Once some scientists made a magnetic resonance imaging at a salmon. If you think that the goal was to find out how these animals perceive the earth’s magnetic field, you are way off track, also because the salmon in question had long since died.
Our story begins in a small American supermarket, where an equally American man has gone to buy a large (probably Norwegian) salmon. The man’s name is Craig Bennett, he is a neuroscience student at Dartmouth College and intends to make a functional magnetic resonance imaging (fMRI) to his new dead fish.
After placing the salmon in the machine and asking it to recognize emotions in a series of images of human faces, Bennett archived the obtained data without analyzing it. On the other hand, his purpose was only to test the protocols of fMRI imaging, he could easily analyze the brain activity of the dead fish another day.
That day came in 2008, when a conversation about the statistical interpretation of fMRI data led Bennett to take a look at the salmon scans. “I ran the fish data through my statistical processing algorithms and I couldn’t believe what I was seeing”writes Bennett. Remember the salmon being shown the images while he was in the machine? Well, according to the data, the salmon (which was dead, we can’t stress this enough) he was actually thinking to those images.
Bennett’s findings have not invalidated fMRI, a foundational neuroscience technique that continues to generate breakthroughs. Rather, the study drew scientists’ attention to the importance of correctly interpret the large amount of data that this technique produces. Although it does not reach the levels of strangeness of the ferret in the particle accelerator, the episode of the dead salmon was therefore important for the history of neuroscience, so much so that Bennett and colleagues were awarded an IgNobel prize in 2012.