Hi Ranila

the different versions contain slightly different algorithms, so why would you expect them to be the same? With 20 subjects you may be at the edge of having enough power to see a significant effect.

cheers Bruce On Wed, 13 Feb 2019, Ranila Bhoyroo wrote:

External Email - Use Caution

Hello Experts,

I was working with some data pre-processed using freesurfer v5.3 - i think this was done on linux. Recently I pre-processed the dataset using v6.0 on an imac. I extracted cortical thickness and surface area from both hemispheres. The data is different for each version. What worries me is that I am correlating the neural metrics with some behavioral data and this gives me different results. My sample size is only 20.

Some advice would be very helpful.

Thanks.

Cheers,

Ranila