Dead Salmon's "Brain Activity" Cautions fMRI Researchers 287
AthanasiusKircher sends in a Wired writeup on what should surely be a contender in the next Improbable Research competition: wiring a dead salmon into an fMRI machine and showing it pictures of humans designed to evoke various emotions. "When they got around to analyzing the voxel... data, the voxels representing the area where the salmon's tiny brain sat showed evidence of activity. In the fMRI scan, it looked like the dead salmon was actually thinking about the pictures it had been shown. ... The result is completely nuts — but that's actually exactly the point. [Neuroscientist Craig] Bennett... and his adviser, George Wolford, wrote up the work as a warning about the dangers of false positives in fMRI data. They wanted to call attention to ways the field could improve its statistical methods. ... Bennett notes: 'We could set our threshold [of significance] so high that we have no false positives, but we have no legitimate results.... We could also set it so low that we end up getting voxels in the fish's brain. It's the fine line that we walk.'" The research has been turned down by several publications, according to Wired, but a poster is available (PDF).
Re:Not a relevant comment (Score:5, Insightful)
The point of the experiment was not to prove the type of fish.
Classical case of Arrogantitis Scientificus? (Score:5, Insightful)
[..] it looked like the dead salmon was actually thinking about the pictures it had been shown. ... The result is completely nuts -- [...] as a warning about the dangers of false positives [...]
Looks to me like the dark matter syndrome: "Our theories wrong? Our calculations off by an insane amount? Unpossible! That can never be. Nature must be lying!"
Has anyone even checked if a dead brain can still have flows of energy through its brain? I mean light patterns still reach the retinas, and can still trigger signals, depending on the state of the neurons there. How long was that salmon dead? I know that pigs can be frozen to be clinically dead for long times (90+ minutes), and still be revived without much damage.
I'd at least check if there are actual signals of current going trough the brain (with an OTHER (better) instrument, before dismissing it. Every unchecked assumption is a good chance for flaw in your study. You wouldn't want it to be dismissed by peer review, because of a faulty assumption.
Discussion (Score:3, Insightful)
Can we conclude from this data that the salmon is engaging in the perspective-taking task? Certainly not. What we can determine is that random noise in the EPI timeseries may yield spurious results if multiple comparisons are not controlled for. Adaptive methods for controlling the FDR and FWER are excellent options and are widely available in all major fMRI analysis packages. We argue that relying on standard statistical thresholds (p 8) is an ineffective control for multiple comparisons. We further argue that the vast majority of fMRI studies should be utilizing multiple comparisons correction as standard practice in the computation of their statistics.
And why wasn't this published? The very conclusion is that we should be more careful when trusting fMRI results and conduct more testing before jumping to conclusion.
Re:Terri was alive (Score:1, Insightful)
Terri Schiavo's not dead, she's pining for the fjords!
Re:Classic case of idiotus not understandus (Score:2, Insightful)
What do you actually know about dark matter and the current state of the evidence? Do you even understand it at a layman's level let alone understand the insanely complex math? Have you heard of the bullet cluster? Do you know about the rotation curve of galaxies? Do you understand anything about the cosmic microwave background and its fluctuations? Do you understand the background theories you're ridiculing? Do you know why General Relativity fits the data we have collected so well? Have you even bothered to find out why scientists believe in these things?
NARRATOR: Tune in next week when the physicist says, "Oh shit! I forgot to divide by two!That changes EVERYTHING!"
Sorry, I couldn't resist.
Re:spoooooky (Score:4, Insightful)
"It's OK to eat the barrel of a gun cause then you don't have any feelings."
-- K. Cobain
Re:Discussion (Score:3, Insightful)
It's always easier to jump to conclusions than to jump from them.
~X~
Vegetarians got another argument (Score:3, Insightful)
Re:igNobel on the way! (Score:5, Insightful)
I thought igNobel prizes were for genuinely useless research. This research is very useful for highlighting some of the problems with fMRI research.
Re:spoooooky (Score:2, Insightful)
Re:Classic case of idiotus not understandus (Score:3, Insightful)
"Do you know why General Relativity fits the data we have collected so well? Have you even bothered to find out why scientists believe in these things?"
