Musicophilia

by Oliver Sacks

  Yet it seems to make little difference whether catchy songs have lyrics or not— the wordless themes of Mission: Impossible or Beethoven’s Fifth can be just as irresistible as an advertising jingle in which the words are almost inseparable from the music (as in Alka-Seltzer’s “Plop, plop, fizz, fizz” or Kit Kat’s “Gimme a break, gimme a break…”).

  For those with certain neurological conditions, brainworms or allied phenomena— the echoic or automatic or compulsive repetition of tones or words— may take on additional force. Rose R., one of the post-encephalitic parkinsonian patients I described in Awakenings, told me how during her frozen states she had often been “confined,” as she put it, in “a musical paddock”— seven pairs of notes (the fourteen notes of “Povero Rigoletto”) which would repeat themselves irresistibly in her mind. She also spoke of these forming “a musical quadrangle” whose four sides she would have to perambulate, mentally, endlessly. This might go on for hours on end, and did so at intervals throughout the entire forty-three years of her illness, prior to her being “awakened” by L-dopa.

  The phenomenon of brainworms seems similar, too, to the way in which people with autism or Tourette’s syndrome or obsessive-compulsive disorder may become hooked by a sound or a word or a noise and repeat it, or echo it, aloud or to themselves, for weeks at a time. This was very striking with Carl Bennett, the surgeon with Tourette’s syndrome whom I described in An Anthropologist on Mars. “One cannot always find sense in these words,” he said. “Often it is just the sound that attracts me. Any odd sound, any odd name, may start repeating itself, get me going. I get hung up with a word for two or three months. Then, one morning, it’s gone, and there’s another one in its place.” But while the involuntary repetition of movements, sounds, or words tends to occur in people with Tourette’s or OCD or damage to the frontal lobes of the brain, the automatic or compulsive internal repetition of musical phrases is almost universal— the clearest sign of the overwhelming, and at times helpless, sensitivity of our brains to music.

  There may be a continuum here between the pathological and the normal, for while brainworms may appear suddenly, full-blown, taking instant and entire possession of one, they may also develop by a sort of contraction, from previously normal musical imagery. I have lately been enjoying mental replays of Beethoven’s Third and Fourth Piano Concertos, as recorded by Leon Fleisher in the 1960s. These “replays” tend to last ten or fifteen minutes and to consist of entire movements. They come, unbidden but always welcome, two or three times a day. But on one very tense and insomniac night, they changed character, so that I heard only a single rapid run on the piano (near the beginning of the Third Piano Concerto), lasting ten or fifteen seconds and repeated hundreds of times. It was as if the music was now trapped in a sort of loop, a tight neural circuit from which it could not escape. Towards morning, mercifully, the looping ceased, and I was able to enjoy entire movements once again.

  Brainworms are usually stereotyped and invariant in character. They tend to have a certain life expectancy, going full blast for hours or days and then dying away, apart from occasional after spurts. But even when they have apparently faded, they tend to lie in wait; a heightened sensitivity remains, so that a noise, an association, a reference to them is apt to set them off again, sometimes years later. And they are nearly always fragmentary. These are all qualities that epileptologists might find familiar, for they are strongly reminiscent of the behavior of a small, sudden-onset seizure focus, erupting and convulsing, then subsiding, but always ready to re-ignite.

  Some of my correspondents compare brainworms to visual afterimages, and as someone who is prone to both, I feel their similarity, too. (We are using “afterimage” in a special sense here, to denote a much more prolonged effect than the fleeting afterimages we all have for a few seconds following, for instance, exposure to a bright light.) After reading EEGs intently for several hours, I may have to stop because I start to see EEG squiggles all over the walls and ceiling. After driving all day, I may see fields and hedgerows and trees moving past me in a steady stream, keeping me awake at night. After a day on a boat, I feel the rocking for hours after I am back on dry land. And astronauts, returning from a week spent in the near-zero-gravity conditions of space, need several days to regain their “earth legs” once again. All of these are simple sensory effects, persistent activations in low-level sensory systems, due to sensory over-stimulation. Brainworms, by contrast, are perceptual constructions, created at a much higher level in the brain. And yet both reflect the fact that certain stimuli, from EEG lines to music to obsessive thoughts, can set off persistent activities in the brain.

