« previous post | next post »

Mariska Mantione, Martijn Figee, and Damiaan Denys, "A case of musical preference for Johnny Cash following deep brain stimulation of the nucleus accumbens", Frontiers in Behavioral Neuroscience, 5/6/2014.

Mr. B., a 59-year old married man, was referred to the department of Anxiety disorders, suffering from obsessive-compulsive disorder (OCD) for 46 years. He reported obsessions about fear for uncertainty and illogical things, and compulsions about seeking reassurance and hoarding. At the moment of referral, Mr. B. scored a total of 33 points on the Yale-Brown obsessive-compulsive scale (Y-BOCS; Goodman et al., 1989a,b), corresponding with extremely severe OCD. He scored 18 points on the Hamilton Anxiety Scale (HAM-A; Hamilton, 1959), corresponding with moderate anxiety and 14 points on the Hamilton Depression Scale (HAM-D; Hamilton, 1960), corresponding with mild depression. In spite of extensive treatment with pharmacotherapy and cognitive behavioral therapy, symptoms were still overpowering and Mr. B. remained extremely hindered in daily living.  

Suffering from treatment-refractory OCD, Mr. B., was included for treatment with deep brain stimulation (DBS) targeted at the nucleus accumbens (NAcc). After signing informed consent, Mr. B., was implanted in December 2006 with two four-contact electrodes (Model 3389, Medtronics, Minneapolis). The electrodes were connected via subcutaneous extensions to Soletra stimulators (Medtronic, Minneapolis) placed bilaterally in an infraclavicular pocket under general anesthesia. The center of the deepest contacts was positioned 3 mm in front of the anterior border of the anterior commissure (AC), 7 mm lateral and 4 mm below the line connecting the anterior and posterior commissure (PC). This was verified by post-operative CT, fused with preoperative MR.

Mr. B., had never been a huge music lover. His musical taste was broad, covering Dutch-language songs, the Beatles and the Rolling Stones, with a preference for the last named. While music did not occupy an important position in his live, his taste in music had always been very fixed and his preferences stayed the same throughout decades. [A] half year after DBS surgery, Mr. B. stated that he was turning into a Johnny Cash fan. […] Mr. B. reported that he felt good following treatment with DBS and that the songs of Johnny Cash made him feel even better. From this moment on, Mr. B. kept listening simply and solely to Johnny Cash and bought all his CD’s and DVD’s. When listening to his favorite songs he walks back and forth through the room and feels like he finds himself in a movie in which he plays the hero’s part. He reports that there is a Johnny Cash song for every emotion and every situation, feeling happy or feeling sad and although Mr. B. played almost simply and solely Johnny Cash songs for the following years, the music never starts to annoy him. From the first time Mr. B. heard a Johnny Cash song, the Dutch-language songs, the Beatles and the Rolling Stones have been banned. Except when the stimulators run down or accidentally go out. Then, Johnny Cash is unconsciously ignored and his old favorites are played once again, just as it was for the past 40 years.

We've been unable to verify the rumors that Frontiers in Neuroscience has been acquired by Onion, Inc.


  1. Jerry Friedman said,

    May 20, 2014 @ 6:56 am

    This is very exciting, as it might eventually lead to a cure for liking Johnny Cash.

  2. MattF said,

    May 20, 2014 @ 7:18 am

    So, do you think it's a Johnny Cash gene or a Johnny Cash neuron?

    [(myl) The exciting thing is that it's a liking-Johnny-Cash neuron, which is one step up the logical ladder from (say) the recognizing-Jennifer-Aniston neuron. Next we need things like the liking-people-who-like-Johnny-Cash neuron, and the wondering-why-people-like-Johnny-Cash neuron, and the preferring-Hank-Williams-to-Johnny-Cash neuron, and other refutations of the outmoded hypotheses of compositional semantics.

    But it seems improbable that there can be a gene for each such neuron, since current belief has it that there are about 8.6 x 10^10 neurons in the human brain — plenty to assign one to every proposition that each of us is likely to entertain, since 86000000000/(365*70) = 3,365,949 propositions per day over 70 years. In contrast, there are only about 3 x 10^9 base pairs – and only 2.5 x 10^4 protein-coding genes — in the human genome.