One might well wonder, because it's certainly not because GR is philosophically compatible with the rest of 20th century science.
As a matter of hobbyist curiosity, I'm reading up on the life of Einstein and his arguments with the QM people at the moment, and the curious thing that jumps out at me is how much Einstein believed that GR was only a provisional theory, and that the 'true' description of the universe had to be a geometrical theory of continuous fields. Which led him to various configurations of Unified Field Theories, and increasing isolation from the quantum hackers who believed that reality had to be fundamentally discontinuous.
John Wheeler tried to push UFT with geometrodynamics and gave up.
Today, UFT has a sort of funky steampunk aura about it, like quaternions. If it weren't for GR still holding a place in cosmology, Einstein's whole geometrical approach would be considered clever and ambitious but fatally flawed, just as his UFT is.
So yes - why *does* it fit the data? It's not necessarily because it's a literally correct representation of reality. At best it must be an approximation, because about the only thing we know for sure about modern physics is that neither GR nor QFT can be 'true' in a final sense.
Re:spoooooky (Score:5, Insightful)
The mechanisms are the most important thing. What is fMRI actually measuring? It doesn't measure activity directly, since it's not built into the brain. Ergo, it measures activity indirectly by measuring something else entirely. But anything which also generates that something else will also be detected.
This is less a false positive than it is a complete confusion between direct and indirect observations. The falseness is not in the measurement but in the observer.
Sir Arthur Conan Doyle would have loved this finding, as he often has his most famous creation of Sherlock Holmes make snide remarks about the folly of poor observation and the absurdities that follow.
Re:Terri was alive (Score:5, Insightful)
But this study showed that dead salmon can show just as much brain activity as Terri Schaivo...This study just shows that a "dead" organism with a brain that hasn't yet decomposed can still support some processes.
Bzzzt. Wrong. The entire point the write up was to warn about the danger of false positives. Your attributing of brain activity to random, natural noise is exactly the danger they want to avoid.
Re:Classic case of idiotus not understandus (Score:2, Insightful)
You should read through The Feynman Lectures, Vol. 1, where Feynman reminds us that there is no such thing as a 'true' theory in science. All science has ever done is to describe natural phenomena as well as possible. This isn't a flaw of science, and it's not even a shortcoming. The best theories we have are at best approximations, and any new theory we'll ever cook up will only be better, more sophisticated approximations. GR and QFT are as 'true' as we need them to be in the overwhelming majority of cases, in that they will let us predict what how nature will behave (with remarkable accuracy).
To sum up: Dark Matter, GR, QFT, etc. are only as wrong as F = ma.
Re:Classic case of idiotus not understandus (Score:3, Insightful)
We have absolutely no direct evidence of either.
Define "direct evidence". There's no "direct evidence" that the wind exists, but you accept it does because you see it's effects. The Bullet Cluster results provide equivalent evidence for the existence of Dark Matter. I mean, you *did* look up the BC results, didn't you?
Similarly, the expansion rate of the universe is accelerating. Period. Of this there is absolutely no doubt, as we actually do have direct evidence demonstrating it. What's causing it? No one knows. So DE is the term that's used as a placeholder.
Either way, we don't have scientists just making shit up and refusing to believe their theories are wrong. We have data which fits certain theories, but not others. For example, the BC results disprove any theory that *doesn't* include some sort of weakly interacting matter (such as pure MOND-style theories).
Re:spoooooky (Score:4, Insightful)
* Dave Lister: Sometimes I think it's cruel giving machines a personality. My mate Petersen once brought a pair of shoes with artificial intelligence. Smart Shoes, they were called. It was a neat idea. No matter how blind drunk you were, they would always get you home. Then he got ratted one night in Oslo, and woke up the next morning in Burma. See, the shoes got bored just going from his local to the flat. They wanted to see the world, man, y'know? He had a helluva job getting rid of them. No matter who he sold them to, they'd show up again the next day! He tried to shut them out, but they just kicked the door down, y'know?
