alt Hacker Newsthat... isn't... satire. It's written in a funny way because it was a "silly"[0] experiment, but it showed a real issue with fMRI.
> What we can conclude is that random noise in the EPI timeseries may yield spurious results if multiple testing is not controlled for. In a functional image volume of 60,000-130,000 voxels the probability of a false discovery is almost certain.
[0] silly as in "there's no way this fMRI will show the frozen salmon as alive, right?"
I think we run into the definition of satire and parody, here.
Satire is "the truth, in the most extreme way", so I think it definitely qualifies to very, very seriously examine a dead salmon with an fMRI to see if it has brain activity.
It's not a parody - they actually did the study, and the results were as described, not an imitation journal article ala The Onion.