Dear experts, 

I know that in a within-subject design there's no need to control for age and gender but should I control for these variables in a within and between-subject design?

The conn toolbox allows to do it but I'm not sure is it statistically correct. 

Thank you, 

M. Cano