* Arnold Rimmer: Is this true?
* Dave Lister: Yeah! Last thing he heard, they'd sort of, erm, robbed a car and drove it into a canal. They couldn't steer, y'see.
* Arnold Rimmer: Really?!
* Dave Lister: Yeah. Petersen was really, really blown away by it. He went to see a priest. The priest told him, he said, it was alright, and all that, and the shoes were happy, and they'd gone to heaven. Y'see, it turns out shoes have soles.
Re:Classic case of idiotus not understandus (Score:1, Insightful)
Keep in mind that until last week, we had no direct evidence of something so basic to modern physics as the Bohr model
We still have no direct evidence of the Bohr model, because the Bohr model is basically wrong. This has been know for over 80 years. I don't know what you are referring to that happened last week, but as far as the quantum theory of electronic structure, we have had vast quantities of direct evidence for decades.
e have numerous alternative theories that explain, without resorting to saying the universe consists of 96% invisible voodoo, various anomalies such as gravitational rotation and the implied anisotropy of the CMB.
Competing theories have been proposed, vetted and discarded. For a relatively recent data-oriented paper, see: N. Spergel et al 2007 ApJS 170 377-408 10.1086/513700 [doi.org], arXiv [arxiv.org].
Selected quote:
1. Cold dark matter serves as a signiïcant forcing term that
changes the acoustic peak structure. Alternative gravity models
(e.g., MOND), and all baryons-only models, lack this forcing
term so they predict a much lower third peak than is observed by
WMAP and small-scale CMB experiments (McGaugh 2004;
Skordis et al. 2006). Models without dark matter (even if we
allow for a cosmological constant) are very poor ïts to the data.
For a more far ranging, theoretical discussion: Bertone,Hooper,Silk. Physics Reports. 405 2005 279-390 [doi.org], arXiv [arxiv.org].
As well as a comparison of alternative theories, this includes a few sections that concern efforts to determine the precise identity of dark matter, as well as future attempts to detect it directly.
Somewhere you got the idea that scientists sort of waved their hands, said "It must be dark matter," and then stopped thinking about it. Scientists have eliminated well known explanations, and so are exploring the unknown explanations, trying to make them known. This is not as scandalous as everyone seems to think.
Theories come and go, and without reproducible, experimental evidence, we have at best a model that fits the data - NOT, as far too many people seem to believe, a necessarily accurate description of objective reality.
Strictly speaking, all science can do is provide models that fit the data. All "reproducible, experimental evidence" is but more data. The argument that science actually reveals and describes objective reality is largely a philosophical one
(and to certain extent a semantic one). For an excellent discussion (unfortunately only for those who have access to Physics Today), see:
N. David Mermin, Phys. Today 62, 8 (2009). What's bad about this habit [doi.org]
Re:Classic case of idiotus not understandus (Score:3, Insightful)
Keep in mind that until last week, we had no direct evidence of something so basic to modern physics as the Bohr model
Sorry, but this statement alone indicate that you don't know what you are talking about.
First off, the Bohr model is wrong, we already knew that. But if you really mean the model of electron orbits, the means Quantum Electrodynamics, then it has been measured and tested and is correct to umpteenth decimal places, that you would have a hard time finding another theory that was tested even more than QED.
If you insist that only pretty pictures could mean "direct evidence" then you know nothing about actual science.
Re:Classic case of idiotus not understandus (Score:4, Insightful)
We have numerous alternative theories that explain, without resorting to saying the universe consists of 96% invisible voodoo, various anomalies such as gravitational rotation and the implied anisotropy of the CMB.
No. We don't. That's the whole point. Dark matter wasn't invented for the hell of it. Astronomers resisted it for decades. It was ultimately accepted precisely because it continued to pass observational tests and other theories didn't.
It's possible to cook up alternative theories to explain individual phenomena such as galactic rotation curves (e.g., MOND). But they all fail when you try to simultaneously explain multiple phenomena such as galactic rotation curves and CMBR anisotropies and early universe structure formation and galaxy cluster dynamics and ... you get the idea.