  There are attributes of musical imagery and musical memory that have no equivalents in the visual sphere, and this may cast light on the fundamentally different way in which the brain treats music and vision. This peculiarity of music may arise in part because we have to construct a visual world for ourselves, and a selective and personal character therefore infuses our visual memories from the start— whereas we are given pieces of music already constructed. A visual or social scene can be constructed or reconstructed in a hundred different ways, but the recall of a musical piece has to be close to the original. We do, of course, listen selectively, with differing interpretations and emotions, but the basic musical characteristics of a piece— its tempo, its rhythm, its melodic contours, even its timbre and pitch— tend to be preserved with remarkable accuracy.

  It is this fidelity— this almost defenseless engraving of music on the brain— which plays a crucial part in predisposing us to certain excesses, or pathologies, of musical imagery and memory, excesses that may even occur in relatively unmusical people.

  There are, of course, inherent tendencies to repetition in music itself. Our poetry, our ballads, our songs are full of repetition. Every piece of classical music has its repeat marks or variations on a theme, and our greatest composers are masters of repetition; nursery rhymes and the little chants and songs we use to teach young children have choruses and refrains. We are attracted to repetition, even as adults; we want the stimulus and the reward again and again, and in music we get it. Perhaps, therefore, we should not be surprised, should not complain if the balance sometimes shifts too far and our musical sensitivity becomes a vulnerability.

  Is it possible that earworms are, to some extent, a modern phenomenon, at least a phenomenon not only more clearly recognized, but vastly more common now than ever before? Although earworms have no doubt existed since our forebears first blew tunes on bone flutes or beat tattoos on fallen logs, it is significant that the term has come into common use only in the past few decades. When Mark Twain was writing in the 1870s, there was plenty of music to be had, but it was not ubiquitous. One had to seek out other people to hear (and participate in) singing— at church, family gatherings, parties. To hear instrumental music, unless one had a piano or other instrument at home, one would have to go to church or to a concert. With recording and broadcasting and films, all this changed radically. Suddenly music was everywhere for the asking, and this has increased by orders of magnitude in the last couple of decades, so that we are now enveloped by a ceaseless musical bombardment whether we want it or not.

  Half of us are plugged into iPods, immersed in daylong concerts of our own choosing, virtually oblivious to the environment— and for those who are not plugged in, there is nonstop music, unavoidable and often of deafening intensity, in restaurants, bars, shops, and gyms. This barrage of music puts a certain strain on our exquisitely sensitive auditory systems, which cannot be overloaded without dire consequences. One such consequence is the ever-increasing prevalence of serious hearing loss, even among young people, and particularly among musicians. Another is the omnipresence of annoyingly catchy tunes, the brainworms that arrive unbidden and leave only in their own time— catchy tunes that may, in fact, be nothing more than advertisements for toothpaste but are, neurologically, completely irresistible.

  6

  Musical Hallucinatio
ns

  In December of 2002, I was consulted by Sheryl C., an intelligent and friendly woman of seventy. Mrs. C. had had progressive nerve deafness for more than fifteen years, and now had profound hearing loss on both sides. Until a few months earlier, she had managed to get by with lip-reading and the use of sophisticated hearing aids, but then her hearing had suddenly deteriorated further. Her otolaryngologist suggested a trial of prednisone. Mrs. C. took a gradually rising dose of this for a week, and during this time she felt fine. But then, she said, “on the seventh or eighth day— I was up to sixty milligrams by then— I woke up in the night with dreadful noises. Terrible, horrific, like trolley cars, bells clanging. I covered my ears, but it made no difference. It was so loud, I wanted to run out of the house.” Her first thought, indeed, was that a fire engine had stopped outside the house, but when she went to the window and looked out, the street was completely empty. It was only then that she realized that the noise was in her head, that she was hallucinating for the first time in her life.

