Print

Print
Dear SPM List,
   I have important fMRI data from 3 subjects with a very rare condition. All 3 showed similar regional effects in both native and MNI-space SPMs, but our statistician advised that a 2nd level analysis could not be generalized from only 3 subjects, so I am combining the 3 individual SPMs in a fixed effects model (15 blocks per subject). Since there are 4-conditions per block, this yields 225 columns (4 x 45 + 45 block constants).  The design matrix looks fine, but each time I estimate, SPM5 (in Matlab 7.7.0) writes the expected 225 beta images, but then gives the following warning:

Variable 'SPM' cannot be saved to a MAT-file whose version is older than 7.3. To save this variable, use the -  v7.3 switch

and the SPM.mat is emptied (goes from 1GB to 128 bytes). Can anybody tell a very basic programmer what to do?

Thanks so much,

          Steven Berman, Ph.D.
        UCLA Dept. of Psychiatry & Biobehavioral Sciences,
          Brain Research Institute, and Center for Addictive Behaviors
          Tel. (310) 825-0616     Fax: (310) 825-0812

          http://ibs.med.ucla.edu
          http://uclamindbody.org
          http://www.bri.ucla.edu
          http://www.semel.ucla.edu/cab/