afni.nimh.nih.gov

The situation would be similar to a task-based analysis: build one model for the whole analysis at the individual level. So just directly add the EEG regressor to the model with 3dDeconvolve (your solution #2).

Gang