  After about an hour, this clangor was replaced by music: tunes from The Sound of Music and part of “Michael, Row Your Boat Ashore”— three or four bars of one or the other, repeating themselves with deafening intensity in her mind. “I was well aware that there was no orchestra playing, that it was me,” she emphasized. “I was afraid I was going mad.”

  Mrs. C.’s physician suggested that she taper off the prednisone, and a few days later the neurologist whom she had now consulted suggested a trial of Valium. Mrs. C.’s hearing, meanwhile, had returned to its previous level, but neither this nor the Valium nor the tapering of the prednisone had any effect at all on her hallucinations. Her “music” continued to be extremely loud and intrusive, stopping only when she was “intellectually engaged,” as in conversation or in playing bridge. Her hallucinatory repertoire increased somewhat but remained fairly limited and stereotyped, confined mostly to Christmas carols, songs from musicals, and patriotic songs. All of these were songs she knew well— musically gifted and a good pianist, she had often played them in her college days and at parties.

  I asked her why she spoke of musical “hallucinations” rather than musical “imagery.”

  “They are completely unlike each other!” she exclaimed. “They are as different as thinking of music and actually hearing it.” Her hallucinations, she emphasized, were unlike anything she had ever experienced before. They tended to be fragmentary— a few bars of this, a few bars of that— and to switch at random, sometimes even in mid-bar, as if broken records were being turned on and off in her brain. All of this was quite unlike her normal, coherent, and usually “obedient” imagery— though it did have a little resemblance, she granted, to the catchy tunes that she, like everyone, sometimes heard in her head. But unlike catchy tunes, and unlike anything in her normal imagery, the hallucinations had the startling quality of actual perception.

  At one point, sick of carols and popular songs, Mrs. C. had tried to replace the hallucinations by practicing a Chopin étude on the piano. “That stayed in my mind a couple of days,” she said. “And one of the notes, that high F, played over and over again.” She started to fear that all of her hallucinations would become like this— two or three notes, or perhaps a single note, high, piercing, unbearably loud, “like the high A Schumann heard at the end of his life.”1 Mrs. C. was fond of Charles Ives, and another worry she had was that she might have “an Ives hallucination.” (Ives’s compositions often contain two or more simultaneous melodies, sometimes completely different in character.) She had never yet heard two hallucinatory tunes simultaneously, but she started to fear that she would.

  She was not kept awake by her musical hallucinations or prone to musical dreams, and when she awoke in the morning, there would be an inner silence for a few seconds, during which she would wonder what the “tune du jour” was going to be.

  When I examined Mrs. C. neurologically, I found nothing amiss. She had had EEG and MRI studies to rule out epilepsy or brain lesions, and these had been normal. The only abnormality was her rather loud and poorly modulated voice, a consequence of her deafness and impaired auditory feedback. She needed to look at me when I spoke, so that she could lip-read. She seemed neurologically and psychiatrically normal, though understandably upset by the feeling that something was going on inside her that was beyond her control. She had been upset, too, by the idea that these hallucinations might be a sign of mental illness.

  “But why only music?” Mrs. C. asked me. “If these were psychotic, wouldn’t I be hearing voices, too?”

  Her hallucinations, I replied, were not psychotic but neurological, so-called “release” hallucinations. Given her deafness, the auditory part of the brain, deprived of its usual input, had started to generate a spontaneous activity of its own, and this took the form of musical hallucinations, mostly musical memories from her earlier life. The brain needed to stay incessantly active, and if it was not getting its usual stimulation, whether auditory or visual, it would create its own stimulation in the form of hallucinations. Perhaps the prednisone or the sudden decline in hearing for which it was given had pushed her over some threshold, so that release hallucinations suddenly appeared.

