Neuro-realism

October 30th, 2010 by Ben Goldacre in bad science, neuro nonsense | 36 Comments »

Ben Goldacre, The Guardian, Saturday 30 October 2010

When the BBC tells you, in a headline, that libido problems are in the brain and not in the mind, then you might find yourself wondering what the difference between the two is supposed to be, and whether a science article can really be assuming – in 2010 – that its readers buy into some strange form of cartesian dualism, in which the self is contained by a funny little spirit entity in constant pneumatic connection with the corporeal realm.

But first let’s consider the experiment they’re reporting on.

As far as we know (this experiment has not yet been published, only presented at a conference) some researchers took seven women with a “normal” sex drive, and 19 women diagnosed with “hypoactive sexual desire disorder”.  Participants watched a series of erotic films in a scanner while an MRI machine took images of blood flow in their brains: the women with a normal sex drive had an increased flow of blood to some parts of their brain associated with emotion, while those with low libido did not.

Dr Michael Diamond, one of the researchers, tells the Mail: “Being able to identify physiological changes, to me provides significant evidence that it’s a true disorder as opposed to a societal construct.” In the Metro, he goes further: “researcher Dr Michael Diamond said the findings offer ‘significant evidence’ that persistent low sex drive – known as hypoactive sexual desire disorder (HSDD) – is a genuine physiological disorder and not made up.”

This strikes me as an unusual world view. All mental states have physical correlates, if you believe that the physical activity of the brain is what underlies our sensations, beliefs and experiences: so while different mental states will be associated with different physical states, that doesn’t tell you which caused which. If I do not have the horn, you may well fail to see any increased activity in the part of my brain that lights up when I do have the horn. That doesn’t tell you why I don’t have the horn: maybe I’ve got a lot on my plate, maybe I have a physical problem in my brain, maybe I was raped last year. There could be any number of reasons.

But far stranger is the idea that a subjective experience must be shown to have a measurable physical correlate in the brain before we can agree that the subjective experience is real, even for matters that are plainly experiential. If someone is complaining of persistent low sex drive, then they have persistent low sex drive, and even if you could find no physical correlate in the brain whatsoever, that wouldn’t matter, they do still have low sex drive.

Interestingly, this odd interpretation is far from new: in fact it’s part of a whole series of recurring themes in popular misinterpretations of neuroscience, first described formally in a paper from Nature Reviews Neuroscience called “fMRI in the public eye”.  To examine how fMRI brain imaging research was depicted in mainstream media, they conducted a systematic search for every news story about it over a 12 year period, and then conducted content analysis to identify any recurring themes.

The first theme they identified was the idea that a brain imaging experiment “can make a phenomenon uncritically real, objective or effective in the eyes of the public”. They described this phenomenon as “neuro-realism”, and the idea is best explained through their examples, which mirror these new claims about libido perfectly.

So an article in the Washington Post takes a view on pain, and whether the subjective experience of it is enough: “patients have long reported that acupuncture helps relieve their pain, but scientists don’t know why. Could it be an illusion?” They have an answer. “Now brain imaging technology has indicated that the perception of pain relief is accurate.”

Another says that brain imaging “provides visual proof that acupuncture alleviates pain”. The reality, of course, is much simpler: for your own personal experience of pain, which is all that matters, if you say that your pain is relieved, then your pain is relieved (and I wish good luck to any doctor who tells his patient their pain has gone, when it hasn’t, just because some magical scan says it has).

The New York Times takes a similarly strange tack in a brain imaging study on fear: “Now scientists say the feeling is not only real, but they can show what happens in the brain to cause it.” Many people find fatty food to be pleasurable, for the taste, the calories, and any number of other reasons. When a brain imaging study showed that the reward centres in the brain had increased blood flow after subjects in an experiment ate high-fat foods, the Boston Globe explained: “Fat really does bring pleasure.”

They’re right, it does. But it’s a slightly strange world when a scan of blood flow in the brain is taken as vindication of a subjective mental state, and a way to validate our experience of the world.

And another thing:

No space to weave it in (the column was over length already, I can’t bring myself to look at what went in the paper) but the other relevant theme here is the medicalisation of sex, if you’re interested in more there’s a good interview with Ray Moynihan on Pharmalot, I wrote about the medicalisation of everyday life here and in these, and Petra is often good on it here. Lastly Neuroethics wrote on this study and dualism.


