Paired t-test on sourses: FDR correction

Hi
After running paired t-test on sourses for the time frequency analysis, for the FDR correction is it necessary to slect all:

signal, time and frequency?
Thanks

This would results in different types of corrections for multiple comparisons.
Do you consider that successive time points are multiple comparisons? That adjacent sources are multiple comparisons? I have no simple answer to your question.
https://neuroimage.usc.edu/brainstorm/Tutorials/Statistics#FDR_correction

Thanks for the reply. Experimental paradigme is that, I compared brain responses evoked by laser stimulation of the dorsum of both hands. In 1 case, attention is paid to right hand, in another case, attention is paid to left hand.