    (Of course, the implant is really stimulating a large number of neurons, not just one. And the region being stimulated is in the brain stem, and not closely implicated in the formation or maintenance of concepts, recursive or otherwise. So the earlier part of this comment is a not-very-good joke…) ]

  3. GH said,

    May 20, 2014 @ 7:55 am

    Well, the one point in the summary that seems to distinguish this from a non-story about "some guy became a huge Johnny Cash fan" is that his former music habit returns whenever the brain stimulation device is off.

    If that happens consistently, it seems like strong evidence that the change in musical preference is caused by the implant. What we learn from this beyond "interestingly specific and subtle changes can be induced in our thinking, emotions and personality by running electricity through the brain" isn't clear to me, but perhaps the details have some scientific value to neuroscientists. In any case it makes for a cool case study, no? Somewhere, Oliver Sacks is taking notes for his next book…

    [(myl) Exactly.]

  4. Rod Johnson said,

    May 20, 2014 @ 9:03 am

    It would be interesting to find musicians with voices or musical styles more or less like Johnny Cash along various dimensions—say Merle Haggard, Waylon Jennings, Johnny Paycheck, Kris Kristofferson and Tennesee Ernie Ford—and see if you could isolate the perceptual cues Mr. B. is responding to. There's doesn't have to be a neuron for every musician, etc.—there could be a neuron for liking basses, or major pentatonic scales, or tic-tac bass, etc. (Not that I believe there is any kind of specific-neuron-to-specific-taste mapping.)

  5. Dan Lufkin said,

    May 20, 2014 @ 9:45 am

    It's too late to arrange for the appropriate placebo testing by having Johnny Cash sing M'n Vissermeisje, for instance, but we could repurpose a Dutch Elvis impersonator (of which there are plenty) to sing Folsom Prison Blues. I wonder how much voltage it would take on the stimulator to get the patient to like Luciano Pavarotti.

  6. Rod Johnson said,

    May 20, 2014 @ 10:18 am

    I endorse Dan's research proposal.

  7. J. W. Brewer said,

    May 20, 2014 @ 1:22 pm

    Does this guy have children? Because if so we could test the hypothesis that stimulating the same neuron in them might result in them becoming Rosanne Cash fans.

  8. Chris C. said,

    May 20, 2014 @ 4:32 pm

    My son, who has extensive brain damage subsequent to a severe intracranial hemorrhage and hydrocephaly after having been born very prematurely, has very specific musical tastes. He's indifferent to a good deal of it. Music he enjoys he really, REALLY likes. Music he dislikes makes him react as if he were in severe pain, and there's nothing for it but to turn it off or get him away from it.

    It's not always obvious how he will react to any given piece of music beforehand. I tried playing a CSN album for him once. He was perfectly happy with the first few songs, but then burst into tears about halfway through, even though there was nothing I could hear that was all that unusual about that particular number. For years we've known of exactly one album we could play all the way through where he enjoyed most of it and reacted with pain to none of it: "The Best of The Cars". He seems to enjoy country — so he finds a George Strait album he got for his birthday last month at least tolerable — and a wide selection of '80s rock, and they tell me at school that he enjoys the Rolling Stones.

    Unfortunately he's not verbal, so he's not able to explain what a given piece of music makes him feel. We have to judge from his physical reactions. How relevant this is to the subject I don't know. His brain is far from typical, as most of the cerebral cortex is missing.

  9. Tim Morris said,

    May 20, 2014 @ 5:32 pm

    Associated symptoms include hypochondriacal concern about cardiac function, hyperextension of the eyelids, and obsessive focalization on the tips of shoelaces.

  10. Keith M Ellis said,

    May 20, 2014 @ 8:55 pm

    "If that happens consistently, it seems like strong evidence that the change in musical preference is caused by the implant."

    Evidence, I agree. Not necessarily "strong".

    What I suspect is happening in this case, assuming it's reported correctly, is that the subject's interest in Johnny Cash developed subsequent to the implant and under the conditions of the implant, and that absent the implant being on, his cognitive state is sufficiently altered as to interfere with this taste he acquired subsequent to the implant.

    The point is that the implant probably didn't cause his enjoyment of Johnny Cash's music, but that this enjoyment is in some complex way dependent upon the implant, because his experiences and evolving cognitive state have been conditioned upon that implant being on. What I'd expect is that there would be a cluster of things that would be similarly, though probably variably and erratically, affected by the implant being "off", where all those things are significant changes in his personality that occurred subsequent to the implant — not limited to those changes which were caused by the implant, but also including the kinds of (large-ish) changes in personality that everyone experiences over time, those interests and preferences that do tend to change over time but which, while they're present, tend to be moderately strong.

