> I am trying to run a multibock pls on PET data. I have 12
> subjects, 2 days and two conditions have 48 observations for the
> behavioral data.  
> I keep getting the following error message during the bootstrap
> procedure.  What is this about?

First, the error you mentioned at the bottom is caused by a bug in bootstrap test for Multiblock PLS. Second, the statement below is a new feature for Behavior & Multiblock PLS. It means that the value of your behavior data are too close, and we need more variablility in order to run analysis. So when you see this message, please check your behavior data to make sure that it provides enough variability (i.e. the values are not too close in behavior data)

> For at least one behavior measure, the minimum unique values of
> resampled behavior data does not exceed 50% of its total.

The following error message is due to a bug in bootstrap test for Multiblock PLS, and it has been fixed.

> ??? Error using ==> mtimes
> Inner matrix dimensions must agree.
> Error in ==> rri_bootprocrust at 5
> temp=origlv'*bootlv;
> Error in ==> pet_multiblock_boot at 416
>         rotatemat = rri_bootprocrust(v,pbehavlv);
> Error in ==> pet_analysis at 181
>         [brainlv2,s2,designlv2,behavlv2,brainscores2,designscores2, ...
> Error in ==> pet_analysis_ui>ExecutePLS at 1881
>    [resultFile, elapsed_time] = pet_analysis(PLSoptions.isbehav, ...
> Error in ==> pet_analysis_ui at 114
>            ExecutePLS; 
> ??? Error using ==> waitfor
> Error while evaluating uicontrol Callback.