Dear list-members,

I submitted a paper regarding cortical thickness in relation to body fat distribution. The analysis was based on the mri_glmfit procedure implemented in freesurfer 5.1. Version. Gender was defined as a categorical or group factor and age, Body Mass Index, total surface are, body fat distribution and education as continuous factors. In a first step, we tested whether the slope of males and females were different regarding the continuous covariates. We found no significant gender specific slopes and used therefore the simpler model with one regressor for each covariate. Now I received the comments of the reviewers. One reviewer asked: "How are results affected if mixed effects modelling is used to control for the random effects of individuals?" Does it mean I need a subject specific error term in my GLM? Is there an option in freesurfer to perform such an analysis?

Best Ralf

Ralf Veit PhD Institute of Medical Psychology and Behavioral Neurobiology Otfried-Müller str. 47 72076 Tübingen phone: +49-7071-2987703 fax: +49-7071-295706