++++++++++++++++++++++++++++++++++++++++++
If you like what I do, and you want me to do more, you can: buy my books Bad Science and Bad Pharma, give them to your friends, put them on your reading list, employ me to do a talk, or tweet this article to your friends. Thanks! ++++++++++++++++++++++++++++++++++++++++++

36 Responses



  1. deevybee said,

    October 30, 2010 at 1:19 am

    I wonder what their conclusion would be if they’d measured blood supply to the genitals rather than the brain.
    Seriously.

  2. Jon_s said,

    October 30, 2010 at 3:02 am

    @deevybee This paper measures erection size:
    www.ncbi.nlm.nih.gov/pubmed/18598769
    In my opinion it makes the same errors Ben outlines above; a physical correlate of a mental phenomena is still a correlate.
    I think TMS can help with some of these issues because it allows you to interfere with, or cause, physical states and observe the mental consequences… but again this is really just sophisticated correlation- what isn’t?

    Interesting stuff.

  3. Jon_s said,

    October 30, 2010 at 3:16 am

    Oh and – this also reminds me of a quote from quinning qualia by Dennett (ase.tufts.edu/cogstud/papers/quinqual.htm):

    “Everything real has properties, and since I don’t deny the reality of conscious experience, I grant that conscious experience has properties.”

  4. oldandrew said,

    October 30, 2010 at 8:59 am

    The variation on this that I keep encountering is that not only must complaints be real if they have neural correlates, but that they must have a neurological cause. People keep telling me that dyslexia, ADHD, ASD *must* be “neurological conditions” because they have neurological correlates. (It is usually then followed by the claim that in the future behaviourial or learning conditions will be diagnosed entirely by brain scans.)

    The possibility that our behaviour and experiences shape our brains, rather than our brains shape our behaviour is never considered. It is a strange one-way dualism, that declares that while our brain might shape our minds, we cannot expect our minds to affect our brains.

  5. fontwell said,

    October 30, 2010 at 9:42 am

    People only make this error of logic because of how their brains work.

  6. basilisk said,

    October 30, 2010 at 10:53 am

    So if they did a MRI scan of someone writing a Daily Mail health article, would that reveal whether the content was real or imaginary?

  7. beeflin said,

    October 30, 2010 at 3:03 pm

    Obviously everything that is sensed must physically exist too. There’s no point going to the trouble of positing a split – why bother unless you’ve got an obsession with doing so. The only difference is between those sensations that are actually measurable and those that aren’t because the measurement isn’t (yet) possible.

  8. Esther Israel said,

    October 30, 2010 at 3:37 pm

    Research on brains and sexuality seems straightforward, but it often has hidden assumptions and stereotypes. The diagnosis of HSDD is controversial in and of itself. (When is lack of interest in sex a disorder?). Trying to understand what the brain does while people look at sexy videos in a MRI is scientifically complex. People with different backgrounds are not likely to react the same way, and the equipment is technical. Also, reporting this kind of research needs smart science reporters. Rather than having smoke blown in the public’s eyes, the reader of these research studies should know that one such study cannot determine that low desire is a “real” problem.
    Esther Israel, on behalf of the New View Campaign, newviewcampaign.org

  9. J.McGuinness said,

    October 30, 2010 at 6:08 pm

    I’m currently working towards my PhD in a Neuroscience field… theoretical (or computational) Neuroscience though, not experimental, and, although I do look at a lot of experimental work, I only look at imaging studies on the periphery of my work (they wouldn’t tell me anything about the sort of things I am investigating in neurons). However, I’ve been quite wary of imaging for a while now… a lot of the time there is either over zealous or faulty interpretation by the researchers themselves, or pretty horrible reporting by the media (they seem like 2 distinct problems to me). For the 1st, I seem to remember there was a study done that showed, using the same statistical analysis and interpretation of brain activity used in ‘normal’ imaging studies, that demonstrated a frozen, quite, quite dead Salmon (Atlantic) was emotionally aroused when viewing certain images. The researchers had decided to use the dead Salmon for calibration of the scanner, but when they analysed the data, it ‘showed’ activity. This the first Google hit for “Dead Salmon fmri”, which has more on that particular study. www.wired.com/wiredscience/2009/09/fmrisalmon/

    I was also reading, quite recently, an article from Slate (2 years old now, but still worth a look… www.slate.com/id/2204590) detailing pretty hapless reporting of an imaging study in The New York Times… this is a good example of the 2nd problem I mentioned above, when you just consider the NYT work… but I think the study itself was pretty poor (the subjects are described as ‘bullies’, when ‘sociopaths’ would be a better term… very confident interpretations of activity in certain brain areas when the consensus is contrary to interpreting that response).