  I added that brain imaging had recently shown that the “hearing” of musical hallucinations was associated with striking activity in several parts of the brain: the temporal lobes, the frontal lobes, the basal ganglia, and the cerebellum— all parts of the brain normally activated in the perception of “real” music. So, in this sense, I concluded to Mrs. C., her hallucinations were not imaginary, not psychotic, but real and physiological.

  “That’s very interesting,” said Mrs. C., “but rather academic. What can you do to stop my hallucinations? Do I have to live with them forever? It’s a dreadful way to live!”

  I said we had no “cure” for musical hallucinations, but perhaps we could make them less intrusive. We agreed to start a trial of gabapentin (Neurontin), a drug that was developed as an antiepileptic but is sometimes useful in damping down abnormal brain activity, whether epileptic or not.

  The gabapentin, Mrs. C. reported at her next appointment, actually exacerbated her condition and had added a loud tinnitus, a ringing of the ears, to the musical hallucinations. Despite this, she was considerably reassured. She knew now that there was a physiological basis for her hallucinations, that she was not going mad, and she was learning to adapt to them.

  What did upset her was when she heard fragments repeated again and again. She instanced hearing snatches of “America the Beautiful” ten times in six minutes (her husband had timed this), and parts of “O Come, All Ye Faithful” nineteen and a half times in ten minutes. On one occasion, the iterating fragment was reduced to just two notes.2 “If I can hear a whole verse, I’m very happy,” she said.

  Mrs. C. was now finding that though certain tunes seemed to repeat themselves at random, suggestion and environment and context played an increasing part in stimulating or shaping her hallucinations. Thus, once as she was approaching a church, she heard a huge rendering of “O Come, All Ye Faithful” and thought at first that it was coming from the church. After baking a French apple cake, she hallucinated bits of “Frère Jacques” the next day.

  There was one more medication that I felt might be worth a trial: quetiapine (Seroquel), which had been successfully used in one case to treat musical hallucinations.3 Though we only knew of this single report, the potential side effects of quetiapine were minimal, and Mrs. C. agreed to try a small dose. But it had no clear effect.

  Mrs. C. had been trying, in the meantime, to enlarge her hallucinatory repertoire, feeling that if she did not make a conscious effort it would contract to three or four endlessly repeated songs. One hallucinatory addition was “Ol’ Man River” sung with extreme slowness, almost a parody of the song. She did not think she had ever heard the song performed in this “ludicrous” way, so this was not so much a “recording” from the past as a memory that had been revamped, recategorized in a humorous wa
y. This, then, represented a further degree of control, not merely switching from one hallucination to another, but modifying one creatively, if involuntarily. And though she could not stop the music, she could sometimes switch it now by an effort of will. She no longer felt so helpless, so passive, so put upon; she had a greater sense of control. “I still hear music all day long,” she said, “but either it has become softer or I’m handling it better. I haven’t been getting as upset.”

  Mrs. C. had been thinking about a cochlear implant for her deafness for years but had postponed this when the musical hallucinations began. Then she learned that one surgeon in New York had performed a cochlear implant in a severely hard-of-hearing patient with musical hallucinations and found that it not only provided good hearing, but had eliminated the musical hallucinations. Mrs. C. was excited by this news and decided to go ahead.

  After her implant had been inserted and, a month later, activated, I phoned Mrs. C. to see how she was doing. I found her very excited and voluble over the phone. “I’m terrific! I hear every word you say! The implant is the best decision I ever made in my life.”

  I saw Mrs. C. again two months after her implant was activated. Her voice before had been loud and unmodulated, but now that she could hear herself speak, she spoke in a normal, well-modulated voice, with all the subtle tones and overtones that were absent before. She was able to look around the room as we spoke, where previously her eyes had always been fixed on my lips and face. She was manifestly thrilled with this development. When I asked how she was, she responded, “Very, very well. I can hear my grandchildren, I can distinguish male from female voices on the telephone…. It’s made a world of difference.”