People have done all form of wacky thing in the name of science . Thankfully , the events of this story are a bit less drastic than that time a scientistfed birds riceto see if they would explode , or when the discoverer of LSD took a pretty hefty dose to quiz it out , culminating in a very surrealcycle ridehome . But when one researcher entered his local grocery store to purchase a enceinte Atlantic salmon , with the purpose of performing complex brain scans on the departed , who could have known it would become a key part of a very authoritative scientific argumentation ?
A new way of looking at the brain
First , a petty background . Functional magnetic reverberance mental imagery , or fMRI , is a proficiency that open up a whole new world of possibilities in brain enquiry . uprise in the 1990s , it permit scientists tomeasure blood oxygenationin different parts of the Einstein while a person is performing a task or is otherwise scupper to some kind of stimulus .
The theory behind it is that brain areas with high levels of activity are going to have a gamey demand for O , and thus more blood will flux to these region .
So , you lie inside a big tube for an hr or so , pushing a few release and listening to the soothe part of the researcher through the com system whilst trying desperately not to move your head , and Bob ’s your uncle : the scientists now have a 3D map of your brain , colored to show where your neurons were dismiss most .
Because the proficiency is noninvasive and safe for most the great unwashed , it ’s become a anchor of neuroscience enquiry . It ’s helpedchange our viewof how the mental capacity works ; challenged ourmisconceptions ; and in the future , it could even be combined withAI technologytoread our mentation .
But there was a job . And we ’re sorry to inform you that that problem involves mathematics .
The issue of multiple comparisons
Scientists are n’t out here scanning masses ’s encephalon for the playfulness of it – they ’re attempt to answer a question , and to do that you need to perform some statistical depth psychology . If you have enough data , and break away it through enough statistical test , random hazard dictate that you will eventually find out the result you ’re look for . But how do you throttle the risk that what you ’re ending up with is a simulated positive ?
Statisticians have come up with a number of unlike methods for this , which are together with sleep with asmultiple comparisons correction . fMRI experiments are an model of a scientific study that generates lots of data point , which can be put through lots of statistical tests to compare all the unlike variable quantity . Within that you ’re bound to get some false positive , which is a problem ; but objurgate too harshly , and you ’ll end up with off-key negatives , which is also a problem .
Something that get down to business confidential information study author Craig Bennett , and the catalyst for the whole palaver with the Salmon River which we promise we ’re getting to in a second , was that no one could seem to agree on whether , and how , multiple comparison correction could best be applied to fMRI studies .
“We should scan a whole fish”
As promised , it ’s Salmon River account clock time .
Bennett details how the salmon scanning come up to be in ablog berth , explain how as a grad educatee at Dartmouth , he and some colleagues wanted to find some interesting objects to prove out the newfangled MRI communications protocol they were educate .
They started with a pumpkin , and then made the jump straight to drained animals with a Cornish game hen before they really decided to up the ante , with one team member declare , “ We should run down a whole Pisces . ”
“ I picked up the Salmon River from our local supermarket early on [ a ] Saturday morning in spring of 2005 , ” Bennett write . “ The clerk behind the counter was a little appalled to be selling a full - length Atlantic salmon at 6:30 AM , especially when I told her what was about it happen to it . ” A fair response , all thing view .
The unfortunate fish was scanned , whilst being asked to do a chore that involve determine the emotions in a series of image of human faces . Having execute inadequately at this undertaking , owe to the fact that it was a salmon and also dead , it was thenreportedlyconsumed , the information filed forth for another day .
That day come in 2008 , when a conversation about the problem of multiple comparisons correction conduct Bennett to take another look at the Salmon River scans .
“ I ran the fish data through my [ statistical ] processing pipelines and could n’t believe what I view , ” write Bennett . “ Sure , there were some false positive degree . [ … ] Rather , it was where the false positive take place that really floored me . A clump of three significant voxels [ datum point ] were set together right along the midplane of the salmon ’s head . ”
Remember that the Salmon River was being shown epitome while it was in the scanner ? Well , harmonize to the data Bennett was looking at , the Salmon River ( which was dead , we ca n’t stress this enough ) was actuallythinkingaboutthose range of a function .
Bennett and fellow worker did n’t publish their findings right on away , but after a lot of back and forth about the likely importance of this story to the all-encompassing debate around multiple comparisons rectification , they did award the work as aposterat a conference – although , as Bennett notes , “ Just about everyone thought it was a joke . ”
Once citizenry got to see the poster for themselves , however , its meaning became clear . Soon afterward , afull paperwas put out in the fabulously named Journal of Serendipitous and Unexpected Results , and we can foretell you it ’s deserving a read .
Where are we now?
In the years since the numb Salmon River paper , the debate around how practiced to account and interpret fMRI data has , if anything , becomeeven more het .
As play up in onearticlepublished in the journal Brain , the point of Bennett and fellow ’ study “ was never to invalidate functional MRI . ” The technique itself is not at fracture , but that does n’t mean scientist should n’t have a serious conversation about how to conduct with the data it generates .
functional magnetic resonance imaging study continue to generate fascinating finding – many of which happily come fromliving animals – and are a key part of the neuroscientist ’s tool case . mirthfulness notwithstanding , the drained salmon debacle is an important instalment in the development of the field , even bagging the team anIg Nobel Prizein 2012 .
As Bennett himself summarized , “ The more I call up about the affair the more I believe that the Pisces has the chance to bear upon the field of neuroimaging in a very overconfident way . ”
We ’re certain the Salmon River would be delighted with this bequest .