    I think imaging studies can be useful, but I think we have to be very careful in how (as researchers and scientists) they are interpreted and the conclusions that are drawn… as well as overcoming the problems inherent in all science journalism, and how those conclusions and interpretations are reported to the public. Unfortunately, there is, as you state in the article, this idea or belief that activity shown by imaging is some sort of concrete evidence for whatever findings are reported.

    While we’re on the Neuro-scepticism trip, I think I am developing into a very sceptical Neuroscientist… My current mantra is “Neuro-plasticity is a dirty word” (mindhacks.com/2010/06/07/neuroplasticity-is-a-dirty-word/), and my favourite article from the last year has been David Parker’s “Neuronal network analyses: premises, promises and uncertainties.” (in a special issue of the Philosophical Transactions of Royal Society of London (Biology))… I think it’s as critical, and as scathing, an article I’ve ever seen, while still being constructive… highlighting the problem of over confidence, spurious conclusions, and the culture of citation and support that allows these problems to go un-rectified.

  10. J.McGuinness said,

    October 30, 2010 at 6:09 pm

    Sorry… that Slate link above should be www.slate.com/id/2204590 . I let a bracket get involved there.

  11. William Levack said,

    October 30, 2010 at 7:34 pm

    RE: “Obviously everything that is sensed must physically exist too. There’s no point going to the trouble of positing a split – why bother unless you’ve got an obsession with doing so.”

    I have difficulty with just accepting, without critique, a purely physical explanation for everything in the world – including one of the universe’s greatest puzzles: consciousness. Arguments like that stated above, or Ben’s quick dismissal of cartesian duality, is equally based on unfalsiable assumptions about the nature of world-stuff. And no, I’m not a spiritualist, a ghost-wrangler, or even religious. I’m just saying that sentences which state ‘everything that is sensed must physically exist too’, shouldn’t start with the word ‘Obviously’. Take the example of pain for example. No one would doubt the existence of pain, and yes nociception can be demonstrated with objective methods – but pain is not just nociception. Ben’s discomfort with the idea of a doctor stating that pain has gone because the physical representation of pain on a brain scan has gone is (I would argue) in part due to an appreciation that we are far from accepting that pain is a purely physical experience. I would agree that one day it might be POSSIBLE to have some scan that correlates 100% with the lived experience of pain, but at present this type of scan doesn’t exist. Conversely, perhaps it is NOT possible to ever make such a scan, because some aspect of pain does not physically exist to be measured and observed.

    Just saying…

  12. SheriffFatman said,

    October 30, 2010 at 8:34 pm

    @Esther Israel: “When is a lack of interest in sex a disorder?”

    When it’s causing distress to the person concerned?

  13. William Levack said,

    October 31, 2010 at 12:12 am

    FYI: unfalsiable = unfalsifiable in my post above by the way.

  14. earcatching said,

    October 31, 2010 at 12:16 am

    I agree with the main thrust of the argument, that people are too willing to believe anything associated with brain scanning. But I think there is potentially more to the acupuncture brain-imaging study than this post suggests. There are two reasons why someone might report less pain after acupuncture: (1) they feel less pain or (2) they change the way they rate their pain (perhaps to humour the acupuncturist or experimenter, or maybe out of wishful thinking). A brain-imaging study could help distinguish these two options, because different parts of the brain would be expected to be affected by each possibility.

  15. SimonW said,

    October 31, 2010 at 12:27 am

    “People only make this error of logic because of how their brains work.”

    Do you have any imaging to support that claim ;)

  16. Dan Kimberg said,

    October 31, 2010 at 5:46 am

    @J.McGuinness: As an occasional fMRI guy, I can’t help commenting on your comment. I’m certainly in agreement that misreporting byscientists and by the media are separate and serious problems (I occasionally revisit an old article of Ben’s on the latter point, I’ll paste the link below). For every decent scientific article I read, I read ten more than convince me there’s no hope; and virtually every popular report of anything I know anything about seems to be at best misleading. So I’m not surprised that both researchers and reporters would say some odd things about what imaging can tell you about the reality of mental states. And the two interact in an unfortunate way — many scientists get a little crazy when their work comes to the attention of the popular press, or when they think it might. Some get a lot crazy.

