Single-electrode models and clustering of coefficients?

[palday]

Two things we could try:

• [ ] a priori topography and clustering / mass-univariate in time
• [ ] a priori time windows and clustering / mass-univariate in space

In terms of current data, clustering in space is a better fit. In terms of obvious programming, mass-univariate in time is easier.