    A habit, or a liking for a certain food, or something similar that he developed post implant might vary depending upon the implant's state, for example.

    But I'd expect that this would be limited, and perhaps unpredictably so, to things that happen to have strong dependencies in the parts of his brain most affected by the implant. So a preference that is widely distributed throughout the brain, such as a friendship or other personal relationship, might be unlikely to be affected because it's not going to be so limited to such a narrow dependency. And for narrower things, it seems to me that we understand the brain so poorly that it's hard to know what is or isn't likely to just happen to heavily be involved in the part of the brain affected by the implant.

    In other words, this could be a very well-behaved and reproducible phenomenon in this particular subject, but we'll never see anything precisely like it again. Although we very likely will see things that are loosely similar. There's not a neuron or cluster of neurons in the human brain in general which will control musical preferences. There might be in one individual brain, for a certain preference or set of preferences.

    But this is really the sorts of things we see in brain injuries and tumors: there are a number of places in the brain which are anatomically and functionally regular and so pathologies are relatively predictable and well-characterized. But then there's a whole bunch of other things that are just almost random, though bizarre from our perspective. If you stimulate a particular cluster of neurons in a particular person's brain, you can reliably elicit a certain memory. That obviously doesn't mean that you could elicit that memory by stimulating the corresponding cluster or neurons in someone else's brain, or even a similar kind of memory or a similarly aged memory, or even a memory at all. Because brains aren't that regular.

    So what I'm saying is that this isn't a ridiculous story, it really shouldn't be that surprising. You needn't assume that the implant itself created a susceptibility to like Johnny Cash's music, nor that there's anything specific to even that area of that one person's brain with regard to music (and especially not to people's brains in general) in order to find this particular account plausible.

  11. Rebecca said,

    May 20, 2014 @ 9:38 pm

    Too bad Johnny's Germanic period didn't include Dutch. It's hard to imagine what state of mind this would put him in:

  12. Milan said,

    May 21, 2014 @ 11:06 am

    Rebecca: I'm afraid that it wasn't so much a period of Johnny Cash, but rather the relentaless German market that forced him to record in German. In the 50s and 60s the German were much less tolerant of English language music than most other nations, and therefore many artists recorded their songs in German. Or rather: "German", because quite often they just sang some text written down phonetically without understanding it, leading to predictably awful results. To us German millennials, those version are a source of mild amusement, or indeed existential shock: "Is our revered Johnny Cash really that similar to the abominations of the volkstümlicher Schlager?!"

    This trend even extended to the Beatles:
    "Komm gib mir deine Hand":
    "Sie liebt dich":
    "Geh raus": – the last one really has some kind of anarchic, pythonesque charm to ti.
    Generally however, the greatest atrocities were committed by German artists covering English songs for similar economic reasons, like "Cindy & Bert" who managed to turn Black Sabbath's "Paranoid" into some kind of goosebumps story: "Der Hund von Baskerville":

  13. GH said,

    May 21, 2014 @ 2:28 pm

    @Keith M Ellis:

    I don't mean to be too insistent — I have no expertise in this area, and if you are better informed I will gladly defer — but I wonder why you think that is a more probable explanation?

    If I understand you correctly, you think the implant didn't cause his devotion to Johnny Cash's music (by inducing some susceptibility that we cannot precisely identify), but that it's the sort of development that might have happened anyway, just the same as people's tastes sometimes change even when they don't have electrodes stuck in their brain. Only since this change happened while under the stimulation of the implant, once it is turned off his brain reverts to something more like its old patterns, and the "Johnny Cash fandom circuits" fall dormant.

    Perhaps. But the paper's proposed explanation seems equally plausible to me. I think it comes down to how likely we think it would be for a change like this to occur naturally. The writeup clearly attempts to portray it as out of character (for his former personality), since he had been relatively indifferent to music and his preferences had been stable for decades, but it's hard to tell without a fuller patient history.

    People do sometimes experience radical, unlikely-seeming changes in interests or personality, of course (though I'm not sure how common it is, after adulthood), but couldn't all those naturally-occurring instances very well be caused by exactly the kind of changes to the brain that this implant is apparently inducing? I know that if I were to develop a gambling habit, or suddenly became an extremely meticulous planner, it would be so out of character that "brain tumor" would be pretty high up the list of explanations.

