Dear Bianca,
This is a sign that correction for non-sphericity went wrong. Are you
using the default options Independence:Yes and Variance:Unequal? Setting
Variance to Equal will solve your problem but is not very satisfactory.
I would be curious to look at the data (there's usually something funny
going on with them) as we have recently tried to make ReML even more
robust - could you send me the SPM.mat file?
Best regards,
Guillaume.
On 06/12/16 18:13, Bianca De Blasi wrote:
> Dear experts,
>
> I am running a two sample t-test in SPM8, but I am getting the following error when I am selecting the SPM.mat file in the Result module and I am not able to set the contrast and have a look at the results:
>
> Error using image
> Invalid datatype for Image CData. Numeric or logical matrix required for image CData.
>
> Error in spm_conman (line 630)
> hDesMtxIm = image((varargin{2}.nKX+1)*32);
>
> Error in spm_conman (line 599)
> spm_conman('ImDesMtx',xX,H.hDesMtxAx) %-Depict DesMtx
>
> Error in spm_conman (line 467)
> [F,cF] = spm_conman('Initialise','on',SPM,STATmode,n,Prompt,Mcstr,OK2chg);
>
> Error in spm_getSPM (line 260)
> [Ic,xCon] = spm_conman(SPM,'T&F',Inf,...
>
> Error in spm_results_ui
>
> Error while evaluating UIControl Callback
>
>
>
> The setting of the model and estimation run with no problems.
>
> Please advise on how to solve this problem,
>
> Thank you,
> Kind regards,
> Bianca
>
--
Guillaume Flandin, PhD
Wellcome Trust Centre for Neuroimaging
University College London
12 Queen Square
London WC1N 3BG
|