    But I can’t help ranting a bit about the dead salmon study. The perpetrators of that poster intentionally botched the statistics, and predictably got ridiculous results. However, they did not in fact botch the statistics in a way that’s common in neuroimaging — virtually no one fails entirely to correct for multiple comparisons. (Many researchers fail to correct correctly, but that’s a separate issue.) The media for the most part mis-interpreted this as some kind of revelation, but it’s no surprise to anyone who has even the most rudimentary science education. In fact, there’s no reason to carry out the analysis with dead salmon data, except for showmanship — no one pays attention if you simply put up a poster that says “Failure to correct for multiple comparisons increases false positive rate.” This would be a much more honest, if less newsworthy, presentation. (And I’m not sure why a dead salmon would be used to “calibrate” the scanner, we have phantoms for that purpose, or in a pinch, graduate students.) Your impression of it, I gather, was that the analysis used on the salmon data was the same as used in normal studies. But I believe this is only true for the most extremely incompetent researchers, who only rarely sneak this kind of thing into the better journals anymore. More typically incompetent researchers just botch the correction, resulting in a false positive rate that’s elevated but not dramatically so. So I really hate to see the salmon study given positive press, because it’s often used to make fMRI seem more bogus than it is. fMRI has enough bogus-ness on its own without having to add on this made-up kind (which, after all, could apply as well to virtually any scientific field that uses inferential statistics).

    Here’s that link:

    www.badscience.net/2005/09/dont-dumb-me-down/

    It’s a great article that never fails to upset me.

  17. parrhesiast82 said,

    October 31, 2010 at 12:33 pm

    If women aren’t interested in sex, it’s a disorder?

    Nothing to do with men?

  18. tialaramex said,

    October 31, 2010 at 4:04 pm

    “When it’s causing distress to the person concerned?”

    Exactly. Most new (and indeed existing) psychiatric disorders are characterised as symptoms specific to that disorder plus “Causes harm to the patient or others”.

    Person A, who spends 20 hours a week writing stories celebrating gruesome murders and cannibalism, but doesn’t see any problem with this and shows no intent of actually murdering anyone or eating them, is healthy.

    Person B, who occasionally shouts “Boo!” for no reason they can explain, but is so worried by this phenomenon that they refuse to go out, can’t get a job, etc. has a psychiatric disorder.

    Person C has tripped and fallen down a flight of stairs, they have a broken leg and possible concussion.

    Person A needs nothing from us, and should be allowed to carry on their interests without harassment. Person B needs our help and deserves our sympathy just as much as Person C.

  19. noahmotion said,

    October 31, 2010 at 6:39 pm

    …it’s part of a whole series of recurring themes in popular misinterpretations of neuroscience…

    The quote from the researcher involved in the libido study indicates that the misinterpretations aren’t only popular.

  20. Dan Kimberg said,

    October 31, 2010 at 7:49 pm

    As an occasional fMRI guy, I’m certainly in agreement that misreporting byscientists and by the media are separate and serious problems. It’s sad but not surprising that both researchers and reporters would say some indefensible things about what imaging can tell you about the reality of mental states. But there’s nothing special in this regard about neuro-realism — so far as I can tell, no area of science is generally well-reported. The errors of researchers and reporters interact in an unfortunate way — many scientists get a little crazy when their work comes to the attention of the popular press, or when they think it might. Some get a lot crazy.

    @J.McGuinness: Be careful not to over-interpret the dead salmon “experiment,” which was just a bit of showmanship piled on top of a simple demonstration of a very trivial statistical point (that if you don’t correct for multiple comparisons, you get an inflated family-wise error rate). Only the most incompetent researchers make that error in all its glory (although many do botch the correction). However, the lessons and details of that exercise seem to be mis-reported just about everywhere. It really tells us nothing about fMRI whatsoever.

  21. anon3455 said,

    October 31, 2010 at 11:51 pm

    Seems a bit weird thing to bitch about, really.

    I mean come on, physiological vs. psychological. There is a world of difference.
    Sure, all psychological phenomena probably have a physiological reason at some level, but there is a divide in that problems at certain level of abstraction (for want of a better word) can be corrected through psychiatric therapy or psychotropic drugs, whereas some lower abstraction levels are clearly physiological and may only corrected through non-psychoactive drugs, surgery, other physiological manipulations or, in case of brains, possibly not at all.

