This is from chatgpt, which actually gave a pretty good answer this time

Why ICV is used as a covariate

• ICV reflects head/brain size: It is established early in life and remains relatively stable in adulthood.

• Controls for global scaling effects: If you’re comparing cortical or subcortical structures across individuals, larger heads naturally tend to have larger absolute volumes. ICV adjustment removes this confound so differences reflect local effects, not just overall size.

• Standard practice: ICV is widely used in structural MRI studies for this purpose because it is a stable “baseline” measure.

Why cortex volume is not a good substitute

• Not independent: Cortex volume itself is strongly influenced by disease, development, and aging. Using it as a covariate risks removing real biological signal related to the outcome you care about.

• Collinearity problems: If your dependent variable is cortical thickness, surface area, or subcortical volume, adjusting for cortex volume introduces statistical dependencies (since they’re not independent measures). This can distort your estimates.

• Less stable than ICV: Cortex volume can shrink over time due to atrophy or expand during development, making it unsuitable as a baseline scaling factor.

When might cortex volume be useful?

• If your research question is specifically about relative differences among cortical subregions (e.g., “is region X disproportionately reduced relative to total cortex?”). In that case, cortex volume could serve as a denominator or scaling reference, but not a general-purpose covariate.

• Even then, most researchers would report both absolute and relative measures, with ICV adjustment kept separate.

        External Email - Use Caution        

Hi Freesurfer group:

I have been using ICV as a covariate in all my analyses. Can CORTEX VOLUME be a better measure of covariance? If not, why?

Thanks,

Alan

--

|~|~|~|~|~|~|~|~|~|~|~|~|~|~|~|~|~|~|~|~|~|~|~|~|~|~|~|~|~|~|~|

Dr. Alan N. Francis
Assistant Professor of Neuroscience

Director: Neuroimaging & Neuromodulation Research Lab

Institute of Neuroscience

University of Texas, School of Medicine

Rio Grande Valley, Texas.

                                                                                                          

_______________________________________________
Freesurfer mailing list -- freesurfer@nmr.mgh.harvard.edu
To unsubscribe send an email to freesurfer-leave@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman3/lists/freesurfer@nmr.mgh.harvard.edu/