    Thanks very much!

    Xu Wang

---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,  Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222726  (fax 222717)
[log in to unmask]    http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------