    So, mind => psychological vs. brains => physiological.

  22. Oldfart said,

    November 1, 2010 at 10:55 am

    Surely you all are being a little obtuse here. English grew and evolved through centuries during which no one ever KNEW that what another person said they felt was not just an excuse for getting out of work or not doing something they should be doing. After all, for how long were germs unknown? So concepts of mind were based on deduction from observed behavior and interviews, not on objective facts tying that behavior and anecdotal evidence to actual changes in the brain. Those paradigms are still among we lay people. It will take much time for the concept that mind and brain are one to take hold.

  23. anonymous coward said,

    November 1, 2010 at 12:29 pm

    “if they did a MRI [brain] scan of someone writing a Daily Mail health article” they’d have to conclude, other evidence to the contrary, that they were scanning a corpse.

  24. Szwagier said,

    November 1, 2010 at 12:57 pm

    Perhaps I’m being obtuse, but I really can’t understand what’s going on here. If we assume that all mental states are somehow reflected physically in the brain (and why wouldn’t we), then whether there is “real” low sex drive or “imagined” low sex drive, these would have correlates in brain operation.

    I’d think there would be cause to be a lot more worried if there were *no* physical correlation between “thinking” and physical brain function.

  25. Mike Lisieski said,

    November 1, 2010 at 6:23 pm

    If you believe that neuroscience can work, you almost have to be a monist. There is no way to approach neuroscience as a dualist, except to declare that it’s futile. A basic assumption of biology is that an organism’s behavior is determined by the interaction between that organism and the environment – a basic tenet of neuroscience holds that the nervous system is one of the primary mediators of this relationship. Every organism-wide behavioral state in animals that have nervous systems involves neurophysiological phenomena.

    The history of drive theories in psychology might have something to do with the reason that researchers are so eager to find neurophysiological correlates for purported states of desire. Once upon a time, psychologists came up with the idea that internal drives cause behavior in an organism – it eats because it is hungry, it drinks because it is thirsty, it has sex because it has sex drive. These turn out to be theoretical nightmares. For one, they are circular explanations – an animal eats because it is hungry, but how do we tell if an animal is hungry? Because it eats, of course! Secondly, the general scheme, if it’s accepted as valid, can be used to “explain” anything; for example: why do people commit crimes? People must have a criminal drive which drives them to it! How do we know if someone has this drive? Well, they’ll commit crimes, of course! (Believe it or not, I’ve read psychology papers, ableit old ones, that use the concept of a “drive to be criminal” to explain the behavior of career criminals.)

    The reliance on such vague concepts as “sex drive” doesn’t get one very far when theorizing about what actually causes behavior – at the most, it’s a way of labeling the black box. A more accurate name for the things were referring to might be “all the factors that influence sexual behavior and cognitive events that we haven’t identified yet.” Showing neural correlates of reported sex drive potentially identifies part of the black box, and so opens the way to start actually theorizing about how sex drive works biologically.

    It is interesting to me that showing that something has neural correlates allows it to be more comfortably classified as a disorder. I’ll be that people refusing and offer to have sex and people accepting an offer to have sex have different patterns of cortical activation during that event – yet this still wouldn’t prove that people who refuse the offer have a disorder of brain function, or that we should treat them as victims.

    What I mean is: nice post.

  26. andrewwyld said,

    November 1, 2010 at 8:38 pm

    This does remind me of the Raymond Smullyan story “An Epistemological Nightmare”:

    www.mit.edu/people/dpolicar/writing/prose/text/epistemologicalNightmare.html

    but surely the phrase “a societal construct” is also significant: perhaps people would be happy with their low sex drive if the rest of society weren’t constantly bombarding them with the message that normal people are constantly enjoying uncomplicated and delightful coition with almost pendulum-like regularity.

    Put another way, I have a low watching-football-matches drive, in that most of the time I don’t mind missing football matches, even though when I do watch one I enjoy it. Nobody would wonder whether this has a physical or a psychological cause though.

    However, should I solve the problem with a psychological solution—stop caring whether people think I am not masculine for not caring much about football—or a physical, by undergoing Clockwork Orange-style conditioning?

    Less idiotically, would some with low sex drive prefer to embrace a comparatively less sexual lifestyle where others would prefer to reignite their libido? Is the problem perhaps that we assume there should be one correct answer?

