Motor-cortex sanity check using a single button-press impulse regressor against the implicit baseline (ses-01+ses-02). Three variants are shown: with task regressors that (should) absorb cognitive/visual variance, without task regressors, and without task regressors restricted to runs 1–5 (may have switched hands after run 5).
Button-press activation without task boxcar.
Button-press activation without task boxcar, runs 1–5 only (consistent hand).