Multi-way model comparison maps.

<p>The maps show voxels where pre- versus post-stimulus activity is best discriminated by each model, for V1 (left) and FFG (right). (A) V1 view from [2 −96 −5]mm (B) FFG view from [36 −47 −19]mm (Talairach coordinates). Models 1, 7, 9 and 11 were not best at any voxel. Note that the null model is best for anterior FFG, but model 3 is best for posterior FFG.