  27. cm said,

    November 2, 2010 at 10:34 am

    @Dan Kimberg @JMcGuinness – the dead salmon poster is fun, you’ve both taken it the wrong way. Dan, it’s a joke, and almost everyone gets that (and in it is nested a cautionary tale that is worth telling). JMcGuinness, the erroneous analytical technique used on the salmon is (extremely) uncommon. There are other statistical issues in fMRI, such as non-independence, as covered by Ed Vul, Hal Pashler and colleagues in the voodoo correlations (—>implausibly high correlations post editing) paper. Getting past the basic analysis, the more interesting and important point is that fMRI has a very limited capacity to tell us anything about cognition and computation. It measures blood flow, it can’t necessarily distinguish inhibition and excitation, it’s scale is vastly too large to make many inferences about what is happening at the level of the neuron, and much more. And then, what inferences can be drawn about cognition from neural data, if any? Max Coltheardt has written extensively on this as have many others. At the end of the day, I’m much more interested in cognition that neural blood flow. Chapeau, Ben!

  28. fontwell said,

    November 3, 2010 at 10:08 am

    @William Levack

    If something can be detected but doesn’t physically exist, what other options are there?

  29. walks with tench said,

    November 3, 2010 at 8:06 pm

    sorry to change the subject, but if you’re quick, there’s a brilliant use-of-science-to-give-authority on the BBC news webpage (via the google home-page gadget thing)(wensdy eve); a 2for 1 in fact.
    On the Deep Impact Calculator story they state that it is “underpinned by scientifically accurate equations” as if scientifically accurate is more accurate than accurate and if there are non-accurate equations (inequations?) (ok spose there are approxomations but they’re not really equations are they?)Sorry to break protocol, just thought you’d all find it amusing!

  30. ronmurp said,

    November 4, 2010 at 8:17 am

    Accepting the main criticisms of science reporting I think Ben answers one of his questions simply by asking it, in the paragraph beginning “When the BBC tells you, in a headline…”

    Isn’t it the case that in 2010 many people do believe in a mind-brain dualism (or soul-brain, or agency-brain)? And, that the fMRI reporting is simply pointing out that the data shows activity in parts of the brain that can be studied (however crudely and however tentative the results may be) and related to behaviour, and may be sufficient to better understand behaviour, without considering the extra mind, soul, agent?

    The significant point is that brain activity, no matter how correlated to mind activity, doesn’t actually refute mind-brain dualism. A dualist can easily hypothesize (or claim) that the brain activity is simply what the mind is doing in the brain to enact the mind’s intentions (though without saying how this is acheived).

    The problem for dualism isn’t addressed by fMRI work – so fMRI data isn’t a problem for dualism. The problem for dualism is simply the total lack of any evidence for it. It’s a speculative hypothesis with no data of its own.

  31. MedicLewis1 said,

    November 5, 2010 at 1:52 am

    The BBC have been at it again with the old poorly reporting neuroscience. Do you think it would be asking too much of Ben to have a look at this cure for dyscalculia that made it into their news reports? It was truly terrible journalism. A milli amp in the correct direction (poor theoretical knowledge of electricity shown by the BBC, direction of flow of electrons or the pre atomic model of + to – flow?) improves ability to perform in maths where the opposite makes you as poor as a six year old. A trial where n=15.

    I’m sure I’ve seen a suggest topic to Ben link somewhere once but long since forgotten where it was. Maybe that is due to a milli amp or two in the wrong directions who knows. For anyone who missed it;
    www.bbc.co.uk/blogs/thereporters/ferguswalsh/2010/11/_a_shocking_idea_or.html

  32. incitatus said,

    November 5, 2010 at 4:08 pm

    “Isn’t it the case that in 2010 many people do believe in a mind-brain dualism (or soul-brain, or agency-brain)? ”

    Oh I do hope not. But you may be right.

    The real difficulty in all of this mess isnt anything to do with brains, sex, “souls” (shudders and washes out brain) or even fmri (which keeps lots of biospods employed in a relatively non-harmful way) but in correlation and interpretation.

    a correlates with b therefore b presupposes a is logically void. Cholera infections correlate with a lack of sewage smell in the local environment, therefore cholera is caused by the smell of sewage?