  14. Keith M Ellis said,

    May 21, 2014 @ 3:13 pm


    "I think it comes down to how likely we think it would be for a change like this to occur naturally. The writeup clearly attempts to portray it as out of character (for his former personality), since he had been relatively indifferent to music and his preferences had been stable for decades, but it's hard to tell without a fuller patient history."

    Yours are good points.

    I guess that I agree that the implant could have altered his cognition in such a way that his preferences themselves were changed. But the narrower the focus of the preference, the less likely I find this. If he developed a strong and out-of-character preference for a variety of similarly sounding music, then I'd be more receptive to the claim that the implant directly affected his preference for Johnny Cash's music.

    But his interest isn't just with the music, it's with a bunch of things associated with Johnny Cash, with the music as his primary focus. Okay, maybe the other things came along more naturally with a musical preference that is artificial, but I still think that you'd see some musical preference changes that aren't just specific to Johnny Cash.

    Staying within the realm of the implant positively affecting his musical preference, a possibility that seems more plausible to me is that the implant didn't affect his musical preference directly, but did directly affect something that was involved in a positive feedback loop when he first heard Cash's music post-implant. This is sort of a hybrid explanation between the one I previously made and the strong, direct argument. It's not just that the implant creates broad conditions upon which his (naturally occurring) discovery of Cash is dependent, it's that the implant created a somewhat narrower condition that influenced his initial response to Cash's music. Really, I'm just talking about some kind of positive conditioning, like that which occurs naturally when we acquire some preference that's related to a circumstance that was very positive for us in a wider sense. The implant created the conditions for that positive conditioning. And somehow, when the implant is off, that association is disrupted.

    The comparison to genetics is apt because while there are certainly specific single genes that control some big functions, and so it makes sense to talk about such-and-such a gene as the gene for that thing, with many more things it's gene complexes, at the least, and gene complexes and epigenetics, and then developmental factors and culture not at the most, but quite commonly, especially with regard to cognition.

    The point is that there's a similar naivete in thinking that such-and-such an area of this patient's brain directly controls his response to Johnny Cash's music. It's possible, but it's more likely that the preference is a complex phenomenon that is dependent upon quite a few different areas and functioning in the brain and that the implant moved the system, as represented by the preference, into a different state. Either a state that is/was more receptive to Cash's music, or a state upon which the patient's acquisition of this preference is dependent simply because all cognition is path-dependent.

  15. GH said,

    May 21, 2014 @ 7:09 pm

    Thanks for an interesting response. I don't think we disagree much on the substance here. I just think that if he wouldn't have formed such a strong attachment to Johnny Cash without the implant, then it's fair enough to say that the implant "caused" the change in musical preference.

    I'm not sure I see why the specificity of his Johnny Cash preference makes it more likely to be "natural" than implant-induced. Isn't the fact that he is so absorbed by a single artist itself one of the striking personality changes that can probably be attributed to the implant's effect? How that might relate to the OCD he was originally treated for is intriguing, but I'll bow out of the discussion at this point, since I'm in way over my head in speculation and amateur diagnoses.

  16. exackerly said,

    May 21, 2014 @ 8:22 pm

    >>>From the first time Mr. B. heard a Johnny Cash song, the Dutch-language songs, the Beatles and the Rolling Stones have been banned.

    Couldn't this be an indication that his OCD hasn't actually been cured, only redirected (while the stimulator is on)?

  17. Mr Punch said,

    May 22, 2014 @ 12:55 pm

    @Milan – The German-language thing is a bit more complicated in the case of the Beatles, who really got their start as professionals in Germany. I don't know if Johnny Cash picked up much German during his time there in the Air Force.

  18. Dave Harmon said,

    May 22, 2014 @ 9:05 pm

    The thing is, we do have considerable prior evidence that local brain stimulation can trigger quite specific thoughts and images. I'm not too surprised that it could include a musical preference.

    The idea that Keith M Ellis floats is interesting, but if the implants were only involved in formation of the liking, I wouldn't expect the preference to switch on and off with them. I think it more likely that it's affecting some low-level circuitry that has non-local effects, which accidentally causes the preference. It's worth noting that the electrodes might well be stimulating more cells than the experimenters think, perhaps due to signal leakage.

    The brain is a classic example of a principle I've been settling into: Mother Nature doesn't care if we watch what she does, but she feels no obligation to explain herself.

  19. Yuval said,

    May 23, 2014 @ 3:54 am

    What's with all the commas following "Mr. B."? My recognizing-redundant-punctuation neuron was furiously acting up while reading this.

RSS feed for comments on this post