A famous demonstration found significant fMRI activation in the brain of a dead salmon shown photographs. What was the point being made?

D. That with tens of thousands of voxels, some pass 0.05 by chance.

Bennett and colleagues ran an ordinary uncorrected analysis on a dead salmon and found significant voxels in its brain cavity, nothing was malfunctioning. Tens of thousands of independent tests at alpha 0.05 guarantee a cluster somewhere. It changed practice quickly, largely because a dead fish is harder to ignore than a formula.