    The issue with correlates in real world situations is, as pointed out many times by mine host, the near impossibility of removing artefacts from the data. exactly the same with the “people who eat x have less cancer therefore x cures cancer” tripe the Mail trots out. The fact that people eating organic blueberries dipped in Tibettan Yak vomit (or whatever the cure du jour is) also happen to be wealthy, leisured, unstressed and called Tarquin or Miranda is never mentioned.

  33. ALHolland said,

    November 7, 2010 at 5:37 pm

    …and did the experiment take into account that sometimes, some people get turned on by visual stimulus which would involve, i’ll wager, blood flow to the brain, whereas sometimes, some people get turned on by having their gentitals stimulated (somehow) and therefore, not so much blood flow (maybe) and that sometimes, some people get turned on by having ridiculous expensive experiments done on them by men in white coats (yeah baby) and sometimes, maybe, not so much….

  34. doraphillips90 said,

    November 8, 2010 at 11:18 am

    As a psychology student this article definitely riled me. First of all, researchers are fully aware that their results rarely (if ever) prove or disprove anything. We are encouraged to say that the evidence found from a piece of research supports a hypothesis and in fact lose marks for writing that any of our experiements prove or disprove a hypothesis. All confounding variables are taken into account and in fact are included in all (good) papers.

    Funnily enough it is the media who take scientific research out of context and it is the media who make the claims, not the scientists. I actually did a whole module last year on psychology in the media. Interestingly we were able to fill a whole eleven week course on how the media often portrays scientific research in the wrong light.

    Secondly, the rather negative view of research that has been undertaken is unwelcome. Nearly all scientific research increases our knowledge and provides us with further insight into ourselves and the world around us.

    I agree that science is often reductionist in nature however this piece of media in particular is not taking into account that scientists openly hold their hands up to this.

    And in regards to whether scientists believe something is real or not based on physiology – psychologists are scientists and are also well aware of the discrepency between what a person feels or experiences and physiological changes (phantom pain in amputated limbs for example).
    That said, trying to understand how these phenomenons occur through analysis of physiology is of undeniably great importance.

    I think it is thus necessary to consider the difference between what scientists say about what they have found and how the media reports it.

  35. Dan Kimberg said,

    November 9, 2010 at 4:44 am

    @cm, *I* know that it’s a joke, but the popular media doesn’t. And the authors seem to answer serious questions about it seriously, which compounds the problem. Google for popular reports about it, and you’ll find both the salmon poster authors and some very distinguished researchers saying things that convey the impression that it’s a shocking and troublesome finding, as opposed to a cute demonstration appropriate for a first course in statistics. In fact it’s the latter.

    The Vul article is more complicated, because it’s generated a lot of noise within the field, and somewhat less in the popular media. But I think it’s harmful in a similar way, in that it sensationalizes an issue that ultimately isn’t particularly complicated or particularly worrisome. I think the Nichols/Poline response was fairly on-target. At the end of the day, if we want things to worry about, we’re much better off reading John Ioannidis than Ed Vul.

    In any case, while I agree that the limitations to fMRI are considerable, I don’t consider it as hopeless as you do. Even on days when I do, I still get irked by this kind of unhelpful nonsense. There are plenty of legitimate reasons to worry about how useful fMRI is without having to discuss dead fish and voodoo correlations. But of course dead salmon make for great “science” stories, and so does “neuro-realism.”

  36. daphne said,

    November 19, 2010 at 7:20 pm

    I’m missing something, or misunderstanding. The illogic of acupuncture, or homeopathy or other pseudoscience notwithstanding, the article seems to imply that fear, or pain, leads to the brain image, not the other way round.
    My question is, if the sensation, or lack of it, isn’t generated in the brain (I include the nervous system as part of the ‘brain’) then where does it arise?
    There isn’t fear ‘out there,’ then we react to it. That makes no sense. The brain perceives a threat, a reaction shows up on the fMri. This would work in a similar way to any placebo effect of acupuncture, or sugar pills or otherwise. The brain receives input that the body has received treatment, whether actual or not is immaterial. The brain either accepts what it is told or not. If the pain is real, a real injury, then acupuncture may provide temporary relief just by sticking in needles as a distraction. Once that wears off, the pain will be very real again.
    The bottom line is, if homeopathy and acupuncture worked, nobody would have cancer, Alzheimer’s or anything else wrong.
    The brain lights up first, then the sensation occurs. I fail to see how it can be any other way.